Slide 1

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 2

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 3

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 4

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 5

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 6

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 7

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 8

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 9

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 10

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 11

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 12

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 13

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 14

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 15

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 16

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 17

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 18

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 19

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 20

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 21

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 22

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 23

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 24

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 25

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 26

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 27

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 28

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 29

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 30

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 31

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 32

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 33

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 34

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 35

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 36

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 37

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 38

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 39

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 40

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 41

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 42

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 43

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 44

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 45

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 46

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 47

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 48

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 49

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 50

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 51

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 52

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 53

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 54

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 55

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 56

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.


Slide 57

sustainable agricultural systems

Actionable climate
knowledge – from
analysis to synthesis
Experiences from 20 years of applied
climate risk research in Australia
Holger Meinke, Rohan Nelson, Roger
Stone, Selvaraju, Aline de Holanda,
Walter Baethgen

Why focus on case studies from
Australia?
 has long been at the forefront of
applied climate research
 often regarded (rightly or wrongly) as a
role model for the creation and
maintenance of ‘actionable climate
knowledge’
 has one of the most variable climates
in the world

Why focus on case studies from
Australia?
 has strong ENSO impact and vulnerable
sectors with considerable scope to
improve risk management
 climate change already a reality and
not just a scenario
 public policy focus on self-reliance,
resilience and societal benefits
 involves many agencies and many
stakeholders (farmers, agribusiness,
policy makers)

Climate knowledge vs climate
forecasting
 Climate knowledge is more that ENSO
and more than just forecasting.
 Climate knowledge is the intelligent use
of climate information. This includes
knowledge about climate variability,
climate change AND climate forecasting
used such that it enhances resilience by
increasing profits and reducing
economic/environmental risks.

sustainable agricultural systems

Risk management
 The systematic process of identifying,
analysing and responding to risk. It
includes maximising the probability and
consequences of positive and adverse
events.
(Guide to the Project Management Body of Knowledge)

 ‘It is our competitive advantage that we
show courage after carefully deliberating
our actions. Others, in contrast, are
courageous from ignorance but hesitant
upon reflection’.
(Pericles’ Funeral Oration, 431 AD; Thucydides 2, 40, 3)

sustainable agricultural systems

Risks arise from variability
Australian farmers are excellent risk
managers. They run successful businesses
within the world’s most variable climate
and without subsidies.
…it seems that the 21st century
has a good chance of becoming
‘the climate century’, a century
in which climate-related
concerns will occupy significant
attention of the next generations
of policy makers…
Mickey Glantz, 2003

sustainable agricultural systems

Sources of variability
 Temporal and spatial
weather (hail, frost); climate (at a range of temporal
scales); soils (at a range of spatial scales); economic
conditions (inputs, commodity prices); management

 External and internal
either beyond manager’s control or consequence of
management

sustainable agricultural systems

E x a m p le o f D e c isio n T y p e s

K e y S ta k e h o ld e r

F re q u e n c y

Lo g istic s (e g . sc h e d u lin g o f p la n tin g / h a rve st
o p e ra tio n s)

Fa rm M a n a g e r

M JO , m o n th s

C ro p typ e , w e a th e r d e riva tive s, in su ra n ce , h e rd
m a n a g e m e n t, irrig a tio n s ch e d u lin g , m a rke tin g

Fa rm M a n a g e r,
A g rib u sin e s s

E N S O , se a so n

C ro p se q u e n ce , fa llo w m a n a g e m e n t, sto ckin g
ra te s, w a te r a llo ca tio n , in s u ra n ce

Fa rm M a n a g e r,
A g rib u sin e s s, P o licy

S e a so n to
in te ra n n u a l

C ro p in d u stry (g ra in o r co tto n ; n a tive ve rs u s
im p ro ve d p a stu re s), ru ra l v e rsu s o ff-fa rm
in ve stm e n ts

B u sin e s s M a n a g e r,
A g rib u sin e s s, P o licy

D e ca d a l
(~ 1 0 yr)

A g ricu ltu ra l in d u stry (e g . cro p s, p a stu re s,
fo re stry , h o rticu ltu re ), in ve stm e n ts

A g rib u sin e s s, P o licy

M u lti-d e ca d a l
(1 0 – 2 0 yrs)

La n d u se , co m m u n ity im p a c t a n d a d a p ta tio n o f
cu rre n t s yste m s

P o licy

C lim a te ch a n g e
???

Three important steps to create
climate knowledge
1.
2.
3.

understanding rainfall (climate)
variability (physical measure)
understanding production variability
(bio-physical measure)
understanding farm income variability
(economic measure)

The first step: understanding rainfall
JJA rainfall for Dalby, Queensland

The first step: understanding rainfall
JJA rainfall for
Dalby,
Queensland

How good is the forecast?
Skill vs Discriminatory Ability
S quantifies agreement
between observed and
predicted values
DA represents the additional
knowledge about future
states arising from the
forecast system over and
above the total variability of
the prognostic variable

F o rcast skill an d d iscrim in ato ry ab ility, D alb y, Q ld

1
L E P S p -v a lu e s

K W p -v a lu e s

p -valu e

0 .8
0 .6
0 .4
0 .2
0
JFM

FMA

MAM

AMJ

MJJ

JJA

JAS

ASO

3-m o n th ly p erio d

SON

O ND

ND J

DJF

The first step:
understanding
rainfall
Discriminatory
Ability of the 5phase SOI
forecast system
as quantified by
KW p-values (KW
is a measure of
shift in
distributions)

sustainable agricultural systems

The second step:

understanding production impacts
Simulation models for better risk management
 how do they work?
 are based on our component knowledge
of science
 integrate many sources of variability
 account for management options
 what can they do?
 benchmark, assess and quantify potential,
attainable, economically optimal and
achieved yield or income
 overcome issues related to moral hazards
and ground truthing

sustainable agricultural systems

WhopperCropper
for on-farm decision
making
Y ie ld (k g /h a )

5000

4000

3000

2000

1000
W h eat
120

W h eat
190

S org h u m
120

S org h u m
190

C ro p & PA W C (m m )

WhopperCropper training and
distribution is now through
Nutrient Management Systems.
www.apsru.gov.au/apsru/products/whopper

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Gross Margin (100$ per ha)

G M ($ p e r h a )

SOI effect on gross margins
500
450

Wheat, Dalby, 150mm, 2/3 full, 15 June
sowing, April/May SOI phase

400
350
300

Negative SOI Phase

Positive SOI Phase

250
200
150
100
50
0

0N

0N
N egative

25N
25N
N egative

50N
100N
50N
100N
N egative
N egative

0N

0N
P os itive

25N

25N
P os itive

Applied Nitrogen
and
Phase
A p p lie d N & S O
I P ha sSOI
e

50N

50N
P os itive

100N

100N
P os itive

sustainable agricultural systems

Using field/farm scale models
 Tactical risk management
(which crop to grow when and how)

 Optimising resource use

(how much water / nitrogen to use when and where)

 Estimating crop value

(benchmarking, forward selling, insurance)

 Determine enterprise mix
(rotation planning)

sustainable agricultural systems

Regional Commodity Models (RCM)

Predicted sorghum
shire yield for the
2004/2005
season, ranked
relative to all
years (1901-2003)

July 2001

July 2002
Le g en d :
0-10 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
No d a ta

NT

Le g e nd :
0-1 0 %
10 -2 0 %
20 -3 0 %
30 -4 0 %
40 -5 0 %
50 -6 0 %
60 -7 0 %
70 -8 0 %
80 -9 0 %
90 -1 0 0 %
N o da ta

NT

#
Em e r a ld

WA

WA

Ro m a

#

Da lb y

#

G o o nd i wi n di

#

SA

SA

NSW

(a)

V IC

TAS

NSW

(b)

V IC

TAS

Probabilities of exceeding long-term median wheat yields
for every wheat producing shire (= district) in Australia
issued in July 2001 and July 2002, respectively.

Chance of exceeding median pasture
growth for NSW, April to June 2005

sustainable agricultural systems

Using regional models
 Marketing decisions
(hedging, contract negotiations, logistics)

 Value chain issues
(quality fluctuations, export vs domestic use,
milling operations)

 Anticipating resource use
(water allocations, nitrogen or seed demand,
storage capacity)

0.5
0.0

?

-0.5
-1.0

1.0
0.5
0.0

?

-0.5
-1.0

1992

1985

1978

1971

1964

1957

1950

1943

1936

1929

1922

1915

1908

1901

-1.5

1998

1995

1992

1989

1986

1983

1980

1977

1974

1971

1968

1965

1962

1959

-1.5

5-year running mean - Wentworth, 1884 to 1998

1894

Standard Deviations from the mean

1.0

1956

Simulated Wheat Yield 1890+
1.5

5-year running mean - Wentworth, 1950 to 1998

1950

When is a drought
a drought?

1.5

1953

Simulated Wheat Yield 1950+

Standard Deviations from the mean

sustainable agricultural systems

sustainable agricultural systems

Using models for public and
private policy decisions
 When is a drought a drought
(Exceptional circumstances, drought relief,
structural adjustments etc).

 Investment / disinvestment
(portfolio balance; cotton, grain or pastures)

 Structural adjustment
(diversification, industry mix eg. sugar industry)

The policy relevance gap
1.
2.

3.

no policy mechanisms for influencing
rainfall (step 1),
few policy options to affect crop or
pasture yields (step 2),
but strong community demand for
policies to anticipate and moderate
the effects of climate variability on
farm incomes (step 3).

The third step ( ‘the big stumble’):
making science relevant
Drought
“The defining feature of drought is its
impact on human activity – it is essentially
socially constructed.
It is about the mismatch between the
availability of water and the uses to which
human communities wish to put it.”
Linda Courtenay Botterill 2003
Exposure to risk does not equal
vulnerability

Climate is often ‘important but not
urgent’
 Many problems are the result of
applying narrow, specialised knowledge
to complex systems
 Modern science has been described as
‘islands of understanding in oceans of
ignorance’
 Scientists and practitioners need to
work together to produce trustworthy
knowledge that combines scientific
excellence with social relevance

Hayman (2001); Lowe (2001)

The multiple dimensions of vulnerability
Human
Carney, 1998; Ellis, 2000

Financial

Social
Exposure to
risk does not
equal
vulnerability

Physical

Natural

Vulnerability of Australian agriculture:
Exposure vs Coping Capacity

10% (most extreme)
10 to 25% (extreme)
below 25% (least extreme)

(Nelson et al. 2005)

Vulnerability includes
 exposure to climate risk
 exposure to other sources of risk
 capacity of rural households to cope
with risk

Why is coping capacity so
important?
 Farming systems have evolved to
effectively manage the risks of farming
in a highly variable climates – without
science intervention.
 While climate synthesis tools might
have contributed to the development of
more effective on-farm risk
management, there is little or no
connection to policy.

Why is coping capacity so
important?
 Greater diversity of income sources
facilitates substitution between activities
and assets in response to shocks such as
drought.
 Policies that enhance diversity of farm
income include investment in production,
transport and marketing infrastructure,
education and training, regional
development, and policies that impact on
the cost and availability of rural credit.

Why is coping capacity so
important?
 We need to distinguish the effects of
climate from other sources of income
risk.
 Without a capacity to distinguish
between sources of income variability,
policies directed toward reducing the
impact of climate risk may
inadvertently reduce incentives to
better manage other sources of risk.

A tool for bridging the policy
relevance gap
 The Agricultural Farm Income Risk
Model (AgFIRM) combines regional,
biophysical models of Australian crop
and pasture yield with an econometric
model of farm incomes.
 AgFIRM simulates regional impacts of
climate variability on farm incomes.

2002/3

2002-03
Forecasting
farm
incomes

2001/2

2001-02
1982/3

Probability of
exceeding median
farm income

1982-83

(Nelson et al. 2005)

2002-03

2001-02

Better
drought
assistance

Probability of 1-in-20
worst farm income

1982-83
1982-83

(Nelson et al. 2005)

Tools for bridging the policy
relevance gap
 Policies aimed at increasing the
capacity of rural communities to cope
with climate risk need to be informed
by measures of the multiple socioeconomic dimensions of resilience.
 Current emphasis on rainfall and
production variability only informs
policy makers of the exposure to
drought, for which there is no policy
solution.

Public versus private policy
development
 Risk managers must decide which risks
should be retained and managed
adaptively and which risks should be
shared through risk sharing contracts.
 It requires financial markets to device
and price risk sharing contracts in a
manner that create benefits for all
stakeholders involved, a process that
has only just begun in Australia.

sustainable agricultural systems

Real options, insurance and
other financial products
retained risks

shared risks

Farm

Reinsurer

Community

Weather/
climate
derivatives

Business

Insurer

Financial
Derivatives

courtesy of Greg Hertzler, Uni of WA

Climate knowledge or seasonal
rainfall forecasting?
 Applied climate knowledge is generated by
synthesising scientific insights across
disciplinary boundaries, often through the
use of models and always jointly with
stakeholders.
 Climate risk management in rural
industries is not solely the responsibility of
farmers. Likewise, it is not the role of
Governments to absorb these risks.
 Risk managers, policy makers and private
sector companies all play important roles
in this process.

The case for institutional
realignment
 Rainfall and production are not what
policy makers are interested in. They
are interested in the social and
economic wellbeing of rural
communities.
 Analytical support for drought policy
that focuses on exposure to climate
risk is largely irrelevant  climate
variability cannot be altered by policy
in the short term.

Failures and risks
 The artificial division of climate
variability and climate change gets in
the way of better decision making.
 The focus of the climate change
community on mitigation bears the
danger of overlooking some obvious
and immediate adaptation strategies
that should from part of any sound
climate risk management approach.

Failures and risks
 A problem rather than a disciplinary
focus will require some scientists to
stop doing what comes naturally
(addressing simple issues such as
rainfall variability, with increasingly
complex analytical tools).
 Instead, they need to take a broader
perspective to addresses not only
exposure to risk, but also the people’s
ability to cope and the system’s ability
to bounce back after times of stress
(resilience)

Other impediments
 institutional and disciplinary
fragmentation prevails
 difficult to ‘gain simplicity on the far
side of complexity’
 R&D funding agencies reluctant to
resource genuinely multi-disciplinary,
cross institutional projects

Some suggestions
 public / private partnership models
need to be explored further in order to
‘mainstream’ climate risk management
 public / private policy concerns need to
be explicitly addressed
 communicate climate risk management
knowledge through functional, existing
communication networks of farmers
and other landholders

First key lesson from several
decades of experience
 Climate knowledge needs to deliver true
societal benefits.
 We need to expand the systems
boundaries and fully explore the scientific
and socio-economic tensions and
interactions - the system is bigger than
most of us thought.
 We need to include the socio-economic
dimensions important to rural
communities and policy makers, but
without abandoning science.
 We need to achieve true integration of
disciplinary knowledge, rather than
focusing on certain aspects of the system
at the exclusion of others.

Second key lesson from several
decades of experience







True integration without abandoning
science takes real resourcing.
The capacity to think and act
beyond disciplinary boundaries is
rare and difficult to nurture in the
established institutional context.
Existing institutional arrangements
often act as a disincentive to true
integration.
Strong leadership is required to
induce cultural change in established
institutional arrangements.

sustainable agricultural systems

Modelling for a purpose

Climate
Warning

Adapt
TEMPORAL

now
field

Mitigate

farm

enterprise

future

SPATIAL
catchment region state
ECONOMICAL
business industry sector

sustainable agricultural systems

Modelling for a purpose
Value of adaptation to the grain industry

PROBABILITY

0.25
0.2

“Increased efficiencies
have outweighed all
expenditure involved.
The costs of tackling
climate change are
clearly lower than
many feared. This is a
manageable problem.”

0.15

Lord Browne, CEO of BP,
announcing that BP had
reached it’s target of reduce
carbon emissions to 10%
below 1990 levels eight years
ahead of schedule

0.1
0.05
0

Values in Millions

The Economist, 9 Oct 2004

Failures and risks
Why do institutional arrangements need to be realigned
in order to implement advances in climate risk
management policy?
1.

Rainfall and production are not what policy makers
are ultimately interested in. They are interested in
the social and economic wellbeing of rural
communities. There should be a natural evolution
from analytical support at certain scales to synthesis
tools that integrate the analysis of rainfall right
through production, farm incomes and sustainability
indicators. So far, institutional and funding structures
have largely prevented this from happening in
Australia, and probably anywhere else.

Policy options for managing climate
variability
 income smoothing and price stabilisation
 emergency relief
 undermines self reliance
 enhanced diversity of income sources
 investment in
 infrastructure
 human and social capital
 outsourcing risk
 enhances self reliance

(Nelson et al. 2005)

Public versus private policy
development
 Underpinned and informed by
quantitative systems analysis, such
policy development should go hand-inhand with the establishment of novel
financial risk management tools such
as ‘real options’ (a right, but not the
obligation, to take action).
 Real options are property rights created
by investments.

The case for institutional
realignment
 Institutional and funding structures
have largely prevented this from
happening in Australia, and probably
anywhere else.
 There should be a natural evolution
from analytical support at certain
scales to synthesis tools that integrate
the analysis of climate right through
production, farm incomes and
sustainability indicators.

Failures and risks
 Climate science and agricultural
systems science has to become more
policy relevant.
 To some extent this has happened with
climate change research.
 Not so with climate variability research
that must also inform policy
development to assist stakeholders to
better cope and adapt.