Transcript Fire effects on vegetation recovery Summary of Results and Project Deliverables Jill Johnstone, Teresa Hollingsworth, Emily Bernhart & Katie Villano.

Fire effects on vegetation
recovery
Summary of Results and
Project Deliverables
Jill Johnstone, Teresa Hollingsworth, Emily Bernhart & Katie Villano
All sites grouped by severity
Burn severity is HIGHLY correlated with the Axis 1
Significant changes key species
pre to post-fire
• Increase in Salix pulchra
• Increase in Salix diversity (4 Salix sps.
that only occur post-fire (Feltleaf,
Bebb’s, Firmleaf, and Scouler’s)
• Decrease in moss abundance and
diversity including Sphagnum sps.
• Decrease in lichen abundance and
diversity including 3 Cladonia species
Wildfire as a
conduit for
invasive species
colonization
Are black spruce forests more
susceptible to invasion after fire?
45
a
40
35
30
% survival
In the
greenhouse,
invasive plants
in burned soil
cores showed
higher survival
than in unburned
cores.
a
Unreproductive
Reproductive
25
20
15
10
5
b
0
2 year old burns
7-19 year old burns
Unburned
0.50
0.40
0.35
log total biomass (g)
Does burn severity and
soil moisture influence
the success of invasive
plant establishment?
M. alba
0.45
0.30
0.25
0.20
0.15
0.10
0.05
0.50
0.00
0.45
H. aurantiacum
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.50
B. inermis
0.45
0.40
log total biomass (g)
No significant differences
between invasive
biomass (or survival) in
different burn site types
at any point in the
experiment.
log total biomass (g)
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
Low Severity/
High Moisture
Low Severity/
Low Moisture
High Severity/
High Moisture
Burn Type
High Severity/
Low Moisture
MANAGEMENT IMPLICATIONS
• Burned areas with both high
invasive propogule pressure
and most suitable soils
should take highest
management priority (i.e.
Dalton Highway).
• Even low severity burns are
susceptible to invasive
species
Fire and Site Effects on
Tree Regeneration
4
6
8
0
1
2 3
4
5
6 7
0
20
40
60
80
1
2
3
4
5
6
6
0
4
5 6
7
0
2
4
Conif.sqrt
2
4
6
8
2
8
0
60
80
0
1
2 3
Decid.sqrt
4
5
6
0
20
40
org.cov
1
2
3
drainage
1
2
3
4
5
6
Modeling Seedling Recruitment
0.98
0.91
0.87
BS
WS
LP
0.50
drainage
R2=0.48
conifer
-0.42
resid.org
-0.48
0.31
0.36
elevation
-0.35
Model χ2=19.2, df=15, p=0.2
(no significant lack of fit)
deciduous
R2=0.57
TA
PB
0.86
0.85
Drivers of forest regeneration
• Differential sensitivity of
functional groups
– Deciduous most sensitive to
post-fire seedbeds
– Conifers more responsive to
site moisture
• Important role of fire severity
in potentially tipping the
balance between deciduous
and conifer dominance
Successional trajectories modelling
• How might changes in fire severity affect
landscape forest composition?
• Simulation experiments:
– ALFRESCO (Alaska frame-based ecosystem
model)
– 2004 Boundary Fire
– Examine potential patterns of forest recovery
2004 Boundary Fire
• Start with prefire vegetation
white spruce
deciduous
black spruce
2004 Boundary Fire
• Start with pre-fire
vegetation high crown
low surface
• Add on fire
severity (NBR)
• Include scenarios
for low or high
surface severity
moderate or low
crown severity
high crown
high surface
2004 Boundary Fire
• Start with pre-fire
vegetation high crown
low surface
• Add on fire
severity (NBR)
• Include scenarios
for low or high
surface severity
moderate or low
crown severity
high crown
high surface
2004 Boundary Fire
• Start with pre-fire
vegetation
• Add on fire
severity (NBR)
• Include scenarios
for low or high
surface severity
• Model black spruce
recovery trajectories
– extended deciduous
phase under high
surface severity
6
8
10
0
log(fire.hss.severe.100R[, 1, 1])
2
4
6
8
10
8
10
70
0 30
12
0
log(fire.lss.severe.100R[, 1, 1])
2
4
6
8
10
12
0
log(fire.lss.severe.100R[, 2, 1])
2
4
6
8
10
12
log(fire.mix.severe.100R[, 1, 1])
2
4
6
8
12
4
6
8
10
12
10
12
log(fire.mix.severe.100R[, 2, 1])
40
0
Frequency
0
10
mix|t=100|sev
0 10
Frequency
10 20
Frequency
0
2
8
log(fire.lss.severe.100R[, 3, 1])
Scenario mix|t=50|sev
3: 50% High-low surface fire severity
mix|t=75|sev
0
6
log(fire.hss.severe.100R[, 3, 1])
Frequency
0
4
lss|t=100|sev
50
12
2
lss|t=75|sev
0 20
Frequency
10
Frequency
0
2
6
log(fire.hss.severe.100R[, 2, 1])
Scenario lss|t=50|sev
2: Low surface fire severity
0
4
Frequency
15 30
0
12
80
4
hss|t=100|sev
100 years post-fire
40
2
hss|t=75|sev
75 years post-fire
0
0
Frequency
15
50 years post-fire
0
Frequency
Scenariohss|t=50|sev
1: High surface fire severity
0
2
4
6
8
10
12
log(fire.mix.severe.100R[, 3, 1])
Boundary Fire Simulations
• Possible to use real landscape and fire
data to project future forest composition
• Including variations in surface fire severity
– Alters projections of future forest cover
– Leads to vegetation effects on fire
propagation – even in conservative scenarios
• Can this knowledge be made useful to
managers?
Successional trajectories workbook
• Aim: to predict potential
changes in post-fire
trajectories
– Focused on black
spruce forest
– Integrates moisture and
severity effects
– Rapid assessment of
post-fire stands
– Identify conditions
leading to change
Subhygri
c
Very Considerable Moisture;
saturated by with less than
<5% standing water <10cm
deep
Mesic to
subhygric
Considerable moisture;
depressions
Subxeric
to mesic
Very noticable moisture; flat
to gently sloping
Subxeric
Noticable moisture; welldrained slopes ridges
Xeric
Little moisture; stabilized
sand dunes, dry ridge tops
% Gravel of soil
Moderate moisture; flat or
shallow depressions
Shallow permafrost
Mesic
Moist site with intact organic layer
• Rapid assessment:
– thick organics but good moisture
– trees standing with cones intact
• Predicted recruitment:
– very little deciduous
– low to high black spruce (depending on
quality of organics)
• Trajectory: open to closed black spruce
Moist site with low residual organics
• Rapid assessment:
– shallow organics with good moisture
– trees fallen (reduced seed dispersal)
– deciduous seed source within 1-2 km
• Predicted recruitment:
– high deciduous
– low to moderate black spruce
• Trajectory: mixed deciduous – black
spruce
Well drained site with intact organics
• Rapid assessment:
– thick organics with low surface moisture
– trees standing with cones intact
– no deciduous seed source nearby
• Predicted recruitment:
– very little deciduous
– moderate black spruce
• Trajectory: open to intermediate black
spruce
Well-drained site with low organics
• Rapid assessment:
– shallow organics with low surface moisture
– trees fallen
– strong nearby deciduous seed source
• Predicted recruitment:
– dense deciduous
– very low black spruce
• Trajectory: deciduous forest
Conditions leading to trajectory
shifts
• Substantial combustion of surface organic
layers => exposed mineral soil
– favors deciduous recruitment
• Deep charring of cone balls or toppling of
trees
– reduces black spruce seed rain
• Adjacent unburned stands of alternative
species
– provides strong propagule pressure