Hubbert’s Peak, The Question of Coal, and Climate Change

Dave Rutledge Chair, Division of Engineering and Applied Science Caltech “There is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact.” Mark Twain

Life on the Mississippi

slides (.ppt) and spreadsheets (.xls) at http://rutledge.caltech.edu/ 1

The UN Panel on Climate Change

• The UN Intergovernmental Panel on Climate Change publishes assessment reports that reflect the consensus on climate change • The 4 th report is being released this year – Over one thousand authors – Over one thousand reviewers • Updated measurements show that the temperature is rising 0.013

 C per year (1956-2005) 2

IPCC Climate-Change Predictions

• Report discusses climate simulations for fossil-fuel carbon-emission scenarios • There are 40 scenarios, each considered to be equally valid, with story lines and different government policies, population projections, and economic models 3

The 40 UN IPCC Scenarios

40 30 20 10 A1C AIM B1T Message 0 1980 2000 2020 2040 2060 2080 2100 Carbon Emitted A1 ASF A1 Message A1 Maria A1C Message A1G AIM A1G Minicam A1V2 Minicam A1T Message A2 ASF A2G Image A2 Minicam B1 Image B1 ASF B1 Maria B1T Message B1High Minicam B2 AIM B2 Image B2 Minicam B2C Maria A1 AIM A1 Image A1 Minicam A1C AIM A1C Minicam A1G Message A1V1 Minicam A1T AIM A1T Maria A2 AIM A2 Message A2-A1 Minicam B1 AIM B1 Message B1 Minicam B1High Message B2 Message B2 ASF B2 Maria B2High Minicam • • • • Data from the EIA (open symbols, 1980 to 2004) Emissions have increased 18% since the Kyoto Agreement was negotiated in 1997 Large differences in emissions among scenarios Oil production in 17 of the scenarios is greater in 2100 than in 2005 4

The Wall Street Journal

 April 5  Collapse of the World’s Second-Highest Producing Oil Field 5 World crude-oil production fell in 2006 by roughly the amount of this drop

Outline

• The 4 th UN IPCC Assessment Report • Hubbert’s peak – The history of US oil production – How much oil do the Saudis have?

– The future of world hydrocarbons – The Canadian oil sands • The coal question – British coal, a nearly complete history – Chinese coal – American coal – The future of world coal, by regions • Climate change – Simulations of future temperature and sea level – Carbon capture – Wind and sun • Concluding thoughts 6

King Hubbert

• Geophysicist at the Shell lab in Houston • In 1956, he presented a paper “Nuclear Energy and Fossil Fuels” at a meeting of the American Petroleum Institute in San Antonio • He made predictions of the peak year of US oil production based on two estimates of the ultimate production 7

Hubbert’s Peak

• • • From his 1956 paper Hubbert drew these by hand, and integrated by counting squares For the larger estimate, Hubbert predicted a peak in 1970 8

3

What Actually Happened?

1970 Hubbert’s Peak Alaskan oil 2 1 0 1900 1920 1940 1960 1980 2000 • • Data from the DOE’s Energy Information Administration (EIA) Production has dropped 15 years in a row 9

US Crude-Oil Production

200 29Gb remaining 100 0 1900 1950 2000 2050 2100 • • • EIA data (1859-2006) Cumulative normal (lms fit for ultimate of 225Gb, mean of 1975, and sd of 28 years) Hubbert’s larger ultimate was 200 billion barrels (the Alaska trend is 19 billion barrels) 10

The Largest US Oil Field Prudhoe Bay, Alaska Discovered 1968

11

600

Prudhoe Bay Oil Production

400 200 Trend for ultimate is 12 billion barrels 0 0 5 Cumulative Production, billions of barrels 10 • • FY1977-2006 data from the Alaska Department of Revenue, Tax Division Initially considered as 8 billion barrels of reserves 12

Estimating Remaining Production from Reserves is Challenging

•

Reserves

refer to fossil fuels that are appropriate to produce, taking the price into account • Reserves may be listed conservatively, as for Prudhoe Bay • Coal reserves have been too high, and they are often not properly distinguished from volume estimates for coal seams of a minimum thickness and a maximum depth • Often reserves are not adjusted for production • New discoveries are important for oil and natural gas • In most countries, the details of oil reserves are secret, and this means that the published reserves are political statements

resources,

which are 13

OPEC Reserves Go Up When the Price Goes Down!

40 100 30 50 0 1975 1980 1985 1990 1995 Iran Iraq Kuwait UAE Price 20 10 2000 2005 0 • • Data from the

2006 BP Statistical Review

269Gb rise in reserves, no adjustment for 65Gb produced since 1986 14

Saudi Reserves

Saudi control 264Gb reserves 200 Nehring RAND study 176Gb reserves 100 0 1975 1980 1985 1990 1995 2000 2005 • • Data from the

2006 BP Statistical Review

95Gb rise in reserves, no adjustment for 53Gb of production since 1988 15

Estimating Remaining Production from a Graph

• • In plots of annual production vs cumulative production – We can estimate the remaining production from a trend line – Advantage is that we can identify points on the trend line – Disadvantage is that we cannot make an estimate until the production drops Alternative is to plot the growth rate of the cumulative production (annual production over cumulative production) instead of the annual production – First applied to

Daphnia

production in 1982 populations in biology in 1963 – King Hubbert introduced this approach for estimating remaining oil – Advantage is that we can make an estimate before the peak – Disadvantage is that we need to know the cumulative production 16

Growth-Rate Plot for US Crude Oil

10% 5% Trend line is for normal fit (225 billion barrels) 0% 0 100 Cumulative Production, billions of barrels 200 • EIA data (cumulative from 1859, open symbols 1900-1930, closed symbols 1931-2006) 17

How Much Oil do the Saudis Have?

10% Trend line is for 1978 RAND study (90Gb remaining) Official Saudi reserves are 264 billion barrels 5% 0% 0 50 100 150 Cumulative Production, billions of barrels 200 • • EIA data (open 1975-1990, closed 1991-2006), 1975 cumulative from Richard Nehring Matt Simmons was the first to call attention to this anomalous situation in his book,

Twilight in the Desert

18

Growth-Rate Plot for World Hydrocarbons

6% 4% Trend line for 3Tboe remaining 2% • • • 0% 0 1 2 Cumulative Production, trillion barrels of oil equivalent 3 Oil + natural gas + natural gas liquids like propane and butane Data 1965, 1972, 1981, 2006

BP Statistical Review

(open 1960-1982, closed 1983-2005) The German resources agency BGR gives hydrocarbon reserves as 2.7Tboe – Expectation of future discoveries and future OPEC oil reserve reductions – Includes 500Gboe for non-conventional sources like Canadian oil sands 19

World Hydrocarbon Production

2 1 4 3 3Tboe remaining 0 1960 1980 2000 2020 2040 2060 2080 2100 • • Cumulative normal (ultimate 4.6Tboe, lms fit for mean 2018, sd 35 years) IPCC scenarios assume that 11 to 15Tboe is available 20

Fort McMurray, Alberta Oil Sands

21

Canadian Oil Sands

• 1.0 Mb per day in 2005, increasing 8% per year • 35Gb reserves for mining (comparable to one year of world oil production) • 140Gb reserves for wells – Production with a steam process – Production and upgrading to synthetic crude oil use 25% of the oil energy equivalent in natural gas – Canadian gas reserves are 10Gboe (end of 2005) – Annual gas production is 12% of reserves per year – Challenges in meeting obligations under the Kyoto agreement • The Uppsala Hydrocarbon Depletion Group were the first to call attention to these limitations 22

Outline

• The 4 th UN IPCC Assessment Report • Hubbert’s peak – The history of US oil production – How much oil do the Saudis have?

– The future of world hydrocarbons – The Canadian oil sands • The coal question – British coal, a nearly complete history – Chinese coal – American coal – The future of world coal, by regions • Climate change – Simulations of future temperature and sea level – Carbon capture – Wind and sun • Concluding thoughts 23

British Coal

24

300

British Coal Production

200 100 0 1850 1900 1950 2000 • • Data from the US National Bureau of Economic Research (1854-1876), the Durham Coal Mining Museum (1877-1956), and the British Department of Trade and Industry (1957-2006) In the peak production year, 1913, there were 3,024 mines 25

Growth-Rate Plot for British Coal

4% 2% 0% 0 5 10 15 Cumulative Production, Gt 20 25 • • 1854-2006, 1853 cumulative from William Jevons,

The Coal Question

Already near the trend line in 1854 26

Remaining Production for British Coal

0.2% 10% per year 0.1% Trend line for 200Mt remaining 0.0% 26.2

26.4

26.6

Cumulative Production, Gt • • • Data from the UK Department of Trade and Industry (1993-2006) 6 producing underground mines  several with less than ten years of coal

Cumulative British Coal Production

Pre-war fit 20 Post-war fit 10 0 1850 1900 1950 2000 • • Pre-war lms fit (1854-1945, ultimate 25.6Gt, mean 1920, sd 41 years) Post-war lms fit (1946-2006, ultimate 27.2Gt, mean 1927, sd 39 years) 28

Reserves-to-Production Ratio for UK Coal

900 600 300 0 0 50 100 Years since Edward Hull's Reserve Survey in 1864 • • • 1864 reserves from Edward Hull of the Geological Survey Other data from the World Energy Council Surveys Current R/P ratio is 7 years 29

Reserves vs Remaining Production

1000 100 Hull Resources + Reserves Reserves 10 Remaining Production 1 0 1850 1900 1950 2000 • • • 1864 reserves from Edward Hull of the Geological Survey Other data from the World Energy Council Surveys of Energy Resources Resources include seams of 2ft or more at depths of 4000ft or less 30

Fraction of Reserves Eventually Produced

40% Hull Hull 20% 0% 1850 1900 1950 2000 • • • 1864 reserves from Edward Hull of the Geological Survey Other data from the World Energy Council Surveys of Energy Resources Will use trends if they exist, reserves otherwise 31

Why Are Coal Reserves Too High?

• • • It seems likely that there are many social, environmental, and technical hindrances that are not fully taken into account in the reserve estimates The German Energy Watch Group was early in pointing out that there is a problem with reserves worldwide Here are some technical restrictions from the USGS 2000 National Coal Assessment for the Illinois basin 32

Production and Reserves

China USA India Australia Russia South Africa World Production, Gt 2.38

1.05

0.45

0.37

0.31

0.26

6.20

Reserves, Gt 189 247 92 79 157 29 963 • • 2005 Production numbers from the BP 2006 Statistical Review Reserves from the World Energy Council surveys of resources (2006/2007 South Africa Yearbook for South Africa, and the Chinese Ministry of Land and Resources 2001 by way of Sandro Schmidt at the BGR) 33

Chinese Coal

34

15%

Growth-Rate Plot for China

10% 5% Trend line for 70Gt remaining Reserves are 189Gt 0% 0 10 20 30 Cumulative Production, Gt 40 50 • • Data from Tim Wright, D.W. Dwyer, and BP 2006 Statistical Review (cumulative from 1896, open symbols 1918-1961, closed symbols 1962-2005), corrections by Jianjun Tu Reserves from the Chinese Ministry of Land and Resources 2001 by way of Sandro Schmidt at the BGR 35

Cumulative Production for China

100 50 0 1950 2000 2050 2100 • Cumulative normal (ultimate 111Gt, lms fit for mean 2015 and sd 27 years) 36

American Coal

37

1,000

US Coal Production

Total 500 0 1850 • • 1900 1950 Data from the USGS (Robert Milici) Will consider the East and the West separately West of the Mississippi 2000 38

Anthracite in Pennsylvania

80 60 40 20 0 1850 1900 1950 • • Data from the USGS (Robert Milici) Anthracite is a grade of coal used for home heating that burns with little smoke 39

Growth-Rate Plot for PA Anthracite

6% 4% 2% 0% 0 1 2 3 Cumulative Production, Gt 4 5 • • Data from the USGS (Robert Milici) cumulative from 1800, symbols 1875-1995 16% of the 1913 reserves have been produced 40

Cumulative PA Anthracite Production

5 2 1 4 3 0 1850 • 1900 1950 2000 Normal lms fit for ultimate 5.00Gt, mean 1916, and sd 27 years 41

40

Bituminous Coal in Virginia

20 0 1900 1950 2000 • • Data from the USGS (Robert Milici) and the EIA Virginia has coal with high energy content, and much of it is used for metallurgy 42

Growth-Rate Plot for VA Bituminous

10% 8% 6% Pre-war Trend WWII 4% Trend is for 800Mt remaining Reserves are 2.8Gt

2% • • 0% 0 1 2 Cumulative Production, Gt 3 Data from the USGS (Robert Milici) cumulative from 1800, closed 1900-1940, open 1941-1945, closed 1946-2006, reserves from the EIA Trend is for 16% of the 1924 reserves to eventually be produced 43

Cumulative VA Bituminous Production

3 Post-war fit 2 1 Pre-war fit 0 1900 1950 2000 2050 • • Pre-war normal (ultimate 0.40Gt, lms fit for mean 1926 and sd 16 years) Post-war normal (ultimate 3.03Gt, lms fit for mean 1984 and sd 34 years) 44

Coal East of the Mississippi

8% 6% 4% 2% Pre-war Trend WWII Trend is for 40Gt remaining Reserves are 96Gt • • 0% 0 20 40 Cumulative Production, Gt 60 Does not include Pennsylvania anthracite Data from the USGS (Robert Milici) cumulative from 1800, closed 1900 1940, open 1941-1948, closed 1949-2005, reserves from the EIA 45

Cumulative Production for the East

Post-war fit 40 20 Pre-war fit 0 1850 1900 1950 2000 • • • Does not include Pennsylvania anthracite Pre-war normal (ultimate 20Gt, lms fit for mean 1924 and sd 20 years) Post-war normal (ultimate 86Gt, lms fit for mean 1999 and sd 67 years) 46

Western Coal

47

Coal West of the Mississippi

10% 5% Trend is for 25Gt remaining Reserves are 79Gt without Montana Pre 70’s trend 0% 0 5 10 Cumulative Production, Gt 15 • • • Data from the USGS (Robert Milici) closed 1800-1970, open 1971-1978, closed 1979-2005 Reserves from the EIA

Cumulative Production for the West

15 Post 70’s fit 10 5 Pre 70’s fit 0 1900 1950 2000 • • Pre 70’s normal (ultimate 1.6Gt, lms fit for mean 1929 and sd 23 years) Post 70’s normal (ultimate 38Gt, lms fit for mean 2016 and sd 25 years) 49

Growth-Rate Plot for Australia and New Zealand 6% 5% Trend line for 50Gt remaining Reserves are 79Gt 4% 3% 3 10 Cumulative Production, Gt • • • Data (1981-2005) from the 2006 BP Statistical Review 1990 Australia cumulative from the

History of Coal Mining in Australia

, A.J. Hargraves Reserves from the 2004 World Energy Council survey 50

Growth-Rate Plot for Europe

2% 1% Trend line for 23Gt remaining Reserves are 55Gt 0% 50 75 Cumulative Production, Gt 100 • • • Data (1981-2005) from the 2006 BP Statistical Review 2005 cumulative from the 2005 BGR Energy Resources Report Reserves from the 2004 World Energy Council survey, down from 171Gt in 1990 51

8%

Growth-Rate Plot for Africa

6% 4% 2% Trend line for 10Gt remaining Reserves are 30Gt 0% 0 5 Cumulative Production, Gt 10 • • • Data (open 1981-1990, closed 1991-2005) from the 2006 BP Statistical Review 2005 cumulative from the 2005 BGR Energy Resources Report South African reserves were recently reduced by 20Gt (2006/2007 South Africa Yearbook) 52

4%

Former Soviet Union

Trend line for 18Gt remaining 1996 reserves are 157Gt 2% • • • • 0% 15 25 Cumulative Production, Gt 35 Data from BP (closed 1981-1988, open 1989-2005) 2005 cumulative from the 2005 BGR Energy Resources Report Drop that started in 1989 is from the collapse of the Soviet Union Reserves from World Energy Council surveys, unchanged since the collapse of the Soviet Union 53

Growth-Rate Plot for South Asia

10% Exponential Growth Reserves are 111Gt 5% 0% 0 5 10 Cumulative Production, Gt • • • Data (1965-2005) from the 2006 BP Statistical Review Earlier production from World Energy Council Surveys Reserves from the 2004 World Energy Council survey 15 54

Growth-Rate Plot for Central and South America 10% 5% Exponential Growth Reserves are 20Gt 0% 0.0

0.5

1.0

Cumulative Production, Gt • • • Data (1981-2005) from the 2006 BP Statistical Review 2005 Cumulative from the BGR Resources Report Reserves from the 2004 World Energy Council survey 1.5

55

Reserves vs Trends for Remaining Production Region North America East Asia Australia and New Zealand Europe Africa Former Soviet Union South Asia Central and South America World (at 3.6boe/t) Reserves Gt 255 190 79 55 30 223 Trends Gt 135 70 50 23 10 18 111 20 963 (3.5Tboe) 437 (1.6Tboe) • • North America includes trends for the East (40Gt), the West (25Gt), reserves for Montana (68Gt), and trends for Canada and Mexico (2Gt) IPCC scenarios assume 18Tboe is available for production 56

Future Fossil-Fuels Production

4 3 2 3.0Tboe hydrocarbons remaining 1 1.6Tboe coal remaining 0 1960 1980 2000 2020 2040 2060 2080 2100 • • • • Hydrocarbons cumulative normal (ultimate 4.6Tboe, lms fit for mean 2018, sd 35 years) 2005 coal cumulative from the 2005 BGR Energy Resources Report (USGS for US) Coal cumulative normal (ultimate 2.6Tboe, lms fit for mean 2024, sd 48 years) The standard deviations of 35 and 48 years can be compared to time constants for temperature and sea level 57

Outline

• The 4 th UN IPCC Assessment Report • Hubbert’s peak – The history of US oil production – How much oil do the Saudis have?

– The future of world hydrocarbons – The Canadian oil sands • The coal question – British coal, a nearly complete history – Chinese coal – American coal – The future of world coal, by regions • Climate change – Simulations of future temperature and sea level – Carbon capture – Wind and sun • Concluding thoughts 58

Fossil-Fuel Carbon Emissions

800 600 400 Producer-Limited Profile Super-Kyoto Profile 520Gt remaining 200 0 1960 1980 2000 2020 2040 2060 2080 2100 • • • Total fossil-fuel carbon is an input for climate-change models Carbon coefficients from the EIA: oil (110kg/boe), gas (79kg/boe), coal (141kg/boe), and future hydrocarbons weighted by BGR reserves (98kg/boe)

Comparing with the IPCC Scenarios

2,000 1,000 Producer-Limited Profile 0 2000 2050 • • Jean Laherrere was the first to point out this anomalous situation 2100

10

Simulated CO

2

Levels

430 380 5 Producer Limited Carbon Super Kyoto Carbon Producer Limited CO2 Super Kyoto CO2 330 • • • 0 2000 2100 2200 Year 2300 280 2400 Predictions using the program MAGICC from Tom Wigley at the National Center for Atmospheric Research in Boulder with a modified WRE profile The Producer-Limited Profile gives a peak CO 2 The Super-Kyoto Profile gives a 440ppm peak concentration of 460ppm in 2070 61

Temperature Rises Associated with Future Fossil-Fuel Use

0.8

0.4

• • • • • 0.0

2000 2100 2200 Year 2300 2400 Predictions from Tom Wigley’s MAGICC (no mechanical ice model) The temperature rise is a maximum of 0.8

 C in 2100 The Super-Kyoto Profile (dashed lines) reduces the maxima by 0.04

 C Time constant is of the order of a thousand years (an integrator) Sensitivity to errors is 0.0012

 C/Gt carbon 62

CO

2

Capture and Storage for Coal Power Plants

• MIT has just completed an outstanding study,

The Future of Coal

, that gives a cost of $150/t of carbon avoided • To reduce the temperature in 2100 by 0.001

 C, the cost would be 100 billion dollars • Additional cost for transportation and burial – A distribution system is needed that is comparable to our present natural gas pipeline system – Cannot have leaks on the time scale of a thousand years 63

Wind and Sun

• • • • The time constants of around 50 years for fossil-fuel exhaustion imply that a transition to renewable sources of energy is likely Wind is the fastest growing renewable – Current world capacity is 74GW, increasing at 25% per year – 19% of new US capacity last year – Advantage is a production learning curve Solar photovoltaics for the home and business – World production in 2006 was 2.2GW, up 33% from 2005 – Advantage is that there is no need for new transmission lines – Caltech is installing a 230-kW plant on top of a parking structure Concentrating solar – Current capacity is 350MW, built in the 80s in the Mojave Desert – New Nevada Solar One with 64MW near Las Vegas – Advantages are that it uses the direct sunlight available in the Southwest, and the possibility of thermal storage – The major California utilities, Southern California Edison, San Diego Gas and Electric, and Pacific Gas and Electric, are each planning to spend a billion dollars on concentrating solar plants 64

Kramer Junction, California

65

 • • From Schott Glass Area in red circle in California could supply sufficient energy to replace the entire US grid

Nevada Solar One June 2, 2007

67

Concluding Thoughts

• • • Results – Estimate for future hydrocarbon production (3Tboe) is consistent with reserves – Estimate for future coal production (1.6Tboe) is about half of reserves – The time constants for fossil-fuel exhaustion are of the order of 50 years – The time constant for temperature is of the order of 1,000 years Implications – Since estimate for future fossil-fuel production is less than all 40 UN IPCC scenarios, producer limitations could provide useful constraints in climate modeling – A transition to renewable sources of energy is likely – To lessen the effects of climate change associated with future fossil-fuel use, reducing ultimate production is more important than slowing it down Opportunities – One-third of US fossil-fuel reserves are on federal lands, so ultimate production could be reduced substantially by limits on new leases for mining and drilling – The US has an outstanding resource in its direct sunlight 68

Thanks for Advice, Criticism, Discussion, and Slides • Tom Wigley and Steve Smith at the National Center for Atmospheric Research in Boulder • Bill Bridges, Dave Goodstein, Melany Hunt, John Ledyard, Ken Pickar, Tapio Schneider, John Seinfeld, and Tom Tombrello at Caltech • Dimitri Antsos at the Jet Propulsion Laboratory • John Rutledge at Freese and Nichols, Inc. in Fort Worth • Charlie Kennel at the University of California at San Diego • Sandro Schmidt at the BGR • Juha Karhu at the University of Helsinki Special thanks to Sandy Garstang in the Caltech Library and Dale Yee in the Caltech Engineering Division for their ingenuity in locating mining records 69