UC Berkeley

There are two commonly-used approaches to modeling the future supply of mineral resources. One is to estimate reserves and compare the result to extraction rates, and the other is to project from historical time series of extraction rates. Perceptions of abundant oil supplies in the Middle East and abundant coal supplies in the United States are based on the former approach. In both of these cases, an approach based on historical production series results in a much smaller resource estimate than aggregate reserve numbers. This difference is not systematic; natural gas production in the United States shows a strong increasing trend even though modest reserve estimates have resulted in three decades of worry about the gas supply. The implication of a future decline in Middle East oil production is that the market for transportation fuels is facing major changes, and that alternative fuels should be analyzed in this light. Because the U.S. holds very large coal reserves, synthesizing liquid hydrocarbons from coal has been suggested as an alternative fuel supply. To assess the potential of this process, one has to look at both the resource base and the net efficiency. The three states with the largest coal production declines in the 1996 to 2006 period are among the top 5 coal reserve holders, suggesting that gross coal reserves are a poor indicator of future production. Of the three categories of coal reserves reported by the U.S. Energy Information Administration, reserves at existing mines is the narrowest category and is approximately the equivalent of proved developed oil reserves. By this measure, Wyoming has the largest coal reserves in the U.S., and it accounted for all of U.S. coal production growth over the 1996 to 2006 time period.

In Chapter 2, multi-cycle Hubbert curve analysis of historical data of coal production from 1850 to 2007 demonstrates that U.S. anthracite and bituminous coal are past their production peak. This result contradicts estimates based on aggregated reserve numbers. In Chapter 4, a similar analysis of world coal results in a peak production year of 2011. Electric power generation consumes 92 percent of U.S. coal production. Natural gas competes with coal as a baseload power generation fuel with similar or slightly better generation e±ciency. Fischer-Tropsch synthesis, described in Chapter 2, creates transportation fuel from coal with an e±ciency of less than 45 percent. Claims of higher e±ciencies are based on waste heat recovery, since this is a highly exothermic process. The yield of fuel as a proportion of the energy content of the coal input is always less than 45 percent. Compressed natural gas can be used for vehicle fuel with e±ciency greater than 98 percent. If we view Fischer-Tropsch synthesis as a form of arbitrage between markets

for electricity and transportation fuel, coal cannot simultaneously compete with natural gas for both transportation fuel and electric power. This is because Fischer-Tropsch synthesis is a way to turn power generation fuel into transportation fuel with low efficiency, while natural gas can be converted to transportation fuel with much greater efficiency. For this reason, Fischer-Tropsch synthesis will be an uneconomic source of transportation fuel as long as natural gas is economic for power generation. This conclusion holds even without the very high capital cost of coal-to-liquids plants.

The Intergovernmental Panel on Climate Change (IPCC) has generated forty carbon production and emissions scenarios, see the IPCC Special Report on Emissions Scenarios (2000). Chapter 4 develops a base-case scenario for global coal production based on the physical multi-cycle Hubbert analysis of historical production data. Areas with large resources but little production history, such as Alaska or Eastern Siberia, can be treated as sensitivities on top of this base case. The value of our approach is that it provides a reality check on the magnitude of carbon emissions in a business-as-usual (BAU) scenario. The resulting base case is significantly below 36 of the 40 carbon emission scenarios from the IPCC, and the global peak of coal production from existing coalfields is predicted to occur about the year 2011. The peak coal production rate calculated here is 160 EJ/y, and the associated peak carbon emissions from coal burning are 4.5 Gt C per year. After 2011, the production rates of coal and CO2 decline, reaching 1990 levels by the year 2037, and reaching 50 percent of the peak value in the year 2047. It is unlikely that future mines will reverse the trend predicted in the base case scenario here, and current efforts to sequester carbon or to convert coal into liquid fuels should be reexamined in light of resource limits.

California provides a good example of the implications of scarcity-driven changes in the fuel supply, and the limited policy options available to deal with them. Once a great oil producer, California is becoming increasingly dependent on oil from the Middle East and Ecuador. Middle East production is not increasing, yet oil cargoes from the Middle East have to pass growing Asian markets to reach California. Alaska and Mexico also supply oil to the Pacific Basin, but are facing production declines. The effect of rising Asian demand on Pacific Basin oil markets is already visible, with significant amounts of oil coming to California from Atlantic Basin sources such as Angola, Brazil and Argentina. Policy options that could affect California's oil supply security include increased oil development in Alaska or offshore California, development of additional heavy oil pipeline outlets on Canada's Pacific Coast or substituting natural gas for oil when possible. The proposed low-carbon fuel standard will negatively impact California's energy supply security.