Thomas S. Ahlbrandt, a high-level USGS petroleum-supply analyst, and Kennneth S. Deffeyes, author of Hubbert's Peak: The Impending World Oil Shortage, and professor at Princeton geology department, will debate on the nature of the evidence for and against the thesis of Deffeyes' book: that world oil production will--contrary to USGS projections--soon reach a peak and thereafter go into a decline, never to rise again.

The authors' abstracts follow:

Future Oil and Gas Resources of the World: A Coming Supply Crisis?
Thomas S. Ahlbrandt (303-236-5776; ahlbrandt@usgs.gov)
U.S. Geological Survey, P.O. Box 25046 MS 939, Denver, CO 80225, United States

Is the world running out of oil? Where will future oil and gas supplies come from? To help answer these questions, the U.S. Geological Survey completed in 2000 a new assessment of the undiscovered conventional oil and gas resources and potential additions to reserves from field growth. One hundred and twenty-eight provinces were assessed in a 100 man-year effort from 1995-2000. The assessed provinces included 76 priority provinces containing 95 percent of the world's discovered oil and gas and an additional 52 "boutique" provinces, many of which may be highly prospective. Total Petroleum Systems (TPS) were identified and described for each of these provinces along with associated Assessment Units (AU) that are the basic units for assessing undiscovered petroleum. The assessment process coupled geologic analysis with a probabilistic methodology to estimate remaining potential. Within the 128 assessed provinces, were 159 TPS and 274 AU. For these provinces, the endowment of recoverable oil, which includes cumulative production, remaining reserves, reserve growth, and undiscovered resources is estimated at about 3 trillion barrels of oil (TBO). The natural gas endowment is estimated at 2.6 trillion barrels of oil equivalent (TBOE). Oil reserves are currently 1.1 TBO; world consumption is about .028 TBO per year. Natural gas reserves are about .8 TBOE; world consumption is about .014 TBOE. Thus, without any additional discoveries of oil, gas or natural gas liquids, we have about 2 TBOE of proved petroleum reserves. Of the oil and gas endowment of about 5.6 TBOE, we estimate that the world has consumed about 1 TBOE, or 18 percent leaving about 82 percent of endowment to be utilized or found. Half of the world's undiscovered potential is offshore. Arctic basins with about 25 percent of undiscovered petroleum resources make up the next great frontier. An additional 279 provinces contain some oil and gas and, if considered, would increase the oil and gas endowment estimates. While petroleum resources in the world appear to be significant, certain countries such as the U.S. may run into import deficits particularly oil imports from Mexico and natural gas from Canada. The new assessment has been used as the reference supply case in energy supply models by the International Energy Agency and the Energy Information Agency of the Department of Energy. Climate energy modeling groups such as Stanford, Massachusetts Institute of Technology and others have also used the estimates in global climate models. Many of these models using the USGS estimates, converge on potential oil shortfalls in 2036-2040. A transition to increased use of natural gas is expected, but gas in turn may experience similar supply concerns in the 2050-2060 time frame. A coal bridge-to-the-future model as well a realistic view of non-renewable resources in the future will be discussed. Non-conventional oil and gas are quite common in the petroleum provinces of the world and represent a significant resource yet to be fully studied and developed. Seventeen non-conventional AU, including coal-bed methane, basin-center gas, continuous oil, and gas hydrate occurrences, have been preliminarily identified for future assessment. Initial efforts to assess heavy oil deposits and other non-conventional oil and gas deposits are also underway. Digital products from the World Energy Project may be downloaded at (http://energy.cr.usgs.gov/energy/WorldEnergy/WEnergy.html).

After M. King Hubbert initially predicted in 1956 that U.S. oil production would peak in the early 1970s, he continued to refine his methodology. In addition to annual oil production, Hubbert used cumulative "discoveries", defined as the cumulative amount of oil produced up to a given year plus the known reserves as of that same year. Because oil is added to the reserves several years before it is produced, the use of discoveries allowed Hubbert to peek farther into the future. Hubbert utilized the simplest hypothesis that might fit the observations: 1) annual production would fit a symmetrical bell-shaped curve, and 2) discoveries would follow an identical bell curve, earlier in time by a fixed number of years. There were four free parameters in the fit: the annual production at the peak, the width (at half height) of the curve, the year of peak production, and the peak year of discoveries. In 1982, when he was 79 years old, Hubbert published a long paper justifying his choice of the logistic bell curve, rather than the more familiar Gaussian. He pointed out that the differential equation underlying the logistic curve could be interpreted as an annual growth rate that depends linearly on the fraction of oil remaining undiscovered. The logistic differential equation could be rearranged into a linear form that turned a logistic curve into a straight line. (Hubbert did not mention that population biologists had been using the same straight-line graph.) In 2001, I showed that both the production and discovery data could be plotted together on the same graph. An interesting puzzle arose: fitting the bell-shaped curves required four free parameters, but a straight line has only two free parameters. It turns out that the center year for production, and for discoveries, are found only indirectly from the straight-line graph. Almost all economists reject the entire Hubbert analysis because the effect of oil prices on exploration effort and production methods is ignored. The straight-line logistic graph takes a small step towards satisfying the economists' objections. At times of high prices, we move rapidly down the straight line but success continues to depend linearly on the fraction of undiscovered oil. The graph for world production and discoveries, updated to include the year 2001 production and reserves, predicts a production peak around the year 2005. U.S. policy initiatives now under discussion focus on targets in the year 2011 through 2015.