GEOLOGY:
The Next Oil Crisis Looms Large--and Perhaps Close

Nature took half a billion years to create the world's oil, but observers agree that humankind will consume it all in a 2-century binge of profligate energy use. For now, as we continue to enjoy the geologically brief golden age of oil, the conventional outlook for oil supply is bright: In real dollars, gasoline has never been cheaper at the pump in the United States--and by some estimates there are a hefty trillion barrels of readily extractable oil left in known fields. Thanks to new high-tech tricks for finding and extracting oil, at the moment explorationists are adding to oil reserves far faster than oil is being consumed. So, many who monitor oil resources, especially economists, see production meeting rising demand until about 50 years from now--plenty of time for the development of alternatives.

Comforting thinking--but wrong, according to an increasingly vociferous contingent, mainly geologists. They predict that the world will begin to run short of oil in perhaps only 10 years, 20 at the outside. These pessimists gained a powerful ally this spring when the Paris-based International Energy Agency (IEA) of the Organization for Economic Cooperation and Development (OECD) reported for the first time that the peak of world oil production is in sight. Even taking into account the best efforts of the explorationists and the discovery of new fields in frontier areas like the Caspian Sea (see sidebar on p. 1130), sometime between 2010 and 2020 the gush of oil from wells around the world will peak at 80 million barrels per day, then begin a steady, inevitable decline, the report says.

"From then on," says consulting geologist L. F. Ivanhoe of Novum Corp. in Ojai, California, "there will be less oil available in the next year than there was in the previous year. We're not used to that." Scarce supply, of course, means a higher price, especially because optimists and pessimists alike agree that the Organization of Petroleum Exporting Countries (OPEC), which triggered the oil crises of 1973 and 1979, will once again dominate the world oil market even before world oil production peaks (see sidebar on p. 1129). At the peak and shortly thereafter, as more expensive fuel sources such as hard-to-extract oil deposits, the tarry sands of Canada, and synfuels from coal are brought on line, prices could soar. "In the 5 to 10 years during the switch, there could be some very considerable price fluctuations," says an IEA official. "Then we will plateau out at a higher but not enormous price level." In other words, gas lines like those of the Arab oil embargo 25 years ago could return temporarily, followed by permanently expensive oil.

The down side of the curve
The debate over just when the end of cheap oil will arrive pivots on an interplay of geology and technology. There's only so much oil in the ground, geologists and technology-loving economists agree, but how much of it geologists can find and engineers can extract at a reasonable cost is much in contention. Geologists considering the past record of finding and extracting oil see a fixed, roughly predictable amount left to be produced and put the production peak sooner rather than later. Their case for the past being the best predictor of the future depends heavily on their success in predicting the oil production peak of the lower 48 states of the United States, the only major province whose oil production has already peaked.

For projections of future oil production, many geologists rely on a kind of analysis pioneered by the late geologist M. King Hubbert. In 1956, when he was at Shell Oil Co., he published a paper predicting that, based on the amount and rate of past production, output in the lower 48 states--which was then increasing rapidly from year to year--would peak between 1965 and 1970 and then inexorably decline. "The initial reaction to this conclusion was one of incredulity--'The man must be crazy!' " Hubbert later recalled. But production peaked right on schedule in 1970 and has declined since.

Hubbert based his successful prediction on what seemed to him a fundamental law governing the exploitation of a finite resource--that production will rise, peak, and then fall in a bell-shaped curve. He constructed his curve by noting that extraction of oil begins slowly and then accelerates as exploration finds more of the huge fields that are too big to miss and that hold most of the oil. That's the ascending side of his bell-shaped curve.

After this fast start, production begins to stall. By this point, exploration has turned up most of the easy-to-find huge fields. The smaller fields, although far more numerous, are harder to find, more expensive to drain, and can't match the volume of the big fields. At the same time, the gush of oil from the big fields slows. Oil in a reservoir lies in pores whose surfaces hold onto it like a sponge, so that wells first gush, then slow toward a trickle. The declining rate of oil discoveries and slowing production from big, early finds combine to force overall production to peak--the top of Hubbert's curve--at about the time that half of all the oil that will ever be recovered has been pumped. From then on, production drops as fast as it rose, creating Hubbert's idealized symmetrical bell-shaped curve.

When applied to world oil production, Hubbert's curve traces out a relatively grim future. During the oil crisis of 1979, Hubbert himself made a rough estimate of a turn-of-the-century world peak. At that time, though, geologists were slightly underestimating just how much oil Earth contains, and Hubbert's forecast was too gloomy--but perhaps not by much. In recent years, a half-dozen Hubbert-style estimates have been made, and they all cluster around a world oil production peak near 2010 (see table). Half the world's conventional oil has already been pumped, these geologists say, so the beginning of the end is in sight.

PREDICTED PEAK IN WORLD OIL PRODUCTION
SOURCE	PEAK DATE
F. Bernabe, ENI SpA (1998)	2000-2005
C. Campbell and J. Laherrére, Petroconsultants (1998)	2000-2010
J. MacKenzie, World Resources Institute (1996)	2007-2014
OECD's International Energy Agency (1998)	2010-2020
J. Edwards, University of Colorado, Boulder (1997)	2020
DoE's Energy Information Administration (1998)	>2020

SOURCE: C. CAMPBELL AND J. LAHERRERE

One of the most pessimistic recent analyses comes from former international oil geologists Colin Campbell and Jean Laherrére, who are associated with Petroconsultants in Geneva; Campbell was an adviser to the IEA on its latest estimate. "Barring a global recession, it seems most likely that world production of conventional [easily extracted] oil will peak during the first decade of the 21st century," they wrote in the March issue of Scientific American.

Campbell and Laherrére's early peak prediction is drawn in part from their low estimates of existing reserves. Of the trillion barrels of oil that countries have reported finding but not yet extracted--their proven reserves--Campbell and Laherrére accept only 850 billion barrels. Much of the rest they view as "political reserves"--overly generous estimates made for political reasons. For example, reserves jumped by 50% to 200% overnight in many OPEC countries in the late 1980s--perhaps because OPEC rules allow countries with more declared reserves to pump more oil and so make more money, says Campbell.

Not all Hubbert-type estimates are quite so pessimistic. "I'm an optimist," says former oil industry geologist John Edwards of the University of Colorado, Boulder. "I think there's a lot more oil to be found. I used optimistic numbers [near the high end of estimated reserves], but I'm still at 2020" for the world production peak. "Conventional oil is an exhaustible resource. That's just the bottom line."

Technology to the rescue?
But other geologists and many economists put more faith in technology. Oil will eventually run out, these self-described optimists agree, but not so soon. "We're 30, maybe even 40, years before the peak," says oil geologist William Fisher of the University of Texas, Austin. Fisher has lots of support from the latest international energy outlook prepared by the U.S. Department of Energy's Energy Information Administration (EIA). "We don't see the peak happening until after the limit of our outlook," in 2020, says the EIA's Linda Doman. "We think technology and developing Middle East production capacity will provide the oil."

In the optimists' view, it doesn't matter that there are few if any huge new fields left out there to find. What does matter, they say, is how much more oil the industry can find and extract in and around known fields. Even as the world consumes 26 billion barrels a year, in their opinion reserves are growing rapidly. They argue that much of OPEC's reserve growth is real, and that OPEC and others are boosting reserves not so much through the discovery of new fields as through the growth of existing fields--and technology is the key. Technology might double the yield from an established field, they say. "Technology has managed to offset the increasing cost of finding and retrieving new resources," says economist Douglas Bohi of Charles River Associates in Washington, D.C. "The prospect is out there for an amazing increase in the [oil] reserve base."

Three currently used technologies are helping drive this boost in reserves, Bohi and others say. Aided by supercomputers, explorationists are using the latest three-dimensional seismic surveying to identify likely oil-containing geologic structures, yielding a sharp picture of potential oil reservoirs. A second technology involves first drilling down and then sideways, punching horizontally through a reservoir so as to reduce the number of wells needed, and therefore the expense, by a factor of 10. Finally, technology that allows wells to be operated on the sea floor many hundreds of meters down is opening up new areas in the Gulf of Mexico, off West Africa, and in the North Sea.

All these new technologies can slow or delay what Hubbert saw as an inexorable production drop in older fields, the optimists say. Indeed, such technological achievements have already helped arrest the decline of U.S. oil production during the past 3 to 4 years, says Edwards.

But the pessimists are unmoved. "Much of the technology is aimed to increase production rates," says Campbell. "It doesn't do much for the reserves themselves." And what new technology does do for reserves, it has been doing since the oil industry began in the 19th century, he says. New technologies for better drilling equipment and seismic probing have been developed continually rather than in a sudden leap and so have been boosting the Hubbert curves all along. The shape of the curve therefore already incorporates steady technology development, he and other pessimists note.

As a result, they argue that today's technological fixes will make only slight changes to the curve. "All these things the economists talk about are just jiggling in a minor way with the curve," says Albert Bartlett, a physicist at the University of Colorado, Boulder, who calculates a 2004 world peak. "You can get some bumps on the [U.S.] curve by breaking your back, but the trend is down." For example, when oil hit $40 a barrel in the early 1980s, the U.S. production curve leveled out in response to a drilling frenzy--but it soon went right back down again. And besides, the pessimists note, when high prices drive increased production, the oil pumped is not cheap oil. Economist Cutler Cleveland of Boston University has found that the price-driven drilling frenzy of the late 1970s and early '80s produced the most expensive oil in the history of the industry. So, such production is a hallmark of the end of the golden age and the beginning of the transition stage of expensive oil.

The next few years should put each side's theory to the test. If technology can greatly boost reserves, then the U.S. production curve should at least stabilize, while if the pessimists are right, it will soon resume its steep downward slide. Production from the North Sea should tell how middle-aged oil provinces will fare; pessimists expect it will peak in the next few years. But it is the world production curve that will finally reveal whether the world is due for an imminent shortfall or decades more of unbounded oil.