You wouldn't know it by sliding oil prices these days, but, if a new book suggesting crude supplies are running out turns out to be right, oil prices could be sky-high in less than a decade.

While the end-of-oil-soon idea has its share of critics, the stakes are high if they're wrong: The world economy could plunge into recession unless alternative energy sources are lined up to replace it.

In the new book, "Hubbert's Peak: The Impending World Oil Shortage" (Princeton University Press), author Kenneth Deffeyes argues that global crude oil production will reach its zenith in one to seven years, then begin a long and slow decline.

Saudi Arabia has come to the rescue in past crises. But the day could be coming when even they can't be counted on to tap their reserves when a shortage occurs.

"One of the first pieces of hard news that you've reached the peak is when the rest of the world declines enough to the point where it locks up the Saudis' remaining capacity," Deffeyes told Reuters in a recent interview.

While Deffeyes argues in his book that the world hasn't properly prepared to avert a new energy crisis, he is no prophet of doom. Nor is he an environmentalist grinding an ax in the name of climate change or alternative energy.

He is a retired petroleum geologist and professor emeritus from Princeton University who says viable alternatives to oil, like natural gas for powering cars and wind and nuclear energy for generating electricity, won't be fully ready to pick up the slack when the production peak arrives.

Interestingly, his vast oilfield experience is precisely what detractors -- usually economists -- seize on. They argue that being so steeped in geology blinds experts like Deffeyes to improved technology and lower tax rates that add to efficiency in the energy business.

"My initial response is that one to seven years (until the oil decline) is a really short time," said Sarah Emerson, an analyst at Energy Security Analysis Inc., or ESAI, in Boston.

"The thing that always strikes me about geologists and engineers is that they kind of pay lip service to the economic arguments, but I don't think they've ever effectively countered them," Emerson added.

HUBBERT'S CHILDREN

Deffeyes book is the latest study based on the work of M. King Hubbert, the late Shell Oil geologist and former Deffeyes colleague, who in 1956 predicted, successfully, it turned out, that U.S. oil production would peak in the early 1970s.

While the United States is still vying with Saudi Arabia for the crown of top world oil producer, U.S. output has been on a slow decline since it peaked 1970, and U.S. imports now total over half of the 19.5 million barrels consumed each day.

Hubbert reckoned that the dual histories of oil discovery and production might conform to a more-or-less symmetrical rise and fall of a bell-shaped curve.

His critics acknowledge that finding oil might soon be a declining industry, but insist its extraction from existing fields will remain dynamic and innovative in the coming years and thus defy the timing of Deffeyes' prediction. They see a curve that's much gentler and longer than the "bell curve."

But Deffeyes stresses that the key to Hubbert's method is watching the course of oil discoveries for clues as to when production will hit its peak and begin falling.

"The thing that tells you production is really going to roll over is that discoveries worldwide since 1975 have not been all that great," Deffeyes said, adding that since then most discoveries have been of natural gas.

Deffeyes and others -- notably Colin Campbell and Gene Laherrere, two executives with Petroconsultants in Geneva who published the article "The End of Cheap Oil" in the March 1998 edition of Scientific American magazine -- have taken methods Hubbert used in his U.S. study and applied it to the whole planet.

Campbell and Laherrere concluded, like Deffeyes, that the world's oil production will start to fall for good sometime during this decade.

"We're right on track," Deffeyes added, citing Oil & Gas Journal's 2001 report, which he says shows world output rose last year, but conformed to his expectation that as 'Hubbert's Peak' nears, output increases should become increasingly slim.

Deffeyes concedes that his one-to-seven year prediction may be off the mark, and there are several factors that could definitely alter his thinking. The first is a big war in the Middle East that would cause a spike in oil demand and bring 'Hubbert's Peak' sooner. In an opposing scenario, oil demand might drop in a serious, protracted economic downturn that would slash demand for oil and buy the world precious time to line up alternative energy sources.

CAVEAT BONANZA

There is also the possibility that more huge "elephant" fields are hiding somewhere, waiting to be discovered and developed to delay the arrival of 'Hubbert's Peak." Some analysts, for example, have seized on the Caspian region in central Asia as a highly "prospective" area, to use the oil industry term.

But Emerson, like Deffeyes, isn't holding her breath for more elephant fields. She and others are looking rather to existing oil provinces with optimism that their full potential can be better unlocked through technology, economics and politics.

"I'm not saying there are a lot of elephants left. I've read enough geologists to understand that there are only so many dead dinosaurs under the earth's crust. I'm not disputing that. What I'm disputing is the time horizon," Emerson said.

"There are so many things outside of geologic potential that could change: the application of technology; the application of capital; changing tax regimes; change in foreign investment rules. All these things still have a long way to go, which will extend the life of the world's oil reserves."

Emerson also believes output in oil-rich Middle Eastern countries like Iran, Libya and Iraq has been seriously crimped by economic sanctions imposed on them by the United States or, in the case of Iraq, the United Nations. Lifting these sanctions would surely delay the production peak, she said.

Aside from supplies, the other huge question is demand, and many say it may well be a more important variable than supply.

What would happen to oil demand -- and the timing of Hubbert's Peak -- if conservation caught on, or if the use of still-plentiful natural gas spread to entire fleets of cars instead of just those in the public sector, like buses?

And what about fuel cells? A deal struck last week by U.S. automakers and the U.S. government to develop hydrogen-powered vehicles demonstrates that the days of gas-guzzling sport utility vehicles are probably numbered. But how numbered?

Even auto executives and government officials admit hydrogen-powered cars are years, if not decades, away, and that reality has a lot to do with the immediacy of Deffeyes' worry.

He doesn't buy that oil production around the world can be ramped up enough to delay the arrival of 'Hubbert's Peak."

But he is sanguine that eventually -- by using gas, wind, nuclear energy, fuel cells or even by 'mining' petroleum from oil-sand pits -- the world economy will thrive even as oil production tails off. Timing, though, is everything.

"I'm not terribly scared about our 15- and 20-year position. I'm scared about the four-year position sneaking up on us."