End of the pipeline for oil?

That made sense. Hubbert knew some oil fields, especially the big ones, were easier to find than others. Those big finds would come first, and then the pace of discovery would decline as the remaining pool of oil resided in progressively smaller and more elusive deposits.

The production figures followed a similar pattern, but it looked like they would peak a few years later than reserves.

That made sense too. After all, oil can't be pumped out of the ground the instant it is discovered. Lease agreements have to be negotiated, wells drilled, pipelines built; the development process can take years.

When Hubbert extended the production curve into the future it looked like it would peak around 1970. Every year after that, America would pump less oil than it had the year before.

If that prognostication wasn't daring enough, Hubbert had yet another mathematical trick up his sleeve. Assuming that the reserves decline was going to be a mirror image of the rise, geologists would have found exactly half of the oil in the Lower 48 when the curve peaked. Doubling that number gave Hubbert the grand total of all recoverable oil under the continental United States: 170 billion barrels.

At first, critics objected to Hubbert's analysis, arguing that technological improvements in exploration and recovery would increase the amount of available oil.

They did, but not enough to extend production beyond the limits Hubbert had projected. Even if you throw in the unexpected discovery of oil in Alaska, America's petroleum production history has proceeded almost exactly as Hubbert predicted it would.

Critics claim that Hubbert simply got lucky.

“When it pretty much worked,” Lynch said, “he decided, aha, it has to be a bell curve.”

But many experts see no reason global oil production has to peak at all. It could plateau and then gradually fall as the economy converts to other forms of energy.

“Even in 30 to 40 years there's still going to be huge amounts of oil in the Middle East,” said Daniel Sperling, director of the Institute of Transportation Studies at the University of California, Davis.

A few years ago, geologists began applying Hubbert's methods to the entire world's oil production. Their analyses indicated that global oil production would peak some time during the first decade of the 21st century.

Deffeyes thinks the peak will be in late 2005 or early 2006. Houston investment banker Matthew Simmons puts it at 2007 to 2009. California Institute of Technology physicist David Goodstein, whose book “The End of Oil” was published last year, predicts it will arrive before 2010.

The exact date doesn't really matter, said Hirsch, because he believes it's already too late. In an analysis he did for the U.S. Department of Energy in February, Hirsch concluded that it will take more than a decade for the U.S. economy to adapt to declining oil production.

“You've got to do really big things in order to dent the problem. And if you're on the backside of the supply curve you're chasing the train after it's already left the station,” he said.

For example, the median lifetime of an American automobile is 17 years. That means even if the government immediately mandated a drastic increase in fuel efficiency standards, the conservation benefits wouldn't fully take effect for almost two decades.

And though conservation would certainly be necessary in a crisis, it wouldn't be enough. Fully mitigating the sting of decreasing oil supplies would require developing alternate sources of energy — and not the kind that politicians and environmentalists wax rhapsodic about when they promise pollution-free hydrogen cars and too-cheap-to-meter solar power.

If oil supplies really do decline in the next few decades, America's energy survival will hinge on the last century's technology, not the next one's. Hirsch's report concludes that compensating for a long-term oil shortfall would require building a massive infrastructure to convert coal, natural gas and other fossil fuels into combustible liquids.

Proponents of coal liquefaction, which creates synthetic oil by heating coal in the presence of hydrogen gas, refer to the process as “clean coal” technology. It is clean, but only to the extent that the synthetic oil it produces burns cleaner than raw coal. Synthetic oil still produces carbon dioxide, the main greenhouse warming gas, during both production and combustion (though in some scenarios some of that pollution could be kept out of the atmosphere). And the coal that goes into the liquefaction process still has to be mined, which means tailing piles, acid runoff and other toxic ills.

And then there's the fact that nobody wants a “clean coal” plant in the backyard. Shifting to new forms of energy will require building new refineries, pipelines, transportation terminals and other infrastructure at a time when virtually every new project faces intense local opposition.

Energy analysts say coal liquefaction can produce synthetic oil at a cost of $32 a barrel, well below the $50 range where oil has been trading for the past year or so. But before they invest billions of dollars in coal liquefaction, investors want to be sure that oil prices will remain high.

Investors are similarly wary about tar sands and heavy oil deposits in Canada and Venezuela. Though they are too gooey to be pumped from the ground like conventional oil, engineers have developed ways of liquefying the deposits with injections of hot water and other means. Already, about 8 percent of Canada's oil production comes from tar sands.

Unfortunately, it costs energy to recover energy from tar sands. Most Canadian operations use natural gas to heat water for oil recovery; and like oil, natural gas has gotten dramatically more expensive in the past few years.

“The reality is, this thing is extremely complicated,” Hirsch said. “My honest view is that anybody who tells you that they have a clear picture probably doesn't understand the problem.”

MORE FROM OIL & ENERGY

Oil & Energy Section Front

Oil markets could settle after OPEC inaction OPEC unlikely to change production targets Cost of gas falls nearly 4 cents in last two weeks Exxon Mobil to buy XTO Energy for $31 billion Iraq hails oil auction but risky sites shunned Iraq opens big slice of oil riches to bidders Saudi oil minister says oil is at ‘perfect’ price Newsweek: New market bubble is brewing Chevron 3Q profit slumps 51 percent BusinessWeek: The coming energy revolution Oil & Energy Section Front   Add Oil & Energy headlines to your news reader: