The subject I teach -- human ecology -- is a discipline that largely concerns population and resources. Over the past few years I have chosen to study oil, because it is the most important energy resource of the modern world.

Only 150 years ago, 85 percent of all work being accomplished in the U.S. economy was done by muscle power -- most of that by animal muscle, about a quarter of it by human muscle. Today, that percentage is effectively zero; virtually all of the physical work supporting our economy is done by fuel-fed machines. What caused this transformation? Quite simply, it was oil's comparative cheapness and versatility. Perhaps you have had the experience of running out of gas and having to push your car a few feet to get it off the road. That's hard work. Now imagine pushing your car 20 or 30 miles. That is the service performed for us by a single gallon of gasoline, for which we currently pay $2.65. That gallon of fuel is the energy equivalent of roughly six weeks of hard human labor.

It was inevitable that we would become addicted to this stuff, once we had developed a few tools for using it and for extracting it. Today petroleum provides 97 percent of our transportation fuel, and is also a feedstock for chemicals and plastics.

It is no exaggeration to say that we live in a world that runs on oil.

However, oil is a finite resource. Therefore the peaking and decline of world oil production are inevitable events -- and on that there is scarcely any debate; only the timing is uncertain. Forecast dates for the peak range from this year to 2035.

The peaking phenomenon itself has been observed again and again in individual oil fields and in entire producing nations. One of the first countries to hit its peak was the U.S.. During the 1930s and '40s, half the world’s production of petroleum came from Texas and Oklahoma. However, U.S. production reached its all-time maximum in 1970 and has been declining ever since. Currently the U.S. imports 60 percent of its oil.

Concern over the likelihood of an impending world peak has increased markedly in recent months as global spare production capacity has dwindled and as prices have achieved what seems to be a new baseline of over $50 per barrel.

Evidence that we are approaching peak includes the following:

ExxonMobil documents that global oil discoveries peaked in 1964. Declining rates of discovery are therefore a long-established trend.

Chevron notes in recent advertisements that 33 of 48 nations are in decline. We have thus seen the peaking of production in a majority of individual nations, including some important producers such as Indonesia, Norway, Great Britain, and Venezuela. Mexico will reach its peak within the next two years.

As noted by the International Energy Agency, there is evidence that a substantial amount of "proven reserves" in OPEC countries are illusory, the result of a scramble for market share within a cartel that allocates export quotas based on stated reserves.

With regard to this last point it should be noted that reserves figures, even when accurate, have historically given little warning of peaking. The U.S. instance is once again emblematic: in 1970, U.S. oil reserves were higher than ever; so were production rates. But only a year later, American production began its terminal decline. The study of discovery rates and depletion rates gives us a much better idea of when the global peak is likely to occur.

Optimistic estimates of future discovery and production issued by Cambridge Energy Research Associates and the U.S. Geological Survey have been criticized by several analysts. The optimists have generally failed to anticipate peaks, first in the U.S. and repeatedly in the case of other nations around the world.

This morning the International Energy Agency (IEA) issued a statement saying that the world will have sufficient energy supplies for the next quarter century. However, the statement noted the necessity of the investment of $17 trillion in the supply train in order to maintain sufficiency for so long. Also, the IEA anticipates Saudi Arabian production expanding to 18 million barrels per day by 2030—a figure considerably higher than the maximum possible rate of production from that country cited not long ago by Sadad al Husseini, the recently retired head of exploration for Saudi Aramco.

Expressions of concern have been voiced by corporations, prominent organizations, and knowledgeable individuals, including ChevronTexaco, the Royal Swedish Academy of Sciences, Volvo, Ford Motor Company Executive Vice President Mark Fields, the Chinese Offshore Oil Corporation’s chief economist, and numerous petroleum scientists and oil industry analysts.

The question immediately arises: Will alternative sources be able to make up the difference?

Alternative sources often discussed include oil sands from Canada, shale oil in Colorado, coal-to-liquids, gas-to-liquids, nuclear, and renewables such as solar and wind. Each of these will require immense investment and well over a decade of intense effort in order to produce substantial quantities of energy to offset declines from fossil fuels. And in most cases, rates of production are and will be constrained by non-economic factors. Take the oil sands, for example. Currently Canada produces one million barrels of synthetic crude per day from that source. There is expectation of two mb/d by 2010, and perhaps as much as four mb/d by 2025. We are unlikely to see higher numbers than that even with extraordinary capital investment, because the production process requires large amounts of natural gas and fresh water, both in short supply in Alberta. Moreover, according to the IEA, the world needs six mb/d of new production capacity each year (and that number is growing) to meet new demand and to offset depletion from existing fields.

How about increased efficiency -- surely that can offset any potential oil supply problems. In principle, yes, but most efficiency strategies will likewise require significant lead times. For example, we have the technology now to enable all of us who own cars to be driving ones that get up to 100 miles per gallon. If we were, that would obviously save an enormous amount of fuel. But how long would it take to implement that strategy? It would certainly take four or five years for Detroit to begin producing such high-efficiency cars in large numbers.

Then, not everyone buys a new car every year. In fact, it takes about 15 years to change out nearly the entire U.S. car and truck fleet. So, altogether, it would take about 20 years to fully implement this particular efficiency strategy.

Will the market be able to respond quickly enough to forestall serious economic, social, and political impacts? It is often said that the Stone Age did not end for lack of stones, nor will the Oil Age end because we run out of petroleum -- but instead because we find a cheaper source of energy. However, as we have just seen, that cheaper source of energy has yet to be identified.

Early this year a report was released, prepared for the U.S. Department of Energy by a team led by Robert L. Hirsch, who has a distinguished background in the oil industry and is a senior energy analyst at SAIC and the Rand Corporation. The Hirsch Report (titled "Peaking of World Oil Production: Impacts, Mitigation and Risk Management") concludes that price signals will arrive at least ten years too late to enable a gentle, market-led transition away from oil to other energy sources. The report describes Peak Oil as an "unprecedented" challenge for modern societies, and describes economic, social, and political risks if preparation is not undertaken soon enough, or on adequate scale.

Let me read you a few sentences from the Hirsch Report:

The problems associated with world oil production peaking will not be temporary, and past "energy crisis" experience will provide relatively little guidance. The challenge of oil peaking deserves immediate, serious attention, if risks are to be fully understood and mitigation begun on a timely basis. Mitigation will require a minimum of a decade of intense, expensive effort, because the scale of liquid fuels mitigation is inherently extremely large. Intervention by governments will be required, because the economic and social implications of oil peaking would otherwise be chaotic.

The report also concludes that the costs of preparing too late for global oil peak would far outweigh those of preparing too early.

The worst-case scenario for the impact of global production peak is very bad indeed. As I mentioned earlier, we are extremely dependent on oil for transportation, agriculture, plastics, and chemicals. In each area, we are already seeing serious impacts resulting from current prices in the $60-per-barrel range. For example,

Currently tens of thousands of farmers are agonizing over whether they can afford to plant next year’s crop, given high fuel and fertilizer costs.

Chemicals and plastics industries are already hard hit: In the chemistry industry alone, more than 100 plants have closed and more than 100,000 jobs have been lost just this year.

In the airline industry, 40 percent of revenues go to pay for jet fuel; most U.S. air carriers are already in bankruptcy or nearing that situation.

Home heating costs are projected to be 40-50% higher this winter than last.

As prices go even higher, and with actual scarcities of fuel, people will experience difficulties commuting, and the maintenance of our far-flung food distribution systems may become problematic.

On top of all this, oil is a strategic resource: as supplies become scarce, there is increasing likelihood of international conflict.

To avoid the worst-case scenario we must begin today to reduce our dependence on oil. The effort must have top priority. It must focus primarily on reducing demand, and only secondarily on producing large quantities of alternative transportation fuels.

A global Oil Depletion Protocol would reduce price volatility and competition for remaining supplies, while encouraging nations to move quickly to wean themselves from petroleum. In essence, the Protocol would be an agreement whereby producing nations would plan to produce less oil with each passing year (and that will not be so difficult, because few are still capable of maintaining their current rates in any case); and importing nations would agree to import less each year. That may seem a bitter pill to swallow.

However, without a Protocol -- essentially a system for global oil rationing -- we will see extremely volatile prices that will undermine the economies of all nations, and all industries and businesses. We will also see increasing international competition for oil likely leading to conflict; and if a general oil war were to break out, everyone would lose. Given the alternatives, the Protocol clearly seems preferable.

National governments, local municipalities, corporations, and private individuals will all need to contribute to the effort to wean ourselves from oil, an effort that must quickly expand to include a reduction in dependence on other fossil fuels as well.

All of this will constitute an immense challenge for our species in the coming century. We will meet that challenge successfully only if we begin immediately.