Well, the peak-oil issue has finally made it to the front page of the Wall Street Journal. Yesterday's article presented both sides of the issue, focusing on Colin Campbell as a leading exponent of the "peak is imminent" faction, and Michael Lynch as the chief naysayer, backed up by statements from ExxonMobil. The Journal did a fine job of presenting enough of each side's argument that their audience can do some evaluating of its own, rather than having to rely on the writer's conclusions.
The theory of peak oil is built on the observation that when you have produced half of the oil that was there before you started pumping it, then production will reach a plateau and then begin to fall. This has been characteristic at the oilfield level, at the national level (particularly in the case of the US), and should be true for the globe as a whole, or so the argument goes. The article correctly differentiates this from the fallacy that we are "running out of oil." The succession of Mr. Campbell's previous incorrect estimates of a peak in production (e.g. in 1995) gets chalked up to inaccuracies in estimating the total amount of oil, not to failures of the underlying theory.
Mr. Lynch's counterargument is a bit more complicated, and possibly less appealing but likelier to be right for that very reason. He suggests that ultimate production is a function of geology, economics, and geopolitics. Thus any projection of future production based on historical levels will be skewed by changes in price, prevailing contractual terms, and access to resources. He also has some specific concerns about the data used by Campbell and others. As you might guess, he does not see a peak occurring anytime soon.
One of the interesting contrasts highlighted by the article is the extreme range of estimates for how much oil is actually still available. Mr. Campbell assumes that there is about as much left as we have pumped to date (roughly 900 billion barrels), while ExxonMobil sees up to 14 trillion barrels of recoverable oil, including non-conventional oil such as tar sands (my blog topic for tomorrow.)
As I've indicated previously (see my posting of May 7, 2004) I think that both sides in this debate are on to something. While much less skeptical about how much oil can ultimtely be extracted than Dr. Campbell, I think that many mainstream analysts underestimate the difficulties of getting timely access to resources and queuing up the financial and engineering means to harvest them, far enough in advance of the need. My retort to industry colleagues skeptical of this is simply this: show me the field-by-field production profile that in aggregate yields a steadily rising supply curve, and I will banish all thoughts of an impending gap between supply and demand. No takers, yet.