Deffeyes has a new book out: Beyond Oil: The view from Hubbert's peak. Read on for my review.
While studded with gems of information and insight, overall I found this book somewhat disappointing. As a former colleague of M. King Hubbert at Shell, geologist Kenneth Deffeyes gives an excellent account of the theory behind "Peak Oil" - in particular the compelling straight-line plots comparing current to cumulative production that are the surest foundation of the predictions. Deffeyes derives from these plots that the effective peak in oil production will be within one month of November, 2005.
But the insight after that point seems lacking - what does the peak actually mean, why does the decline happen? Is the price volatility Deffeyes identifies, rather than just high prices, the key to the failings of economic analysis of the problem? Does that come about from substitution, or is there really going to be a decline in energy resources available? The fact that oil companies have not been building new refineries recently suggests they are well aware they will need fewer, not more, in future.
The bulk of the book, as the title suggests, looks at what comes after oil; Deffeyes notes that this reflects his own expertise in the issues of extracting things from the ground, but that means he neglects renewables. From Deffeyes one would think that what comes after peak oil is little different from what came before, just in more limited quantities; past centuries' experience with technological change suggest that's a very short-sighted viewpoint. There is also a rather inconsistent stance throughout the book on the near-term problem vs. long-term solutions: some long-term solutions are dismissed because they will not help with the immediate crisis, while others of equally long delay are discussed at length.
Deffeyes seems to see carbon sequestration as a helpful step, but of his seven conditions explaining why oil deposits are rare, at least four would also be required for safely storing carbon dioxide underground; are there so many likely locations in the world?
Deffeyes has some interesting insight on the resource issue for nuclear power - is there enough uranium available to satisfy world energy needs for a good length of time? Deffeyes once had access to restricted uranium discovery data from the Department of Energy that was enough to indicate there is a continuum of uranium deposits available, of gradually declining quality, with tens of trillions of tons in totol. At ore grades below 100 parts per million uranium it's been argued that the energy required for mining exceeds the energy payback from a fission reactor (at least for the once-through cycle). From Deffeye's chart there seem to be 100 million to 1 billion metric tons of uranium in higher-grade ores, enough to power the world using a standard once-through fission cycle for hundreds to thousands of years.
Deffeyes' discussion of unconventional fossil fuels - oil shales, heavy oils, and so forth, is also very interesting. From the perspective of a petroleum geologist, these deposits have been known for a very long time (the most recent discovery being ocean-floor methane hydrates) and billions of dollars in research devoted to them still hasn't made them successful, at least to any large degree.
The final summary seems to be a future of diminished circumstances for the human race, a dreary future with more pollution from coal use and nuclear power, and little real hope for change. Deffeyes pushes half-heartedly for investment in efficiency technologies, but mostly seems to think we just have to weather the coming storm somehow and hope for the best.