Oiling the Wheels of Debate
Last month, I read most of the recent books by Ken Deffeyes and Paul Roberts on oil depletion. Deffeyes, an academic geologist, isn't a good a writer as Roberts the Harpers journalist, but he does KNOW the technical details. However, he explained Hubbert's bell curve application using a single Pennsylvalia anthracite mine; It might be worth getting the original paper by Hubbert sometime, as I don't know exactly how the method worked.
All of these peak oil forecasts I've read have the weakness that, whatever the strength of their geology, they assume an oversimple linkage between the peak and the end of the world as we know it, whereas various resource, metal and energy transitions in the past have been largely smooth replacements over 10-50 year cycles.
I've read somewhere that every time it's used it forecasts a peak in the near future, which accords with the overly pessimistic forecasts made in the past and proven wrong.
According to Smil's Energy at the Crossroads, Hubbert himself predicted a global peak between 1993 and 2000. The CIA came out with a prediction in 1979 at the height of the Iranian revolution saying that "[global] output must fall within a decade ahead". BP forecast the peak in production of crude outside the USSR by 1985. The IEA WEO 2001 report highlights (pp.52-53) how 1983 forecasts of North Sea production peaking later that decade at under 4 million barrels per day, compared with observed production in 1999 near to 7 million. Also, the USGS estimates seem to be revised upwards into the region of the upper 5% of their previous forecasts.
Robert Kaufmann, a Boston U energy economist and IPCC Working Group One member, insists that Hubbert's model needs to be supplemented with other variables, including the oil price, production costs and cartel power of the Texas Railroad Commission. Otherwise, its success in forecasting the peak in American oil production was just dumb luck and attempts to apply it globally futile.
Also today, Charles Hymans pointed me to a good paper by the USGS summarising the reserves debate.
All of these peak oil forecasts I've read have the weakness that, whatever the strength of their geology, they assume an oversimple linkage between the peak and the end of the world as we know it, whereas various resource, metal and energy transitions in the past have been largely smooth replacements over 10-50 year cycles.
I've read somewhere that every time it's used it forecasts a peak in the near future, which accords with the overly pessimistic forecasts made in the past and proven wrong.
According to Smil's Energy at the Crossroads, Hubbert himself predicted a global peak between 1993 and 2000. The CIA came out with a prediction in 1979 at the height of the Iranian revolution saying that "[global] output must fall within a decade ahead". BP forecast the peak in production of crude outside the USSR by 1985. The IEA WEO 2001 report highlights (pp.52-53) how 1983 forecasts of North Sea production peaking later that decade at under 4 million barrels per day, compared with observed production in 1999 near to 7 million. Also, the USGS estimates seem to be revised upwards into the region of the upper 5% of their previous forecasts.
Robert Kaufmann, a Boston U energy economist and IPCC Working Group One member, insists that Hubbert's model needs to be supplemented with other variables, including the oil price, production costs and cartel power of the Texas Railroad Commission. Otherwise, its success in forecasting the peak in American oil production was just dumb luck and attempts to apply it globally futile.
Also today, Charles Hymans pointed me to a good paper by the USGS summarising the reserves debate.
<< Home