ABSTRACTS
 2nd International Workshop on Oil Depletion
Paris, France, May 26-27 2003
Organised by the Association for the Study of Peak Oil and Gas
The workshop was held at the  Institut Francais du Pétrole , Rueil Malmaison, Paris.

  Proceedings of the 2nd International Workshop on Oil Depletion, Paris, France, May 26-27 2003,
Edited by K. Aleklett, C. Campbell and J. Meyer, www.peakoil.net/iwood2003 .

Modelling Oil Production, Energy Consumption, Population and Economy Jean Laherrère Most published data on energy, population and the economy are unreliable.  In many cases, authors have political motives, selectively choosing data from a wide range of uncertainty to give a desired image.  In addition to the uncertainty of the measurements themselves, as in the case of population or the confidentiality of the oil reserves, they often indulge in manipulation.  A so-called hedonistic factor distorts the calculation of GDP in the United States; and the definition of the Proved Reserves by the Securities and Exchange Commission gives rise to “reserve growth”. OPEC misreports its oil reserves because its quotas depend upon the reported reserves, and the reserves were overestimated in the Soviet Union because economic and technical constraints were ignored. Our present culture of eternal growth makes the word "decline" politically incorrect, but constant growth is unsustainable in a finite world.  Growth is the Santa Claus of the modern age who is supposed to provide welfare and retirement for our children and us. All natural events, when measured over their full life, can be modelled under one or more cycles, as in the Fourier analysis.  This cyclical nature corresponds with the finite nature of the Universe; everything that is born will die, whether we speak of the solar system, the Earth, or human species.  What goes up must come down. The Russian population is already declining and Europe’s will soon do so too.  This basic understanding was recognised by the celebrated King Hubbert when he made his famous prediction in 1956 that US oil production would peak in 1970.  But, in fact, he oversimplified by showing a single peak.  In reality, US oil production had a secondary peak (93% of the first one) in 1985, reflecting the entry of Alaskan production, which itself peaked in 1988.  A symmetrical oil cycle reflects a large number of independent producers, acting randomly, but in many cases economic and political factors disturb the pattern, giving one or more new cycles. To model an event made up of several cycles extending into the future calls for an estimate of the ultimate value, which corresponds with the area under the curve up to the end of the event.  For oil, the best tool to determine an ultimate value is the creaming curve that plots cumulative discovery versus the cumulative number of new field wildcats, the result being modelled by one or more hyperbolas.  Another method is to plot the ratio of annual to cumulative production versus cumulative production, and extrapolate the trend to zero.  When the trend is linear, it represents the derivative of the logistic curve.  The fractal distribution of sizes (field reserves, incomes, urban agglomerations plotted against decreasing rank) can also be extrapolated to an ultimate value. Population can be well modelled with two cycles, distinguishing countries with high and low fertility rates.  Previous UN forecasts were too high for different reasons. Economic parameters, such as unemployment or inflation, can be correlated with oil price after a certain time-shift.  Income distribution is well described by a fractal plot of population versus income.  The income fractal distribution in France is in fact the same as that in the United States, although the total of the latter is higher because of a larger population. Many graphs are shown for each domain using the same tools.  The goal is that the reader may be able to draw his own conclusions, and make his own forecast.  Ironically, it appears that the modelling is more reliable than the input data. Accordingly, the main challenge is to secure better data, but that will be achieved only if and when political influences can be removed.  A neutral agency is needed, but neither the UN nor national agencies are neutral.  It is hard to see how to force the actors to tell the truth, or know who would run and finance such an organisation. A step in the right direction would be to make official organisations liable to prosecution for releasing false data, as is already supposed to apply in the United States under Public Law 106-554.    Back to Proceedings