Estimating long-term world coal production with logit and probit transforms

An estimate for world coal production in the long run would be helpful for developing policies for alternative energy sources and for climate change. This production has often been estimated from reserves that are calculated from measurements of coal seams. We show that where the estimates based on reserves can be tested in mature coal regions, they have been too high, and that more accurate estimates can be made by curve fits to the production history. These curve fits indicate that total world production, including past and future production, will be 680 Gt. The historical range for these fits made on an annual basis from 1995 to 2009 is 653 Gt to 749 Gt, 14% in percentage terms. The curve fits also indicate that 90% of the total production will have taken place by 2070. This gives the time scale for considering alternatives. This estimate for total production is somewhat less than the current reserves plus cumulative production, 1163 Gt, and very much less than the amount of coal that the UN Intergovernmental Panel on Climate Change, or IPCC, assumes is available for its scenarios. The maximum cumulative coal production through 2100 in an IPCC scenario is 3500 Gt.