The problem with using Hubbert peak analyses to predict the future with coal is that the assumption that only traditional coal mining methods will be used in future doesn't stand up too well to close examination:-

http://www.coal.gov.uk/resources/cleanercoaltechnologies/ucgintro.cfm

Hi ES,

My guess is that UCG would be treated like coal-bed methane in statistics, and would show up as natural gas production, rather than coal production. The situation is somewhat like Canadian oil sands, which are strip mined but count as oil production. The output would naturally be included in the Hubbert linearization for natural gas for Uzbekistan, according to your link.

Dave

Can you tell me where the largest mines for coal are in South Africa. A map or just general area would be fine.