the drop in reserve estimate is not due to re-evaluation of the amount of coal by geologists but to an economical decision at the level of EU. Such decision could be reversed

And that's exactly why the "Reserves" figure is not an "Ultimate Reserves" figure, much less an overestimate of URR.

Take a look at Gail's graph for coal at one large field; a 10% increase in the retail cost of electricity would be enough to make reserves in that field increase by six times.

Such massive price-based variability demonstrates that not only are Reserves not the same as URR, they're not even a good estimate for it. (For coal, anyway.)

Shale particularly that mined in Estonia (but also Brazil) is included in our work on unconventional oil. (hopefully gets accepted at some point in time).
Coal insitu-gasification is treated as unconventional gas (or will be if I get round to it).

evidently I hit a nerve, I'm not particularly interested in the URR of coal, I am interested in creating better and more accurate models to predict energy and other resources production. The URR value to me, is a necessary evil I need to get from somewhere - I'd be much happier if there was a general consenses on what the URR value is, and just use high low estimates from that. I hope the real URR is somewhere contained in the interval indicated, but I do not know, and cannot know that (again, I'm a Mathematican and Chemical Engineer, not a Geologist or Economist). I'm happy if I've gotten the URR wrong, its straight forward to change the numbers in the model, again, if there are geological based, recent estimates that indicate the URR is somewhere outside the range (eg a justification for the IPCC estimate of 7000 Gt that Dave mentioned) then please let me know. that way I can modify the model for the phd and use HL trend as a low case, a presumably modified best guess case and some high case based on a better geological URR estimate.

Majorian,

I still think EnergyWatch are producing good work but it may be helpfull to break the post-peak curve into 2 elements, the bulk curve followed by a long slim tail.

Judgments as to what is economically recoverable using current economics and technology seem to support an early peak and a substantial reduction below IPCC SRES scenarios (which IIRC are based on IEA data). But that may not be the whole story.

I'm picturing the situation much later this century (assuming we've not cracked fusion), we may see a situation where people are using the sort of mining technology employed at the start of the industrial revolution (even before) to extract reserves we would not normally consider recoverable. Indeed the technology would not be solely early/pre-industrial because we'd still have some benefit from what we have learned since. It's not necessary for the whole globe to be working at such a low-tech level without heavy machinery, such mining could be confined to regional areas.

In impoverished societies well beyond peak energy labour may well be cheaper than machines. Just look at what the Egyptians achieved at Ghiza with cheap labour (relative to modern standards) and the will of the ruling class.

This reduction of rate of emission might seem to be reassuring. However Dr David Archer in "Millennial Atmospheric Lifetime of Anthropogenic CO2" concludes:
"The [carbon cycle] model studies we reviewed here differ substantially from each other in
parameterizations of oceanic and land carbon uptakes, the deep-sea sediment response,
and emissions pathways. However, despite all these differences, the models agree that the
substantial fraction of projected CO2 emissions will stay in the atmosphere for millennia,
and a part of fossil fuel CO2 will remain in atmosphere forever. Many slowly responding
components of the climate system, such as ice sheets and methane hydrates,
will be affected, and significant sea level rise is inevitable."

In view of which if I am correct in suggesting a "long slim tail" of emissions as humanity scratches out the last reserves of coal, the tail may be long and slim, but it would still be significant in terms of additions to the carbon cycle. However estimating how much coal may be extracted in that tail (time integral) will be virtually impossible with so much uncertainty about the assumptions that need to be used.