Economicobjectovism had this quote from the Sydney Morning Herald:
Dr Wheeler said his data had been compiled from public records and by extrapolating from a commercial database used by the power industry. Where disclosures of CO2 emissions were not publicly available, the researchers used modelling based on the age and size of the power station.
And goes on to ask:
I would like to see which countries/power plants required that inferencing. China’s lack of any sort of pollution controls at the plant-level are well-enough known - do these numbers under-estimate their carbon-load? Method can make a lot of difference. Not that it matters a great deal. For us to even be on such a list is as great a shame as it is entirely not surprising....Good question. CARMA's got a snazzy website, should be easy enough to pinpoint the plants that have been audited vs. those that were estimated.