Independent Assessment of Climate Trace global powerplant CO2 emissions

Accurate estimation of planetary greenhouse gas emissions at the scale of individual emitting activities is a critical need for both scientific and policy applications. Powerplants represent the single largest and most concentrated form of global GHG emissions. Climate Trace, founded and promoted by the former Vice President Al Gore, is a new effort using, in part, machine learning approaches to estimating these emissions with a recently released database of global powerplant CO2 emissions. Here we compare the Climate Trace powerplant CO2 emissions to the Vulcan-power database, an atmospherically calibrated, multi-constraint estimate of powerplant CO2 emissions in the United States. We find that the mean relative difference between the 1,747 paired facilities is -47.6% and the unsigned relative difference is 108.0%. Climate Trace emissions estimates have large numbers of facilities that both overestimate and underestimate power plant emissions with only 10.2% (179 facilities) of the 1747 facilities agreeing to within 20%. The large differences between Climate Trace and Vulcan-power powerplant emission estimates in the US is primarily caused by Climate Trace use of a national-mean power plant capacity factor (CF) which does not reflect the non-gaussian distributed power plant capacity factors of US facilities.