Comparing GOSAT observations of localized CO2 enhancements by large emitters with inventory-based estimates

We employed an atmospheric transport model to attribute column-averaged CO2 mixing ratios (X-CO2) observed by Greenhouse gases Observing SATellite (GOSAT) to emissions due to large sources such as megacities and power plants. X-CO2 enhancements estimated from observations were compared to model simulations implemented at the spatial resolution of the satellite observation footprint (0.1 degrees x 0.1 degrees). We found that the simulated X-CO2 enhancements agree with the observed over several continental regions across the globe, for example, for North America with an observation to simulation ratio of 1.05 +/- 0.38 (p < 0.1), but with a larger ratio over East Asia (1.22 +/- 0.32; p < 0.05). The obtained observation-model discrepancy (22%) for East Asia is comparable to the uncertainties in Chinese emission inventories (similar to 15%) suggested by recent reports. Our results suggest that by increasing the number of observations around emission sources, satellite instruments like GOSAT can provide a tool for detecting biases in reported emission inventories.