Results have been compared with a global gridded and hourly emission inventory (EDGARv6.0). The corresponding emission estimates explain a large part of the inventory variability, despite uncertainty in both datasets (see figure). Consistent variations in the middle emission values are seen at different time scales. These results suggest that the differences between the satellite-based inverse model estimates and the inventory estimates are mostly random. This implies that trends can be calculated robustly in areas of favourable observation conditions, especially when future satellite CO₂ imagers provide an order of magnitude more data. Details of the study can be found in an article that was just published in Geophysical Research Letters.