Mapping high-resolution energy consumption CO emissions in China by integrating nighttime lights and point source locations.

High-resolution CO emission inventories are essential to accurately assess spatiotemporal patterns of carbon emissions, analyze factors affecting carbon emissions, and develop sound emission reduction policies. The top-down approach is often used to map CO emissions from energy consumption due to its simplicity. However, the spatial proxy variables commonly used in this method, such as nighttime light (NL), land use, and population, are difficult to reflect the spatial distribution of CO emissions from large point sources. Therefore, this study uses the active fire product provided by Visible Infrared Imaging Radiometer Suite (VIIRS) sensors on Suomi National Polar-Orbiting Partnership (Suomi-NPP) satellite to extract the location of industrial heat sources in China, and then develops an improved CO emission estimation model by integrating industrial heat sources, Global Energy Monitor (GEM) power plant location and nighttime lights. The model is used to map CO emissions from energy consumption at a resolution of 1 km*1 km from 2012 to 2019 in China. It is found that the overall accuracy of the model is greatly improved at the provincial level, the R value is >0.75, and RMSE is distributed in 40-110 Mt. At the grid level, the improved model allocates more carbon emissions to the grid where the point source is located, which makes the spatial distribution of CO emissions more reasonable.