Spatial representation of a current NOx emission inventory over China: A satellite perspective

The top-down approach provides valuable constraints to bottom-up emission inventories. In this work, we evaluated the spatial representation of a current NO_x emission inventory over China using NO₂ retrievals from the Ozone Monitoring Instrument (OMI) and the nested-grid GEOS-Chem model. Comparison between model simulations and satellite observations showed that modeled NO₂ columns were underestimated in most medium-sized cities, while good agreement was found in megacities such as Beijing, Shanghai. Large discrepancies between model simulations and satellite retrievals were found in grids with high Gross domestic product (GDP) value across different regions of China. We also found that the growth rates of NO₂ columns have slowed in Chinese megacities over recent years, in contrast with medium-sized cities where the NO₂ columns were soaring. Sensitivity simulations using emissions allocated by different spatial surrogates (e.g. population density, location, nighttime light, industrial GDP, road networks) indicated that the uncertainties in spatial representation of the current emission inventory might be caused by inappropriate spatial proxies.