Space-based NO_x emission estimates over remote regions improved in DECSO

Abstract. We improve the emission estimate algorithm DECSO (Daily Emission estimates Constrained by Satellite Observations) to better detect NO_x emissions over remote areas. The new version is referred to as DECSO v5. The error covariance of the sensitivity of NO₂ column observations to gridded NO_x emissions has been better characterized. This reduces the background noise of emission estimates by a factor of 10. An emission update constraint has been added to avoid unrealistic day-to-day fluctuations of emissions. We estimate total NO_x emissions, which include biogenic emissions that often drive the seasonal cycle of the NO_x emissions. We demonstrate the improvements implemented in DECSO v5 for the domain of East Asia in the year 2012 and 2013. The emissions derived by DECSO v5 are in good agreement with other inventories like MIX. In addition, the improved algorithm is able to better capture the seasonality of NO_x emissions and for the first time it reveals ship tracks near the Chinese coasts that are otherwise hidden by the outflow of NO₂ from the Chinese mainland. The precision of monthly emissions derived by DECSO v5 for each grid cell is about 20 %.