Estimating global 0.1° scale gridded anthropogenic CO emissions using TROPOMI NO and a data-driven method.

Satellite remote sensing is a promising approach for monitoring global CO emissions. However, existing satellite-based CO observations are too coarse to meet the requirements of fine-scale global mapping. We propose a novel data-driven method to estimate global anthropogenic CO emissions at a 0.1° scale, which integrates emissions inventories and satellite data while bypassing the inadequate accuracy of CO observations. Due to the co-emitted anthropogenic emissions of nitrogen oxides (NO = NO + NO) and CO, high-resolution NO measurements from the TROPOspheric Monitoring Instrument (TROPOMI) are employed to map the global anthropogenic emissions at a global 0.1° scale. We construct the driving features from NO data and also incorporate gridded CO/NO emission ratios and NO/NO conversion ratios as driving data to describe co-emissions. Both ratios are predicted using a long short-term memory (LSTM) neural network (with an R of 0.984 for the CO/NO emission ratio and an R of 0.980 for the NO/NO conversion ratio). The data-driven model for estimating anthropogenic CO emissions is implemented by random forest regression (RFR) and trained using the Emissions Database for Global Atmospheric Research (EDGAR). The satellite-based anthropogenic CO emission dataset at a global 0.1° scale agrees well with the national CO emission inventories (an R of 0.998 with Global Carbon Budget (GCB) and an R of 0.996 with EDGAR) and consistent with city-level emission estimates from Carbon Monitor Cities (CMC) with the R of 0.824. This data-driven method based on satellite-observed NO provides a new perspective for fine-resolution anthropogenic CO emissions estimation.