Estimating Daily Wet Nitrate Deposition Across China Considering Under-Cloud and In-Cloud Scavenging Effects From Satellite Observations

Satellite remote sensing offers a promising approach to estimate atmospheric nitrate (NO₃⁻) wet deposition with high spatiotemporal resolution. Considering the under-cloud and in-cloud scavenging effects on NO₃⁻ by the rainfall, this study developed a machine learning model to estimate daily wet NO₃⁻ depositions across China based on atmospheric NO2 columns from the Ozone Monitoring Instrument (OMI) observations, NO₃⁻ concentrations in rainfall, and meteorological data. Results demonstrated that the proposed model achieved high accuracy, with cross-validation of R = 0.78 (p < 0.01) and RMSE = 0.09 kg N ha⁻¹ for single-precipitation event and R = 0.90 (p < 0.01) and RMSE = 0.26 kg N ha⁻¹ for monthly NO₃⁻ depositions; rainfall amount, under-cloud nitrate concentrations, and socioeconomic factors collectively dominated the spatial and temporal variations of NO₃⁻ depositions, while in-cloud nitrate concentration exhibited great contributions in less-N polluted areas such as northwestern China; deposition fluxes exhibit an increasing gradient from the northwest to southeastern China, ranging from 0.13 to 81.76 kg N ha⁻¹ yr⁻¹ and concentrated from June to September. This work pioneers the contributions from in-cloud and under-cloud scavenging processes on wet NO₃⁻ depositions, providing a scalable tool for regional nitrogen deposition monitoring and environmental management.