Estimating ground-level CH₄ concentrations inferred from Sentinel-5P

Methane (CH₄) is an important greenhouse gas; however, there is a lack of large-scale studies on ground-level CH₄ concentrations. We estimated global ground-level CH₄ concentrations based on the CH₄ columns from the Copernicus Sentinel-5 precursor satellite (S5P) and vertical profiles of CH₄ concentrations simulated from the Atmospheric Chemical Transport Model (GEOS-Chem). The proposed approach had achieved a high predictive accuracy for monthly ground-level CH₄ concentrations, with a correlation coefficient of 0.93 (p < 0.01) and RMSE of 29.93 ppb between the estimated CH₄ concentrations and those of ground measurements from the World Data Centre for Greenhouse Gases (WDCGG). Compared with the S5P CH₄ columns, the estimated ground-level CH₄ concentration has a close spatial relationship with emissions. The high CH₄ concentrations occurred in eastern China, northern India, western Russia, eastern U.S., and central Europe. Furthermore, the estimated ground-level CH₄ concentrations could reflect the seasonal variations of the observations, with correlation coefficients from 0.14 to 0.92. Our findings highlight the importance of satellite observations on atmospheric CH₄ in understanding the spatial and temporal emissions.