Improved Gaussian regression model for retrieving ground methane levels by considering vertical profile features

Abstract

Atmospheric methane is one of the major greenhouse gases and has a great impact on climate change. To obtain the polluted levels of atmospheric methane in the ground-level range, this study used satellite observations and vertical profile features derived by atmospheric chemistry model to estimate the ground methane concentrations in first. Then, the improved daily ground-level atmospheric methane concentration dataset with full spatial coverage (100%) and 5-km resolution in mainland China from 2019 to 2021 were retrieved by station-based observations and gaussian regression model. The overall estimated deviation between the estimated ground methane concentrations and the WDCGG station-based measurements is less than 10 ppbv. The R by ten-fold cross-validation is 0.93, and the R2 is 0.87. The distribution of the ground-level methane concentrations in the Chinese region is characterized by high in the east and south, and low in the west and north. On the time scale, ground-level methane concentration in the Chinese region is higher in winter and lower in summer. Meanwhile, the spatial and temporal distribution and changes of ground-level methane in local areas have been analyzed using Shandong Province as an example. The results have a potential to detect changes in the distribution of methane concentration.