Hourly mapping of surface air temperature by blending geostationary datasets from the two-satellite system of GOES-R series

Spatio-temporally continuous surface air temperature (SAT) is a valuable input for many research fields such as environmental studies and hydrological models. Substantial efforts have been focused on mapping daily SAT using land (LST) derived from polar-orbiting satellites NASA’s Terra Aqua. However, reconstruction of at very high temporal scales sub-daily or hourly was carried out in few studies, most which are based the Meteosat Second Generation (MSG) small-scale areas with limited stations. In this study, we developed estimation schemes random forest over large-scale region (CONUS) by blending LST datasets two-satellite system Geostationary Operational Environmental Satellite-R (GOES-R) Series aim reconstructing maps spatially uniform resolutions about 2 km CONUS. Estimation different variables were compared to evaluate contributions reanalysis hour day (HOD) variable model predicative performance, impacts station density performance also explored. We found that scheme HOD achieved higher mean RMSE 2.0 K 2.5 K, respectively. Systematic predictive errors across synoptic hours diurnal cycle significantly reduced when considering modeling SAT. The both highest 1.9 K. observed influence related types cross-validation schemes, uncertainty levels associated evaluation decrease density, increases our confidence results schemes. Therefore, accuracy area CONUS could be conservatively estimated 1.9–2.5 site-based cross-validation. expect study will facilitate further high-temporal abundant sustained observations global systems meteorological geostationary satellites.