A 40-year 1-km daily seamless near surface air temperature product over Yellow River Basin of China

Near-surface air temperature (NSAT) is an important environmental parameter; however, there a lack of long-term, fine-scale NSAT products that offer complete spatiotemporal continuity. Although several available cover different periods and regions, obtaining gridded data outside the provided regions remains challenging. Using multiple source (ERA5-Land, GLDAS, remotely-sensed auxiliary factors), new high-efficient framework for estimating near-surface presented using Random Forest algorithm implemented on Google Earth Engine platform. Thereby, 1-km daily seamless product (containing maximum, mean minimum) from January 1st 1981 to December 31st 2020 over Yellow River Basin in China generated. To our knowledge, this first satisfies conditions high resolution, coverage, long-term continuity simultaneously Basin. Tenfold cross validation shows RMSE, MAE R2 maximum are 1.746-1.932 K, 1.351-1.486 K 0.968-0.974; 1.219-1.354 0.940-1.035 0.984-0.987; minimum 1.663-1.732 K,1.280-1.322 0.975-0.997. Additionally, user-friendly estimation tool was developed time, enabling users derive specific areas time interest. Evaluation indicates accuracy temporally extended satisfactory, while spatially lower compared original region. The generated finer-scale hold potential benefits ecoclimate researchers policy-makers.