Evaluation of the RF-Based Downscaled SMAP and SMOS Products Using Multi-Source Data over an Alpine Mountains Basin, Northwest China

Passive microwave surface soil moisture (SSM) products tend to have very low resolution, which massively limits their application and validation in regional or local-scale areas. Many climate hydrological studies are urgently needed evaluate the suitability of satellite SSM products, especially alpine mountain areas where plays a key role terrestrial atmospheric exchanges. Aiming overcome this limitation, downscaling method based on random forest (RF) was proposed disaggregate products. We compared ability downscaled active passive (SMAP) ocean salinity (SMOS) capture information upstream Heihe River Basin by using situ measurements, triple collocation (TC) temperature vegetation dryness index (TVDI). The results showed that RF has strong applicability study area, two after residual correction more details, can better represent spatial distribution moisture. with measurements indicates correlation between SMAP is than SMOS at both point watershed scales Babaohe Basin. From TC method, root mean square error (RMSE) CLDAS (CMA land data assimilation system), were 0.0265, 0.0255 0.0317, respectively, indicating smaller errors area others. However, shown closer degree drought reflected TVDI. Overall, suggests RF-based variation well, perform significantly accuracy verification analysis results, it should be helpful facilitate applications for small scales.