Comparative Study on the Different Downscaling Methods to GPM Products in Complex Precipitation Area

Fine spatial information of precipitation plays a significant role in regional eco-hydrological studies, but is hardly obtained from satellite data, especially in complex precipitation areas. The Sichuan Province, located in the southwest of China, has a highly variable terrain and the spatial distribution of precipitation exhibits extremely heterogeneity and significant autocorrelation. Multi-scale Geographically Weighted Regression (MGWR) and Random Forest (RF), were employed for downscaling the Global Precipitation Measurement Mission (GPM) data in Sichuan province, and windwardness and diurnal surface temperature range were creatively utilized to reflect the influence of localized environments. The results show those, the influence of each environment factor on the distribution of precipitation at different scales was well represented in the MGWR model, the downscaled data showed good spatial sharpening effects, additionally, the biased in the overestimated region were well corrected after downscaling. However, when based on spatial autocorrelation and considering adjacent influences, the MGWR performed poorly in correcting outlier sites where distributed around the high-high clusters. Compared with MGWR, relying on independently constructed decision trees and powerful regression capabilities, better correction for outlier sites has been achieved in RF. Nevertheless, the influence of environmental variables reflected in RF differs from actual conditions, and detailed characteristics of precipitation spatial distribution have been lost in the downscaled results. MGWR and RF demonstrate varying applicability when downscaling GPM products in complex terrain areas, as they both improve the ability to finely depict spatial information, but differ in terms of texture property expression and precipitation bias correction.