Statistical Learning-Based Spatial Downscaling Models for Precipitation Distribution

The downscaling technique produces high spatial resolution precipitation distribution in order to analyze impacts of climate change data-scarce regions or local scales. In this study, based on three statistical learning algorithms, such as support vector machine (SVM), random forest regression (RF), and gradient boosting regressor (GBR), we proposed an efficient approach produce precipitation. demonstrate efficiency accuracy our models over traditional multilinear (MLR) models, did a analysis for daily observed data from 34 monitoring sites Bangladesh. Validation revealed that R 2 GBR could reach 0.98, compared with RF (0.94), SVM (0.88), (0.69) so the GBR-based model had best performance among all four models. We suggest should be used replace MLR more accurate map high-resolution flood disaster management, drought forecasting, long-term planning land water resources.