An Assimilating Model Using Broad Learning System for Incorporating Multi‐Source Precipitation Data With Environmental Factors Over Southeast China

Remote sensing technique is beneficial for rainfall data retrievals, however, enhancing the accuracy remains a challenge. In this study, novel framework based on broad learning system (BLS) was proposed to assimilate multi-source data. The dataset includes six satellite-based products (3B42V7, 3B42RT, IMERG, CBLD, GSMaP, and PCDR), gauge-based rainfall, environmental (temperature, specific humidity, wind speed, locations) from 1 March 2014 31 December 2017 over southeast China (SEC). Leave-one-year-out cross-validation (LOYOCV) independent validation were used evaluate BLS assimilating model. model outperformed original Pearson's correlation coefficient (CC), root-mean-square error (RMSE), Nash-Sutcliffe of efficiency (NSE) in each test year LOYOCV. considering factors performed better CC, RMSE, NSE compared that without factors. Seasonal variations daily precipitation accurately captured by BLS-based estimates. method satellites at most sites low altitudes (0–1000 m). According validation, more accurate estimates could be obtained than half using source datasets. has potential improve SEC expected applied other regions.