All-weather instantaneous and daily average surface temperature estimation method cooperating with multi-source data

The invention discloses an all-weather instantaneous and daily average surface temperature estimation method cooperating with multi-source data, and the method comprises the steps: coupling mode data, site actual measurement data, atmosphere top observation, surface remote sensing parameters and other multi-source data, and cooperatively creating an instantaneous and daily average surface temperature model based on the multi-source data; the model is used for estimating large-area scale and all-weather instantaneous and daily average surface temperature data. According to the method, radiation, vegetation, observation angles and other related factors influencing the instantaneous and daily average surface temperature are comprehensively considered, the nonlinear relation between multi-source data and the surface temperature is fitted through a learning model, and the remote sensing-mode coupling space-time continuous instantaneous and daily average surface temperature model facing the MODIS polar orbit satell