Runoff Estimation in the Upper Reaches of the Heihe River Using an LSTM Model with Remote Sensing Data

Runoff estimations play an important role in water resource planning and management. Many accomplishments have been made runoff based on data recorded at meteorological stations; however, the advantages of use remotely sensed estimating watersheds for which are lacking remain to be investigated. In this study, MOD13A2 normalized difference vegetation index (NDVI), TRMM3B43 precipitation (P), MOD11A2 land–surface temperature (LST), MOD16A2 evapotranspiration (ET) hydrological station were used as sources with estimate monthly through application a fully connected long short–term memory (LSTM) model upstream reach Heihe River basin China from 2001 2016. The results showed that inputting multiple remote sensing parameters improved quality estimation compared rain gauge observations; increase R2 0.91 0.94 was observed implementation process, Nash–Sutcliffe efficiency (NSE) improvement 0.89 0.93. incorporation well satellite provided slight respective value 0.95 NSE 0.94. This indicates LSTM has great potential estimation, obtained by technology provide alternative approach areas available lacking.