Forecasting Water Temperature in Cascade Reservoir Operation-Influenced River with Machine Learning Models

Water temperature (WT) is a critical control for various physical and biochemical processes in riverine systems. Although the prediction of river water has been subject extensive research, very few studies have examined relative importance elements affecting WT how to accurately estimate under effects cascaded dams. In this study, series potential influencing variables, such as air temperature, dew discharge, day year, wind speed precipitation, were used forecast daily downstream First, permutation variables was ranked six different machine learning models, including decision tree (DT), random forest (RF), gradient boosting (GB), adaptive (AB), support vector regression (SVR) multilayer perceptron neural network (MLPNN) models. The results showed that year (DOY) plays most important role each model WT, followed by flow which are two commonly factors unregulated rivers. Then, combinations three inputs develop parsimonious based on where their performance compared according statistical metrics. demonstrated GB3 RF3 gave accurate forecasts training dataset test dataset, respectively. Overall, could be effectively applied predict regulation