Forecasting of Suspended Sediment in River Using Artificial Intelligence Methods

Accurate estimation of the amount of suspended sediment in the river is important for protecting and managing water structures. By accurately estimating the amount of suspended matter in streams, important information is obtained by determining the life of water structures, water pollution, and river transportation. In this study, the West Branch Neversink River near Claryville, USA, was chosen as the study area. 567 real-daily data measured between 2021-2023 were used as the study data. As input parameters of the model; air temperature, precipitation, stream flow and turbidity were selected and sediment amount was estimated. Multi Linear Regression (MLR), M5 Decision Tree (M5 Tree) and Artificial Neural Network (ANN) models were used in sediment estimation. Model results were evaluated according to statistical criteria. As a result, ANN model showed best performance.