Suspended Sediment Amount Estimation in River Using Anfis and Conventional Methods

The amount of sediment in rivers is of great importance in terms of determining the amount of pollution, river transport, dam life and other factors. In this study, the sediment matter in river was estimated using Multiple Linear Regression (MLR), Sediment Rating Curve (SRC) and Adaptive Neuro-Fuzzy Inference System (ANFIS) methods. For sediment estimation, models were developed using the flow and precipitation data between the years 2014-2022 as input parameters from a station on the Yahara River near Catonsville, United States. The models were evaluated using both graphical and statistical analyses. Performance comparison between prediction models was conducted using three key criteria: coefficient of correlation (R), root mean square error (RMSE), and mean absolute error (MAE). Based on these evaluation metrics, the ANFIS model demonstrated superior performance compared to other models in predicting suspended sediment amount in the river.