Assessing machine learning models for streamflow estimation: A case study in Oued Sebaou watershed (Northern Algeria)

Abstract

This paper proposes rainfall-runoff models based on machine learning to estimate daily streamflows in Oued Sebaou Watershed, a Mediterranean Coastal Basin located in northern Algeria. Therefore, we applied Random Forest (RF), Artificial Neural Networks (ANN) - under different training algorithms -, and Local Weighted Linear Regression (LWLR) using as input combinations of current and past amounts of rainfalls and previous values of streamflow. We selected streamflow and rainfall records to calibrate and validate the stated approaches. The study considered Root Mean Square Error (RMSE) and Correlation Coefficient (R) to evaluate the accuracy of the models. Analyses of the results show that RF provided the best outcomes for both training (RMSE = 4.7458 and R = 0.9834) and validation (RMSE = 2.3617 and R = 0.9719). The ANN calibrated with the Levenberg-Marquardt algorithm presented the second-best result, outperforming its counterparts and LWLR.