Predicting Daily Suspended Sediment Load Using Machine Learning and NARX Hydro-Climatic Inputs in Semi-Arid Environment

Sediment transport in basins disturbs the ecological systems of water bodies and leads to reservoir siltation. Its evaluation is crucial for managing resources. The practical application process-based model can confront some limitations noticed lower accuracy during validation process due lack reliable physical datasets. In this study, we attempt apply machine-learning-based modeling (ML) predict suspended sediment load, using hydro-climatic data as input variables semi-arid Bouregreg basin, Morocco. To that end, years 2016 2020 were used training process, was performed with 2021 data. results showed most ML models have good accuracy, a Nash–Schiff efficiency (NSE) ranging from 0.47 0.80 phase, which indicates satisfactory performances predicting SSL. Furthermore, ranked against their generalization ability (GA), revealed developed are excellent terms GA. Overall, present study provides new insight into SSL environment, such basin.