Predicting the impacts of climate change on nonpoint source pollutant loads from agricultural small watershed using artificial neural network

This study described the development and validation of an artificial neural network (ANN) for the purpose of analyzing the e ects of climate change on nonpoint source (NPS) pollutant loads from agricultural small watershed. The runo discharge was estimated using ANN algorithm. The performance of ANN model was examined using observed data from study watershed. The simulation results agreed well with observed values during calibration and validation periods. NPS pollutant loads were calculated from load-discharge relationship driven by long-term monitoring data. LARS-WG (Long Ashton Research Station-Weather Generator) model was used to generate rainfall data. The calibrated ANN model and load-discharge relationship with the generated data from LARS-WG were applied to analyze the e ects of climate change on NPS pollutant loads from the agricultural small watershed. The results showed that the ANN model provided valuable approach in estimating future runo discharge, and the NPS pollutant loads.