SALP SWARM OPTIMIZATION NEURAL NETWORK FOR DAILY WATER LEVEL FORECASTING WITH THE IMPACTS OF CLIMATE CHANGE

Forecasted daily water level data is essential in water resource planning and management. Proper water resource planning and management based on accurate water level forecasting, considering the impact of climate change, can help minimize flooding damage and achieve optimum use of water resources. Thus, this paper proposed applying the Salp Swarm Optimization Neural Network (SSONN) model to forecast daily water levels at Batu Kitang River under the impact of climate change. This study was conducted using seven years of rainfall and water level data from Batu Kitang Station and Global Climate Model (GCM) predictors from Institut Pierre Simon Laplace – Climate Model 5A – Medium Resolution (IPSL-CM5A-MR) under the Representative Concentration Pathway (RCP) 4.5 scenario. The performance of the SSONN model for daily water level forecasting was evaluated based on Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Coefficient of Correlation (r). The reliability of the SSONN model was compared with the performance of the Levenberg-Marquardt Neural Network (LMNN) and Scale Conjugate Gradient Neural Network (SCGNN). Results obtained from this study indicate that the performance of SSONN was superior to LMNN and SCGNN.