Rainfall-Runoff relationship for streamflow discharge forecasting by ANN modelling

Abstract

Rainfall-runoff modeling has been considered as one of the major problems in water resources management, especially in most developing countries such as Thailand. Artificial Neural Network (ANN) models are powerful prediction tools for the relation between rainfall and runoff parameters. Lam Phachi watershed is located in Western Thailand. In each year, people usually undergo drought problem in dry season or flooding problem in wet season due to the influence of the monsoon leading to soil erosion and sediment deposition in the watershed. The goal of this work is to implement ANN for daily streamflow discharge forecasting in Lam Phachi watershed, Suan Phung, Rachaburi, Thailand. For model calibration and validation, two time series of rainfall and discharge are daily recorded from only one hydrologic station (K. 17) in water years 2009–2012. The data from the first three years are used as the training dataset and the last year are used as the test dataset. The results showed that the coefficient of determination (R2) of ANN equal to 0.88. On the other hand, these results could be applied to solve the problems in water resource studies and management.