Prediction of evaporation in tropical climate using artificial neural network and climate based models

摘要

Malaysia is a tropical country with high rainfall rate. But there is increasing demand on water due to development. Also, global weather change makes the dry season longer and evaporation rate from impounding reservoirs higher. So, evaporation as a natural phenomenon contributes in reducing the availability of water for various uses. Estimation of evaporation can give an idea about the water losses from storage reservoirs and help in water management. In this study, the evaporation from Batu Dam Reservoir which is located at the Selangor state, Malaysia is estimated using selected models namely, the artificial neural networks (ANN) models and climate based models (Penman and Priestley-Taylor). Data acquired from dam authority was used to run theses models. Statistical tests conducted on the models output reveal that ANN-4 model is the best among the tested ANN models with the coefficient of efficiency (CE) of 90%. However, Priestley-Taylor model give better accuracy than Penman model with CE of 82%.