Prediction of Seasonal Monthly Rainfall Using Back Propagation Neural Network

Abstract

AbstractRainfall is a principle component in hydrological cycle and is a considerable source for both surface and ground water. Rainfall affects the maximum livelihood and all major activities that connected with it; therefore, prediction of rainfall is very important. The early prediction of rainfall is more significant to ascertain the effective use of water resources, productivity of crop, and preplanning of water structures. Due to nonlinear behavior of rainfall and complex nature of climatic system, prediction of rainfall is a challenging task. Many approaches have been practiced for the purpose of rainfall prediction. A nonlinear model such as artificial neural network (ANN) approach is considered in this work. The main objective of this study is to predict the seasonal monthly rainfall in Mysuru taluk, southern part of Karnataka state, India. Back propagation neural network (BPNN) technique has been attempted using previous 8 years datasets for climatic parameters; relative humidity, atmospheric temperature, wind speed, and rainfall. In the present study, ten different networks using BPNN were created and analyzed based on their regression results. The network result with higher regression relationship has been considered for rainfall prediction.KeywordsRainfall predictionArtificial neural networkBack propagation neural networkRegressionDenormalizationSimulation