Prediction of annual runoff using Artificial Neural Network and Regression approaches

Abstract

Prediction of runoff is often important for optimal design of water storage and drainage works and management of extreme events like floods and droughts. Rainfall-runoff (RR) models are considered to be most effective and expedient tool for runoff prediction. Number of models like stochastic, conceptual, deterministic, black-box, etc. is commonly available for RR modelling. This paper details a study involving the use of Artificial Neural Network (ANN) and Regression (REG) approaches for prediction of runoff for Betwa and Chambal regions. Model performance indicators such as model efficiency, correlation coefficient, root mean square error and root mean absolute error are used to evaluate the performance of ANN and REG for runoff prediction. Statistical parameters are employed to find the accuracy in prediction by ANN and REG for the data under study. The paper presents that ANN approach is found to be suitable for prediction of runoff for Betwa and Chambal regions.