Comparative Evaluation of ANNs and Hargreaves Method to Model ETo

Abstract

This review goes for creating counterfeit neural system (ANN) based reference evapotranspiration (ET_o) models comparing to Hargreaves (HG) strategy. The ANN models were created utilizing pooled atmosphere information of various areas under four agro-environmental locales (semi-dry, dry, sub-damp, and muggy) in India. The inputs for the development of ANN models include daily climate data of minimum, maximum air temperatures and extra terrestrial radiation and the target consists of the FAO-56 PM estimated ET_o. Comparison of developed ANN models with the conventional HG method. The performance indices used for comparison include root mean squared error (RMSE) and coefficient of determination (R²). Based on the comparisons, it is concluded that the ANN models performed better than conventional HG method.