Artificial neural network modelling of reference evapotranspiration for Garhwal region of Uttarakhand

In the present study, an Artificial Neural Network (ANN) technique was used for modelling of reference evapotranspiration for Nagini watershed located near Chamba-Ranichauri, on Rishikesh-Uttarkashi route, in Tehri Garhwal district of Uttarakhand, India. Using pattern matching capability of ANN, the network was trained to recognize the pattern of daily weather data and its corresponding ETo obtained by Penman-Monteith method. More than 250 architectures were tried and 3 −6 −4 −1 was observed to be the best architecture. The developed model was validated by testing its performance using correlation coefficient, root mean square error and coefficient of variation of the residual error. On the basis of these performance indicators the developed ANN model performed satisfactory and it can be utilized effectively for prediction of ETo for the study area.