| description abstract | In this study, radial basis neural network (RBNN), self-organizing map neural network (SOMNN), and multiple linear regression (MLR) are used for the estimation of daily pan evaporation at Pantnagar, situated in the foothills of the Himalayas, Uttarakhand, India. Daily climatic data include minimum and maximum air temperatures, relative humidity measured in the morning and afternoon, wind speed, and sunshine hours. Pan evaporation data were used for model calibration and validation. The combination of significant input variables for RBNN, SOMNN, and MLR models were decided using the gamma test. The results obtained by RBNN, SOMNN, and MLR models were compared with climate-based empirical models such as Penman, Stephens-Stewart, Griffiths, Christiansen, Priestley-Taylor, and Jensen-Burman-Allen models on the basis of root-mean squared error (RMSE), coefficient of efficiency (CE), and correlation coefficient (r). The results indicate that the performance of RBNN model (RMSE=1.24 mm/day, CE=.874, and r=.934) with six input variables was found to be superior in estimating daily pan evaporation at Pantnagar. | |
