Performance evaluation of evapotranspiration estimation methods in Sultanpur, Uttar Pradesh, India

Abstract

Accurate estimation of evapotranspiration plays a vital role in judicious irrigation scheduling. Keeping this in view, a study was undertaken to compare the performance of five methods, viz. Penman method (PM), FAO Penman Monteith (FAO-PM), Priestley-Taylor (PT), FAO Radiation method (FAO-RM) and Hargreaves methods (HM) for Sultanpur District of Uttar Pradesh, India using the daily weather data acquired from automatic weather station during 2016-17. The performance evaluation of selected methods was carried out using linear regression and simple statistical analysis to suggest a substitute of FAO-PM for estimation of reference evapotranspiration using minimal climatic parameters available at regional scales. It was observed that the PM method performed the best and was in line with estimated ET0 by FAO-PM method with coefficient of determination (R2), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Modified Nash-Sutcliffe Coefficient (EMNS), Percent Bias (PBIAS) estimated as 0.94, 0.54, 0.40, 0.73, 3.74 for the year 2016 and 0.95, 0.44, 0.36, 0.73, 6.01 in 2017, respectively. Nonetheless, it was observed form this study that the PM method which requires four parameters, viz. daily solar radiation, maximum and minimum air temperature, and wind speed can be a substitute to FAO-PM that require more parameters for ET0 estimation in data scarce situations.

Keywords

• Irrigation scheduling
• Performance evaluation
• Reference evapotranspiration models
• Weather parameters