Comparative study of various reference evapotranspiration models under different agroclimatic zones

Reference evapotranspiration (ET_o) study is important for climatically diverse Himachal Pradesh to ensure sustainable water resource management. The present study aims to establish the best alternative ET_o method among the combination-, radiation-, and temperature-based methods. The study area consists of four agroclimatic zones (zone I: subtropical, zone II: humid, zone III: wet temperate, zone IV: dry temperate) of Himachal Pradesh, state of India. The statistical performance indicators, i.e., root mean square error (RMSE), coefficient of determination (<i>R</i>²), percentage error (PE), and scatter index (SI), were used to evaluate model performance in each agroclimatic zone over the period of 2012-2021. In combination methods, corrected Penman (CPEN), and modified Penman (MPEN) performed well; however, the results of CPEN were found to be significantly closer to 56PM for all agroclimatic zones. Among the radiation methods, the Priestly-Taylor (P-T) model was found to be a better predictor than the other three methods, i.e., 24-radiation (24RAD), Turc (T-C), and Makkink (MAK) for agroclimatic zone I and zone II. For zones III and IV, the T-C method outperformed all other radiation methods, followed by MAK. The results of the study established that H-S (RMSE = 0.87, <i>R</i>² = 0.98, PE = 2.63, SI = 0.13) as a temperature method, outperformed other methods in the study area, followed by P-T and T-C methods.