Abstract
Abstract— In this paper, the performance of Hargreaves -Samani equation compared with Penman Monteith equation of FAO as a reference evapotranspiration (ET0) prediction method was investigated. Furthermore, the Hargreaves-Samani equation was locally calibrated. Averages of monthly meteorological data from ten stations for 30 years in northwest Libya were used. In order to obtain best evaluation for the of Hargreaves-Samani equation, the study area was divided into three sub climatic regions, which are coastal, inland, and mountainous regions. The performance ofHargreaves-Samani equation was evaluated using a number of statistical indices, namely, the mean bias error (MBE), the root means square error (RMSE), the mean absolute error (MAE),Nash-Sutcliffe efficiency (η), coefficient of determination (R2), and the ratio between both averageestimations of ET0. The results showed that Hargreaves-Samani equation underestimates the ET0 inall the data used compared with Penman Monteith equation. The variation was most significant in the mountainous stations. This was evident from the average values of MBE, RMSE, MAE, which were-1.13, 1.2, and 1.14 mm/day, respectively, while the values of the same indicators for the coastal area were -0.4, 0.48, and 0.41 mm/day, correspondingly, and for the inland area were -0.35, 0.42, and 0.37mm/day, respectively. Since R2 and (η) values were good, the improvement of the local prediction of HS equation was performed. Corrections of HS equation factor (i.e. 0.0023) were obtained for the tree regions using the slop of the linear regression analysis. The new values of the calibrated factor of the Hargreaves Samani equation were 0.0026, 0.0024 and 0.0032 for coastal, inland and mountainous regions, respectively. These values are greater than the value of the original factor (i. e. 0.0023). In order to obtain better prediction of the monthly ET0 in the study area, it is recommended to modify the HS equation factor to the new values