ESTIMATION OF REFERENCE EVAPOTRANSPIRATION BY HARGREAVES-SAMANI AND PENMAN-MONTEITH METHODS IN MEDITERRANEAN ENVIRONMENT (PORTUGAL)

Abstract

Accurately estimating reference evapotranspiration (ETo) is fundamental for calculating crop water needs, supporting irrigation programs, climate change studies, and many analysis and mapping applications. Whenever data are lacking for calculating ETo by the Penman-Monteith (PM) method, FAO-UNESCO advises using the Hargreaves-Samani (HS) method. These two methods were assessed for ETo estimation at 20 locations distributed over the Portuguese territory. Normal monthly data (1971-2000) and monthly and daily data from two years (2019 and 2020) were the datasets used. Monthly and daily estimates of ETo obtained by the HS method were very good predictors of those obtained by the reference method (PM method), even when linear regressions were forced through the origin. Even so, correlations were always better with monthly data (R2≥0.97) than with daily data (R20.90). These results thus make the HS equation easy to calibrate, regardless of the time base used. The corrected values of the adjustment coefficient (Krs) depend more on the annual mean values of wind speed (𝑉𝑎̅) or thermal amplitude (𝑇𝐴𝑎̅̅̅̅̅) than on elevation. Good correlations (R2 0.80) were also obtained when Krs-values were plotted against pairs of 𝑇𝐴𝑎̅̅̅̅̅ and 𝑉𝑎̅ allowing the calibration of the HS equation as a function of these variables (multiple regression). This approach can be useful in the local calibration of the Hargreaves-Samani equation whenever it replaces the PM method.