A wind-based qualitative calibration of the Hargreaves ET0 estimation equation in semiarid regions

Abstract This study evaluates the Hargreaves (HARG) equation for estimation of monthly ET 0 under the semiarid conditions of the middle Ebro River Valley (NE Spain). First, the Hargreaves equation was compared against measured lysimeter ET 0 values at Zaragoza for the period May 1997–October 2000. The average of estimated values was only 5.6% above the average of measured values. Later, the Hargreaves equation was compared against the FAO Penman–Monteith equation for monthly ET 0 estimation at nine locations. These locations can be grouped as non-windy (Alcaniz, Daroca and Tamarite) and windy (Almudevar, Ejea, Gallocanta, Monflorite, Sarinena and Zaragoza). Simple linear regression and error analysis statistics suggest that agreement between the two estimation methods was quite good for the windy locations. Average errors ranged between 2 and 5% for Almudevar, Ejea, Sarinena and Zaragoza, and between 7 and 10% for Gallocanta and Monflorite where some underestimation was observed. However, the agreement between the Hargreaves and FAO Penman–Monteith equations was lower for the non-windy locations. In this case, the Hargreaves equation overestimated ET 0 and average errors varied between 14 and 20%. According to these results, it is proposed that, under the semiarid conditions of this study, no local calibration would be required for windy locations (those where monthly average windspeeds above 2.0 m s −1 are frequent), while a value of 0.0020 instead of the original 0.0023 should be used in the Hargreaves equation for non-windy locations. Further research should be undertaken to evaluate whether these results can be extended to other semiarid regions of the world.