Fixed versus variable bulk canopy resistance for reference evapotranspiration estimation using the Penman-Monteith equation under semiarid conditions

In this paper, daily ET0 estimates at two semiarid locations, Zaragoza and Cordoba, were obtained from the Penman-Monteith equation using either fixed (70 s m(-1)) or variable r(c) values. Variable r(c) values were computed with two models, Katerji and Perrier, and Todorovic. Daily ET0 estimates were computed from 24-h meteorological averages or from the sum of hourly estimates. Daily ET0 measured values were obtained from a weighing lysimeter (Zaragoza) and an eddy covariance system (Cordoba). There was a good agreement at both locations between estimated and measured ET0 values using a fixed r(c) value and 24-h meteorological averages. Estimates obtained from the sum of hourly estimates were somewhat worse. When 24-h meteorological averages were used, the Katerji and Perrier model for variable r(c) slightly improved ET0 estimates at both locations. But that improvement does not support the effort to locally calibrate that model. When daily ET0 estimates were obtained from the sum of hourly estimates, the Todorovic model improved the estimation at Zaragoza and, at a lesser degree, at Cordoba. Under the semiarid conditions of the two studied locations, the use of the Todorovic model is recommended to get hourly ET0 estimates from which daily estimates can be obtained. If 24-h meteorological averages are used, a fixed r(c) value as proposed by Allen et al. [Crop evapotranspiration: guidelines for computing crop water requirements, FAO Irrigation and Drainage Paper No. 56, FAO, Rome, 1998] should be enough for accurate ET0 estimates. (C) 2002 Elsevier Science B.V. All rights reserved.