干旱区生态环境知识资源中心

Two methods, indirect and direct, for simulating the actual evapotranspiration (E) were applied to an irrigated overhead table grape vineyard during summer, situated in the Mediterranean region (south Italy), over two successive years. The first method, indirect but more practical, uses the crop coefficient (K-c) approach and requires determination of the reference evapotranspiration E-0 (FAO (Food and Agriculture Organization) method). This method underestimated on average by 17% the daily values of the actual evapotranspiration E. The analysis in this paper shows that the values of K-c for the table grapes determined by the FAO method seem to not be valid in our experimental conditions. Similar conclusions can be found in the literature for the table grape cultivated under different experimental conditions and using different training systems. The second method, is a direct method for estimating the evapotranspiration. It requires development of a model for the overhead table grape vineyard E, following the Penman-Monteith one-step approach, and using standard meteorological variables as inputs for the determination of the canopy resistance. This method, which needs a particularly simple calibration, provided a better simulation of the hourly and daily evapotranspiration than the indirect method. In additon, the standard error of the daily values for the direct method (+/- 0.41 mm) was about 50% lower than that obtained for the indirect method, also when the indirect method used a locally calibrated coefficient K, instead of a generic K-c. Both, for practical application and theoretical issues, the advantages and disadvantages linked to the use of each tested method are discussed in detail. Copyright (c) 2007 John Wiley & Sons, Ltd.