Using the dual approach of FAO - 56 for partitioning ET into soil and plant 1 components for olive orchards in a semi - arid region

24 The main goal of this research was to evaluate the potential of the dual approach of 25 FAO-56 for estimating actual crop evapotranspiration (AET) and its components (crop 26 transpiration and soil evaporation) of an olive (Olea europaea L.) orchard in the semi arid 27 region of Tensift-basin (central of Morocco). Two years (2003 and 2004) of continuous 28 measurements of AET with the eddy covariance technique were used to test the performance 29 of the model. The results showed that, by using the local values of basal crop coefficients, the 30 approach simulates reasonably well AET over two growing seasons. The Root Mean Square 31 Error (RMSE) between measured and simulated AET values during 2003 and 2004 were 32 respectively about 0.54 and 0.71 mm per day. The obtained value of basal crop coefficient 33 (cb K) for the olive orchard was similar in both seasons with an average of 0.54. This value 34 was lower than that suggested by the FAO-56 (0.62). Similarly, the single approach of FAO-35 56 has been tested in the previous work (Er-Raki et al., 2008) over the same study site and it 36 has been shown that this approach also simulates correctly AET when using the local crop 37 coefficient and under no stress conditions. 38 Since the dual approach predicts separately soil evaporation and plant transpiration, an 39 attempt was made to compare the simulated components of AET with measurements obtained 40 through a combination of eddy covariance and scaled-up sap flow measurements. The results 41 showed that the model gives an acceptable estimate of plant transpiration and soil 42 evaporation. The associated RMSE of plant transpiration and soil evaporation were 0.59 and 43 0.73 mm per day, respectively. 44 Additionally, the irrigation efficiency was investigated by comparing the irrigation 45 scheduling design used by the farmer to those recommended by the FAO model. It was found 46 that although the amount of irrigation applied by the farmer (800 mm) during the growing 47 season of olives was twice that recommended one by the FAO model (411 mm), the 48 3 vegetation suffered from water stress during the summer. Such behaviour can be explained by 49 inadequate distribution of irrigation. Consequently, the FAO model can be considered as a 50 potentially useful tool for planning irrigation schedules on an operational basis.