Evaluating the Performance of the AquaCrop Model to Soil Salinity in Jordan Valley

The demand to apply a decision support system to simulate salinity and drought is increasing with time, particularly in arid and semi-arid regions like Jordan, where the threat of land degradation by salinization is of high concern.   The main objective of this research was to evaluate the response of the AquaCrop model to soil salinity in Jordan Valley. Two experiments were conducted, one at the field and the other at the greenhouse. Three water salinity levels (S): S1 (control = 2 dS m−1), S2 (4 dS m−1), and S3 (8 dS m−1) with three irrigation amounts (R): R1 (control = 120%), R2 (100%), and R3 (70%) were used in the field. Four levels of saline water (S):S1 (control =0.65), (4) S2, (8) S3, and (10) dS m-1) S4 were used in the greenhouse. In both experiments, grain yield, and aboveground biomass were measured after harvesting. Soil salinity and pH were measured every three weeks during the growing season to monitor soil salinization. Results showed that the final field grain yield was good in calibration and validation, with a 0.96 agreement index (d). The efficiency factor (E) was 0.86 and 0.87 for calibration and validation, respectively, while the normalized root mean square error (NRMSE) was less than 4 %.  Field biomass d-index of 0.87 and 0.71 and E of 0.65 and 0.45 for Calibration and Validation were found, respectively. In the greenhouse experiment, the results were less satisfactory.  Grain yield showed d-index of 0.84 and 0.88 in calibration and validation, respectively, while biomass showed poor results. All statistical criteria used in this research indicated that the model can simulate grain yield and biomass properly in the field, however, biomass statistical results were less accurate. Overall it is recommended, to use AquaCrop for soil salinity management in Jordan Valley.