Calibration and Validation of the STICS Crop Model for Managing Wheat Irrigation in the Semi-Arid Marrakech/Al Haouz Plain

Abstract : In the first part of this work, the shoot growth module and grain yield of the STICS crop model were calibrated and validated by using field data which was collected from irrigated winter wheat fields in the Haouz plain near Marrakech. The calibration was performed on the thermal units between the four phenological stages that control the dynamics of leaf area index and the thermal unit between emergence and the beginning of grain filling. The plant phenology was calibrated for three fields monitored during the 2002/03 season. Evaluation of the grain yields and the temporal evolution of leaf area index were done for six validation fields during 2003/04. The results showed the significant accuracy of the model in simulating these variables, and also indicated that the plants mainly suffered from lack of nitrogen. The results in the second part show the potential of crop modeling to schedule irrigation water, on the assumption that the plants were growing under optimal conditions of fertilization. In this case, the model was used to manage the time of irrigation according to a threshold for water deficit. Various simulations displayed logical trends in the relationship between the grain yield and both the amount and timing of irrigation water. These results were finally compared with those obtained from real irrigation practices. For the particular climate of 2003/04, the comparison showed that 70 mm and 40 mm of water could be saved in case of early and late sowing, respectively.