Analise da resposta do feiljão-caupi a distintas laminas de irrigação usando o modelo aquacrop

Abstract

The Northeastern part of Para State has been the main productive pole of cowpea in the state regarding high investments that were applied to reduce current low yield (788 kg ha”), which is due to poor management and partly due to climatic variability. Thereby, several ways have been adopted to increase its production, as an example, the use of models that intend to anticipate the use of different managements, aiding in the decision making and the obtaining of better productions. The crop models are still used in simulation of biomass (B) and yield (Y) under different climatic conditions and applied. water blades. Therefore, this work aimed to calibrate (evaluate) and apply a generic model of culture (AquaCrop) for cowpea in Castanhal, in attempt to define an agricultural calendar and zoning in the state of Para. Thus, this thesis addressed the use of the AquaCrop model in canopy cover (CC), B and Y estimations, as well as its application in the estimation of soil water content (SWC). After validation, the model was used to identify the optimum seeding window (OSW) for Castanhal, and finally an agropedoclimatic zoning of the areas suitable for the cultivation of cowpea in the state, taking into account the diverse characteristics of the region, through the spatialization of the simulated results. The results showed that AquaCrop, after its calibration for the CC, according to the leaf area index (LAI), obtained excellent estimates of B and Y, presenting mean normalized quadratic errors (NRMSE) of less than 20% and 6% for all the treatments adopted and during the calibration and validation stages, respectively. These values of NRMSE obtained demonstrate the efficiency of the model (EF) in the yield simulation of the cowpea, presenting values higher than 0.88, which indicates a very good performance. The model can also help in water management, because when comparing the SWC simulated by the model with the data obtained in the field, its efficiency (EF) was always higher than 0.70, which indicates a good performance in the simulation of this parameter. The municipality of Castanhal has an OSW that runs from April 10 to 20", due to the average yield loss of less than 10%, to a 90% probability for yields exceeding 1300 kg ha’', and to the low rain risk in its harvesting period. These results show that the model can help in the management of irrigation by cowpea cultivation and that can contribute to the development of this crop in the state.