SILVA, R. A.; http://lattes.cnpq.br/3806476992851925; SILVA, Roberta Araújo e.
Résumé:
This study evaluated the current conditions and effects of climate change on the productivity and water footprint of soybean cultivated in Matopiba region. To simulate crop productivity, the FAO AquaCrop version 5.0 model was used, calibrated with data from 2016 and validated with 2014, using climate, soil, crop and management parameters collected in two experimental campaigns conducted between the months of June and October in the years 2014 and 2016 in Palmas, TO. The performance of the model was evaluated using the statistical indicators: prediction error (Ep), coefficient of determination (R2), square root mean error (RMSE), mean absolute error (EMA), Nash efficiency and Sutcliffe (NSE) and Willmontt's agreement index (d). Calibration and validation of soybean crop productivity estimated by the AquaCrop model presented satisfactory results, illustrating the robustness and general applicability of the model. The AquaCrop model underestimates soybean grain yield for severe water stress conditions throughout the growing cycle. After calibration and validation, AquaCrop was used as a simulation tool for soybean crop productivity for the current scenario (2016) and medium-term (2045/2046; 2055/2056) and long-term (2075/2076; 2064/2095), fed by data from two climatic models (HadGEM2-ES and MIROC5) and considering RCPs 4.5 and 8.5. Then, the water footprint (green, blue and gray) of the current soybean of the main producing municipalities of each state that compose the Matopiba region was calculated. Subsequently, the effects of possible climatic changes under soybean productivity and water footprint, considering the climatic variations with focus on temperature, precipitation and CO2, were evaluated. The climatic models projected increase of productivity in both RCP considered, but more accentuated under RCP 8.5, due to the increase in temperature and concentration of CO2 and precipitation, that even undergoing a reduction in rainfall totals over time, still taking into account water requirement of soybeans. The current PHsoja of the Matopiba region, ranged from 2036.60 m³t-1 to 2584.12 m³t-1, values similar to those found in the literature. Under scenarios of climate change, the PHsoja decreases over the years. The future PHsoja decreases, especially the green component, due to the less accentuated increase of the evapotranspiration, resulting in greater final yield. PHverde decreases over the years, PHazul increases in the same proportion and PHcinza show practically continuous behavior. The results of this study can be used to quantify future soybean yield, water demand and utilization, as well as to obtain useful information for the management of water resources in the study region.