Modelagem agrometeorológica com dados de reanálise para a cultura da soja na região oeste do Paraná

Abstract

The determination of crop productive potentials allows identifying the impact of climate and management on agricultural production, contributing to improvements in management practices and agricultural planning. Agricultural productivity can be determined using agrometeorological models, which, when regionally calibrated, increase the accuracy of estimates. These models rely on meteorological data to determine productive thresholds. However, the scarcity of meteorological stations in Brazil limits the use of these models. An alternative is the use of reanalysis meteorological databases, such as NASA/POWER and ERA5-Land. This research was based on the hypothesis that reanalysis data are satisfactory for estimating Potential Productivity (PP), Achievable Productivity (PA), and Yield Gap (YG), allowing the quantification and separation of losses due to water deficit and crop management. The objective was to determine the YG due to water deficit (YGW) and agricultural management (YGM) for 50 municipalities in the western mesoregion of Paraná, using reanalysis meteorological data. The research was conducted in four stages. The first stage involved the calibration of soybean's Ky, obtained from the relationship between actual and maximum evapotranspiration and the relationship between PA and maximum productivity. The second stage validated the reanalysis data by comparing them with data from meteorological stations at the same points. The validation used statistical indices such as Mean Absolute Error, Root Mean Square Error, Pearson correlation, Wilmott's agreement, and Nash-Sutcliffe efficiency coefficient. The third stage validated the estimates of PP, PA, and YG with reanalysis data using the MZA-FAO model and the regionally calibrated Ky. The fourth stage involved making estimates of PP, PA, and YG using the NASA/POWER reanalysis database, which showed the best performance in the third stage. The estimates were made for each of the 50 municipalities in the western mesoregion of Paraná, from 2007 to 2022. The YG due to water deficit and agricultural management, as well as the Crop Climate Efficiency (EC) and Agricultural Efficiency (EA), were determined. Soybean in the western region of Paraná showed moderate sensitivity to water stress, with a Ky of 0.57. The Ky values for the phenological stages were: 0.0 for establishment; 0.49 for the vegetative stage; 0.52 for the reproductive stage; and 0.32 for maturation. ERA5-Land data showed better performance for maximum and minimum temperature, relative humidity, and wind speed, while NASA/POWER was superior in simulating net radiation and potential evapotranspiration. NASA/POWER showed the best performance in estimating PP and PA. The PP in the western region varied from 7200 to 10800 kg ha⁻¹, with an average of 8432.7±766.3 kg ha⁻¹. The PA ranged from 3400 to 5200 kg ha⁻¹, with an average of 4237.5±429.0 kg ha⁻¹. The Yield Gap Water (YGW) ranged from 3300 to 5300 kg ha⁻¹, with an average of 4195.2±528.3 kg ha⁻¹. The Yield Gap Management (YGM) ranged from 500 to 2000 kg ha⁻¹, with an average of 1074.6±455.7 kg ha⁻¹. The EC ranged from 46% to 56% and the EA varied between 58% and 88%.