Calibration and validation of the AquaCrop model to estimate soybean production in the Campos Gerais, Parana State, Brazil

Abstract

The Campos Gerais region, located in the Subtropical Zone in Southern Brazil, stands out for presenting agricultural yields above the national average, especially for soybean. This study aimed to calibrate and validate the AquaCrop model for soybean crop in the edaphoclimatic conditions of Campos Gerais region. The data used were from the experimental stations of ABC Foundation in Arapoti, Castro and Ponta Grossa, Parana State, and Itaberá, São Paulo State. The input data (climate, crop, soil and soil management) from the 2006/07 to 2015/16 harvests were collected at the respective experimental stations and entered in the AquaCrop model for yield simulations. The data for model calibration were different from those used in the validation. The observed and simulated productivities were evaluated by simple linear regression analysis, mean absolute and relative errors, Pearson correlation coefficient (r), agreement (d) and performance (c) indexes. The model calibration was satisfactory in the studied localities, with agreement indices ranging from 0.87 < d < 0.99. In the validation, the model performance index ranged from “terrible” to “excellent”. The results showed a good relationship between the observed and simulated yields, indicating that the AquaCrop model is an option to plan and investigate alternatives that improve soybean crop productivity in the Campos Gerais region.