Maize (Zea mays L.) yields in southern Mexico calculated by the Decision Support System for Agrotechnology Transfer

Abstract

Simulation models -based on biophysical processes- applied to agriculture allow knowing the dynamics of biological and environmental variables, but its use involves a large amount of soil information, climate, crop management and phenology. This study calibrated and validated, with grain yield, the genetic coefficients of the Crop Environment Resource Syntesis model (CERES-Maize) (DSSAT, version 4.7) of hybrid plants and corn creoles (Zea mays L.) The objective was to determine the scope and limitations of the model for estimating grain yields in the Mixteca (southern Mexico) region of Oaxaca, Mexico. For the calibration process, the setting for hybrid maize showed r2 = 0.94 and RMSE = 567.11 and r2 = 0.86 and RMSE = 601.58 for the Creole. The validation correlation process of hybrid maize showed r2 = 0.73 and RMSE = 976.65, values greater than Creole cultivation (r2 = 0.62, RMSE = 698.74). The CERES model underestimated in average 10% of grain yield. Hybrid cultivars showed 10% greater fit than Creole. In general, the CERES-Maize model adequately estimated maize yields.