Modelos de regressão aplicados ao rendimento de grãos do milho em função dos atributos químicos do solo

Abstract

In the field, it is important to know the factors influencing the yield of cultivars, so that soil use and management techniques can be improved, aiming at sustainability and increased productivity. The chemical attributes of the soil are generally taken into account when purchasing fertilizer. In this sense, knowing how the chemical attributes jointly influence the grain yield of a crop, becomes relevant, since this knowledge can provide more accurate methods for soil management. Thus, this work seeks to obtain a model through mathematical modeling that describes the behavior of corn grain yield, as a function of the values of chemical attributes of a clayey oxisol. The models were obtained through regression using the least squares method and, after the construction and presentation of the models, some validation tests will be applied to determine the most appropriate model and, according to the chosen validation criteria, some improvements in the land use and management practices. Through these criteria, the linear model and the Gompertz model were considered to be the best.