Linear and spatial correlation of the yield components and soybean yield

Abstract

The soybean is the crop most cultivated in Brazil, with great socioeconomic importance. In the agriculture year 2008/09 in Selvíria County, Mato Grosso do Sul State, in the Brazilian Savannah, was analyzed the production components and the soybean yield cultivated in a Typic Acrustox on no-tillage. The main purpose objective was select among the production components number of pods per plant, number of grains per pod, number of grains per plant, mass of a thousand grains, mass of grains per plant and population of plants, which of the best linear and spatial correlation aiming explain the soybean yield variability. The irregular geostatistical grid was installed to collect of data, with 120 sampling points, in an area of 8.34 ha. The values of spatial dependence range to be utilized should be among 38.1 and 114.7 meters. The model of the adjusted semivariograma was predominantly the spherical. Of the lineal and spatial point of view, the number of pods per plant and the mass of grains per plant they were correlated in a direct way with the soybean yield, demonstrating be the best components to esteem her.