Atributos físicos do solo e produtividade da soja sob plantas de cobertura.

Abstract

The benefits of a no-tillage system and crop rotation in order to mitigate soil compaction problems show that the use of cover crops is an interesting management to improve soil quality and therefore crop yield. However, there is a spatial dependence for soil properties, thus it is an important geostatistic analysis to detect the structure of dependence in the studied area. This study aimed at evaluating the cover crops performance in winter management and the spatial variability of porosity, bulk density, water content of soil and soybean yield. The experiment was carried out in a randomized block design with three blocks, five treatments and two replications per treatment. Each plot was averagely 5.1 m wide and 133 m of length. There were five winter managements as treatments: black oat; Consortium 1 (consortium of forage turnip and black oats); Consortium 2 (forage turnip, black oat and common vetch), wheat, and control (remained fallow during the winter). Data were collected in a two hectare area, in two periods: before each sowing and after management of winter treatments to evaluate macro and microporosity, total porosity, bulk/ soil density and water content of soil from 0 - 0.1 m depth. The cover crops management was carried out with a roller-knife and wheat was cropped by a harvester machine. Soybean was sown in the whole area during summertime and determined its yield for each treatment. After exploratory and geostatistic analyses, the theoretical models were set to semivariograms for each attribute by cross-validation. For each map drawing, the sampling points were considered for each winter treatment. Then an interpolation of values for the whole area was done, as if the respective treatment were used in it. For the comparison of maps made for different treatments, it was used the relative deviation coefficient (RDC) by taking the control as standard. Pearson's correlation and Spearman non-parametric correlation coefficients were used to evaluate the correlation degree among variables. In the first soil analysis, all variables showed spatial dependence. This fact, however, was not observed in the second soil analysis, where, except for soybean yield, all variables showed a nugget effect in someone treatment (except for Consortium 2). It was concluded that the reduction of porosity raised high values for the soil density, which, in its turn, decreased soil ability to retain water. The studied cover crops improved macroporosity and total porosity in some regions of this area. Besides, the use of black oats as cover crop was more efficient and kept higher soil water content, so it can be suggested in rotation with soybeans to decrease bulk density.