Bivariate spatial correlation between soil attributes and soybean productivity in an agricultural area with Dystroferric Red Latosol

Abstract

In agricultural soils with low cation exchange capacity, it is essential to analyze the bivariate spatial correlation of soybean productivity and organic matter with the soil chemical attributes. Using bivariate spatial correlation makes it possible to identify patterns and behaviors that suggest a spatial association between two soil attributes, thus enabling better soil management and more efficient use of resources. The main objective of this study was to analyze bivariate spatial correlation considering variables with different spatial dependence structures. The bivariate Lee index was also calculated for this purpose. To model and describe the spatial pattern of two spatially correlated variables, the Bivariate Gaussian Common Component Model was used. In addition to calculating the bivariate spatial correlation of soil chemical attributes with soybean productivity and organic matter, the Lee index was also calculated for pairs of simulated variables with different weight matrices and geographic distance functions. It was observed that the greater the common practical range, the higher the Lee index value, indicating a higher bivariate spatial correlation. Furthermore, shorter distances between neighboring point pairs caused higher Lee index values. The distance function to calculate the distance between the point pairs was more relevant than the weight matrix in estimating the spatial dependence radius and the Lee index value. Soybean productivity showed a direct spatial correlation with the sum of bases, as well as with the calcium and magnesium contents. Organic matter had a direct spatial correlation with the sum of bases and an inverse one with the phosphorus content