Optimization of management zone delineation by using spatial principal components

Definition of management zones is the delimitation of sub-areas with similar topographic, soil and crop characteristics within a field. Among the many variables that can be used for this definition, those that are stable and spatially correlated with yield are more often recommended for use. Clustering algorithms such as Fuzzy C-means are also frequently applied to define management zones. Three variable selection techniques that can be applied with Fuzzy C-means are spatial correlation analysis, principal component analysis (PCA), and multivariate spatial analysis based on Moran's index PCA (MULTISPATI-PCA). In this study, the efficiency of each of these three techniques used in conjunction with the clustering method was assessed. Furthermore, a new variable selection approach, named MPCA-SC, based on the combined use of Moran's bivariate spatial autocorrelation statistic and MULTISPATI-PCA, was proposed and tested. The evaluation was performed by using data collected from 2010 to 2014 from three agricultural areas in Paraná State, Brazil, with corn and soybean crops, generating two, three, and four classes. The delineated management zones were different according to the method used, and MPCA-SC provided the best performance for the Fuzzy C-means algorithm and the best variance reduction values of the data after the delimitation of the sub-areas. Furthermore, MPCA-SC provided management zones with greater internal homogeneity, making them more viable for implementation from the viewpoint of field operations.