Delineation of management zones in integrated crop–livestock systems

Abstract

The lack of studies on spatial variability in integrated crop–livestock systems (ICLS) hinders understanding how to increase their efficiency by implementing precision agriculture (PA) practices. As such, little is known about how grain and forage crops interact and how to improve the decision-making process on fertilization and forage management. One technique that can help manage such systems is the delineation of management zones (MZs), regions with similar yield potential and soil and topography characteristics. Thus, this paper assesses the spatial correlation between yield and potential factors affecting it, and identifies whether it is possible to establish MZs for field management of grain and forage crops in succession in ICLS. Bivariate Moran's index was used to identify the attributes most spatially correlated with the yields. Elevation, soil apparent electrical conductivity, and clay content were the most spatially correlated variables with soybean [Glycine max (L.) Merr.] yield, while soil organic matter content and elevation were the most spatially correlated with the forage yield. Spatial principal components analysis and fuzzy c-means clustering algorithm were combined to delineate MZs for each crop. The MZs created for soybean were statistically different in grain yield, available phosphorus (P) in the 0-to-0.40-m layer and pH in the 0-to-0.20-m layer. The forage MZs showed significant differences in terms of available P in the 0-to-0.40-m layer. We conclude that MZs for ICLS tends to be crop specific, demanding different MZs to characterize soybean and forage spatial variability.