Cluster Analysis of Spatiotemporal Corn Yield Patterns in an Iowa Field

Abstract

Crop yields are frequently heterogeneous across space and time. We performed this study to determine if cluster analysis could be used to decipher the temporal and spatial patterns of corn (Zea mays L.) yield within a field. Nonhierarchal cluster analysis was applied to 6 yr of corn yield data collected for 224 yield plots on a regular grid on the southern half of a 32‐ha field. We were able to group the yield observations into five temporal yield patterns or clusters. The clusters were not randomly distributed across the field but instead formed contiguous areas roughly equivalent to landscape positions. Cluster membership was determined primarily by yield differences in years with growing season precipitation greater than the 40‐yr average. A multiple discriminant analysis was used to predict the spatial occurrence of the clusters from easily determined field attributes: soil electrical conductivity, elevation, slope, and plan and profile curvature. The multiple discriminant functions were unable to distinguish between the two clusters located on the lowest portions of the landscape. Because of similar temporal yield patterns in these two clusters, they were combined and the multiple discriminant analysis repeated for four clusters. Using a holdout sample approach, we achieved 76 and 80% success rates in classifying the yield plots into the correct yield clusters. If response curves for inputs such as N prove to be unique for the different yield clusters, then clustering of multiple‐year yield data may prove an effective method for determining management zones within fields.