CLUSTER ANALYSIS OF SPATIOTEMPORAL CORN YIELD PATTERNS IN AN IOWA FIELD

Authors: Jaynes, Dan; Kaspar, Thomas; Colvin, Thomas; James, David

Submitted to: Agronomy Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/31/2001
Publication Date: 5/1/2003
Citation: JAYNES,D.B., KASPAR,T.C., COLVIN,T.S., JAMES,D.E., CLUSTER ANALYSIS OF SPATIOTEMPORAL CORN YIELD PATTERNS IN AN IOWA FIELD, AGRONOMY JOURNAL, 2003. v. 95. p. 574-586.

Interpretive Summary: Varying the rate of fertilizer inputs across a field according to the site-specific needs of a crop holds promise for producing crops with fewer inputs and thus improved profitability for farmers and reduced impacts of farming on the environment. To be successful, variable rate applications require that production fields be divided into smaller areas or management zones where inputs such as fertilizer can be tailor applied. It has become obvious over the past decade that identifying management zones within non-irrigated corn fields is not an easy task because of the great seasonal variability in yield patterns within fields. In this research, we demonstrated that precise records of corn grain yield from several years can be used to classify the field into a few zones that reflect stable yield behavior within a field. We also demonstrated that easily measured auxiliary field information such as elevation, slope, and soil electrical conductivity can be used to predict the distribution of these stable yield zones within a field. Using these auxiliary data should make it relatively inexpensive to identify stable yield zones within other fields. The next step in the research is to quantify whether these yield response zones can be used as management units for adjusting the application rate of inputs such as fertilizer. These findings will be of value to researchers, farmers, agricultural consultants, and others interested in applying variable rate techniques to corn production.

Technical Abstract: Defining management zones within a field is the first step in varying agricultural inputs to improve crop production efficiency. 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. Non-hierarchal cluster analysis was applied to six year of corn yield data collected for 224 yield plots on a regular grid. We were able to group yield observations into five temporal yield patterns. The clusters were not randomly distributed across the field but instead formed contiguous areas roughly equivalent to landscape positions. A multiple discriminant analysis was used to predict the spatial occurrence of the clusters from easily determined field attributes; soil electrical conductivity, elevation, slope, plan curvature, and profile curvature. Gray-scale soil color and aspect did not improve the multiple discriminant functions in descriminating between clusters. The multiple discriminant functions were unable to distinguish between the two clusters located on the wettest 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. Finally, the discriminant functions allowed us to partition the entire 32-ha field into four yield clusters by using field attributes averaged for 2-m grid cells across the field.