Skip to main content
ARS Home » Research » Publications at this Location » Publication #157336
Title: CORN SOYBEAN YIELD ZONES FROM CLUSTER ANALYSIS OF MULTI-YEAR YIELD DATA

Author
item Jaynes, Dan
item Kaspar, Thomas
item Colvin, Thomas

Submitted to: ASA-CSSA-SSSA Annual Meeting Abstracts
Publication Type: Abstract Only
Publication Acceptance Date: 11/6/2003
Publication Date: 11/6/2003
Citation: JAYNES, D.B., KASPAR, T.C., COLVIN, T.S. CORN SOYBEAN YIELD ZONES FROM CLUSTER ANALYSIS OF MULTI-YEAR YIELD DATA. ASA-CSSA-SSSA ANNUAL MEETING ABSTRACTS. 2003. CD-ROM. MADISON, WI.

Interpretive Summary:

Technical Abstract: We applied cluster analysis to determine if corn (Zea mays L.) and soybean (Glycine max [L.] Merr.) yield within a field could be classified into a limited number of groups or clusters and if cluster characteristics and location within the field were similar for the two crops. We further tested if the spatial distribution of these yield clusters could be determined from easily measured soil and terraine properties. The analysis was applied to yield data from 224 plots arranged along eight transects spanning a 16-ha field collected for all consecutive years for a corn and soybean rotation. The partitioning phase of cluster analysis rvealed that the 224 locations could best be grouped into five clusters for either corn or soybean. Clusters for the two crops were similar and spatially distributed roughly according to landscape position. Clusters were characterized by yield response to variations in growing-season precipitation. Discriminant functions constructed from simple terrain attributes and apparent soil electrical conductivity were able to assign about 80 percent of the 224 plots into the corrct corn and soybean cluster. While not perfect, the discriminant functions did capture the major characteristics of the yield cluster pattern across the field. Applying the functions with high spatial resolution terrain and EC attributes, we were able to map grain yield zones for the entire field.