Submitted to: International Conference on Precision Agriculture Abstracts & Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 6/26/1996
Publication Date: N/A
Technical Abstract: Detailed maps delineating the spatial pattern of crop yield are now possible as a result of recent advances in computer technology coupled with state-of-the-art machinery for planting and harvesting. A more quantitative understanding of soil attribute spatial pattern is needed in order to make informed decisions about how to change agricultural management practices in response to yield patterns. The goal of this study was to functionally relate crop yield patterns observed over a seven-year period (1989-1995) with patterns in soil properties measured in 1994 for a field in Boone County, Iowa. Multiple linear regression analysis was used to analyze the effect of soil properties on yield variability on a year-by-year basis. In all cases, the models did a better job of simulating soybean yield than corn yield. Yield simulations differed most from observed values for the sites where yield was consistently low for the seven-year period. The models simulated yield fairly accurately for all other sites. Information from this analysis will be used in the development of soil quality estimates relative to crop production for this field site.