Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Analysis of Multi Year Yield Data for Delineating Yield Response Zones

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  Paper number  031086,  2003 ASAE Annual Meeting . (doi: 10.13031/2013.13974) @2003
Authors:   Kenan Diker, Dale F. Heermann, Gerald W. Buchleiter
Keywords:   Three-state frequency analysis, yield response zones, corn, precision agriculture

Producers have been collecting yield maps for several years now. The use of multi year yield maps is limited. Two-state frequency analysis overcame pre-set number of classes in the field and allowed users to simultaneously visualize the spatial and temporal variability. The technique is simple and easily understood. The objective of this paper is to expand the technique to three-states by introducing a middle range bounded by mean +/- a fraction of standard deviation (). The zone delineation was assessed. The fractions of that would be most suitable to use was determined. The yield data for each of the 5 years were interpolated by using Inverse Distance Weighted technique using 6.1m cell size. The yield data was reclassified into three classes: low yield (L), high yield (H) and medium yield (M). Medium yield represented the range bounded within the mean +/- (0.25, 0.5, 0.75 or 1 ). L was assigned to below the middle range and H was assigned to above the middle range. An ArcView script was developed to determine the majority of the yield class (L, M, H) in each cell between the years and to assign it to that cell. A medium range of mean yield +/- 0.5 standard deviation seemed to be the most representative. The frequency analysis seemed to be better in delineating stable low and high yielding zones than the averaging technique in which the percent area of the yield classes was highly related to the number of classes. The overall agreement between two and three state frequency analysis was 64%. It is hypothesized that the three-state frequency analysis would provide a tool to delineate similar response zones for site-specific management.

(Download PDF)    (Export to EndNotes)