Using Corn and Soybean Yield History to Predict Subfield Yield Response

Authors:
Publisher website
Google Scholar
Abstract
Site‐specific management often uses past grain yield data to delineate management zones in a field. Questions remain about the type and amount of data required to classify subfields into management zones. The objective of this study was to test the hypothesis that yield history could predict yield patterns of subfields. Past grain yield data were collected from two experiments conducted on a Plano silt loam (fine‐silty, mixed, mesic Typic Argiudoll) near Arlington, WI. Experiment 1 included 26 yr and Exp. 2 included 10 yr of data for continuous corn (Zea mays L.), continuous soybean (Glycine max L. Merr.), and rotated corn–soybean in conventional‐till and no‐till. Average yield range among subfields within a year was 1.51 to 3.67 Mg ha−1 for corn and 0.54 to 0.81 Mg ha−1 for soybean. Across both experiments, spatial yield standard deviation of corn was 0.68 to 1.69 Mg ha−1; temporal yield standard deviation was 1.87 to 2.73 Mg ha−1. Spatial yield standard deviation of soybean was 0.24 to 0.37 Mg ha−1; temporal yield standard deviation was 0.56 to 0.85 Mg ha−1. The length of time required to detect yield differences among subfields was 2 to 18 yr and to detect consistent subfield yield patterns was 4 to 24 yr. High corn yield predicted high soybean yield and low corn yield predicted low soybean yield in 87% of subfields. Since temporal yield variation was greater than spatial yield variation, classifying subfields into management zones for input prescriptions remains challenging.

This publication has 15 references indexed in Scilit: