Variability in Soil Test Potassium and Crop Yield

The primary objectives of this research are (1) to study the variability in soil test potassium (K) and corn response to K fertilization and (2) to evaluate soil tests for K currently used in the Midwest as well as a new soil test for K with potential to improve the predictability of response. This research is based on evaluation of long-term trials at three research farms and several on-farm replicated strip trials. The conventional trials compared several rates of K fertilizer and the strip trials compared fixed-rate and variable-rate fertilization. The various K treatments at research farms were replicated 3-6 times depending on the farm. At the on-farm trials, treatments were applied to field-length, 60 ft. wide strips replicated 3-4 times across each field. Initial soil-test K levels before applying fertilizer treatments were measured on soil samples collected from 0.75-acre cells. After harvest, soil samples were collected from each strip and cell. Grain was harvested with plot combines at the research farms trials and with yield monitors at producers’ fields.

Results indicate that higher soil K levels than normally recommended are needed to optimize crop yields in many fields. In the early 1990s, a switch was made from wet to dry samples in the analytical procedures. This required conversion of calibration data from wet sample results to dry sample results. This project has revealed that conversion factors used up to this point have not been appropriate. The data suggest that the overestimation of soil K supply is worse in soils of the Clarion-Nicollet-Webster soil association, which dominates a major part of central and northern Iowa and southern Minnesota. Overestimations of soil K supplies have been found for other associations as well. IA-09F