Soil test phosphorus predicts field-level, but not sub-field level corn yield response

Authors: REED, VAUGHN - Mississippi State University; FRIDGEN, JENNI - University Of Kentucky; FINCH, BRONC - University Of Arkansas; SPARGO, JOHN - Pennsylvania State University; MCGRATH, JOSHUA - University Of Kentucky; BOWEN, JAMES - University Of Kentucky; HAHN, GENE - University Of Maryland; Smith, Douglas; RITCHEY, EDWIN - University Of Kentucky

Submitted to: Agronomy Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/25/2025
Publication Date: 2/24/2025
Citation: Reed, V., Fridgen, J., Finch, B., Spargo, J., McGrath, J., Bowen, J., Hahn, G., Smith, D.R., Ritchey, E. 2025. Soil test phosphorus predicts field-level, but not sub-field level corn yield response. Agronomy Journal. 117(1). Article e270028. https://doi.org/10.1002/agj2.70028.
DOI: https://doi.org/10.1002/agj2.70028

Interpretive Summary: Soil test-based fertilizer recommendations are compiled from a collection of many site years and locations, which allow for a robust model to provide accurate fertilizer recommendations on average, but lack site-specific precision. Our objective was to determine if soil phosphorus concentrations could predict crop nutrient response at specific locations within a field, and to determine if the critical level for soil test phosphorus varied within field boundaries. Over seven growing seasons, research was conducted in two Kentucky fields. Each field was divided into over 150 paired response plots, with sub-plots receiving no phosphorus or phosphorus fertilizer. On average, the Mehlich 3 soil test for phosphorus accurately predicted corn yield response at the field level; however, at the individual plot level the soil test failed to consistently predict where within the field phosphorus fertilizer would affect crop yield. These results suggest that the current soil test recommendations are accurate on average but lack precision. With the data collected in this study we were unable to adequately predict where yield response to phosphorus will occur. We conclude that additional factors beyond or in addition to soil phosphorus concentration were needed to support variable rate phosphorus management.

Technical Abstract: Public and private sector soil test-based fertilizer recommendations are compiled from a collection of many site years and locations, which allow for a robust correlation model adequate for accurate fertilizer recommendations on average, but not necessarily for site-specific precision. Our objectives were to determine if soil phosphorus concentrations predicted crop nutrient response at sub-field levels, and if soil test critical levels varied within field boundaries. We conducted research over seven growing seasons at two locations in Kentucky. Mehlich 3 extractable soil phosphorus ranged from 0.8 to 63 mg kg-1, with 96% of sample points falling below 30 mg kg-1 Mehlich 3 phosphorus, the cutoff or minimum concentration of Mehlich 3 phosphorus that the University of Kentucky recommends fertilizer P for corn. We divided each field into over 150 paired response plots, with subplots receiving 0 or 29.5 kg ha-1 phosphorus. On average, Mehlich 3 phosphorus accurately predicted corn yield response at the field level but failed at the plot scale (10-2 ha), with yield increasing in only 51% of plots. These results suggest that current soil test recommendations are accurate on average but lack precision. With the data collected in this study we were unable to adequately predict where yield response to phosphorus would occur. We conclude that additional factors beyond or in addition to soil phosphorus concentration were needed to support variable rate phosphorus management.