Validation of Soil-Test Based Phosphorus and Potassium Fertilizer Recommendations for Soybean.

Soil-test-based fertilizer recommendations are used to determine how much P and K fertilizer is needed to maximize yields and maintain soil fertility. Our objectives were to validate the accuracy of the University of Arkansas’ fertilizer recommendations in predicting irrigated soybean [Glycine max (L) Merr.] yield response to existing soil-test P (STP), and K (STK) interpretations, and assess the accuracy of trifoliate leaf-P and -K concentrations at the R2 stage for identifying fields that respond positively to fertilization. Fertilization trials were established at 17 sites and each contained six treatments including a combination of two P rates (0 or 32 kg P ha^-1) and four K rates (0, 56, 112, and 148 kg K ha^-1). The recommended P and K rates were based on the mean soil-test values from the 0-10 cm soil depth. Soil-test P and K levels were categorized as Suboptimal [<26 mg P (10 sites) and 91 mg K kg^-1 (6 sites)], Medium [1 (P) and 3 sites (K)], or Optimal [>35 mg P (6 sites) and 130 mg K kg^-1 (8 sites)] for interpreting the expected yield response to fertilization. Significant (P≤0.10) yield increases from P fertilization occurred at only 1 site (Medium STP level) and from K fertilization at 5 sites (Suboptimal and Medium STK levels).  The existing STP and STK recommendations accurately predicted yield response to fertilization at 60% and 71% (P≤0.10) of the fields, respectively. Soil-test interpretations were least accurate for P in fields with a Suboptimal P level and for fields with a Medium K level. Leaf-K concentrations were 48% accurate in predicting yield response to K fertilization. The utility of leaf-P concentrations taken at the R2 stage could not be evaluated as no site had deficient leaf-P concentrations.