The soils used in this study are from University of Tennessee (UT)’s Research and Education Center at Milan (35.9, -88.73333) and UT’s Highland Rim Research and Education Center at Springfield (36.466667, -86.816667). The soil at Milan is a moderately well-drained Grenada silt loam (fine-silty, mixed, active, thermic Oxyaquic Fraglossudalfs) and the soil at Springfield is a well-drained Mountview silt loam (fine-silty, siliceous, semiactive, thermic Oxyaquic Paleudults).

These soil samples are from field experiments initiated in 2009. There were two separate fertilizer rate trials at each location, one for P and one for K, and each was implemented in a corn-winter wheat-soybean rotation. Phosphorus trials received five P fertilizer rates (0, 29, 59, 88, and 118 kg P ha–1 yr–1) and an annual maintenance rate of 149 kg K ha–1 yr–1. The K trials received five K fertilizer rates (0, 47, 93, 139, and 186 kg K ha–1 yr–1) and an annual maintenance rate of 79 kg P ha–1 yr–1. Nitrogen fertilizer was applied in both trials according to University of Tennessee’s recommendations for corn and wheat. No synthetic N was applied for soybean. Sources of N, P, and K were ammonium nitrate, triple superphosphate, and muriate of potash, respectively. Fertilizers were broadcast between mid-December and early March. Nitrogen was applied at planting in April for corn and between mid-February and mid-March for wheat. The treatments were arranged in a Latin Square Design (five treatments with five replications) in the corn–winter wheat–soybean rotation. Plots were 4.6 by 9.1 m with a 1.5-m alley between them. No-tillage management was followed at both locations throughout the trial.

Soil samples were taken at the end of each crop year, from mid-December to as late as the end of February. The year associated with each soil value refers to the following or new crop year and should reflect the extractable P and K present before additional P and K fertilizer are applied for that new crop year (Savoy, 2013). Six, 0 to 15 cm depth soil cores (2.5 cm diameter) were randomly collected from each plot and composited. All soil samples were air-dried and ground to pass through a 2-mm sieve for analysis at the UT Soil, Plant and Pest Center (SPPC) in Nashville, TN. Mehlich-1 extraction (Mehlich, 1953) was also conducted at SPPC followed by determination of extractable P and K using Inductively coupled plasma optical emission spectroscopy (5300 DV ICP-OES, Perkin-Elmer, Waltham, MA). The remaining samples were transported to UT-Knoxville’s Department of Biosystems Engineering and Soil Science, where extractable P and K based on Mehlich-3 extraction (Zhang et al., 2014), Lancaster extraction (Oldham 2014), and Haney H3A extraction were conducted. The H3A extractant was prepared by combining 1.1 g malic acid, 0.7 g citric acid, and 0.45 g oxalic acid in 2 liters of water (personal communication with Richard Haney). Soil extraction was done by adding 40 ml of H3A to 4 g of soil, shaking for 10 minutes at 200 rpm, and filtering the extractant through 8 micron filter paper (Whatman 2V) into 50 ml plastic tubes. Inductively coupled plasma optical emission spectroscopy (5300 DV ICP-OES, Perkin-Elmer, Waltham, MA) was used for all analyses.

It is to be noted that similar data is reported for wheat and soybeans in 2015 because soil sampling was done only once during this double-cropping season.

The creation of this dataset was funded through University of Tennessee startup funds for the Jagadamma laboratory. The University of Tennessee is a land grant institution that receives NIFA funding through the USDA.