Spatial-Temporal Assessment of Soil Exchangeable K Concentrations As Affected by K Applications in a Long Term Experiment.

Improving the ability to predict spatial and temporal variations in K bioavailability is critical to optimize fertilizer use. This study assessed the effect of K fertilization/residual K fertilizer on plant available K at several depth increments in a corn-soybean rotation grown in a no-till Toronto-Millbrook silt loam, tilled Tracy sandy loam, and no-till Cobbsfork silt loam in west-central, north-central, and south-eastern Indiana, respectively. Cumulative fertilizer rates applied ranged from 0 to 900 kg ha^-1 (1997-2002). Following 2002, no K fertilizer was applied. Soil samples were obtained from three fall samplings (2009, 2010, and 2011). Exchangeable soil K (1 mol ammonium acetate-extractable (NH₄-ext)) was determined at depths 0 to 10, 10 to 20, 20 to 30, 30 to 40, 40 to 60, and 60+ cm after harvest for both crops.

Soil K concentrations pattern with depth was unique for each soil series. For the Tracy soil, K concentrations beyond depth the interval of 10 to 20 cm remained relatively constant between 32 and 42 mg kg^-1across depths, while K concentrations increased with depth for the Cobbsfork soil, with the highest concentrations at depth 60cm+ ranging from 53 to 66 mg kg^-1. However, for the Toronto-Millbrook series soil K concentrations with depth have an S-like pattern, with relatively high K concentrations ranging from 98 to 161 mg kg^-1 and 110 to 114 mg kg^-1 at depths intervals 0 to 10 cm and 30 to 60 cm, respectively. Soil K concentrations differences within each depth increment were more pronounced in 2009 for the Tracy series, with significant treatment effects down to depth interval of 30 to 40 cm; in 2011 significant treatment effects were only observed on upper soil layers (0 to 10, and 10 to 20 cm). Soil K concentration treatment differences within each depth increment for the Toronto-Millbrook series were limited to the uppermost soil layer for all years, while no treatment differences within each depth interval were observed for the Cobbsfork soil at any year. Overall, it was more likely to find significant differences on soil K concentrations between depths than between treatments for the Toronto-Millbrook and Cobbsfork soils; for the Tracy soil, treatment*depth interaction was highly significant for all years. Furthermore, while soils for the Toronto-Millbrook and Tracy series showed a slight reduction in soil K concentrations from 2009 to 2011, at the Cobbsfork site K concentrations remained relatively the same with time. While K fertilizer recommendations are generalized, soil specific response to K applied must be considered as they differ greatly in space and time between locations.