Predicting Mehlich-1 Soil Test Phosphorus Changes On Pastures in West Virginia.

Successful management of soil test phosphorus (STP) is dependant on the ability to predict responses to specific amendments at agronomically relevant levels.  Absent of timely soil testing the most practical way to predict a soil’s future fertility is to model a post-harvest soil test response.  Conceptually, this is similar to predicting residually available nitrogen (N) from organic fertilizer applications.  With N, cultural practices and weather dictate how various fractions are accounted for followed by calculations to estimate the rate of mineralization and subsequent availability over one or more growing seasons.  Phosphorus (P) management is at times over simplified.  Often, P is added as organic and inorganic P from organic fertilizers applications made on a N basis.  When the soil is not tested annually, STP levels are assumed to increase at one unit per unit of total P added, less crop removal.  In instances of low STP levels, high P sorption capacity or fertilizer sources with high percentages of stable organic P, such an overly simplified model could be overestimating STP levels.  To develop a more effective model for estimating STP change, greenhouse trials were conducted to evaluate STP levels relative to chemical and organic P input in the presence of actively growing vegetation, across a taxonomically diverse group of benchmark soils from West Virginia pastures.  The data collected was used to develop a series of predictive models for Mehlic 1 STP response.  The constructed models were an overall model, a taxonomically segregated model, and a pedotransfer function model.