A comparison of five extraction methods for determining available soil phosphorus in Nova Scotia blueberry soils

A project was initiated to determine the best predictor of plant available phosphorus for the Nova Scotia lowbush blueberry from the following extractants: Mehlich 1, Mehlich 3, Bray 1, Modified Morgan and the Anion Exchange Membrane (AEM). The project was initiated in 1999 with three newly established sites; two loamy soils located in South Branch (South Branch and White Fields), and one sandy loam soil in Debert. Two additional sites, both sandy loams, one located in Parrsboro and the other at the Wild Blueberry Institute were also used. Soil samples were taken in the spring and at tip dieback (July) or harvest (August) from all sites during the 1999 and 2000 growing seasons. Tissue samples were obtained, with the July and August soil samples, to enable correlation of extracted soil phosphorus with the phosphorus present in the tissue. All data obtained were statistically analyzed with the SAS System software package. Results indicate that the complexity with which the blueberry plant takes in P affects the ability of the extractants to predict leaf P concentrations. It would appear that at low levels of extractable P (<10 mg extractable P kg-1 soil) none of the five extractants can accurately predict leaf P concentrations. However, the White Field site did have significant correlations (r > 0.60) for all five extractants, due to the high levels of extractable P, which were higher than those obtained at the other sites. The remaining sites had levels of P which were too low to accurately predict the P concentrations in the leaf. Further work on the uptake of P by the blueberry plant under conditions of low and high levels of extractable P is required.