The relationship between the phosphorus requirement and ambic 1 phosphorus content of selected South African soils

Abstract

The phosphorus requirement (P0.11) of low to moderate P-fixing soils from the South African tobacco industry and KwaZulu-Natal was estimated from P sorption isotherms. The relationships between P0.11, Ambic 1 P, clay content and various indices of P sorption were investigated, using non-linear regression analysis. The results showed that a high degree of correlation existed between P0.11 and the Ambic 1 P content. Although this showed the Ambic 1 soil-P extractant to be a useful predictor of P0.11, the functions best describing the relationship between both these variables differed between the two regional groups of soils. The variance in P0.11 accounted for by the Ambic 1 P content was improved when supplementing it in a multiple regression equation with an index of the buffer power of the soils. This was especially true for the soils from KwaZulu-Natal or when the soils from the two different ecological regions were combined into a single group. A single point sorption test (SI) readily determined routinely, was shown to relate closely to the (multiple point) isotherm-derived indices of buffer power. Equations, allowing for P0.11 to be routinely predicted from both the Ambic 1 P content and SI, are proposed for each of the two regional groups of soils as well as for the combined group. Although the determination coefficients for the various models were all high, analysis of the predicted mean squared error terms showed that the accuracy with which P0.11 was predicted was less for the combined model than for each of the two regional ones. © 2005, Taylor & Francis Group, LLC.