Relationships between commonly-used South African and international soil phosphorus extraction tests on pristine and cultivated soils

Abstract

The availability of soil phosphorus (P) norms for various plant available P tests is limited due to the cost of establishing these norms. Consequently a need exists for a means of converting between the values of commonly used South-African and international soil P extraction tests for advisory purposes or comparing the results of soil P studies from different regions. Therefore, the aim of this study was to compare and correlate eight widely used soil P extraction tests on a wide variety of South African soils (48 pristine and cultivated samples). It was found that the relative effectiveness of total P extraction was a follows: 1:2 H₂O < Olsen < Colwell < Bray I < Ambic I < Bray II < Mehlich II ≈ citric acid. The strongest linear correlations were observed between Mehlich III, Bray I and Bray II (r = 0.92–0.99, p < 0.001), Ambic I and citric acid (r = 0.96, p < 0.001), and Olsen and Bray I, Bray II and Mehlich III (r = 0.85–0.89, p < 0.001). Thus, it is possible to directly convert between the soil P values of these tests with more confidence using linear regression equations derived in this study.