A comparison of methods used for the extraction of K in soils of the Western Cape

Abstract

A number of common soil extractants, mainly those used for soil P, were used to extract K from a range of widely differing soils which occur in the Western Cape; the objective being to evaluate the simultaneous determination of P and K availability. Soil samples were taken mainly from the upper horizons of 110 profiles representing the dominant soil forms occurring in this region. The soils varied from highly weathered, deep, well-drained acid Oxisols (Hutton form) to younger less developed duplex Alfisols (Swartland, Glenrosa and Kroonstad forms) and deep sandy Inceptisols (Fernwood form). The soils exhibited a wide range of values of clay content, CEC, pH and exchangeable K. Seven soil P extraction methods (Bray No. 1, Bray No. 2, citric acid 1%, Olsen, ISFEI (Na), Ambic and Mehlich No. 3) were compared with two K extraction methods, (0.5 M NH4CI and M NH4OAc). All the methods gave K indices which were significantly correlated with one another (r> 0.9). The extractants containing NH/were better correlated with one another than with the other extractants. Significant linear relationships were obtained with all the methods between extractable K and soil K-values determined using the NH4OAc method over a range of values from < 10 mg kg−1 to over 300 mg kg−1. Apart from the results of the Bray No. 1 and ISFEI methods, which exhibited lower extractability, all R2 values were greater than 0.9. It appears that any of the methods used for extracting soil P are suitable for assessing the K status of the soils of this region. © 1995 Taylor & Francis Group, LLC.