An improved procedure for routine determination of reserve-K in pastoral soils

Non-exchangeable K (reserve K) can be an important source of K for plant uptake in micaceous soils but testing for reserve K is laborious and expensive and involves use of potentially hazardous chemicals. Our objectives were to (1) identify a cheaper and more reliable reserve K test that is amenable for routine use by commercial laboratories, and (2) quantify the response of grass-clover pasture to applied K in soils differing in reserve K. We measured the release of reserve K using different combinations of the K-precipitant, sodium tetraphenyl boron (0.033 to 0.2 M), and extraction time (1 to 16 h). Copper chloride was used to decompose the K-tetraphenyl boron precipitate rather than the more toxic mercuric chloride or acetone. Based on data obtained for soils with a wide range of reserve K values, a test involving extraction of soil (1-g) in 0.1 M Na-tetraphenyl boron for 4 h was selected as the best option for routine use because it generally gave similar values to the current test (thus the same diagnostic criteria may be applicable) and it provided a better fit with operational requirements of commercial laboratories. Pasture K responses, in the presence and absence of applied N, were measured over a single growing season (pasture harvested at intervals of 1-2 months, depending on growth rate) at seven sites in Canterbury, New Zealand with reserve K (measured using the current test) ranging from 0.5 to 2.3 cmol(c)/kg (all sites were low in exchangeable K, < 0.33 cmol(c)/kg). The size of the K response depended on the clover content of the pasture and only one site with a high clover content responded consistently to K throughout the season. Clover content increased at all sites following K addition, confirming that a major benefit of pasture K fertilization is in improving clover production. Our results suggest that there may not be a unique critical threshold for reserve K that can be applied to all pastures; rather, the value below which a production response to K is obtained may depend on the amount of clover in the pasture.