A comparison of 3 soil-testing procedures for estimating the plant availability of phosphorus in soils receiving either superphosphate or phosphate rock

Abstract

The ability of 3 soil-testing procedures to predict the uptake by perennial ryegrass (Lolium perenne) of phosphorus (P) applied as either single superphosphate or phosphate rock (PR) was evaluated in a glasshouse study over 298 days and in the field over 3 years. In' the glasshouse, water extraction provided an accurate estimate of plant-available P in soils to which superphosphate was added. In contrast, water extraction grossly underestimated the initial effectiveness of both Sechura phosphate rock (SPR) and Chatham Rise phosphorite (CRP). The better predictive ability of the Bray test with soils receiving SPR and CRP appears to result from the ability of this procedure to extract both unreacted PR which is likely to dissolve in the short term, and sorbed P. In the field, the bicarbonate procedure was as effective as the Bray in predicting the residual effectiveness ofCRP in the third year, particularly at Ballantrae and Wanganui, where this P source exhibited a marked residual effect. This contrasts with the poorer initial performance of the Olsen procedure in the first year. In situations where unreacted PR contributed significantly to plant-available P in the soil, for example where a PR material is applied annually, the Bray procedure appears to be the most effective soil-testing procedure. © 1984 Taylor & Francis Group, LLC.