Residual value of superphosphate and Queensland rock phosphate for serradella and clover on very sandy soils as assessed by plant growth and bicarbonate-soluble phosphorus

Abstract

Superphosphate (0-0.4 t/ha P) and Queensland rock phosphate (0-20 t/ha P) were incorporated into the top 10 cm of very sandy soil near Esperance, W.A. The effectiveness of the fertilisers for pastures was calculated from dry herbage yields of yellow serradella, slender serradella and subterranean clover. Soil samples were collected just after fertiliser application and at intervals up to 2.5 years for measurement of bicarbonate-extractable phosphorus (soil test), which was related to plant yield. Results for all 3 species were very similar in response to superphosphate. The effectiveness of superphosphate decreased by about 50% between years 1 and 2, and by a further 25% between years 2 and 3. The effectiveness of Queensland rock phosphate was about 7% that offreshly applied superphosphate in the year of application for all 3 species and for the next 2 years, its effectiveness relative to freshly applied superphosphate remained about constant for yellow serradella and approximately doubled for slender serradella and clover. For superphosphate 2.5 years after application, the amount of phosphorus extracted by the soil test compared with freshly applied superphosphate decreased by about 80%. The soil test extracted a very small proportion of the phosphorus applied as Queensland rock phosphate and the amount extracted decreased by about 50% during 2.5 years. Although the plants responded strongly to increasing levels of applied phosphorus, soil test values remained low until yields of about half the maximum yield were attained. Separate calibration curves were required for each fertiliser and species and each calibration curve was best described by 2 component linear spline functions.