Improved phosphorus and potassium fertiliser management: redefining the soil test and lupin response relationships in WA

When phosphorus requirement experiments in lupins were started in the early 1970s, most soils were highly deficient in phosphorus and consequently lupin yields were highly correlated (r2=0.72) with the Colwell soil phosphorus test, with a critical value defined as 20 mg P/kg soil. However, as phosphorus was applied to the soil the phosphorus status improved and this has, in turn, altered the ability of the Colwell soil phosphorus test to predict phosphorus response. Over time, various measurements of soil ability to adsorb phosphorus have been developed. The phosphorus retention index (PRI) was introduced during the early 1990s while the phosphorus buffer index (PBI) was introduced during the early 2000s (Bolland et al., 2006). Incursion of an indicator of phosphorus sorption into the phosphorus calibration curves can improve the ability of the phosphorus soil test procedure to predict phosphorus requirements of pastures (Bolland and Russell, 2010). In general, lupins are grown on sands that have a relatively low capacity to adsorb phosphorus. Within the available data base of phosphorus experiments done with lupins, PRI levels can be grouped into PRI=1, PRI=2 and PRI=3. These different PRI levels are known to have an impact on the shape of the phosphorus calibration curve for pastures (Bolland et al., 2010). The aim of this paper was to examine whether inclusion of PRI levels in the phosphorus soil test procedure could improve the ability of the Colwell soil phosphorus test to predict phosphorus deficiency in lupins. While WA soils initially contained adequate indigenous soil potassium for cropping, removal of potassium over time in harvested grain gradually resulted in some soils becoming potassium deficient for grain production. While studies have been done to determine the potassium requirements of lupins grown on grey sands, a calibration for the relationship between relative yield and Colwell soil potassium test has not been reported. An additional aim of this paper was to examine the ability of the Colwell soil potassium test to predict potassium deficiency in lupin grown on grey sands