Subsoil testing required to detect the rundown of soil potassium to deficient levels for wheat production on loam-textured soils

Abstract

Context Long-term negative potassium (K) balances in crop production have depleted soil K levels in Western Australia (WA). Previous research has focussed on sand-textured soils, but recently, monitoring of crops grown on loam-textured soils has shown deficient or marginal shoot K concentrations where Colwell K 0–10 cm is above current critical levels. Aims The aims were to examine whether grain yield responses to fertiliser K can be detected on loam-textured soils and if soil test calibration curves can be identified for these soils. Methods Eight field trials were conducted with wheat on loam-textured soils. The same experimental design was used at all sites; six levels of K applied at sowing, from 0 to 200 kg K ha−1 with one treatment including a split application. Soil and plant test calibration curves were modelled using measurements from the trials. Key results Grain yield responses of 0.69 to 1.37 t ha−1 to fertiliser K (P < 0.05) occurred in 4 of 8 trials. Relative yield was closely related to soil exchangeable K and the goodness of fit of the soil test calibration curves increased as the depth of sampling increased. The best soil test calibration curve was for sampling 0–40 cm. Conclusions This research confirms that on some loam-textured soils, yield loss is occurring to K deficiency if no K fertiliser is applied. Implications As soil K reserves are run down, soil sampling at 0–40 cm on loam-textured soils will provide the most accurate monitoring of soil K deficiency for wheat production.