Soil and tissue tests to predict pasture yield responses to applications of potassium fertiliser in high-rainfall areas of south-western Australia

Abstract

Dairy and beef pastures in the high (>800 mm annual average) rainfall areas of south-western Australia, based on subterranean clover (Trifolium subterraneum) and annual ryegrass (Lolium rigidum), grow on acidic to neutral deep (>40 cm) sands, up to 40 cm sand over loam or clay, or where loam or clay occur at the surface. Potassium deficiency is common, particularly for the sandy soils, requiring regular applications of fertiliser potassium for profitable pasture production. A large study was undertaken to assess 6 soil-test procedures, and tissue testing of dried herbage, as predictors of when fertiliser potassium was required for these pastures. The 100 field experiments, each conducted for 1 year, measured dried-herbage production separately for clover and ryegrass in response to applied fertiliser potassium (potassium chloride). Significant (P<0.05) increases in yield to applied potassium (yield response) were obtained in 42 experiments for clover and 6 experiments for ryegrass, indicating that grass roots were more able to access potassium from the soil than clover roots. When percentage of the maximum (relative) yield was related to soil-test potassium values for the top 10 cm of soil, the best relationships were obtained for the exchangeable (1 mol/L NH4Cl) and Colwell (0.5 mol/L NaHCO3-extracted) soil-test procedures for potassium. Both procedures accounted for about 42% of the variation for clover, 15% for ryegrass, and 32% for clover + grass. The Colwell procedure for the top 10 cm of soil is now the standard soil-test method for potassium used in Western Australia. No increases in clover yields to applied potassium were obtained for Colwell potassium at >100 mg/kg soil. There was always a clover-yield increase to applied potassium for Colwell potassium at <30 mg/kg soil. Corresponding potassium concentrations for ryegrass were >50 and <30 mg/kg soil. At potassium concentrations 30–100 mg/kg soil for clover and 30–50 mg/kg soil for ryegrass, the Colwell procedure did not reliably predict yield response, because from nil to large yield responses to applied potassium occurred. The Colwell procedure appears to extract the most labile potassium in the soil, including soluble potassium in soil solution and potassium balancing negative charge sites on soil constituents. In some soils, Colwell potassium was low indicating deficiency, yet plant roots may have accessed potassum deeper in the soil profile. Where the Colwell procedure does not reliably predict soil potassium status, tissue testing may help. The relationship between relative yield and tissue-test potassium varied markedly for different harvests in each year of the experiments, and for different experiments. For clover, the concentration of potassium in dried herbage that was related to 90% of the maximum, potassium non-limiting yield (critical potassium) was at the concentration of about 15 g/kg dried herbage for plants up to 8 weeks old, and at <10 g/kg dried herbage for plants older than 10–12 weeks. For ryegrass, there were insufficient data to provide reliable estimates of critical potassium.