Wheat grain-yield response to lime application: relationships with soil pH and aluminium in Western Australia

Abstract

Soil acidity, or more specifically aluminium (Al) toxicity, is a major soil limitation to growing wheat (Triticum aestivum L.) in the south of Western Australia (SWA). Application of calcium carbonate (lime) is used to correct Al toxicity by increasing soil pH and decreasing soluble soil Al3+. Soil testing using a 0.01 m calcium chloride (CaCl2) solution can measure both soil pH (pHCaCl2) and soil Al (AlCaCl2) for recommending rates of lime application. This study aimed to determine which combination of soil pHCaCl2 or soil AlCaCl2 and sampling depth best explains the wheat grain-yield increase (response) when lime is applied. A database of 31 historical lime experiments was compiled with wheat as the indicator crop. Wheat response to lime application was presented as relative yield percentage (grain yield for the no-lime treatment divided by the highest grain yield achieved for lime treatments × 100). Soil sampling depths were 0–10, 10–20 and 20–30 cm and various combinations of these depths. For evidence that lime application had altered soil pHCaCl2, we selected the change in the lowest pHCaCl2 value of the three soil layers to a depth of 30 cm as a result of the highest lime application (ΔpHmin). When ΔpHmin <0.3, the lack of grain-yield response to lime suggested that insufficient lime had leached into the 10–30 cm soil layer to remove the soil Al limitation for these observations. Also, under high fallow-season rainfall (228 and 320 mm) and low growing-season rainfall (GSR) (<140 mm), relative yield was lower for the measured level of soil AlCaCl2 than in the other observations. Hence, after excluding observations with ΔpHmin <0.3 or GSR <140 mm (n = 19), soil AlCaCl2 provided a better definition of the relationship between soil test and wheat response (r2 range 0.48–0.74) than did soil pHCaCl2 (highest r2 0.38). The critical value (defined at relative yield = 90%) ranged from 2.5 mg Al kg–1 (for soil Al calculated according to root distribution by depth within the 0–30 cm layer) to 4.5 mg Al kg–1 (calculated from the highest AlCaCl2 value from the three soil layers to 30 cm depth). We conclude that 0.01 m CaCl2 extractable Al in the 0–30 cm layer will give the more accurate definition of the relationship between soil test and wheat response in SWA.