Effect of pasture composition on lime and phosphorus responses on a dryland soil

Abstract

A field trial was conducted on a yellow-grey earth in the Hawke's Bay (mean annual rainfall 875 mm) over a period of 5 years to measure the effects of lime applications (0, 5, 10 t/ha) on soils high (50 kg P/ha applied annually) or low (5 kg P/ha applied annually) in phosphorus (P). The average annual pasture production on this dryland soil in the absence of applied lime or P was 5060 kg DM/ha (range 3861-6024). The botanical composition of the pasture was variable, average annual legume composition (4/o on DM basis) ranging from 3 to 42%. The predominant. legume was subterranean clover (Trifolium subterranean) with white clover (Trifolium repens) making a small contribution in some years. In the first two years after application responses to lime were large (lo-20%), due entirely to liming stimulating the grass component of the pasture, and consistent with liming enhancing the rate of net mineralisation of soil organic nitrogen (N). In years three and-four-the-dominant treatment effect was to P applications (130% increase in total pasture production), which stimulated legume growth initially, but subsequently increased grass growth. In the fifth year no significant responses to lime or P were observed. The seasonal pattern of pasture responses to either lime or P was related to pasture botanical composition. The size of the annual total DM pasture responses to P was positively related to the legume content of the pasture. Conversely the lime responses were negatively related to pasture legume content. These results suggest that pasture botanical composition could be used to improve the ability to predict lime and P responses based on soil nutrient levels. This may be important where pasture botanical composition is extremely variable over time. Keywords fertiliser, lime, pasture, pasture composition, phosphorus, soil acidity