Pasture yield response to different sulphur fertiliser strategies and its application to modelling

Abstract

A database of sulphur (S) fertiliser trials was analysed to determine the effect of different fertiliser application times, frequencies and forms on S response. Yields were about 6% higher when S was applied in spring than autumn on highleaching sites (free-draining, high-permeable soils with high rainfall), and about 4% higher when S was applied in autumn on South Island soils. Pasture yield was about 6% higher when S was applied as a split application than as a single annual application, with the advantage to split applications also tending to be higher on high-leaching soils. The application of S had a small residual affect on S soil test after one year, the size of the residual being dependent on fertiliser rate and season of application. In contrast, a once-only application of S could increase yield above that in the control treatment for at least 3 years after S was applied such that the response in the second and third years was about 0.5 and 0.25 times that in the first year. The residual effect on yield, the effect of applying different fertiliser rates, and previous fertiliser history have been incorporated into decision support software for calculating fertiliser S responses by calculating an effective fertiliser rate using an exponential weighted moving average. However, a standard response curve for determining the relationship between relative yield and effective fertiliser rate had to be used, as a site specific curve could not be predicted from data in the S database. Keywords: fertiliser, model, pasture yield, sulphur