Amount and form of sulphur fertiliser required to prevent sulphur deficiency in spring oilseed rape

Abstract

Sulphur (S) requirements for preventing S deficiency in spring oilseed rape were tested at eleven sites over a three year period. In 1995 and 1996, seedbed dressings of 0, 10, 20, 40 and 80 kg/ha (S), applied as potassium sulphate, were compared with elemental sulphur in the seedbed and foliar sprays of Epsom salts, both applied at 20 and 40 kg/ha S, at three sites each season. Additional foliar spray treatments of magnesium chloride, applying the same amounts of magnesium as the two rates of Epsom salts, were also applied to test for any magnesium response. In both years, two sites were selected on sandy and shallow chalk soils with moderate to high risk of S deficiency, with a third site on heavy loam representing marginal deficiency risk. In 1997, the experiment design was simplified, to test only seedbed dressings of 0, 10, 20 and 40 kg/ha S applied as potassium sulphate, but with five sites instead on sandy or shallow chalk soils. The cultivar Starlight was used at all sites.

Significant (10% probability level) yield responses of 7-30% were obtained from S application at five sites on sandy or shallow chalk soils out of the eleven sites in total. Finely divided elemental S, incorporated as a seedbed dressing, was as effective as seedbed applied potassium sulphate at the two S-responsive sites in 1995 and 1996, when different forms of S were compared. Foliar spraying with Epsom salts at the 6-7 leaf stage was, however, a less effective method of applying S to prevent deficiency and, at three sites, reduced yield as a result of crop scorch or stress, particularly at the higher rate tested.

Yield responsive sites were mostly associated with total S concentrations below 0.4% (on a dry matter basis) in the leaves at flowering and extractable total S (Inductively Coupled Plasma Atomic Emission Spectrometry determination) contents in the soil shortly before drilling of less than 7 mg/kg in the top 30 cm layer or as an average over the 0-90 cm depth. These results suggest that the soil and leaf analysis guidelines previously developed for predicting or diagnosing S deficiency in winter oilseed rape are also appropriate for the spring sown crop.

An application rate of 20 kg/ha S was generally sufficient for preventing S deficiency in this series of experiments, but 30 kg/ha S may be needed for very deficient sites with otherwise good yield potential. Symptoms of S deficiency were only observed at two sites, of which one was the most yield responsive site in this experiment. At the second site, however, mineralisation of soil S later in the season and a moderate yield potential may largely account for the small, non-significant yield response (+5%) to S application.