Monitoring winter barley, wheat, oilseed rape and spring barley for sulphur in England and Wales to predict fertiliser need


Cereals & Oilseeds
Project code:
01 September 2001 - 31 March 2005
AHDB Cereals & Oilseeds.
AHDB sector cost:
£41,376 from the Home-Grown Cereals Authority (Project 2612)
Total project value:
Project leader:
M F F Carver The Arable Group, Manor Farm, Daglingworth Cirencester, Gloucestershire, GL7 7AH



About this project


The dramatic reductions in sulphur deposition that have been achieved as a result of reduced sulphur emissions in the UK since the 1970s have actually resulted in increases in sulphur deficiencies in arable crops.

Methods of predicting sulphur deficiency in crops have existed for many years but it is now even more important that they are accurate. This project therefore addressed two aspects of the problem of increased sulphur deficiency in crops.

  • The extent of deficiencies across soil types and geographic regions of England and Wales.
  • The accuracy of techniques used to identify sulphur deficiency in crops in identifying beneficial yield responses resulting from sulphur applications.

A total of 183 trials were conducted on four crops, winter wheat, winter barley, winter oilseed rape and spring barley during 2002, 2003 and 2004 at locations stretching from Aberdeen to Trerulefoot in south Cornwall.

Replicated trials, grown at nitrogen input levels suitable for the specific locations, were treated with early season sulphur applications (22kg/ha S for cereals and 44 kg/ha S for oilseed rape) or left without a sulphur treatment.

The results confirmed that significant yield responses to sulphur applications (indicating sulphur-deficient crops) were more frequently recorded on lighter soils rather than medium or heavy soils. Significant yield responses to applications of sulphur were recorded on 16 of the 68 trials conducted on light soils whereas the 115 trials conducted on medium and heavier soils produced 11 significant yield responses to sulphur application.

It has also been previously reported that brassica crops are more likely to be sulphur deficient than cereal crops and this was again confirmed in this project.

Significant yield responses to applications of sulphur were recorded on 32% of the winter oilseed rape trials, 15% of the spring barley trials, 13% of the winter barley trials and 10% of the winter wheat trials. The highest recorded response was 81% yield increase (1.90 t/ha) in a crop of winter oilseed rape grown on light soil at Cirencester. However, in absolute yield terms the highest response was from a crop of winter wheat grown on light soil at Andover which produced a yield response to sulphur application of +2.92 t/ha (+38%).

The 183 trials reported in this project can be viewed as 183 field crop decisions on the potential value of sulphur. Was it possible to predict accurately the 27 trials out of the 183 conducted that produced significant yield responses to sulphur applications?

The malate test, which measures the malate:sulphate ratio in leaf tissue, produced a more successful indication of the likelihood of a significant yield response to a sulphur application than the soil Sppm test. However, the malate results clearly indicate that it overestimated the number of likely significant responses possibly by regularly detecting transient sulphur deficiency. The malate test suggested that 132 of the 183 trials would produce significant yield responses to sulphur applications but it only successfully identified 25 and actually did not predict two significant yield responses. In contrast, the soil test predicted 35 responses and successfully identified 11 of the 27 significant yield responses.

The lack of an accurate predictive test for sulphur deficiency will lead to the continuation of sulphur application in both situations where they are not needed and, in some cases, the lack of applications in situations where they are needed.

It is recommended that a review of the threshold levels used in the malate test is undertaken to determine if the predictive accuracy of the test can be increased.