Restoring biodiversity in winter wheat

Key findings

  • In general, invertebrate abundance declined with distance from the crop edge into the crop; in standard fields, the abundance of skylark food invertebrates was 72% lower at 32 metres than at one metre from the crop edge.
  • Skylark food invertebrates were 31% lower at the crop edge in fields with a wild flower margin than in ones without. These results underline the importance of establishing conservation headlands or insect-rich cover along standard field margins.
  • Wildflower margins and skylark plots did not affect the abundance of grey partridge chick-food invertebrates, so these measures will not enhance grey partridge chick survival.
  • When no skylark plots were present, putting in a wild flower margin led to 23% more skylark food items (compared with standard fields). However, when skylark plots were present, putting in a wild flower margin led to 9% fewer skylark food items.

In 2007 we completed the final phase of the five-year Sustainable Arable Farming for an Improved Environment (SAFFIE) project, in which we collaborated with others to find ways of increasing biodiversity in winter wheat. In this experiment we tested two management options separately and in combination: skylark plots (now adopted as Entry Level Stewardship option: EF8) and six-metre-wide wild flower margins (EF4).

We knew that the wildflower margins supported high numbers of invertebrates but our main aim was to determine whether these spilled over into the adjacent crop and skylark plots, providing a boost in food resources for birds that prefer to forage in more open vegetation.

Figure 1: Mean skylark food items per sweep net (back transformed data) sampled at one metre, 16 metres and 32 metres from the crop edge

Mean skylark food items per sweep net (back transformed data) sampled at one metre, 16 metres and 32 metres from the crop edge

On each of 15 farms there were four treatments in conventional wheat fields:

  • a) no skylark plots;
  • b) skylark plots only;
  • c) wild flower margins only;
  • d) skylark plots and wild flower margins.

Over three years, we sampled the four fields on each farm in both first- and second-year wheat. Using pitfall traps, a suction sampler and sweep net, we collected invertebrates on transects at one metre, 16 metres and 32 metres from the crop edge. We also collected mid-field samples.

Here we present the results, averaged across year, for the grey partridge chick-food index (CFI)1, specifically calibrated for suction samples, and skylark food-items (SFI)2, sampled by both suction sampler and sweep net. The CFI measures invertebrate food in terms of grey partridge chick survival; 0.7 is considered a good score.

In general, there were more invertebrates at the crop edge than in the crop. In standard fields there were 71% fewer skylark food items (sampled by sweep net) at 32 metres than at one metre from the crop edge; in fields with wild flower margins there were 51% fewer at 32 metres than at one metre (see Figure 1).

Although not statistically significant, grey partridge invertebrate food and skylark food items (sampled by suction sampler) were also higher in abundance at the crop edge. There were 31% fewer skylark food items at the crop edge in fields with wild flower margins than at the edge of standard fields; we suggest that either the insects moved into the margins where there was more food or the insects were eaten by predators residing in the margins.

These results emphasise how, in standard fields, the crop edge is important for invertebrates and the birds that feed on them. For this reason we have long recommended selectively sprayed conservation headlands to provide some weed cover for invertebrates in standard fields.

Mid-field, there was an unexpected effect of combining wild flower margins with skylark plots. In fields without skylark plots, wild flower margins led to a 23% increase in SFI (see Figure 2). Conversely, in fields with skylark plots, the presence of wild flower margins led to a 9% decrease. It is possible that this was due to higher predation by birds, as skylark densities were between 1.3 and 2.8 times higher on fields with skylark plots and wild flower margins than in fields without these habitats. However, this pattern was not replicated for grey partridge chick food items or skylark food items sampled by sweep net.

Figure 2: Mean skylark food items per 0.5 square metre (back transformed data) in fields with and without skylark plots and wild flower margins

Mean skylark food items per 0.5 square metre (back transformed data) in fields with and without skylark plots and wild flower margins

1CFI =(0.00614*plant bugs & hoppers) + (0.0832*(leaf beetles&weevils) + 0.000368*aphids) + (0.1199*caterpillars) + (0.1411*ground & click beetles). Potts, GR, Aebischer, NJ (1995) Population-dynamics of the grey partridge Perdix perdix 1793-1993 – monitoring, modeling and management. Ibis, 137, s29-s37 Suppl.

2SFI = Sum of snails + plant bugs + hoppers + sawflies + beetles + weevils + flies. Calculated using evidence from skylark faecal samples collected during the SAFFIE project.

Overall, the abundance of invertebrates in wheat was low; the CFI was frequently lower than 0.5, which indicates poor partridge chick survival. There is some evidence that skylarks, which avoid the edges of fields, may have benefited from skylark plots by gaining access to food, but we have found no robust evidence to suggest that wild flower margins or skylark plots boosted overall abundance of invertebrate birdfood. Indeed, combining these options reduced the abundance of skylark food in the middle of fields.

Conventional herbicide regimes were in place, thereby limiting weed cover and the insects associated with weeds. The utility of measures such as skylark plots and wild flower margins can be maximised only if more selective herbicide regimes are introduced.