Leave arable field margins uncropped with natural regeneration

Nine bee species were recorded on a single naturally regenerated field margin strip established for three years at ADAS Bridgets Research Centre, Hampshire, England in 1998 (Carreck et al. 1999), the same number of species as on three strips sown with a diverse wildflower seed mix in the same study.

A small replicated, controlled trial of field margin management options on two farms in North Yorkshire, England in one summer (Meek et al. 2002) did not find significantly more bumblebees Bombus spp. (species or individuals) on four naturally regenerated 6 m wide margins than on four cropped margins.

A replicated controlled trial of UK arable field margins allowed to regenerate naturally for one year found that they supported significantly more honey bees and bumblebees than unsprayed cropped margins managed as conservation headlands (averages between 10 and 50 bees/transect on naturally regenerated margins compared to <3 bees/transect in conservation headlands; Kells et al. 2001).

In a replicated controlled trial on three arable field margins at one farm in North Yorkshire, Carvell et al. (2004) found 6 m wide naturally regenerated, uncultivated field margin plots supported significantly more foraging bumblebees than margins sown with tussocky grass, or control cropped field margins, but only in one year (2001) of this three year study. In the other two years (2000 and 2002), the naturally regenerated field margins did not support significantly more bumblebees than the control or grass-sown sites. In 2001, the bumblebees were mostly foraging on spear thistle Cirsium vulgare, a pernicious agricultural weed that had to be controlled by cutting at the end of that summer. Naturally regenerated margins were the only treatment that did not support consistent numbers of bumblebees in all three years.

Bumblebee foraging activity and species richness were significantly enhanced on 18 uncropped, regularly cultivated field margins where natural regeneration had been allowed to take place for five years, compared to paired control sites of conventionally managed cereal, in East Anglia and the West Midlands, England (Pywell et al. 2005). The uncropped margins had significantly more plant species than either conservation headlands or uncropped margins sown with a wildflower seed mix. However, two species considered to be pernicious weeds, spear thistle and creeping thistle C. arvense were key forage plants for the bumblebees, so this option may lead to conflict between agricultural and conservation objectives.

In a replicated controlled trial at six sites across central and eastern England, Carvell et al. (2007) found that naturally regenerated field margins supported a greater number and diversity of foraging bumblebees than cropped margins (including conservation headlands), but only in the first year of the study. In subsequent second and third years, bumblebee numbers were not significantly different from cropped treatments, but this may be due to the presence of more attractive floral resources planted on the same field margins for the experiment.