Numbers of bumblebees and bumblebee species enhanced in field margins planted with wildflowers in central and eastern England
Pywell R.F. et al. (2005) Providing foraging resources for bumblebees in intensively farmed landscapes. Biological Conservation, 121, 479-494.
Background
Bumblebees Bombus species across the Europe and North America have undergone considerable declines in both their geographical distribution and abundance. Post-war (1945 onwards) intensification of farming, resulting in the loss of vital foraging and nesting habitats, is thought to be at the root of this decline. Within both semi-natural and agricultural landscapes, bumblebees provide a vital pollination service, directly affecting the composition and survival of wild plant communities, as well as the yields of many key agricultural crops. Reversal of these trends, through maintenance and restoration of appropriate foraging, nest and hibernation sites therefore represents a key priority for both farming and biodiversity conservation in the UK and elsewhere.
The study summarised here was developed to inform the future direction of the Arable Stewardship Pilot Scheme (ASPS), established by Ministry of Agriculture Fisheries and Food (now DEFRA) during 1998. It was borne out of the need to develop practical, cost-effective prescriptions for integrating biodiversity conservation into the agricultural framework. The findings of this survey represent an important step in taking informed decisions towards this goal. They are of potential interest to individual farmers, land managers, rural advisers and conservation organisations.
Action
Research was undertaken within two geographically distinct regions in lowland England selected for their contrasting agri-environmental characteristics (Table 1, attached). The study examines the effects of four contrasting management techniques for arable field margins on: i) vegetation species richness, ii) bumblebee forage resources (flower abundance and species richness) and iii) the numbers, activity and species richness of foraging bumblebees.
Thirty-six farms, each having agreed and implemented the range of Arable Stewardship Pilot Scheme (ASPS) field margin regimes during 1999, provided a total of 120 sample field margins. On each farm, margins were selected using proportionate random sampling, where the number of margins surveyed was in proportion to the number of margins of that type found on the farm. The four field margin types are outlined below:
i) Conservation headland with no fertilisers (Option 3b); 16 sites surveyed.
ii) Naturally regenerated field margins (Option 4c); 18 sites surveyed.
iii) Field margins sown with a wildlife seed mixture (Option 5); 28 sites surveyed.
iv) Control - wherever possible, to act as a control treatment, each site was paired with a conventionally managed cereal field margin with similar aspect and boundary type. 97% of sites were paired in this way; 58 sites surveyed.
Surveys took place during July and August of 2003, 4 years after implementation of ASPS management regimes.
Bumblebee counts: Counts were made along linear transects of 6 m wide by 100 m long, sited along the centre line of each of the established options. For each observation of a foraging bumblebee, the relevant plant species was identified and noted. Each 100 m sampling zone was randomly located, except when randomisation provided a sample within 10 m of a margin�s end. These were re-sited by further random location. Where management options established field blocks of less than 100 m length, two parallel 6 x 50 m transects were surveyed.
Transects were walked between 10.00 and 17.00 hours, and surveys carried out only when weather accorded with Butterfly Monitoring Scheme rules (essentially dry and low windspeeds). Ambient temperature, percentage sunshine and wind speed were recorded at the end of each transect.
Vegetation sampling: Vegatation samples were taken from within the same sampling zone as bumblebee counts, using 20 randomly positioned 0.5 x 0.5 m quadrats. Within each, all vascular plant species were recorded on a presence/absence basis, leading to a frequency out of 20 for each species. Percentage frequencies were then used for analysis.
Floral surveys: Plant surveys of each transect followed immediately after each bumblebee count, identifying all flowering dicotyledons (dicots). This was used as an indicator of the foraging resource. A 5-level floristic index was used to score the approximate flower abundance for each species across the entire transect. Abundance scores for individual species were later combined into 18 �flower groups� for analysis.
Consequences
Regional comparisons: Comparison between the two pilot areas, regardless of treatment, found significant differences only within: i) abundance of monocot species, ii) number of flowering dicot species, and iii) total abundance of dicot flowers. In all these cases, the West Midlands pilot area had significantly greater levels. This did not translate, however, into increases in either the abundance or richness of bumblebees, where regional comparison revealed no significant differences. No other intra-regional differences were observed.
Field Margin options: Comparison between the three different field margin management options and their equivalent controls revealed the following trends:
a) Conservation Headlands (Option 3b Sites)
Conservation Headlands were found to contain more plant species per 100 m than the conventional field margins, largely represented by significantly greater numbers of dicot species, most of which were annuals. However, these differences did not translate into significant differences in bumblebee abundance, whether comparisons were made within individual species, across functional groupings, or via overall species richness.
b) Natural Regeneration (Option 4c Sites)
Field margins where natural regeneration had been allowed were found to contain four times the number of plant species than their conventionally managed counterparts, again (as with conservation headlands) attributable to marked increases in dicot species. Annual and perennial species shared this space in roughly equal proportion. Within natural regeneration margins, flowering species were found to be both significantly more abundant and four times more diverse than control sites. With a single exception, all bumblebee species were found to be significantly more abundant on naturally regenerated field margins. Total bumblebee activity and bumblebee species richness were both significantly greater than the control sites.
c) Wildlife Seed Mix Margins (Option 5 Sites)
The second non-crop option - sown with a wildlife seed mixture - contained significantly greater monocot, dicot and perennial species than their control sites, with over double the total number of species. Whilst annuals remained the same between treatments, flowering dicots were both more abundant and diverse. Significantly greater bumblebee species richness and abundance was recorded within these habitats. Option 5 sites represented the most successful of all the treatments in terms of bumblebee foraging activity.
Bumblebee foraging preferences: Non-crop field margins (options 4c and 5) had the greatest number of preferred bumblebee forage plants associated with them. This resulted in significant differences in both bumblebee abundance and species richness when compared with either control sites or conservation headlands (option 3b). In both of the latter sites, a few species of plant contributed to the vast majority of foraging visits by bumblebees, mainly represented by field pansy Viola arvensis, creeping thistle Cirsium arvense and spear thistle C.vulgare.
Importantly, whilst naturally regenerated field margins contained significantly more annuals, dicot and total plant species than the other management options, it was the wildlife seed mixture which contained the greatest abundance and richness of bumblebee species. This is explained by consideration of bumblebee foraging preferences. Different bee species showed marked preferences for individual plant species, and option 5 sites provided the greatest range of forage species. Nectar rich perennial and biennial plant species found only on these sites accounted for 28% of foraging visits. Data analysis particularly highlighted the positive effects of legumes from both the red/purple- and yellow/white-flowered groups on bumblebee abundance across most bee species.
Conclusions: Of the four treatments, both of the non-crop options resulted in marked increases in bee abundance and richness. However, natural regeneration within these margins tended to bring about increases in bee numbers and diversity through increases in those species also considered to be agricultural weeds (creeping and spear thistles). Conflicts therefore clearly exist between agricutural and conservation interests.
Wildlife seed mixtures represent a potentially more integrated means through which to provide bumblebee foraging resources, with beneficial results having been observed in both bee abundance and species richness across the survey area. However, this result is dependent upon key forage species being included in the seed mixture. Such species include: red clover Trifolium pratense, bird�s-foot trefoil Lotus corniculatus and borage Borago officinalis.
Note: If using or referring to this published study, please read and quote the original paper. Please do not quote as a conservationevidence.com case as this is for previously unpublished work only.