Mass-flowering crops: cost or benefit to bumblebees and wild flower pollination?

Lead Research Organisation: Newcastle University
Department Name: Sch of Biology


There is a widespread perception that pollinators are in decline, and that this threatens both natural plant communities and agricultural productivity. Although this decline has recently been questioned, most authorities agree that some important pollinator groups, notably bumblebees, have declined in both abundance and range in Europe. There are also well documented declines in abundance of many plant species in the UK, particularly perennial herbs such as cowslips and red clover. The cultivation of mass-flowering crops (primarily oilseed rape and field beans) in the UK has been arguably the most dramatic change to the floral landscape for centuries. Recent work by the applicants shows that, when in bloom, the amount of pollen and nectar provided by these crops greatly exceeds that provided by all other insect-visited flowering plants combined in arable landscapes. We still have a poor understanding, however, of the impact that this brief glut of floral resources has on pollinator populations or on the reproduction of wildflowers. There is disagreement as to whether mass-flowering crops are of benefit to populations of bumblebees (they provide a lot of resources, but colony success requires a continuous supply of food throughout spring and summer, and some bee species avoid shallow flowers like oilseed rape). Mass-flowering crops are also likely to affect pollination and hence seed set in wildflowers of arable ecosystems. But it is unclear whether wildflowers will suffer adverse affects from competition with the crop for pollinators or through stigma-clogging with pollen from such crops, or conversely whether they will benefit from a boost to local pollinator populations. The answer will depend on the phenology and floral similarity of crop and wildflower. Wild flower population processes and their pollinators' population dynamics in farmland have never been considered together in a single project. This project aims to capitalise on previous BBSRC-funded work by the applicants, to assess how the amount and landscape-scale spatial distribution of mass-flowering crops affect pollinator populations (particularly bumblebees) and reproduction of wildflowers, using a three-pronged approach: A. Studies of the effects of oilseed rape and field beans on bumblebee colony survival, focussing on two contrasting bee species, the longer tongued B. pascuorum and the shorter-tongued B. lapidarius. Molecular approaches pioneered by the applicants will be used to detect nests through the season (Goulson@Stirling). B. Studies of the effects of these same mass-flowering crops on pollinator visitation and seed-set in five target wildflower species, chosen to provide a range of phenologies and flower morphologies (Osborne@Rothamsted) Ci) Expansion, development and testing of an existing GIS-based forage map and associated model to predict how the distribution and phenology of forage in arable landscapes determines pollinator abundance, and in particular which areas act as sinks for bumblebees, and which act as sources (net exporters of queens) (Sanderson@Newcastle) Cii) Use of the same forage map to develop new spatially-articulate models to examine the broader, landscape scale effects of mass-flowering crops on pollinator visitation to, and seed set in, wildflower populations In combination, this work will enable us to evaluate the likely ecological consequences of growing mass-flowering crops on pollinators and wildflowers, and will substantially increase our ability to predict and manipulate the impacts of agricultural practices on farmland biodiversity. The project is of direct relevance to policy makers producing schemes for arable and countryside stewardship, for example those advising on field margin management. Practices that encourage large bumblebee populations will in turn boost pollination of crops and wild plants, underpinning the sustainability of a diverse flora on unfarmed agricultural land.

Technical Summary

The project will focus on a 15 x 15 km sq. of arable farmland in Hertfordshire. Obj A: Determine whether mass-flowering crops (MFCs) affect the growth of colonies of long and short-tongued bumblebee species. We will perform genetic analysis on samples from bees caught on or in the margins of oilseed rape, field bean and cereals (30 sites). Samples of B. lapidarius and B. pascuorum queens will be taken in the spring, and workers in May and July. Molecular markers will be used to determine relatedness, and from this we will estimate how many colonies are foraging in an area, and how colony survival over time varies according to forage availability. Obj B: Determine whether MFCs have a competitive or facilitatory effect on pollination and seed-set of field-margin wildflowers. We will plant wild flowers in the margins used above and assess pollinator visitation, pollen deposition and consequent seed set. Five species of wild plant with different floral shapes and flowering times will be chosen. Thus they will be visited by different pollinators and we predict different levels of interaction with mass-flowering crops. It is currently not possible to predict whether these interactions will be positive or negative. Obj C: Development and testing of a GIS-based forage map and associated model to predict how the distribution and phenology of forage in arable landscapes determines which areas act as sinks or sources for bumblebees. Our existing 10 x 10 km forage map will be enlarged to 15 x 15 km, and resolution improved to 60 cm. We will use the existing map to identify training areas in the new imagery, prior to supervised classification. Predictions of the effects of changing forage availability through the season on bumblebee colony survival will be tested against observed survival (Obj A). Development of new spatially-articulate models based on the same GIS map to predict the effects of the landscape-scale distribution of MFCs on wildflower reproduction. (joint with BBE0009321 and BBE0014911; cofunded by NERC)
Description The three broad areas of the research can be summarised as follows:

A. Nest density and survival of the longer tongued B. pascuorum and the shorter-tongued B. lapidarius was measured using molecular approaches (pioneered by Stirling). These indicated the importance of gardens, and to a lesser extent grasslands, for both survival and density in a relatively intensively managed agricultural landscape.

B. Experimental studies at Rothamsted compared the seedset of two contrasting species of field margin flowers (ground ivy and birdsfoot trefoil), both pollinated by bees. These indicated that there was an increase in the rate of birdsfoot trefoil seedset when grown near oilseed rape, but only when the latter was flowering. Gardens were again indicated to have a positive impact, although the role of woodlands was more complex.

C. These studies were facilitated by the development of high resolution land cover maps of the study area by Newcastle. New methods were used to combine satellite data with existing OS maps in the classification process, to ensure that important but difficult to detect landscape features, especially field margins, were correctly and reliably identified. This will enable maps of impacts on potential forage density and nest suitability to be developed, which can be generalised to similar agricultural landscapes.

This research has indicated the impacts of large-scale use mass-flowering crops in agricultural landscapes on both bumblebee survival and their changing pollinator activity on the seedset of some species wildflowers. It is demonstrated that the scale of this effect (often substantially less than 10km) is such as to have conservation implications for bumblebees and the species they pollinate.
Exploitation Route We now have a better understanding of the small-scale importance of areas of semi-natural vegetation in intensively managed agricultural landscapes to maintain both suitable nesting and foraging areas for bumblebees. Whilst large areas of suitable forage may be temporarily available (e.g. oilseed rape) these do not provide continuity throughout the whole season, nor sufficient range of flower types to maintain an adequate diversity of bumblebee species. Our findings could be taken forward by encouraging farmers to maintain areas of wildflowers (analogous to the approach used for beetle banks) to encourage bumblebee diversity.
Sectors Agriculture, Food and Drink,Environment,Other
Description This research provided a fuller understanding of the interaction between pollinators in the wider agricultural landscape, and land cover type, particularly in relation to mass-flowering crops. It has demonstrated the importance of suburban gardens in maintaining a higher diversity of bumblebees near the edges of intensively managed agricultural systems, and provided a better understanding of the spatial scale over which bees forage. Its findings have fed into to the wider debate about the decline of native pollinators and agricultural pesticide usage.
First Year Of Impact 2011
Sector Agriculture, Food and Drink,Environment,Other
Impact Types Cultural,Societal
Description Leverhulme Research Grant
Amount £92,582 (GBP)
Funding ID F/00 125/AI 
Organisation The Leverhulme Trust 
Sector Academic/University
Country United Kingdom of Great Britain & Northern Ireland (UK)
Start 03/2010 
End 02/2012