Towards 'crop-pollinating' landscapes: quantifying pollen supply and demand to manage wild pollinators for their benefits to food production

Three quarters of the world's important crops, or 35% of all food produced by volume, depends on animal pollinators. This includes many fruits, vegetables, nuts and oils. Recently, scientists have found that most crop pollination in the world is carried out by a small group of bee species, called the 'dominant crop pollinators'. In the UK, no more than 20 species are doing most of the work. One is the honey bee, kept by beekeepers. The others are wild species found commonly in farmland.

Yet numbers of wild bee species found in each place are declining. One of the main reasons for this decline is loss of wild flowers in the countryside, which was dramatic in the twentieth century in the UK. In particular, the loss of flowering plants in the pea and bean family, which have protein-rich pollen, is thought to have caused trouble for wild bee species. Bees rely entirely on food from flowers - nectar and pollen - to survive and raise their young. Nectar is a sugar syrup, and provides them energy to fly and grow. Pollen provides most of the protein, a crucial resource for growing larvae.

One way to help pollinators is by planting extra flowers. This has been shown to increase the numbers of dominant crop pollinators, and can even lead to improvements in crop yield. At the moment scientists don't know how much, or what types of pollen wild bees need to survive at different times of year, and have very little idea how much pollen is already supplied by existing landscapes. Some think there are particular times of year, such as early spring or late summer, when pollen is especially limiting, and adding flowers would really help bees.

This research uses very new scientific techniques to quantify the demand for and supply of pollen in farmland. One exciting development is that scientists have just made a library of DNA sequences from all 1,479 of UK native plants. This makes it possible to identify plant species in bee-collected pollen, by putting the pollen into a sequencing machine. It's called DNA metabarcoding. Early results for honey bees show that pollen from trees could be more important at the start of the year than was previously thought. This research will document all the plants that red-tailed and early bumblebee colonies use over at least two years.

A series of experiments with bumblebees in the laboratory will test how much and what kind of pollen are needed to raise individual larvae, or supply a healthy colony. Combining this information with the pollen sources they choose in real landscapes will allow me to accurately calculate pollen demand by these crop-pollinating bumblebees.

To quantify pollen supply in an area of farmland, I will use image processing techniques borrowed from cell biology to measure how many flowers of each type there are through spring, summer and autumn, using photographs. The idea is to build an easy-to-use piece of software to tell farmers exactly what flowers to plant so their farm supports a thriving wild pollinator community that provides reliable pollination to their crops. The software will compare pollen supply with pollen demand on a monthly basis.

Being able to measure pollen supply and demand in an area of farmland will allow scientists to answer a fundamental question central to ongoing debates about pollinators: What currently limits wild pollinator populations? Is available protein (pollen) the main limiting factor, or is it carbohydrate (nectar), nesting sites, overwinter mortality or negative impacts of pesticides or disease? The overall aim of my research is to test the hypothesis that pollen is the limiting factor in agricultural landscapes. If correct, then the best thing we can do for pollinators is plant flowers. If it is wrong, and something else is limiting pollinator numbers or causing declines, then different strategies will be necessary to retain viable communities of hard-working wild pollinators that support food production.

The main beneficiaries are as follows:
*National governments or government agencies responsible for pollinator conservation (early beneficiaries)
Natural England, the UK Government's nature conservation agency for England, will benefit from this research because it will enable them to monitor and potentially enhance the effectiveness of their recently launched Mid-Tier agri-environment package: the Wild Pollinator and Farm Wildlife Package. The Package focuses largely on flower provision for pollinators, but does not currently take landscape context or existing flower provision into account. The data on pollen dependencies of wild bumblebees (Objectives 1 and 2) and the flower demand model (Objective 4) can be used to improve advice supporting the scheme, and the design of the scheme itself. The automated flower density protocol and flower demand model (Objectives 3 and 4) provide a mechanism for targeting the scheme at farm level. The scientific question of what currently limits wild pollinator populations in farmed landscapes is hugely pertinent to this scheme and other conservation policy around pollinators.
Other national Governments with Pollinator Strategies, including the Welsh and Irish Governments, and the United States Agriculture Department could also benefit from this approach when designing policy interventions for pollinators.
*Farm advisers and agronomists (early beneficiaries)
Farm advisers and agronomists provide advice directly to farmers, including advice on implementing agri-environment schemes and managing natural resources. The decision support tool that advises how to rectify flower deficiencies for core crop pollinators will help them deliver high quality advice. Hutchinsons Ltd, a leading agronomy company in the East of England, has already perceived a market for such a product (see Pathways to Impact). Demand for pollination in agriculture is growing globally, and there is increasing awareness of the need for pollination services as an input. I expect agronomists around the world to increasingly seek information on how to manage wild pollinators for their services.
*Farmers (medium term beneficiaires)
Farmers who take up the Wild Pollinator Package, or plant flowers without incentive, will be able to avoid wasting money planting in areas, or at times of year when existing flower provision is adequate. Growers of insect pollinated crops can expect to benefit from more effective action to manage dominant crop pollinators, increasing the likelihood that their actions will deliver an economic return through enhanced yield or quality of their produce.
*Larger agri-food businesses (medium term beneficiaries)
Supermarkets such as Waitrose and Sainsbury's, and suppliers such as Worldwide Fruit Ltd, have strong interests in protecting pollinators as part of their sustainability strategies. Such companies need simple, straightforward advice on how to manage pollinators for food production. By identifying likely pollen shortfall periods during the year, and establishing whether pollen supply is limiting wild pollinators in agricultural landscapes, this project will provide crucial evidence to support this advice.
*Conservation NGOs (medium term beneficiaries)
Organisations such as the Bumblebee Conservation Trust, and Buglife (and their equivalents around the world) aim to support wild pollinators in all landscapes, and prevent further declines or extinctions. The datasets on pollen resources needed by common bumblebees, and identified likely shortfall periods during the year, will be invaluable to them in shaping their conservation strategies.
*The public (long term beneficiaries)
There is strong public concern about pollinator decline, and many members of the public are motivated to take action. Planting flowers, in the town or the countryside, is a popular response. Outputs of this research will improve the effectiveness of this action, by allowing targeting, and selection of appropriate species.