Pollination modelling in complex landscapes

In this project we are developing a theoretical model for how bees (bumblebees, solitary bees, honeybees) use the landscape. This will help us gain a mechanistic understanding of what landscapes features are good or bad for the bees, and where they can be expected to pollinate. This is important to understand as many species of bees are in decline and many of them are important for the pollination of crops. Thus, with an enhanced understanding of how bees use resources in their environment we will improve our abilities to propose conservation actions for bees, as well as for example agri-environment schemes to enhance pollination from wild bees.

Bees, especially some bumblebees and honeybees, forage over large scales (up to several kilometers). They are central place foragers, as they need to return to the nest with the pollen and nectar, and this restricts their foraging in the landscape. It also means that it should be fruitful to apply central place foraging theory, when modelling bees’ habitat use.

The model we currently develop is behaviorally based, and we are using the concepts from central place foraging theory to develop a modeling framework that can help us estimate visitation rates and pollination service across spatially explicit landscapes. The behavioral strategies in the model are based on our recent paper (Olsson & Bolin 2014). These strategies are applied to rasterized maps of landscapes, where different habitats have floral resource values and nest site values, in ways similar to those applied in a previous model by Lonsdorf and colleagues (2009).

Olsson & Bolin 2014 – springer.com 

Lonsdorf and colleagues 2009 – academic.oup.com

According to our theory, the foraging strategy for central place foragers is determined by two properties of a foraging patch: the distance to the nest and the quality (in terms of how quickly resources can be harvested from it). A patch in the lower right corner has the highest marginal fitness value to the forager (indicated by the red color). The lower the quality or the farther away from the nest is the lower is its value. Only patches below the white curve should be used. A patch of infinite quality would be used if the distance to it was less than τ*.

Pollinating insects also provide an important ecosystem service, they increases yields in for instance oil seed rape, strawberries and apples. Therefore, there is an economic incentive to further develop agri-environmental schemes such as hedgerows and semi-natural pastures.

In summary, the aim of this project is to:

• Estimate pollination services over large spatial scales
• Understand how different agri-environment schemes will affect diversity of wild bees
• Provide recommendations of cost-efficient schemes to stakeholders