SYMP 7-4 - Using species distribution models to predict current and future UK crop pollination

Tuesday, August 7, 2012: 2:30 PM
Portland Blrm 251, Oregon Convention Center
Chiara Polce, IICB Biology, University of Leeds, Leeds, United Kingdom, Jacobus Biesmeijer, Netherlands Centre for Biodiversity Naturalis, and IICB Biology University of Leeds, Leiden, Netherlands, Mette Termansen, National Environmental Research Institute, Arhus University, Roskilde, Denmark and Simon G. Potts, Centre for Agri-environmental Research, University of Reading, Reading, United Kingdom
Background/Question/Methods

Pollination is a key ecosystem service, vital to the maintenance of both wild plant communities and agricultural productivity. Insect pollination, mostly by bees, benefits over three quarters of the major world crops, with an economic value of approximately €150 billion globally and about €500 million in the United Kingdom.

Despite this importance, our current knowledge about the pollination requirements of crops, the pollination potential of different pollinators and the impact of environmental changes on pollination has significant gaps. These gaps prevent progress towards designing effective policies to conserve and manage pollination services. In addition, we have incomplete knowledge of where wild crop pollinators occur in the UK.

We identify the key elements needed to fill these gaps and we construct a modelling framework using high resolution pollination records, environmental information and primary data. Detailed model predictions will help us move forward our understanding of current and future pollination service provision within UK, whilst the general modelling framework may potentially be also applied to other geographical areas.

Results/Conclusions

Using existing literature on plant-visitor links, we first identified a subset of wild bees and hoverflies species that are important for pollination services of UK crops. Test models have been developed and evaluated (by means of randomization approaches) for their capability to predict the current distribution of a representative set of wild crop pollinators. Our results demonstrate satisfactory model performance for wild pollinators characterised by different environmental requirements, geographical ranges and number of available records. We show how this information can be used to identify areas of potential pollinator deficit using the current distribution of UK focal crops, which include members of all major crops categories. We will discuss the limitations and main challenges of the approach and will contemplate how our work informs the area of sustainable pollination service provision.