REPORT The area requirements of an ecosystem service: crop pollination by native bee communities in California Claire Kremen, 1 * Neal M. Williams, 2 Robert L. Bugg, 3 John P. Fay 4 and Robin W. Thorp 5 1 Department of Ecology and Evolutionary Biology, Guyot Hall, Princeton University, Princeton, NJ 08540, USA 2 Department of Biology, Bryn Mawr College, 101 N. Merion Ave, Bryn Mawr, PA 19010, USA 3 University of California Sustainable Agriculture Research and Education Program, University of California, One Shields Avenue, Davis, CA 95616-8716, USA 4 Department of Biological Sciences, Center for Conservation Biology, Stanford University, Stanford, CA 94305, USA 5 Department of Entomology, University of California, One Shields Avenue, Davis, CA 95616-8716, USA *Correspondence: E-mail: ckremen@princeton.edu Abstract Managing ecosystem services is critical to human survival, yet we do not know how large natural areas must be to support these services. We investigated how crop pollination services provided by native, unmanaged, bee communities varied on organic and conventional farms situated along a gradient of isolation from natural habitat. Pollination services from native bees were significantly, positively related to the proportion of upland natural habitat in the vicinity of farm sites, but not to any other factor studied, including farm type, insecticide usage, field size and honeybee abundance. The scale of this relationship matched bee foraging ranges. Stability and predictability of pollination services also increased with increasing natural habitat area. This strong relationship between natural habitat area and pollination services was robust over space and time, allowing prediction of the area needed to produce a given level of pollination services by wild bees within this landscape. Keywords Agriculture, Apis mellifera, Apoidea, bee community, bee foraging distance, conservation planning, landscape ecology, pollination service, scale effects. Ecology Letters (2004) 7: 1109–1119 INTRODUCTION Ecosystem services, including climate regulation, soil pro- duction, water purification, pest control and crop pollination are critical to human survival (Daily 1997). Management of services could also provide incentives for biodiversity conservation (Daily & Ellison 2002), particularly in human- dominated landscapes where such services are most needed (Scheer & NcNeely 2002). Nonetheless, few natural areas are managed or valued for the services they provide, although many are managed to produce ecosystem goods (e.g. wood, wildlife, fish). In large measure, this is because the ecology of ecosystem services is poorly known, limiting our ability to understand their value and to plan their conservation and management (Palmer et al. 2004). Developing such plans require knowledge of the relationship between the services provided and the area of habitat conserved. This relationship has been estimated for services from plant communities like carbon sequestration and storage (Niles et al. 2002) and water flow regulation provided by different vegetation types (Guo et al. 2000), but not for any animal-based ecosystem service. One such service is crop pollination. Thirty per cent of the US food supply by volume depends on animal pollinators (McGregor 1976), of which bee species (Apoidea) are the most important (Roubik 1995; Nabhan & Buchmann 1997). Many farmers rely on colonies of the European honeybee (Apis mellifera) that they import temporarily to crop fields to provide pollination during bloom (Free 1993; Delaplane & Mayer 2000). Honeybees are not always the most effective pollinators of a given crop (Parker et al. 1987; Kevan et al. Ecology Letters, (2004) 7: 1109–1119 doi: 10.1111/j.1461-0248.2004.00662.x Ó2004 Blackwell Publishing Ltd/CNRS