A recent paper in Ecological Applications put a classic theory to the test, with a twist. The theory of island biogeography was one of the first conceptual frameworks to provide guidance for the conservation of biodiversity on a landscape scale. It’s pretty intuitive stuff. Changes in species diversity on ecological time scales (i.e., over decades) in any habitat are the result of colonizing species minus species extinctions. MacArthur and Wilson (1967) first posited that the species diversity of islands (for the moment we’re talking about actual islands in actual oceans, but hold on to your seats) is controlled by both the island’s distance from the mainland and its size. Original tests of insect communities on mangrove islands (Simberloff and Wilson 1969) seemed to support this idea, and the theory gained some traction. Today, it is one of the central ideas in community ecology.
Turns out this idea is useful in today’s fragmented landscape where “islands” of habitat exist in an “ocean” of non-habitat. Habitat fragments closer to large, undisturbed habitats are more likely to receive immigrants and should therefore have more species. Larger habitat fragments can support viable populations of more species, resulting in fewer local extinctions. Therefore, larger fragments that are closer to or more connected to other fragments should be able to support more species.
But, wait! Habitat fragments are not really islands nor are they actually in an ocean. Especially to the bees of the shortgrass steppe in Colorado. The suburban landscape (i.e., the “ocean”) has fragmented these bees’ native prairie (i.e., the “mainland”), but still provides potential nesting sites, nectar, and water sources. Hinners and others (2012, Ecological Applications) recently explored how native bee communities are affected by habitat fragment size AND the quality of the suburban landscape in which they are found. They compared bee communities in prairie fragments of various sizes in the matrix of suburban Denver and Boulder, CO to large patches of continuous grassland outside of suburbia. The following quote sums up their interesting take on island biogeography, and the hypothesis that they tested in this innovative study.
“Organisms in fragments interact with the matrix, and the increased heterogeneity of the fragmented landscape can provide new ecological conditions and opportunities for species. Depending on the taxa, we find it relatively easy to envision fragmented landscapes in which not just individual species, but entire communities, are able to persist, albeit with somewhat altered structure and dynamics.” (page 1924)
To compare bee communities, they looked at both overall bee abundance (all the species combined) and the diversity and composition of bee communities. While bee abundance was highly variable among sites, there was no pattern with respect to either the size of the fragment or between fragments and contiguous habitat. Changes in composition were observed, but no overall pattern emerged. However, an interesting pattern emerged when they examined differences in the diversity of the bee communities.
So, does this pattern of bee diversity in suburban Denver and Boulder support the theory of island biogeography? As with any application of theory to a real-life conservation question, the answer is yes . . . and no. Well, mostly no. Yes, in the sense that larger habitat fragments had more diverse bee communities, although even this effect had its limits. While small (<8 ha) fragments were not very diverse and larger ones (~ 20 ha) were more diverse, there was no difference between large and very large (~ 60 ha) fragments. This is not surprising, as most species-area relationships are asymptotic, so that at some point you will not find more species no matter how large of an area you sample.
But the answer is largely no because most of the fragments (all but the smallest) actually supported more diverse bee communities than the large unbroken expanse of prairie! By being imbedded in suburbia, the fragments are actually part of a more heterogeneous landscape that provides a greater array of resources and therefore a greater diversity of bees. These are probably resources that are lacking in the undisturbed shortgrass steppe, such as more consistent and diverse flower resources (e.g., landscape plants), vertical structure (e.g., trees), and an abundance of water. The ocean of suburbia is not an uninhabitable void for the bees in this study as the theory of island biogeography assumes. And by discussing the resources provided by this ‘suburban ocean’, Hinners and others have made a significant contribution to how a classic ecological theory can be adapted to a novel situation.
Suburbia was a boon to the bees, but not all taxa can move around as freely to access resources, and every organism will respond differently to fragmentation and the ocean they find themselves in. Plants, for example, generally stay put once they put their roots down and may not benefit from the suburban landscape. But, that is why I like this paper so much. It puts a finer point on a classic theory, and helps us to understand exactly why the theory works, when it does work . . . and does not work, when it does not.