Local and regional constraints on diversity of an experimental urban metacommunity

Background/Question/Methods

The complex urban landscape is hypothesized to mediate both the local and regional factors that influence species diversity of ecological communities.  Despite rapidly expanding urban centers, we know far less about the mechanisms that shape urban communities, despite the evidence that highly organized assemblages thrive in the built environment. To begin to understand how urbanization can shape biodiversity patterns, we focused on two regional constraints, source pool composition and dispersal, and one local environmental factor. We hypothesized that the urban regional species pool is relatively depauperate, and therefore constrains the number of species available for colonization.  Furthermore, high fragmentation of the urban landscape should limit dispersal, which should result in a decline in local diversity. Finally, degradation in habitat quality should serve as an environmental filter, eliminating species, and reducing diversity. To test the relative importance of these local and regional factors, we manipulated each in a full-factorial design to estimate the response of zooplankton communities in pond mesocosms. Mesocosms were inoculated from either three rural or three urban stormwater ponds, and dispersal maintained in half of the mesocosms for 12 weeks. Local water quality was degraded by introducing road deicer as sodium chloride, a contaminant of rising local concern.

Results/Conclusions

Our hypotheses were supported overall. Dispersal increased diversity significantly, and the presence of chloride significantly reduced diversity. The interaction between these two factors was highly significant. Biodiversity levels were highest when dispersal was allowed and no chloride was present, but lowest when dispersal was inhibited under the same no chloride treatment.  This suggests dispersal played a strong role in shaping local diversity patterns. Interestingly, the source of colonists interacted strongly with the dispersal treatment, but in the opposite pattern we predicted. Dispersal increased diversity of the urban source pool beyond the rural effect.  We contend that this could be due to a "rescue effect" for some taxa in the urban pool of colonists.  Why this should be more important for the urban regional species pool more so than the rural will require more investigation. How biodiversity is generated and maintained in the urban environment is understudied, yet important to understand from both local and regional perspectives. This study provides one of the first experimental tests of local versus regional factors shaping urban metacommunities.