Tuesday, August 3, 2010 - 9:50 AM

COS 23-6: Interactions lost and gained: A review of how trophic interactions are impacted by the fragmentation of habitats

Holly Martinson, University of Maryland and William F. Fagan, University of Maryland.

Background/Question/Methods

The fragmentation of habitats can have profound effects on species persistence, population density, and species richness.  Many recent studies have also documented how habitat fragmentation and related variables can influence trophic structure and food web interactions, yet this diverse literature has not been brought together and examined for commonalities in mechanism and outcome.  In this study, we investigated how fragmentation, in the broad sense, has been shown to impact trophic interactions, and we tested the prediction that, as for single species, the habitat affinity of interacting species is a major determinant of whether a trophic interaction will be lost or gained in fragmented habitats.  We undertook a systematic, quantitative review of the literature on trophic interactions studied at both the landscape scale (fragmentation per se, proportion habitat in the landscape, matrix composition) and the patch scale (patch size, distance to edge, patch connectivity).  We summarized the findings of these studies with an overall vote-count analysis, followed by meta-analysis for well-studied trophic interactions.

Results/Conclusions

We identified 171 studies of trophic interactions in fragmented or patchy habitats.  The majority (54%) of these studies focused on just a few specific trophic interactions: seed predation, insect parasitoid-host interactions, and bird nest predation and parasitism.  More complex food web modules (sets of interacting species), such as food chains, apparent competition, or those involving indirect interactions, were infrequent in the literature (4% of studies).  Overall, the habitat specificity of the interacting species was a key trait influencing whether the trophic interaction occurred more or less frequently in relation to fragmentation variables.  Across systems, trophic interactions in which the top species was a habitat specialist were more likely to occur on large patches, whereas those involving habitat generalists were more likely to occur on small patches.  Similar differences between interactions involving generalists versus specialists were found for other variables as well, including connectivity, distance from edge, and the proportion of habitat in the landscape.  Although other ecological traits may mediate the strength of these responses, the habitat specificity of interacting species was a consistent predictor of whether diverse trophic interactions were lost or gained in fragmented habitats.