Arthropod diversity and ecosystem services in Manhattan after Super Storm Sandy: Does variation in chronic environmental stress predict the impacts of extreme weather events?

Background/Question/Methods: The frequency and intensity of extreme weather events are increasing globally, while urbanization is rapidly expanding worldwide. However, we have a poor understanding of how these simultaneous global changes interact to influence local community dynamics and ecosystem services. Home to more than half of the world’s humans, cities also have surprisingly diverse communities of other species across a mosaic of habitats which have varying levels of chronic environmental stress. Experimentally mimicking the effects of extreme weather events can be unethical and lacks feasibility. However, we had the unique opportunity to resample Manhattan’s arthropod communities after Super Storm Sandy that we sampled just before the storm struck. We used a Before-After-Control-Impact design to assess the responses of arthropods across Manhattan’s urban habitat mosaic. We established plots across urban forests, parks and medians in Manhattan in which we monitored temperature, relative humidity, leaf litter depth, and arthropod community composition and abundance. We also examined the influence of chronic stress and storm-associated disturbance on the ecosystem services of local arthropod communities. Specifically, we provided human food (hot dogs, cookies, chips) either inside an exclosure (to block vertebrate access) or freely available across all site combinations and tracked removal rates. We also placed mesh bags with either dried leaves or paper bags in the leaf litter of each site to assess decomposition rates. 

Results/Conclusions: Across all sites, habitat type (street median vs. city park), a proxy for chronic environmental stress level, affected within-site arthropod community composition (PerMANOVA: P_{habitat type}<0.0001), but did not significantly influence among-site arthropod composition (PermDisp: P=0.1378). While flooding did not significantly influence community composition in street medians or city parks, there were significantly more accumulated families in flooded than unflooded medians (ANCOVA: P_Flood=0.0069). There were more accumulated arthropod families in parks than medians (P_{Site Type}<0.0001), but this richness was not different in flooded vs. unflooded sites. Arthropods in medians removed as much human food as controls; in contrast, significantly more food was removed in controls than in exclosures in park sites (ANOVA: P_{Site Type x exclosure}=0.0076), suggesting that arthropods may be more important to food removal in high stress habitats. In sum, these findings suggest that urban arthropod communities are remarkably resilient to the short-term disturbances associated with extreme weather events, perhaps because they are exposed to highly variable chronic environmental stressors over longer time spans.