COS 2-1 - Biogeographic responses of ant species and communities to regional climate change

Monday, August 6, 2012: 1:30 PM
B112, Oregon Convention Center
Israel Del Toro, Center for Macroecology Evolution and Climate, University of Copenhagen, Copenhagen, Denmark
Background/Question/Methods

Regional climate change alters species distributions, usually by shifting, expanding and contracting species’ native ranges. Arthropods are especially susceptible to the impacts of regional climate change, since their physiology is dependent on the conditions of their physical and environmental surroundings. Here we use the ants (Hymenoptera: Formicidae) of the northeastern United States to model how terrestrial invertebrate species distributions and community assembly may respond to regional climate change. Ants are ecologically important organisms in the forests of the Northeast because they play key roles in several ecosystem processes, and so it is important to hypothesize how these species and communities may respond to regional climate change.  We modeled the distributions of 50 of the most commonly collected and abundant ant species in New England using two modeling approaches, MaxEnt and Boosted Regression Trees. We mined data from regional entomological collections, literature and online databases to create the species distribution models. We then projected the distribution models onto the future climate scenarios proposed by the IPCC and documented any range contractions and expansions.

Results/Conclusions

We evaluated model performance using independent datasets and found that for 78% of species model prediction was extremely accurate (AUC =0.85 (+/- .06). MaxEnt preformed better than Boosted Regression Trees in 83% of the cases. A comparison of the two methods showed that the distribution models typically agreed on approximately 80% of the of the species range for most species. When projected onto future climate scenarios we see that 37 species show a significant range size reduction with 35 having southern range boundary contractions and 17 having northern range expansions. Our results provide preliminary evidence that ant community structure in New England may change as a result of the predicted regional climate change. Ultimately changes in community structure may lead to changes in ecosystem processes and services mediated by ant, a key terrestrial arthropod.