Jump to Main Content
Swallowtail butterflies show positive edge responses predicted by resource use
- Siu, Jenna C., Koscinski, Daria, Keyghobadi, Nusha
- Landscape ecology 2016 v.31 no.9 pp. 2115-2131
- Lindera benzoin, Papilio glaucus, biodiversity, butterflies, flight, forests, landscapes, meadows, models, nectar plants, oviposition sites, temporal variation, Ontario
- CONTEXT: The prevalence of edges is increasing due to anthropogenic landscape change. Edge responses can vary considerably between and within species. Understanding species’ responses to edges, and the causes of variation in such responses is central to managing biodiversity in contemporary landscapes. OBJECTIVE: A resource distribution model predicts that species that require complementary resources in different land cover types will be most abundant at edges, displaying a positive edge response. Eastern tiger (Papilio glaucus) and spicebush (P. troilus) swallowtail butterflies use forest plant species for oviposition sites but open-habitat plants for nectar. They are excellent models for testing the positive edge response and exploring sources of variability in edge responses, such as species-specific traits or temporal effects. METHODS: In southwestern Ontario, we examined both the abundance and flight orientation of these species in relation to forest/meadow edges and at different times of day. We used a transect method similar to the Pollard walk and a catch and release method, respectively. RESULTS: The distribution and flight behaviour of these butterfly species were overall consistent with a positive edge response. Both species were most abundant at the edge and oriented their flight towards the edge from the forest and meadow. However, P. glaucus demonstrated a much stronger positive edge response, while P. troilus showed temporal variation in its response. CONCLUSIONS: Our results confirm the ability of the resource distribution model to predict species edge responses and movement behaviours, but also indicate that species-specific traits and time of sampling can influence such responses.