Butterfly Responses To Management Of Disturbance-Dependent Ecosystems In North America And Australia.

Abstract

Fire and grazing are primary sources of natural disturbance in grasslands and savannas worldwide, but they are also shaped by human impacts and decision-making. Appropriate management is therefore an essential goal for biodiversity conservation in disturbance-dependent landscapes. The butterflies which inhabit these ecosystems are reliant on the disturbances necessary for grassland and savanna persistence. Regular fire, herbivory, and drought shape and maintain these ecosystems, keeping woody plants from dominating. How, then, do butterflies persist and respond to disturbance in their chosen habitats? What management regimes are required in human-altered landscapes to conserve grassland and savanna butterfly biodiversity? I examine butterfly responses to management of disturbance-dependent ecosystems in North American and Australian using both taxonomic and functional trait lenses. In Chapter 1, I assess the impacts of current fire and grazing management regimes on butterfly communities in Minnesota tallgrass prairie and compare butterfly and bee responses to management. In Chapter 2, I assess the impacts of these same fire and grazing regimes on monarch butterflies (Danaus plexippus) and their milkweed host plants in Minnesota tallgrass prairie. In Chapter 3, I examine butterfly and butterfly resource responses to fire regimes in Australian tropical savanna, and in Chapter 4, I demonstrate the utility of trait-based ecology to explain global and mechanistic patterns in butterfly responses to fire by presenting and comparing butterfly traits associated with fire regimes in U. S. temperate prairie and Australian tropical savanna. I present findings that butterfly and bee abundances in tallgrass prairie are significantly negatively correlated. Butterfly abundance, but not species richness, is nearly twice as high at tallgrass prairie sites managed with fire compared to grazing, and prairie-associated grass-feeding butterflies are more abundant at sites with higher plant species richness. I find that monarch butterflies are also more abundant at tallgrass prairie sites managed with fire than with grazing, and that this association is not related to milkweed or forb frequencies, which are similar between burned and grazed prairies. In Australian tropical savanna, I find that recent, early dry season burning promotes butterfly diversity and abundance by increasing the supply of nectar resources in tropical savanna fire treatments. In my evaluation of butterfly traits associated with fire, diapause strategy, host plant specificity, wingspan, voltinism, and flight period are all associated with at least one fire treatment, but trait associations are not shared across Australian tropical savanna and U.S. temperate prairie. In tallgrass prairie, land managers and conservation practitioners interested in promoting butterfly abundance and diversity may consider increasing plant species richness and maintaining fire in the landscape. Because species composition differs between sites managed with fire and grazing and among sites managed for different numbers of years, I suggest a variety of management strategies is required to support the entire suite of butterfly species. In tropical savanna, land managers and conservation practitioners can ensure some areas of recent, early-season fire are maintained and focus on additional actions that will promote floral resource abundance and diversity, which will benefit butterflies as well as many other animal taxa. Trait-based ecology methods can help us understand the role of traits such as diapause strategy, host plant specificity, wingspan, voltinism, and flight period in explaining general patterns in butterfly responses to fire. However, to inform good conservation and management, trait-based findings should be tied back to the local species and landscapes being managed. The second chapter of this dissertation is published in Frontiers in Ecology and Evolution (Leone et al. 2019) and the first chapter is currently under review with the Journal of Insect Conservation. As a reflection of the collaborative nature of the work I present, I will use the first-person plural voice “we” throughout the rest of this dissertation.