Butterfly conservation in fragmented landscapes

Human activities disturb and fragment many natural areas, thereby altering the way in which species interact with their landscape. In natural settings, species often evolved to disperse through areas unsuitable for habitation, but in a world dominated by human-modified landscapes, species face novel features that may disrupt their movement, increasing the isolation of populations. Consequently, there is a pressing need to understand if animals can persist in small natural areas separated by agricultural or urban development. The way in which species respond to fragmentation will inform if conservation actions are necessary to protect a species and what strategies might mitigate the consequences of fragmentation. My dissertation addresses these conservation problems using two studies of butterflies in fragmented landscapes.;First, I used a novel synthesis of behavioral, mark-recapture, and population genetic studies to identify the effects of urban fragmentation on a rare butterfly. I present the synthesis in the context of a butterfly endemic to barrier islands in North Carolina, Atrytonopsis new species 1. Atrytonopsis sp1 occupies sand dune habitat, which in this region has largely been converted to residential and commercial development. Many studies rely on a single method to measure movement, but combining methods may help elucidate the effects of natural or anthropogenic fragmentation. To test edge permeability, I conducted butterfly releases at four edge types: sand dune - beach, sand dune - maritime forest/shrub, sand dune - urbanization, and beach - ocean. I then used mark-recapture to compare Atrytonopsis sp1 movement rates in continuous sand dune habitat and between habitat patches separated by urban areas. Finally, I conducted a range-wide population genetic study to assess if there was significant population structure across the range of Atrytonopsis sp1, and to determine how the structure was partitioned relative to natural barriers and urbanization. All three studies independently indicate that natural features, not urbanization, are barriers to Atrytonopsis sp1 movement. Analyzed separately, each study would point to the same general conclusion, but could result in different conservation recommendations. Both linear corridors and stepping-stones are conservation strategies that can increase connectivity of fragmented landscapes. For stepping stones to work, however, Atrytonopsis sp1 must be willing to enter the matrix, successfully find new habitat, and integrate its genes into the new population. The edge behavior studies indicate that Atrytonopsis sp1 will fly into urban areas, and the mark-recapture studies confirm that it can successfully disperse over short distances. The population genetic studies indicate that longer distances of urbanization are not intrinsically barriers to movement. Consequently, I recommend conservation strategies that focus on preserving stepping stones, rather than on habitat corridors.;In a separate study, I examined the effects of habitat fragmentation on tropical butterfly communities. Habitat fragmentation is a major driver of biodiversity loss, but its effects may be obscured by the ways in which different species respond to fragmentation. One reason that species display varying responses to fragmentation may be that variability itself is an effect of fragmentation. I tested whether forest fragmentation causes long-term hyperdynamism in butterfly communities, a taxon that naturally displays large variations in species richness and community composition. Using an 11-year dataset from an experimentally fragmented landscape in the central Amazon, I evaluated the effect of fragmentation on changes in species richness and community composition through time. Butterfly species richness did not differ between fragmented forest and intact forest. However, spatial and temporal patterns of butterfly species richness in fragmented forests were significantly more variable than in intact forest. Fragmented forest had a significantly lower proportion of understory shade species and a significantly higher proportion of edge species. I conclude that fragmentation fundamentally alters the natural dynamics of butterfly communities.