A central question in landscape ecology is how to quantify connectivity in fragmented landscapes.  To answer this, we need to determine how animals behave at vegetation borders and how readily they move through matrix vegetation.  The success of efforts to manipulate landscape connectivity, like corridors and stepping stones, is dependent on the behavior of the target organism and relies on a mechanistic understanding of movement decisions.  Improved understanding of the environmental cues used to make movement decisions will allow for better quantification of matrix conductance and landscape connectivity.  This study focuses on how a guild of fruit feeding nymphalid butterflies react to fragmentation in a post agricultural landscape, with emphasis on the behavior and distribution of the Hackberry Emperor Asterocampa celtis.  This species is found in shrubland and forest vegetation in close proximity to its host plant the Hackberry Celtis occidentalis.  Butterflies are ideal organisms to answer questions related to movement behavior, as their behavioral decisions are often less complicated than vertebrates.  Further, the majority of species move distances that are amenable to landscape level studies of fragmentation.  This work was conducted at the University of Virginia’s Blandy Experimental Farm.  Using a combination of observations at habitat edges and large scale trapping across different vegetation types, I measured Nymphalid vegetation preference and how this scales up to landscape level butterfly distribution.  I also conducted controlled screen house experiments to isolate the environmental cues that are important for butterfly vegetation preference.

Results/Conclusions

Initial experiments show that butterflies exhibit vegetation bias at a variety of spatial scales and that movement at vegetation borders is indicative of larger vegetation preferences. Vegetation preference also corresponds with the conductance of different matrix vegetation as measured by mark recapture experiments.  An analysis of different cues indicated that structural complexity of vegetation seems to be an important driver of butterfly vegetation preference.  Better quantification of matrix conductance, combined with knowledge of the cues butterflies use when choosing to move into non-habitat vegetation, will allow for a more complete understanding of landscape connectivity.