Effects of habitat affinities and resource needs on edge responses by small mammals.

Edge effects are the responses of organisms to ecological boundaries. I studied edge responses by small mammals in two prairie communities to test two models: 1) edge responses reflect and map to simple habitat associations, and 2) edge responses reflect differences in resource availability on either side of an edge. I also investigated how edge contrast, the degree of difference between habitats at an edge, could modify edge responses. Pilot sampling in 2009 and 2010 at a Blackland prairie preserve revealed that hispid cotton rats (Sigmodon hispidus Say and Ord 1825) responded negatively to edges of wooded patches and weakly positively to edges between tallgrass and dicot forbs. Deer mice (Peromyscus maniculatus (Wagner 1845)) and fulvous harvest mice (Reithrodontomys fulvescens Allen 1894) did not respond to habitat edges, although this may have been an artifact of low capture rate (especially in 2010). Sampling in 2011 at the Lyndon B. Johnson National Grasslands (LBJNG), a mixed grass preserve, found that S. hispidus responded negatively to edges where dense grass abutted dicot forbs or wooded patches. These edge effects reflected the strong association of S. hispidus with grassland habitat. Sampling of leptin, a hormone secreted by adipose tissue and associated with energy intake in many species, revealed that serum leptin concentration (SLC) was limited by body mass and not associated with any other organismal characteristic. Among male S. hispidus SLC was limited by food availability and among females the lower limit of SLC decreased with increasing grass cover. This suggests that if S. hispidus edge responses are caused by resource mapping, the critical resource is not nutrition. Other species at LBJNG, P. maniculatus and hispid pocket mice (Chaetodipus hispidus (Baird 1858)) were associated with habitats with intermediate ground cover and did not show clear edge responses. I concluded that 1) when prairie dwelling small mammals respond to edges, they are more likely to show matrix effects (response to nonhabitat) than ecotonal effects (emergent properties at boundaries); and 2) small mammal abundances and edge responses are driven by resources other than nutrition.