Responses of Small Mammal Populations too Wild Hog Rooting in the Gray Beech Forests of the Great Smoky Mountains National Park

Lusk, Michael

ABSTRACT: A population of exotic European wild hogs (Sus scrofa) entered the boundaries of Great Smoky Mountains National Park sometime during the 1940s or 1950s. Since that time their population has spread throughout the Park, and they have damaged its rich flora and fauna. An intensive control program by the Park has reduced hog populations in recent years. The upper elevation beech forests have received especially heavy impact due to their small total acreage, rich herbaceous layer, and preferential rooting by hogs in summer. A great deal of research has been conducted on the effects of rooting on flora of these areas, but very little has been done on the effects on the animal communities. I conducted a two-year study to gauge the effects of rooting in these areas on small mammal populations. During the summers of 1989 and 1990, I chose three rooted and three unrooted sites for comparison in the beech forests. A mark-recapture study using Sherman live traps was conducted at each site. In 1990, I added 10 pit-fall traps at each site. I recorded a variety of habitat measurements at each live trap site and analyzed them using discriminant function analysis. Populations of cloudland deer mice (Peromyscus maniculatus nubiterrae) showed no significant differences between rooted and unrooted sites and are apparently unaffected by hog rooting. Captures of shrews were not sufficient for statistical analysis but their capture on rooted sites indicated that shrew populations can recover after cessation or decrease of rooting. Discriminant analysis of P. maniculatus microhabitat variables indicated that deer mice orient towards areas dominated by deciduous trees with heavy midstory cover and light ground cover. I believe that habitat selection among P. maniculatus in the upper elevation beech forests of GSMNP is motivated primarily by predator avoidance. There were no significant differences between rooted and unrooted sites for the habitat variables entered into the model.