Distribution, population structure and habitat use of the endangered Saint Francis Satyr butterfly, Neonympha mitchellii francisci

American Midland Naturalist, The, April, 2008 by Daniel Kuefler, Nick M. Haddad, Stephen Hall, Brian Hudgens, Becky Bartel, Erich Hoffman

SURVEYS FOR DISTRIBUTION

Between 1992-94 we conducted broad surveys for the butterfly across the Inner Coastal Plain physiographic region of the Carolinas to determine the distributional limits of Neonympha mitchellii francisci (Hall, 1993; Hall and Hoffman, 1994). Within North Carolina, these surveys covered 33 sites at Fort Bragg, 17 sites in the neighboring Sandhills region of NC and four Carolina bays in the adjoining, lower portion of the Inner Coastal Plain. In South Carolina, we surveyed six sites in the Sandhills National Wildlife Refuge and seven sites at Fort Jackson. All of these surveys targeted shallow, graminoid-dominated wetlands, based on a habitat profile initially described by Parshall and Kral (1989). These habitats included both natural community types such as swamp forest glades, Carolina bays and vernal pools and artificial impoundments such as old mill ponds. Special attention was given to abandoned beaver ponds and shallow depressions, since they most closely matched the habitats at the original colony sites.

On Fort Bragg, we conducted two additional sets of surveys with increasing resolution. In 2000-2001 we surveyed 110 sites excluding the artillery impact areas but including most of the perennial stream drainages (Hall, 2003). In these surveys, we first identified suitable habitat using aerial photography and maps, giving special attention to abandoned beaver ponds and other herbaceous wetlands. We then visually inspected each site during butterfly flight periods to verify whether Neonympha mitchellii francisci were present. During the course of these surveys, it became clear that many potential sites could not be identified from aerial photographs alone, as many consisted of small glades deeply imbedded within dense growths of riparian forest or pocosin shrubs. So, employing a finer-scale technique, we walked the lengths of watersheds, looking for small open glades by laboriously penetrating through the thick riparian vegetation.

Butterfly subpopulations were identified on the basis of both consistent presence during the flight periods and the presence of suitable habitat. Additionally, transitory sites were identified where butterflies were irregularly seen in low numbers (often a single individual), typically at sites supporting lower-quality habitat.

HABITAT DESCRIPTION

From 2001 to 2004 we conducted quantitative vegetation sampling at 33 of the sites identified during reconnaissance surveys (Hall, 2005). We surveyed all six subpopulations that were active in 2002, one abandoned subpopulation site, six sites where transients had been observed and 20 unoccupied sites supporting a range of potentially suitable habitat types.

We sampled vegetation at each study site using a series of 3 x 3 m quadrats. At the first nine sites we surveyed, we used two intersecting 50 m transects with eight to ten quadrats laid along them. Subsequently, we employed a single 50 m transect with five quadrats laid along it. At four smaller sites, we used a shorter transect and fewer quadrats. Within each of the quadrats; we estimated cover classes for all plant species using methods developed by the NC Vegetation Survey (NCVS) for ease of visual estimation (Peet et al., 1998). For woody species, height was also recorded in three classes following NCVS conventions.