Bird populations and habitat in Bacillus thuringiensis and dimilin-treated and untreated areas of hardwood forest.

INTRODUCTION

The gypsy moth (Lymantria dispar) has infested approximately 25% of the nation's hardwood forests and is steadily moving S and W (Gottschalk, 1991). Eventually the range of the gypsy moth is expected to include most of the eastern United States. Multiple defoliations have additive effects and can result in a significant loss of trees (Herrick and Gansner, 1987). Since 1980, 4.9 million acres of forested land have been sprayed with Bacillus thuringiensis (Bt) and 3.7 million acres have been sprayed with diflubenzuron (Dimilin) to control the gypsy moth (U.S. For. Serv., 1994). Nearly 60% of the area treated with Dimilin occurred in West Virginia and Pennsylvania (U.S. For. Serv., 1994). Dimilin and Bt have been considered safe to use because of their low toxicity to vertebrates (Maas et al., 1981; Hudson et al., 1984). However, these agents have recently come under examination as there is concern about the effects of pesticide use on nontarget Lepidoptera larvae, which have high fat and low chitin content, and are a primary food source for many birds during the breeding season (Cooper, 1988). Both Dimilin and Bt cause a significant reduction in non-target Lepidoptera abundance and species richness (Sample et al., 1993a, b).

Pesticides used to control gypsy moths have had negative impacts on birds. Black-throated blue warblers (Dendroica caerulescens) made fewer nesting attempts in Bt-treated areas than in unsprayed areas (Rodenhouse and Holmes, 1992). Birds foraging in Dimilin-treated areas have experienced dietary shifts to suboptimal food items, reduced food intake (Sample et al., 1993c) and had reduced fat levels when compared to birds in control areas (Whitmore et al., 1993). These findings suggest that the use of Dimilin and Bt may be detrimental to bird populations.

There also has been concern about the effects of gypsy moth defoliation on bird populations. As changes in forest structure occur, changes in bird populations can be expected. Vegetation characteristics are a dominant factor in determining bird community composition and abundance (Holmes et al., 1986). Habitat selection by breeding birds is believed to be largely affected by forest structure (Anderson and Shugart, 1974; Crawford et al., 1981). Gypsy moths alter the habitat, and may reduce necessary resources. Cooper et al. (1987) predicted that areas heavily defoliated by the gypsy moth may be avoided by forest-interior bird species and colonized by open habitat species. Moreover, dead trees may provide better perch sites for avian nest predators and brood parasites (Cooper et al., 1987). In light of the deleterious effects of pesticide use, it is important to determine the effects of gypsy moth impact on bird populations, many of which are of special concern due to reports of population declines (reviewed in Askins et al., 1990). Our data may aid in management decisions for forests threatened with gypsy moth defoliation.

STUDY AREAS

This study was conducted at the Sleepy Creek Public Hunting and Fishing Area (hereafter, Sleepy Creek) located in Morgan and Berkeley counties, and Cacapon State Park (hereafter, Cacapon) in Morgan County in the eastern panhandle of West Virginia. Sleepy Creek Mountain and Cacapon Mountain lie parallel to one another and were chosen for comparison due to similarities at the onset of the study with respect to physiogeographic region, location, soil type, forest age and species of plants present. Both areas are even-aged, second-growth forest approximately 50-65 yr old. The dominant tree species were northern red oak (Quercus rubra), chestnut oak (Q. prinus), white oak (Q. alba) and hickories (Carya spp.). Other common tree species included red maple (Acer rubrum), black birch (Betula lenta), black gum (Nyssa sylvatica), yellow poplar (Liriodendron tulipifera) and sassafras (Sassafras albidum).

The objective of this study was to evaluate the effects of gypsy moth impact on bird populations and habitat by comparing an area that was severely defoliated by the gypsy moth (Sleepy Creek) to an area (Cacapon) that was repeatedly sprayed with Bt and Dimilin to control gypsy moth populations. The vegetation and bird communities were sampled in both areas in 1984, before defoliation or pesticide treatment, and again in 1993. Sleepy Creek, owned by the West Virginia Division of Natural Resources, was designated as a "no-spray" area, and no efforts were taken to control gypsy moth populations. Sleepy Creek was severely defoliated by the gypsy moth in 1987 and 1988, and received additional sporadic defoliation each year afterward to the present. In 1981, a complex of geometrid loopers (Lepidoptera: Geometridae) defoliated approximately 1 million acres of forest in the eastern panhandle of West Virginia (Wygal et al., 1983). These insects were a nuisance to the visitors at Cacapon, and fearing the encroaching gypsy moth populations, the park officials initiated a spray program in 1985. From 1985-1987, the park was treated with Dimilin. In 1992 and 1993, the park was treated with Bt. Each spray treatment encompassed the areas used in this study (J. Hacker, pers. comm.). When the areas were sampled again in 1993, Sleepy Creek was to serve as a gypsy moth-impacted area and, because it had been sprayed, Cacapon was to serve as an example of what forests in the region would be like in the absence of gypsy moth defoliation.

METHODS

In 1984, six transects, approximately 2 km long, were established in both Sleepy Creek and Cacapon along the …