Using simple regression, pooled-sites route-regression, and nonparametric rank-trend analyses, we evaluated trends in counts of hawks migrating past 6 eastern hawk lookouts from 1972 to 1987. The indexing variable was the total count for a season. Bald eagle (Haliaeetus leucocephalus), peregrine falcon (Falco peregrinus), merlin (F. columbarius), osprey (Pandion haliaetus), and Cooper's hawk (Accipiter cooperii) counts increased using route-regression and nonparametric methods (P 0.10). We found no consistent trends (P > 0.10) in counts of sharp-shinned hawks (A. striatus), northern goshawks (A. gentilis), red-shouldered hawks (Buteo lineatus), red-tailed hawks (B. jamaicensis), rough-legged hawks (B. lagopus), and American kestrels (F. sparverius). Broad-winged hawk (B. platypterus) counts declined (P < 0.05) based on the route-regression method. Empirical comparisons of our results with those for well-studied species such as the peregrine falcon, bald eagle, and osprey indicated agreement with nesting surveys. We suggest that counts of migrant hawks are a useful and economical method for detecting long-term trends in species across regions, particularly for species that otherwise cannot be easily surveyed. J. WILDL. MANAGE. 54(3):463-470 Wildlife managers have responsibilities to monitor a wider variety of wildlife species than in the past, and cost-effective and efficient methods are needed to detect trends in populations that might warrant management action. Monitoring populations of nesting raptors over large areas is difficult because most species occur in low densities, and some species are secretive or nest in remote areas (Fuller and Mosher 1987). Although nesting season surveys have been used to monitor populations of endangered raptors such as the peregrine falcon (Cade et al. 1988), such long-term surveys are expensive and labor intensive. Most other raptors have received less study on their breeding grounds. However, migratory species have been counted annually during their migrations at certain concentration points in the eastern United States (e.g., Broun 1935, Allen and Peterson 1936). Ward and Berry (1972) suggested that counts of migrant hawks might provide a useful method for indexing population trends. Assuming that hawk count data can be used to detect trends in numbers of migrants and that these counts are a reliable index to populations, then analyses of hawk counts could be used for raptor management. Qualitative analyses of hawk count data have been used to infer population declines, increases, or stability (e.g., Nagy 1977, Dunne and Sutton 1986). The usefulness of these eyeball analyses and the quality of data on which they are based have been questioned (e.g., Kerlinger 1985). Hussell (1985) used quantitative methods to allay some previous criticisms of the use of migration count data, and Titus et al. (1989a, 1990) and Bednarz and Kerlinger (1989) discussed methods for analyzing count data. Svensson (1978) and others suggested that counts of migrants might be a useful method for detecting population trends. Indeed, counts of migrating hawks have been used for policy decisions about raptor management (e.g., U.S. Dep. Inter. 1983) and the status of raptors. Our objective was to estimate trends in 14 species of migrant hawks using regression and nonparametric techniques with single-site and pooled-sites analyses and to compare our results with information available from other population studies. We thank the observers who counted the migrant raptors. Data collection was supported, in part, by the Office of Migratory Bird Management (OMBM), U.S. Fish and Wildlife Service, through contracts with Cape May Bird Observatory, Derby Hill Bird Observatory-Onondaga Audubon Society, Whitefish Point Bird Observatory, and the Duluth Audubon Society. The Hawk Mountain Sanctuary Association (L. 'Present address: Alaska Department of Fish and Game, Division of Wildlife Conservation, P.O. Box 20, Douglas, AK 99824.
Read full abstract