Movements and wetland selection by brood-rearing black ducks (Anas rubripes) were studied in Maine during 1977-80. Eight radio-marked hens moved their broods an average of 1.2 km from the nest to rearing pond, but only 1 hen initiated secondary brood movements. Half of the 85 broods reared in the study area used only 3 wetlands, and most rearing ponds contained active beaver (Castor canadensis) colonies. Brood-rearing hens preferred Emergent ponds over lakes and Evergreen Scrub-Shrub wetlands, and did not occupy Dead Scrub-Shrub, Unconsolidated Bottom, or Aquatic Bed wetlands. Rearing ponds were large and possessed extensive areas of flooded mountain alder (Alnus incana), willow (Salix spp.), and herbaceous vegetation. Wetlands avoided by brood-rearing hens were those with large areas of open water, submergent aquatics, or ericaceous shrub vegetation. J. WILDL. MANAGE. 46(3):615-621 The ability of female black ducks to select brood-rearing ponds that will fulfill the changing food and habitat requirements of growing ducklings increases the prospects for brood survival. Many researchers (V. D. Stotts, unpubl. rep., Maryland Game and Inland Fish Comm., Fed. Aid Proj. W-30-R-7, 1959; Reed 1970; Hepp and Hair 1977) have recognized the importance of rearing areas and have identified the types of wetlands used by black duck broods. However, comparatively little is known about the wetland habitat components that serve as proximate factors (Hilden 1965) in the selection process. Studies of primary and secondary brood movements (Evans et al. 1952, Berg 1956, Stewart 1958, Beard 1964, Ball 1973) indicate that ducks actively select rearing areas. Wright (1954) observed a black duck hen leading her brood >3.2 km overland to a rearing area, and Young (1967) discussed the overland movements of 2 broods to a large pond >1.6 km away. Detailed data on brood movements of inland-nesting black ducks have been unavailable, but are necessary for an understanding of duckling survival and habitat selection. We investigated the selection of wetlands and habitat components by radio-marked and unmarked female black ducks, and the movements of marked, brood-rearing hens. We acknowledge the work of field assistants J. M. Connolly, J. D. Schoultz, J. Sease, and S. E. Staples. T. J. Dwyer, L. D. Flake, and G. M. Haramis provided comments on early drafts of the manuscript. Financial support for this study was provided in part by contract #14-160008-2125 from the Migratory Bird and Habitat Research Laboratory, U.S. Fish and Wildlife Service, and by the School of Forest Resources, University of Maine, Orono. STUDY AREA AND METHODS The 151-km2 study area was 30 km southwest of Bangor, Maine in the northern hardwoods-spruce ecoregion (Bailey 1978). Forty-four of the 118 study wetlands were created or modified by beaver. Most ponds were classified as EverI Present address: Colorado Division of Wildlife, Research Center, 317 W. Prospect, Fort Collins, CO 80526. J. Wildl. Manage. 46(3):1982 615 This content downloaded from 157.55.39.147 on Sun, 07 Aug 2016 07:03:12 UTC All use subject to http://about.jstor.org/terms 616 BROOD-REARING BLACK DUCKS* Ringelman and Longcore green or Deciduous Scrub-Shrub, Deciduous Forested, or Persistent Emergent wetland types (Cowardin et al. 1979), and ranged in size from 0.1 to 64.2 ha. A detailed description of the study area was presented by Ringelman (1980:3). Data on brood movements were attained by following radio-marked females which were captured before brood-rearing with nest traps (Coulter 1958) or rocket nets. Females were instrumented with 25-g, back-mounted transmitters and monitored 2-3 times per day with a mobile, null-peak tracking system using conventional telemetry techniques. Observations of unmarked broods, supplemented with telemetry records of radio-marked hens, provided data that were used in the analysis of habitat selection. Rearing ponds were defined as wetlands used by a brood for >24 hours. Brood surveys were conducted on all rearing ponds every 6-10 days from the last week in May through the end of July 1977-80. Most observations were made from tree stands 3-15 m above the water's surface. Black duck broods were characterized by number and age of ducklings (J. B. Gollop and W. H. Marshall, unpubl. rep., Miss. Flyway Tech. Sect., 1954), because a unique combination of these characteristics enabled us to re-identify individual broods during subsequent visits. Details of our brood survey methods are described elsewhere (Longcore and Ringelman 1980). The hypothesis that brood use differed significantly from the availability of each wetland class was tested with a z statistic (Neu et al. 1974). A stepwise forward discriminant analysis (Klecka 1975) was performed to identify individual habitat components characteristic of wetlands used vs. those avoided by broods. Twenty-eight habitat components (Table 1) were used as discriminating variables. Preliminary analyses indicated that variable means were proportional to standard deviations, so a logarithmic transformation (log[x + 1]) was applied to all data (Zar 1974:184) to avoid problems created by heteroscedasticity and non-normality (Box and Cox 1964, Eisenbeis and Avery 1972).
Read full abstract