Donald H. Rusch

RESPONSE OF GREAT HORNED OWL POPULATIONS TO CHANGING PREY DENSITIES1

From 1966 through 1969, we measured ie numbers, productivity, and food habits of great horned owls (Bubo virginianus) on an area of 62 square miles near Rochester, Alberta. During these years, there were sevenfold, twofold, and threefold increases in the numbers of snowshoe hares ( Lepus americanus), ruffed grouse (Bonasa umbellus), and sharp-tailed grouse (Pedioecetes phasianellus) re- spectively. The owls responded to these chzanges in numbers of prey by an increase in numbers from io to 18, by an increase from 20 to 100 percent in the proportion nesting, and by an increase in the percentage of snowshoe hare biomass in the diet from 23 percent in 1966 to 50 percent in 1969. The proportion of sharp-tailed grouse in the owl diet also increased from none in 1966 to 10 percent in 1968 but decreased to 3 percent in 1969. The proportion of ruffed grouse in the diet decreased each year, from 23 percent in 1966 to none in 1969. The diets of owl families were related to the vegetative cover surrounding the nests and presumably to the composittion o£ the prey base. Snowshoe hares, waterfowl, and pocket gophers (Thomomys talpoides) were the pnncipal components of the os^71 diet. The quantitative impact of owl predation on spring populations of snowshoe hares and ruffed grouse appeared to be low ( between 0 and 7 percent). Although 60 percent of the hares alive in spring were young-of-the-year, 97 percent of the hares in ie owl diet were adults. Similarly, only 3 of 26 grouse in the owl diet were females, despite balanced sex ratios in the spring grouse populations. The continued increase in numbers of snowshoe hares coincided with decreased predation rates on ruffed grouse and decreased incidence of sharp-tailed grouse in the owl diet, as owls consumed relatively more snowshoe hares. Abstract: From 1966 through 1969, we measured ie numbers, productivity, and food habits of great horned owls (Bubo virginianus) on an area of 62 square miles near Rochester, Alberta. During these years, there were sevenfold, twofold, and threefold increases in the numbers of snowshoe hares ( Lepus americanus), ruffed grouse (Bonasa umbellus), and sharp-tailed grouse (Pedioecetes phasianellus) re- spectively. The owls responded to these chzanges in numbers of prey by an increase in numbers from io to 18, by an increase from 20 to 100 percent in the proportion nesting, and by an increase in the percentage of snowshoe hare biomass in the diet from 23 percent in 1966 to 50 percent in 1969. The proportion of sharp-tailed grouse in the owl diet also increased from none in 1966 to 10 percent in 1968 but decreased to 3 percent in 1969. The proportion of ruffed grouse in the diet decreased each year, from 23 percent in 1966 to none in 1969. The diets of owl families were related to the vegetative cover surrounding the nests and presumably to the composittion o£ the prey base. Snowshoe hares, waterfowl, and pocket gophers (Thomomys talpoides) were the pnncipal components of the os^71 diet. The quantitative impact of owl predation on spring populations of snowshoe hares and ruffed grouse appeared to be low ( between 0 and 7 percent). Although 60 percent of the hares alive in spring were young-of-the-year, 97 percent of the hares in ie owl diet were adults. Similarly, only 3 of 26 grouse in the owl diet were females, despite balanced sex ratios in the spring grouse populations. The continued increase in numbers of snowshoe hares coincided with decreased predation rates on ruffed grouse and decreased incidence of sharp-tailed grouse in the owl diet, as owls consumed relatively more snowshoe hares.

Characteristics of winter feeding aggregations of ruffed grouse in Alberta

Winter feeding aggregations of ruffed grouse (Bonasa umbellus) were studied near Roch- ester, Alberta. The annual onset of budding by grouse in fall commenced with the first snow cover. During two winters, aggregation size increased through mid-January, after which there was a significant linear decrease averaging nearly 50 percent by early April. The decrease in group size after mid-January was probably effected chiefly by fragmentation resulting from increasing intragroup strife. Sex and age structure of aggregations was not significantly different from that expected by random combinations. Ruffed grouse fed in male aspen (Populus tremuloides) or in willow (Salix spp.) for an average of 16 minutes during morning, usually leaving at sunrise. Feeding began in evening about 52 minutes after sunset, and lasted an average of 24 minutes. The mean age of 100 male aspen utilized by grouse near Rochester was 36 years. Tree densities at 12 feeding sites were significantly higher than those at systematic-grid stations, drumming logs, and kill sites within aspen stands at Rochester. Buds and twigs of aspen and willow comprised about 80 percent (by volume) of crop contents of 148 grouse shot while budding. Nutrient analyses of male aspen buds indicated that ruffed grouse ingested those buds having the highest protein and potassium contents. Willow buds from grouse crops contained more protein (14.0 percent) than aspen buds from two other groups of crops (12.9 and 11.7 percent).

Seasonal and Annual Trends in Numbers of Alberta Ruffed Grouse

Estimates of spring populations of ruffed grouse (Bonasa umbellus) on a 4-square-mile area (Main Area) near Rochester, Alberta, increased from 112 grouse in 1966 to 124 in 1967 and to 164 in 1968. Data from smaller study areas and provincewide questionnaires showed similar trends. In 1966 and 1967, grouse populations declined rapidly throughout November, but numbers were stationary between December and the following spring. Approximately 36, 42, and 23 percent of the males present in 1966-68 did not establish territories. The numbers of grouse produced (308 and 330) were similar in 1966 and 1967, and survival of juveniles ( 12 months of age). Predation on grouse, mainly by great horned owls (Bubo ?irginianus), lynxes (Lynx canaderesis), and goshawks (Accipiter gentilis) accounted for more than 25 percent of the annual mortality and for more than 80 percent of fall-to-spring mortality each year. In the fall, predation was selective toward males and young but not toward smaller versus larger grouse. As populations of snowshoe hares (Lepus amertcanus) increased rapidly between 1966 and 1969, great horned owls consumed relatively more snowshoe hares and relatively fewer ruffed grouse. It was this shift in predator food habits that allowed higher survival of young ruffed grouse and was thus responsible for increased numbers of grouse in the spring. Marshall and Gullion 1963, Gullion and Marshall 1968, and others ) have implicated predation as an important proximate source of mortality, but the limiting or regulatory effects of predation on these grouse populations were not clear. In western Canada, where the 10-yearcycle is best known and most pronounced ( Keith 1963), little is known of population dynamics of ruffed grouse and the role of predation in the grouse cycle. IThis paper documents the dynamics of an increasing ruffed grouse population at Rochester, A1berta, during 196S68 and quantifies and evaluates the impact of predation upon numbers of ruffed grouse. We acknowledge D. H. Wood, C. Bitzer, G. Saunders, N. Saunders, R. Saunders, S. Saunders, E. C. Meslow, P. D. Doerr, and C. Cameron for field assistance and thank 803 Islany past studies of ruffed grouse were concerned with periodic or cyclic fluctuations in numbers, and hypotheses concerning these fluctuations are legion (Keith 1963). Two long-term population studies, the ruffed grouse investigation in New York ( Bump et al. 1947 ) and the Minnesota studies at Cloquet (King 1937, Eng 1959, 1 This paper is a contribution of the Department of Wildlife Ecology, University of Wisconsin, and the Research Council of Alberta. The study was supported by the Research Council of Alberta; the University of Wisconsin, College of Agricultural and Life Sciences, and Research Committee of the Graduate School; the Alberta Department of Lands and Forests, Fish and Wildlife Division; the A1berta Department of Agriculture, Veterinaxy Services Branch; the National Science Foundation (Grants GB-615(3 and GB-7744); and the Wildlife Management Institute. 2 Present address: Manitoba Department of Mines, Resollrces and Environmental Management, NVinnipeg. This content downloaded from 157.55.39.223 on Wed, 24 Aug 2016 05:33:10 UTC All use subject to http://about.jstor.org/terms 804 Journal of Wildlife Management, Vol. 35, No. 4, October 1971 was comprised of 9 percent open farmland, 20 percent burn, 27 percent aspen woods, 11 percent spruce woods, and 33 percent open bog. These cover types are largely self-descriptive, but plant communities and history of recent fires on the Main Area are also described in detail by Rusch et al. 1971. A 330x 660-fo+at grid was surveyed and marked on this area, and the grid intersections served as reference points for locations of traps and grouse. Drumming males were trapped on four smaller areas in the springs of 1966-68. The Landing-trail ( 69 acres ), Moores-pinewoods ( 85 acres ), and Halls ( 63 acres ) areas (Fig. 1), also used as study plots for snowshoe hares, are described by Meslow and Keith (1968:813). Upland forest, the primary grouse habitat near Rochester ( Rusch and Keith 1971), covered all of the Landing-trail and Moores-pinewoods areas, and about 40 percent of the Halls Area. The Camp Area (Fig. 1) consisted of 37 acres of mature aspen woods on a northfacing slope of the Tawatinaw River Valley. This area had probably not been burned for 30 years or more, and many aspen, balsam poplar, and scattered white spruce had attained 13 inches or more in diameter. Alder (Alnus crispa and A. rugosa) and willow (Salix spp.) were the dominant shrubs on the Main, Moores-pinewoods, and Halls areas, but beaked hazel (Corylus cornufa) and s askato on ( Ame lanc hi er alnifolia ) were more abundant on the Camp and Landingtrail areas. Drumming males were experimentally removed from Schmidts Area in 1967 and Revols Area in 1968 ( Fig. 1). These areas consisted of 640 acres and 634 acres of predominantly aspen woods. Both were Oll west-facing slopes of the Tawatinaw River Valley, and both had vegetation similar to that on the Camp Area. 9tG STRE M 2 -n g ---O D

Nesting ecology of Rio Grande Turkeys

The nesting ecology of Rio Grande turkeys (Meleagris gallopavo intermedia) was studied on the Rob and Bessie Welder Wildlife Refuge (WWR), Sinton, Texas, during January-August 1983 and 1984. Radio telemetry was used to monitor the nesting activities of turkey hens. Twelve of 31 radio-tagged turkeys died during the study: 8 from predation, 1 from disease, 2 from trapping related injuries, and 1 from unknown causes. An annual hen survival rate of 0.73 was calculated from the total number of observation days and the known number of deaths. Ten (58%) of 17 instrumented hens reached incubation in 1983, but all nests were destroyed by predators. Two turkey broods were observed in 1983, one of 2 hens with 5 poults, the other 2 hens with 3 poults. In 1984, nesting data were gathered on 6 hens which reached incubation; 3 hatched eggs. Seven turkey broods totaling 8 hens with 40 poults were observed in August 1984. Nest sites were located in pastures deferred from livestock grazing. J. WILDL. MANAGE. 51(2):435-439 The population of wild Rio Grande turkeys, which occupies the Aransas River bottoms and surrounding uplands of the WWR, has declined from a high of approximately 700 (Watts 1969) in 1966-67 to <100 birds in 1983. In recent years, no broods have been observed by WWR personnel (J. G. Teer, pers. commun.). Although there have been several studies of the wild turkey on the WWR (Watts 1969, Smith 1977, Baker 1979b), none has directly investigated the nesting ecology of this population. Decreases in nesting attempts, nest success, or poult survival-all of which influence juvenile recruitment-could have played a part in this decline. This study was to determine whether reproductive failures by turkey hens wintering and breeding on the WWR and adjacent areas were responsible for the population decline. Financial support was provided by the Rob and Bessie Welder Wildl. Found. (WWF), Sinton, Tex., and the Coll. Agric. and Life Sci., Univ. Wisconsin, Madison. We gratefully acknowledge the assistance of several WWF students, as well as J. G. Teer, D. L. Drawe, and T. M. Yuill in trapping birds. We thank J. F. Rooke, P. H. Welder, H. Thomas, and N. E. Adams for access to their property. We also wish to thank W. Aschenbeck, of Cuero, Tex., for donating the domestic turkey eggs used in the simulated nest study. This is Welder Wildl. Contrib. 301.

Molt migrant Canada geese in Northern Ontario and Western James Bay

We undertook migration monitoring surveys and analysis of long-term banding data to determine if there was a significant premolt movement of Canada geese (Branta canadensis) from restored and reintro duced populations in southern Canada and midcontinent United States into northern Ontario and western James Bay. We examined migration chronology, origins, and demographic characteristics of molt migration of Canada geese in northern Ontario and on Akimiski Island, Northwest Territories. From 1985 to 1989, a conspicuous northward migration of large Canada geese was documented throughout northern Ontario from midMay to the end of June, well after the April migration of the subarctic nesting subspecies of Canada geese (B. c. interior); most nesting interior Canada geese in the Hudson Bay and James Bay lowlands were incubating eggs at this time. Summer-banded Canada geese originating from populations in 26 states and 6 Canadian provinces were captured in coastal areas of James Bay and Hudson Bay between the borders of Quebec and Manitoba. Morphometric discrimination indicated the presence of molting giant Canada geese (B. c. maxima) Most foreign, summer-banded birds were yearlings (53%) and 2-year-olds (17%), but birds up to 15 years old were captured. Approximately 58% of 2-15-year-old females had brood patches, which indicated a nesting attempt in the year of recapture. We suggest that increasing populations of giant Canada geese and declining habitat availability on northern brood-rearing areas will result in increasing levels of competition between populations of Canada geese. Presence of molt migrants on northern breeding areas will also complicate management of some Arctic and subaretic nesting populations of Canada geese.

Clinal Size Variation in Canada Geese Affects Morphometric Discrimination Techniques

We evaluated morphometric discrimination models designed to use skull length to differentiate between giant Canada geese (Branta canadensis maxima) and interior Canada geese (B. c. interior). We found significant differences in mean skull lengths of interior Canada geese from 3 areas in James Bay (Duncans multiple range test, P < 0.05). Geese decreased in size with increasing latitude, but those from Akimiski Island did not fit this pattern, and were smaller than those from the adjacent mainland. Morphometric discrimination techniques appeared to work adequately in northwest James Bay and on Akimiski Island, but were less reliable in southern James Bay, where the largest interior Canada geese were found. Molt migrant giant Canada geese were apparent in all areas, but most were distinguishable by morphometric methods from interiors at all sites except southern James Bay. Detailed knowledge of geographic variation in body size will improve the usefulness of morphometric discrimination techniques for Canada goose research and management. We recommend the skull length models developed by Moser and Rolley (1990) for use in differentiating giant from interior Canada geese in much of the Mississippi Flyway, except where southern James Bay mainland geese occur. J. WILDL. MANAGE. 61(1):183-190

Broad-Winged Hawk Nesting and Food Habits

COMPARED to the literature on other species of raptors, relatively little has been published on the Broad-winged Hawk (Buteo platypterus). Although clutch size (Burns, 1911) and food habits (May, 1935, and others) are well-documented in this species, little information is available on fledging rates and nesting densities. Similarly, reports on Broadwing ecology in the northern part of their breeding range are scarce. This paper describes the nesting density, productivity, and food habits of a breeding population of Broad-winged Hawks near Rochester, Alberta. Central Alberta, the locale of this study, is on the northwestern fringe of the Broad-winged Hawk range in North America (Burns, 1911; May, 1935; Bent, 1937). Reptiles and amphibians, which are important components of this hawks diet in more southern latitudes, are scarce in central Alberta. We were especially interested in the incidence of these prey items in the Broad-wing diet in our area, and in evaluating the possibility that their scarcity limits the northern distribution of this raptor.

Measurements of Canada Goose Morphology: Sources of Error and Effects on Classification of Subspecies

Subspecific classification of Canada geese (Branta canadensis) based on morphological measurements serves many management and research functions, such as determining harvest pressure on subspecies or estimating the population composition of wintering flocks. Despite this widespread use, the magnitude of error involved in such measurements, the effect of observer experience on measurement error, and the effect of measurement error on classification are not known. To investigate these issues, we carried out a study on Canada geese harvested in Wisconsin involving replicated measurements by observers of different experience levels. Measurement error for experienced observers was half as large as that for inexperienced observers (6-10% vs. 13-21% of all variability for all structures except the tarsus). Experienced observers measured the skull and culmen most precisely, the tarsus least precisely. Consistent differences among observers (observer bias) that could bias classification were smaller for experienced observers. We used reference data and distributional assumptions to estimate that without observer bias or other forms of measurement error, 8-9% of geese measured would be misclassified because of actual size overlap between subspecies. Without observer bias, remaining measurement error among experienced and inexperienced observers increased misclassification by 1% and 2%, respectively. Observer bias can increase misclassification substantially beyond these levels, depending on the magnitude and direction of observer bias and the prevalence of the subspecies. Misclassification of geese resulted in overestimating the prevalence of the less common subspecies in mixed populations, which may be important in developing management strategies. We recommend training observers and standardizing measurement procedures primarily to reduce observer bias that leads to biased classification of geese, and secondarily to reduce other components of measurement error.

Winter Distribution and Affinities of Canada Geese Marked on Hudson and James Bays

Management plans for redistribution of the Mississippi Valley Population of Canada geese (Branta canadensis) in the United States prompted our investigation of the fall and winter distribution, migration chronology, and population affiliation of Canada geese marked on nesting areas in Manitoba and Ontario. During 1976-77, 804 geese were neckbanded. Encounters in 1976 and 1977 (N = 1,418) demonstrated a strong association of geese from the coast of Hudson Bay and the western coast of James Bay with the Mississippi Valley Population. About 50-60% of the marked samples bypassed east central Wisconsin refuges in migrating to wintering areas. Marked individuals from both years were observed at all major goose refuges in east central Wisconsin and southern Illinois. The 8 October mean arrival date in Wisconsin in 1976 was 8 days earlier than that for geese marked in previous years at Horicon. Departure from Wisconsin in 1977 was earlier than in 1976 and was associated with abnormal heavy snowfall and the Horicon goose dispersal program. Age ratios in trapped samples (7.7 immatures/adult female) suggested the inclusion of gang broods. Geese from the same nesting area did not remain together during fall and winter. Although some family groups were identified on migration and winter refuges, the observed mean number of marked immatures per adult female (1.7) was lower than the mean recorded on the nesting areas. J. WILDL. MANAGE. 47(2):307-319 The Mississippi Valley Population (MVP) of Canada geese numerically increased during the past 3 decades, with most of the increased numbers of geese wintering on refuges in Wisconsin, Illinois, and Kentucky (Hankla and Rudolph 1967, Reeves et al. 1968). Large concentrations of geese may have impacts on hunting quality and agricultural operations (K. Klepinger and J. W. Ellis, unpubl. rep., U.S. Dep. Inter., Fish and Wildl. Serv., Minneapolis, Minn., 1975). Because of these impacts and a continuing desire for more geese in southern states, management agencies would like to redistribute some of the MVP from northern wintering areas to smaller, widely scattered southern refuges. If the MVP consists of aggregations of geese with discrete and consistent patterns of migration and refuge use, as suggested by Raveling (1969a) and Kennedy and Arthur (1974) and documented for giant Canada geese (B. c. maxima) by Raveling (1979), selective management could be applied. Development of a management plan for redistribution of the MVP requires data on the migration and distribution of MVP geese. These needs prompted this study. Our objectives were to (1) ascertain the degree of association of geese from sections of the nesting range of the MVP; (2) investigate variation in migration chronology of geese from nesting areas in Manitoba and Ontario, and (3) determine whether samples of geese from the breeding grounds have discrete migration patterns and winter distributions. We thank F. D. Caswell, M. Gillespie, and D. Soprovich of the Manitoba Department of Renewable Resources and K. Brace of the Canadian Wildlife Service for arrangement, support, and assistance in banding operations in 1976. We thank H. G. Lumsden and J. P. Prevett of the Ontario Ministry of Natural Resources for organizing and conducting the banding effort in 1977. J. P. Prevett, D. G. Raveling, and H. C. Hanson critically reviewed the manuscript. J. Wildl. Manage. 47(2):1983 307 This content downloaded from 207.46.13.128 on Fri, 29 Jul 2016 05:17:53 UTC All use subject to http://about.jstor.org/terms 308 MISSISSIPPI VALLEY POPULATION OF CANADA GEESE * Craven and Rusch

Foods of blue-winged teal in two neotropical wetlands

Effective management and conservation of blue-winged teal (Anas discors) require information on foods consumed in main wintering areas. We describe foods eaten by 84 blue-winged teal collected in Cienaga Grande de Santa Marta, Colombia, and Palo Verde Wildlife Refuge, Costa Rica. The volume of foods found in 12 blue-winged teal collected in Palo Verde in 1982-83 consisted of 92% plant and 8% animal material; cultivated rice predominated. In Cienaga Grande, foods found in blue-winged teal included 71% plant material during 1979-80 (n = 10) and 91% animal material (n = 62) in 1985-88 (P < 0.01). Water lily (Nymphaea spp.) seeds were the most common plant item. Snails (Pyrgophorus spp.) and Corixidae insects were the most prevalent animal items