Abstract During the summers of 2000–2002, we used radio telemetry to document Eastern Prairie Population (EPP) Canada goose (Branta canadensis interior) brood movements and use of brood-rearing habitat. We compared these data with similar data collected in 1976–1978 (Didiuk 1979), prior to a significant increase in the size of the midcontinent light goose (lesser snow geese [Chen caerulescens] and Ross's geese [C. rossii]) population and consequent habitat alteration near Cape Churchill, Manitoba. Since the late 1970s, use of traditional EPP Canada goose brood-rearing areas by light geese has increased significantly near Cape Churchill, and the density of nesting EPP Canada geese has declined. Alteration of brood-rearing habitat has been hypothesized as a cause of the decline in EPP breeding density, as natal dispersal to more distant brood-rearing areas may influence future recruitment into the local breeding population. In 1976–1978, 20 (95%) of 21 radio-marked broods nesting in beach ridge/sedge meadow habitat moved to salt marsh brood-rearing areas; however, only 5 (19%) of 27 Canada geese, nesting in the same habitat, made initial movements to these traditional salt marsh brood-rearing areas in 2000–2002. In 2000–2002, 30 (75%) of 40 geese with broods made initial movements to beach ridge/sedge meadow habitat—10 of these broods eventually moved to salt-marsh habitats later in the brood-rearing period (χ̄ date = 22 days postmedian hatch). Mean brood home range size from 2001–2002 in coastal and inland habitats nearly doubled compared to the mean brood home range size during 1976–1978. Eastern Prairie Population Canada geese currently use brood-rearing habitat other than the coastal salt marshes they used prior to habitat alteration resulting from foraging by light geese. A shift in the use of brood-rearing habitat could potentially reduce nest densities on the study area if first-time breeders nest closer to distant brood-rearing areas. The impact of alternative brood-rearing habitat on gosling growth and survival for EPP geese is unknown, but foraging in poorer quality brood-rearing habitat may also contribute to the observed decline in nesting density.

We used radiotelemetry to examine foraging habitat preferences of 17 breeding, male northern goshawks (Accipiter gentilis) in Minnesota from 1998-2000. We assessed habitat preference using radio relocation points and 50-m radius buffers of radio relocation points. Our data Suggested that foraging male goshawks used early-successional upland conifer stands (>= 25 yrs old), early-successional upland deciduous stands (>= 50 yrs old), late-successional upland conifer stands (>= 50 yrs old), and late-successional upland deciduous stands (>= 50 yrs old) more frequently than expected based on the abundance of these vegetation types in the landscape. The 2 most available stand types, early-successional upland deciduous (< 25 yrs old) and all ages of late-successional lowland conifer stands, were used less than expected by foraging goshawks. Late-successional lowland deciduous stands (>= 50 yrs old) were used in proportion to availability. Although analysis of relocation points suggested early-successional upland deciduous stands (25-49 yrs old) and late-successional upland conifer stands (>= 50 yrs old) were used in proportion to availability, analysis of buffers around relocation points indicated that these stand types were also used more than expected by foraging goshawks. Regardless of vegetation community type, stands used by goshawks were structurally similar with high canopy and understory stem densities, high canopy closure, substantial shrub cover, and large amounts of woody debris. Nest stands consisted of taller and larger diameter canopy trees and fewer understory trees than foraging stands, but stands were otherwise similar in structural features, Suggesting goshawks used similar stands for nesting and foraging but that they tended to select the most mature stands for nesting. A commonality among nesting and foraging stands was the presence of open spaces between the canopy and understory foliage, and between understory and shrub layer foliage. In our study area, these spaces may have served as relatively unobstructed flight paths where foraging and nesting stands possessed stem densities at the tipper end of that reported for goshawk habitat.

--We used video-recording systems to collect diet information at 13 Northern Goshawk (Accipiter gentilis) nests in Minnesota during the 2000, 2001, and 2002 breeding seasons. We collected 4871 hr of video footage, from which 652 prey deliveries were recorded. The majority of prey deliveries identified were mammals (62%), whereas birds (38%) composed a smaller proportion of diet. Mammals accounted for 61% of biomass delivered, and avian prey items accounted for 39% of prey biomass. Sciurids and leporids accounted for 70% of the identified prey. Red squirrel (Tamiasciurus hudsonicus), eastern chipmunk (Tamias striatus), and snowshoe hare (Lepus americanus) were the dominant mammals identified in the diet, while American Crow (C0rvus brachyrhynchos) and Ruffed Grouse (Bonasa umbellus) were the dominant avian prey delivered to nests. On average, breeding goshawks delivered 2.12 prey items/d, and each delivery averaged 275 g for a total of 551 g delivered/d. However, daily (P < 0.001) and hourly (P = 0.01) delivery rates varied among nests. Delivery rates (P = 0.01) and biomass delivered (P = 0.038) increased with brood size. Diversity and equitability of prey used was similar among nests and was low throughout the study area, most likely due to the dominance of red squirrel in the diet.

Compared to other regions of North America, little information exists regarding Northern Goshawk (Accipiter gentilis) ecology and population dynamics in the western Great Lakes Region. We examined productivity and nesting habitat characteristics of goshawks in Minnesota from 1998-2001. Apparent nesting success varied annually from as low as 38% to as high as 83%. The Mayfield estimate of daily survival for nests was 0.992 +/- 0.002 (SE). The mean fledgling number across years was 1.85 +/- 0.14 for successful nests and 1.14 +/- 0.17 for all nesting attempts. Twenty-one percent of all nesting attempts failed, primarily due to predation or suspected predation (52%) and inclement weather (35%). Overall, productivity of goshawks in Minnesota was at the lower end of the range reported in other studies across western North America, which is not atypical for peripheral populations. During the 3-yr study, we recorded mortalities of nine (four males and five females; eight radio-marked and one unmarked) adult goshawks-causes of mortality were avian (33%) and mammalian (22%) predation, human persecution (22%), and unknown causes (22%). Fifty-six percent of mortalities occurred during the breeding season, and 44% occurred during the winter. Based on radiotelemetry data, we estimated adult annual survival to be 74 7.8%, which is similar to survival estimated using mark-recapture analysis in three western North America studies.

The U.S. Fish and Wildlife Service (FWS) was petitioned in 1997 to consider listing Northern Goshawks (Accipiter gentilis atricapillus) under the Endangered Species Act of 1973, west of the 1001 meridian of the contiguous United States. In their 12-mo finding issued in June 1998, the FWS determined that listing this population as threatened or endangered was not warranted and based that decision on review of existing population and habitat information. Because the status of goshawks in the western U.S. continues to be contentious and the FWS finding has been challenged, the Raptor Research Foundation, Inc. and The Wildlife Society jointly formed a committee to review information regarding the status of the goshawk population in the contiguous U.S. west of the 100(th), meridian. The committee was requested to: (1) determine if there is evidence of a population trend in goshawks in the western U.S., excluding Alaska; (2) determine if there is evidence that goshawks nesting in the eastern and western U.S. represent distinctive, genetically unique populations; and (3) evaluate evidence for goshawk-habitat relations, including any association with large, mostly-unbroken tracts of old growth and mature forests. Based on existing information, the committee concluded: (1) existing data are not adequate to assess population trend in goshawks west of the 100(th) meridian; (2) existing analyses of phylogeography have not provided evidence of genetic differences among recognized (atricapillus, laingi) or putative (apache) subspecies, and the genetic distinctness of atricapillus goshawks in western and eastern North America is not known; and (3) at present, assessing the status of goshawks solely using distribution of late-successional forests is not appropriate, based on the current understanding of goshawk-habitat relations, although goshawks clearly use and often select late-successional forests for nesting and foraging. We provide recommendations on information needs to assess status and population trend of goshawks in the western U.S.

We used radio-telemetry to estimate breeding season home-range size of 17 male and I I female Northern Goshawks (Accipiter gentilis) and combined home ranges of 10 pairs of breeding goshawks in Minnesota. Home-range sizes for male and female goshawks were 2593 and 2494 ha, respectively, using the minimum convex polygon, and 3927 and 5344 ha, respectively, using the 95% fixed kernel. Home ranges of male and female members of 10 goshawk pairs were smaller than combined home-range size of those pairs (mean difference = 3527 ha; 95% CI = 891 to 6164 ha). Throughout the nonbreeding season, the maximum distance from the nest recorded for all but one goshawk was 12.4 km. Goshawks breeding in Minnesota have home ranges similar to or larger than those reported in most other areas. Home-range overlap between members of breeding pairs was typically less than or equal to50%, and both members of breeding pairs were associated with breeding home ranges year round. Goshawk management plans based on estimated home-range size of individual hawks may substantially underestimate the area actually used by a nesting pair.

Least Flycatchers (Empidonax minimus) often aggregate their territories, even though advantages of clustering are not apparent. In north-central Minnesota we investigated four potential reasons for clustered nesting in Least Flycatchers: (1) predator deterrence, (2) preferred habitat features, (3) greater food availability, and (4) competitive exclusion. In 1995 and 1996 we compared arthropod abundance, forest structure, and plant species composition inside and outside nine Least Flycatcher clusters. We also compared the response of forest birds to the presentation of a predator (Broad-winged Hawk [Buteo platypterus]), and used point counts to detect any exclusion of potential predators or competitors from the clusters. The predator deterrence hypothesis received the strongest support. In response to hawk presentations inside the clusters, 2.8 times more birds (all species combined) gave 5.8 times more alarm calls than outside, with differences due entirely to the presence of Least Flycatchers. In addition, birds responded to the hawk more quickly inside clusters. Minor differences in habitat seemed insufficient to produce such tight clustering. Neither all arthropods combined nor any of the five orders most frequently caught in sticky traps indicated greater food availability inside the clusters. Black-throated Green Warblers (Dendroica virens) were detected 2.2 to 5.4 times less frequently inside the clusters than outside, but they probably do not compete to a large degree with Least Flycatchers. The dramatic response of Least Flycatchers to the hawk presentations provides the first evidence that antipredator behavior may contribute to the clustering of their territories.

We used a remote video recording system and direct observation to quantify provisioning rate and adult and nestling behavior at Bald Eagle (Haliaeetus leucocephalus) nests in north-ceutral Wisconsin in 1992 (N = 5) and 1993 (N = 8). Eagles nesting in this region have a high reproductive rate (greater than or equal to1.3 young/occupied territory), and the number of occupied territories has expanded nearly threefold since 1980. The season-long provisioning rate averaged 5.2 prey deliveries/nest/d and 3.0 prey deliveries/nestling/d, and did not vary by year or with nestling number or age. Fish (Osteichthyes) made up 97% of identified prey deliveries followed by reptiles (Reptilia) (1.5%), birds (Aves) (1.2%), and mammals (Mammalia) (0.6%). Nearly 85% of prey items were >15 cm and <45 cm and 13% were <15 cm in length. Adult attendance (time greater than or equal to1 adult was at the nest) at nestling age 2-4 wk was >90% of the day and was negatively correlated with nestling age. Time adults spent feeding nestlings was negatively correlated with nestling age. Nestlings stood or sat in the nest >30% of the day, began to feed themselves, and exhibited increased mobility in the nest at 6-8 wk. We identified three stages of the nestling period and several benchmarks that may be useful when scheduling data collection for comparison of Bald Eagle nesting behavior. Our results support the hypothesis that food was not limiting this breeding population of Bald Eagles.

Forest fragmentation has been implicated as a cause of population declines of several Neotropical migrant bird species. Fragmentation increases the amount of habitat edge, and reduced nesting success rates near forest edges are well documented in agricultural landscapes ("edge effects"). However, edge effects in predominantly forested landscapes, particularly those related to timber harvest, are poorly understood. This study examines nesting success of ground nesting birds in relation to clearcut edges in a forest-dominated landscape in north-central Minnesota. A total of 383 nests of seven species of ground nesting birds were found and monitored during 1992-1998. Ovenbird (Seiurus aurocapilus; n = 318) and Hermit Thrush (Catharus guttatus; n = 44) nests composed the majority of the sample. Predation accounted for 94% of all nest failures. Brown-headed Cowbird (Molothrus ater) parasitism was low (1.8% for all ground nests). Using proportional hazards regression, distance to nearest clearcut edge was the best predictor of nest failure. For all ground nests, nesting success was 0.18 at 0-100 m, 0.39 at 101-500 m, and 0.52 at 501-954 m from nearest clearcut edge. Source-sink modeling indicated that distances less than or equal to100 m from clearcut edges were sink habitats for Ovenbirds (i.e. recruitment was lower than survival). These results provide strong evidence of a negative edge effect on ground nests, extending 100 m or more from clearcut edges in a forest-dominated area of north-central Minnesota.

The effects of edge and fragmentation on avian nesting success are well documented in agricultural landscapes. However, it is unclear whether these effects are common in predominantly forested landscapes. In particular, edge and fragmentation effects caused by clearcutting are poorly understood. To better understand this problem, we examined the relation of nesting success to clearcut edges in north-central Minnesota. We found elevated predation rates near clearcut edges using artificial and natural ground nests. To aid interpretation of studies conducted in the Northern Hardwood- Conifer Forest Region (NHCE;) of North America, we estimated statistical power for 26 analyses (subsets of 1 1 papers and our own analyses) and, where possible, we reana- lyzed the data and estimated effect size (with associated confidence intervals). In addi- tion, we examined design issues such as presence of pseudoreplication. Statistical power was low for many of the studies and pseudoreplication was evident in several. Without considering power or design issues, 13 of the analyses found edge effects, 12 showed no effects, and one yielded greater predation rates in unfragmented versus frag- mented areas (oc=0.05). When we excluded studies with low statistical power (<0.80) and pseudoreplication from analysis, 10 of the remaining studies reported edge effects, 3 showed no effects, and one showed greater predation rates in unfragmented versus frag- mented areas. Variability in edge-effect results may be due in part to variability in strength of study design. Previous evaluations have suggested that edge effects are found mainly in agricultural landscapes, but our results suggest that these effects also may be common in extensively forested NHCF landscapes. Within extensive but man- aged forests of this region, relatively large, contiguous patches of mature forest, unfrag- mented by clearcutting, may be required to conserve some forest interior bird species.

Determining whether nesting attempts are successful can be difficult. Yet, current protocols for estimating nesting success do not address how uncertain nest fates should be handled. We examined the problem of nest-fate uncertainty as it relates to Mayfield estimation of nesting success and in analyses of factors that influence success. We used data from Minnesota to illustrate the potential effect of uncertain fate; 40% of Ovenbird (Seiurus aurocapillus; n = 127) nests and 30% of Least Flycatcher (Empidonax minimus; n = 144) nests had uncertain fates. How this uncertainty is incorporated into Mayfield estimates of success varied widely among researchers. In a survey of researchers who use the Mayfield method, 9 of 22 respondents (of 40 contacted) excluded nests with uncertain fate. Excluding uncertain fates is counter to how Mayfield first described his estimator and can result in severe downward bias. The remaining respondents (59%) included nests with uncertain fate but varied in how they terminated the exposure period. We developed a simulation model that calculated Mayfield estimates using different approaches and compared them with a known rate of nesting success. Magnitude of bias in Mayfield estimates varied considerably in our simulations. The approach with the least bias terminated exposure with the last observed active date for nests with uncertain fate, and with the midpoint between last observed active and first observed inactive dates for nests with known fate. In addition, information necessary to interpret and compare Mayfield estimates often is not reported. These values, including variance estimates and the period lengths used to estimate survival rates, should be reported with Mayfield estimates. Finally, nest fate is commonly used as a categorical variable in studies of factors affecting nesting success. In this approach, however, nests with uncertain fate must be excluded. An alternative approach is Cox regression, which incorporates nests with uncertain fate.

Abstract We evaluated characteristics at Red-shouldered Hawk (Buteo lineatus) nest sites at two study areas with different topography and forest types in north-central and central Minnesota to identify nest site commonalities across geographically distinct areas. During the breeding seasons of 1994–1995, we located nests of Red-shouldered Hawks at the Camp Ripley Army National Guard Training Site and the Chippewa National Forest using a combination of broadcast surveys, helicopter searches, and systematic foot searches. All 38 nests at Camp Ripley and 18 nests in the Chippewa National Forest were in upland hardwood stands; the remaining two nests in the Chippewa National Forest were in aspen (Populus spp.) stands. We aged cores from 19 nest trees at Camp Ripley and measured habitat characteristics in a 0.04 ha circle centered on each nest tree and at a paired random site within the nest stand. We compared habitat variables at nest and random sites to identify habitat characteristics that were consistent predictors of nest sites versus random sites for each study area and for all nests combined. Compared to random sites, nest sites in the Chippewa National Forest had larger diameters at breast height (dbh) of the nest tree, taller nest tree height, and higher canopy height. At Camp Ripley, nest sites differed from random sites with regard to many more variables; nests were located in portions of the stand with larger trees and closer to surface water. Nest trees ranged in age from 50–89 years. Logistic regression models indicated that, for both study areas combined, nest tree dbh, basal area, canopy height, and distance to water were the most important variables in distinguishing nest sites from random sites.

Previous work has shown that the rate at which Brown-headed Cowbirds (Molothrus ater) parasitize forest nesting birds is affected by the proportion of a local landscape that is forested. However, much of the previous work has been restricted to a relatively small part of the cowbird's range, and has looked at forest coverage in very restricted areas around study plots. We used data from a wider geographical area, the entire width of the United States, and examined forest coverage in relatively large areas (10-km and 50-km radii) around study plots to determine if forest coverage is a generally useful statistic for predicting rates of brood parasitization. As was found in previous studies, we showed that increased amounts of forest coverage within 10 km of an area resulted in lower rates of parasitization by cowbirds. This pattern held not only among widely separated sites, but also within local clusters of study plots. However, we found that increased amounts of forest within 50 km of a study site resulted in slightly increased rates of parasitization in sites west of the Great Plains, contrary to previous research findings. Forest structure, as indicated by the relationship between forest coverage and other measures of forest distribution and abundance, differed across the United States. However, differences in forest structure were not obviously related to differences in the manner that parasitization and forest coverage covaried from east to west across the continent. Even given the variable patterns found, management for higher proportions of forest within 10-km radius areas should result in decreased rates of parasitization of host species; however, the impact of such a management strategy will vary across the continent.

We studied a heavily-hunted population of black bears (Ursus americanus) on the periphery of the bear range in east-central Minnesota in 1991 and 1992. This was one of the few areas in Minnesota where hunting pressure was not controlled by a quota on the number of hunting licenses. We hypothesized that the area supported high harvest levels because it was a population sink supplied by seasonal migrants, dispersing subadult male immigrants, or both from northern Minnesota. However, we captured 7 female and 5 adult male bears during late summer and from radiotracking found that all were residents of the study area. Also, the following evidence indicated that immigration of young males was not sustaining the population: (1) males made up a similar fraction of the harvest in the study area (53.6%) as statewide (53.1%); (2) the rate of decline in the ratio of males to females in each harvested age class, due to high harvest mortality that depleted male numbers faster than females, showed no sign of being retarded by an influx of males; and (3) yearling males, an age group not heavily represented among dispersers, composed a high proportion of the harvest. A simple deterministic model suggested that the population could remain stationary or grow with current harvest pressure. Thus, the area was not a population sink.

Little information is known about the ecology of ferruginous hawks (Buteo regalis) in winter versus the breeding season and less about how the species adapts to fragmented grassland habitats. Accordingly, we studied the behavior of 38 radiotagged ferruginous hawks during 3 winters from 1992 to 1995. We used 2 adjacent sites in Colorado that were characterized by low and high levels of anthropogenic influence and habitat fragmentation: the Rocky Mountain Arsenal National Wildlife Refuge RMANWR; low-level influence), and several adjacent Denver suburbs (high-level influence). Relative abundance of ferruginous hawks differed by treatment area and year (P < 0.001); hawks were most numerous where black-tailed prairie dogs (Cynomys ludovicianus) were most plentiful. Daily Minimum Convex Polygon (MCP) home range areas did not differ (P = 0.28) for RMANWR ((x) over bar = 4.71 km(2), SE = 1.33, n = 25) and suburban hawks ((x) over bar = 2.30 km(2). SE = 0.50, n = 13). The number of perches occupied per day between the sites was not different (P = 0.14), but hawks at RMANWR used pole and ground perches more frequently and for a greater portion of the daily time budget (P < 0.05). Hawks at RMANWR spent less time roosting after sunrise ((x) over bar = 61 min) than did suburban hawks (x) over bar = 138 min: P = 0.004) and spent less time roosting during the day (RMANWR = 100 min; suburb = 189 min; P = 0.009). Prey acquisition and associated intra-and interspecific interactions were not different (P > 0.05) at RMANWR and suburban sites. Ferruginous hawks appear to modify their behavior in fragmented, largely human-altered habitats, provided some foraging habitats with adequate populations of suitable prey species are present.

Forest-nesting raptors are often difficult to detect and monitor because they can be secretive, and their nests can be difficult to locate. Some species, however, respond to broadcasts of taped calls, and these responses may be useful both in monitoring population trends and in locating nests. We conducted broadcast surveys on roads and at active red-shouldered hawk (Buteo lineatus) nests in northcentral Minnesota to determine effects of type of call (conspecific or great horned owl [Bubo virginianus]), time of day, and phase of the breeding cycle on red-shouldered hawk response behavior and to evaluate usefulness of broadcasts as a population monitoring tool using area occupted-probability-of-detection techniques. During the breeding seasons of 1994 and 1995. we surveyed 4 10-station road transects 59 times and conducted 76 surveys at 24 active nests. Results of these surveys indicated conspecific calls broadcast prior to hatch and early in the day were the most effective method of detecting red-shouldered hawks. Probability of detection via conspecific calls averaged 0.25, and area occupied was 100%. Computer simulations using these field data indicated broadcast surveys have the potential to be used as a population monitoring tool.

Reproductive rate (productivity) of bald eagles (Haliaeetus leucocephalus) nesting on the shores of Lake Superior was significantly less than that of neighboring eagles nesting in inland Wisconsin (1.0 vs. 1.3 young per breeding attempt, 1989–1993), and at other inland lake/riverine habitats in the Great Lakes Basin. It is possible that the current causes of low productivity on Lake Superior might include exposure to organochlorine contaminants and/or low food availability. Levels of dichloro-diphenyl-dichloroethylene (DDE) and total polychlorinated biphenyls (PCBs) in addled eggs and eaglet blood from Lake Superior and inland Wisconsin reference sites were measured. Food delivery rates by parent eagles to nestlings, a possible index to food availability, were quantified at both locations. Concentrations of both DDE and total PCBs in addled eggs declined significantly from 1969 to 1993 (p<0.001, p=0.006 respectively), and current concentrations of DDE are at or below the no observable adverse effect level (NOAEL) for reproductive impairment. Concentrations of DDE and total PCBs in plasma were greater in individual nestlings from the shores of Lake Superior than in nestlings at inland locations (18.9 μg/kg vs. 3.0 μg/kg DDE, p<0.001, and 109.1 μg/kg vs. 42.6 μg/kg, p=0.002), but were not correlated to the 5-year average history of productivity for the territory (p>0.05). Food delivery rates by parent eagles to nestlings at Lake Superior were 56% lower than those to inland nestlings (2.16 vs. 4.87 prey items per day, p=0.002). Food delivery rates were significantly correlated to average 5-year productivity for inland Wisconsin reference sites (p<0.001, r2=0.90), although not for Lake Superior sites (p=0.593). It is concluded that it is likely that the current low productivity of Lake Superior eagles is at least partly attributable to low food availability, but some other factor, possibly PCBs, may also contribute to low productivity.

To determine habitat use and movements of male Boreal Owls (Aegolius funereus) in northeast Minnesota, we monitored 10 radio-equipped owls from 1990-1992. We used mist nets, bal-chartris, and the taped playback recording of the primary song of the male Boreal Owl to trap territorial male owls during the springtime breeding season.

We conducted nocturnal auditory surveys from 1987-1992 to determine the distribution, abundance, and habitat use of Boreal Owls (Aegolius funereus) in northeast Minnesota. We concentrated our efforts in areas where documented nesting attempts by the owls had occurred, along roadways maintained for winter-time access by motor vehicles, and by traversing all habitat types found within the study area.

During January 1986-September 1988 we studied the behavioral responses of 71 radiocollared mule deer (Odocoileus hemionus) to military activity on the Pinon Canyon Maneuver Site in southeastern Colorado. Military training was initiated on the site during August 1985 and recurred about 3 times yearly for periods of one month. During a maneuver, 3/7 of the site was used for training in accordance with a rotational land use schedule. During the nonsummer seasons, female seasonal convex polygon and harmonic mean home ranges were larger in maneuver and previous-maneuver areas than nonmaneuver areas (P < 0.002). During summer, female convex polygon home ranges were larger in maneuver than nonmaneuver areas (P = 0.066). Fawn summer home ranges were larger in maneuver than previous-maneuver areas (P < 0.01). Male home range sizes differed only for 50% harmonic mean transformation annual home ranges (P = 0.056); bucks in maneuver areas had larger home ranges than in nonmaneuver areas. Female deer in maneuver areas exhibited significant home area shifts (P = 0.049) between premaneuver and maneuver periods more frequently (40.0%) than did deer in nonmaneuver (control) areas (12.5%). Mule deer in military training areas may have responded to human harassment, alteration of security cover, or destruction of the forage base. We suggest that deer may respond more intensely to unpredictable than predictable human activity.