Research in the Parks (Grizzly Bear Population Studies in Glacier National Park, Montana)

Grizzly Bear Population Studies in Glacier National Park, Montana¹

CLIFFORD J. MARTINKA, Glacier National Park, Montana

A population of grizzly bears (Ursus arctos) inhabits Glacier National Park as a native faunal species. Management of the park as a natural area requires that the integrity of this population be preserved as an integral component of park ecosystems. The presence of a grizzly population also requires certain management practices to provide for the safety of park visitors.

A study was conducted from 1967 through 1971 to obtain quantitative data on status, dynamics, habitat relationships, and management of the park's grizzly population. Preliminary results relating to status of the population and management to protect visitors have been reported elsewhere (Martinka 1971). This paper presents additional data on the status and characteristics of the grizzly population. The effects of management on the bear population and the park's status as a natural area are also discussed.

Glacier National Park includes 1583 miles² of mountainous terrain in northwestern Montana (Fig. 1). Topography is characterized by a central core of precipitous peaks and ridges with glacial cirques, moraines, and lakes as prominent features. Extensive talus, persistent snowfields, and remnant glaciers are common. Streams radiate from the park as headwaters of the Saskatchewan, Missouri and Columbia River systems. Elevations vary from approximately 3100 to 10,500 ft.

Park climate is classified as continental, with decided Pacific maritime modifications on the western slopes (Dightman 1967). Weather records from Summit (5213 ft) along the southern boundary are considered to illustrate the general moisture and temperature regime within the park. Mean annual precipitation was 38.29 inches, of which approximately 60% fell as snow from November through April. Annual snowfall of 251 inches resulted in maximum accumulations which frequently exceeded 60 inches. Mean monthly temperature was 35.9°F, with extremes of 15.3°F (January) and 56.7°F (July). Data from other areas show that precipitation increases and decreases, respectively, at higher and lower elevations.

Vegetation of the park has been described by Habeck (1970). In general, habitats include a complex interspersion of climax and seral plant communities. Alpine types occur at high elevations and include outcrop, talus, meadow, and krummholz communities. Lower elevations are dominated by coniferous forests with alpine fir (Abies lasiocarpa), Engelman spruce (Picea engelmannii), lowland white spruce (Picea glauca), and Douglas fir (Pseudotsuga menziesii) as the principal climax species. The historical influence of wildfire is reflected by extensive successional stands of lodgepole pine (Pinus contorta), western larch (Larix occidentalis), and Douglas fir. Snowslides occur frequently on more precipitous mountain slopes and create disclimax communities characterized by alder (Alnus spp.), willow (Salix spp.), mountain maple (Acer glabrum), and serviceberry (Amelanchier spp.). Grasslands occur in frequently on drier sites, particularly east of the Continental Divide, and are generally dominated by bluegrass (Poa spp.), fescues (Festuca spp.), or bluebunch wheatgrass (Agropyron spicatum).

Characteristics of the park's grizzly population were studied from sightings of grizzlies by the author, park personnel, and others who were considered qualified observers. Bears were observed while traveling the park's trail system on foot or horseback, from a vehicle on main and secondary roads, or during aerial flights. The date, location, number, description, and other pertinent information were recorded for each sighting. Descriptions included size, color, and distinguishing features. Bear locations, sighting dates, and descriptions were used to exclude duplicate sightings and determine the number of individual bears seen each year. These data were used to compute population parameters.

Numerical estimates of the population were obtained from a 290-mile² sample area in the north-central region of the park (Fig. 1). Sample area boundaries were selected to include a distribution of habitats and physiographic features which was representative of the entire park. An extensive trail system within the area permitted more intensive observational coverage than in other parts of the park. The number of individual bears observed within the sample area was determined each year and densities computed. These densities were expanded to estimate the population for the 1583-mile² park area.

The sex and/or age classes in yearly samples of individual bears were used to compute population composition. Unclassified adults (males, females, subadults), productive females, and young classes were recognized. Young accompanied productive females and were distinguishable as cubs (0.5 years) and yearlings (1.5 years) on the basis of relative size. The yearling class contained an undetermined number of subadults as discussed later in the text. Sizes of cub and yearling litters were computed from yearly sightings, and frequently included sightings of the same litters during consecutive years.

Records were maintained for each known grizzly bear mortality which occurred within the park during the study. Carcasses were subjected to routine post-mortem examinations, which included aging according to the tooth-sectioning method described by Mundy and Fuller (1964). Additional data on mortalities adjacent to the park were provided by the Montana Fish and Game Department and the United States Bureau of Sport Fisheries and Wildlife.

The number of different grizzlies observed on the density sample area each year, computed density, and expanded estimate for the park's population are presented in Table 1. The mean computed density of one grizzly per 8.2 mile² was intermediate as compared to that reported for other populations. Troyer and Hensel (1964) and Mundy (1963) reported respective densities of one grizzly per 0.6 and 5 mile² for Kodiak Island, Alaska, and Glacier National Park, Canada. Both of these areas were characterized by extensive alpine, subalpine, and shrub habitats. In contrast, Jonkel (1967) found a density of one grizzly per 15 mile² on an area of extensive coniferous forests immediately adjacent to Glacier National Park, Montana. An intermediate population density in Glacier National Park possibly relates to an intermediate composition and interspersion of important habitat types (Martinka, unpubl. data).

The expanded park population estimates of 175-230 grizzlies were greater than those made prior to 1967. These ranged from 90 to 130 and were generally derived from a combination of reported observations, area counts of grizzlies by park rangers, and estimates from previous years. Higher figures during this study are considered to reflect greater efficiency of the sample-area census technique rather than a population increase.

Annual differences in density and population estimates probably resulted from differences in the observability of bears, sampling bias, and other variables. General difficulty in observing grizzlies, even when they were known to frequent an area, suggested that expanded population figures were minimum estimates. In contrast, inclusion of some bears that had only part of their home range within the sample area may have tended to inflate the estimates. Increased estimates would also have occurred where random wandering of subadults inadvertently contributed to duplicate observations. However, the generally comparable figures obtained during the 5-year study period and the comparisons with other populations suggest that density and population estimates were reasonable.

TABLE 1. Density and population estimates for grizzly bears in Glacier National Park from 1967 through 1971 as determined from sightings on a 290 square mile area within the Park.

Results of classifying grizzly bears according to sex, age, or both are presented by years in Table 2. The data are considered to reflect the general composition of the park's population from year to year and show that unclassified adult grizzlies were the predominant class observed. This class included adult males, nonproductive adult females, and subadults of both sexes.

TABLE 2. Composition of the grizzly bear population in Glacier National Park as determined from classifications of 350 bears from 1967 through 1971.

Productive female grizzlies annually comprised 21-35% of the total adult population segment. Mean proportions of cubs and yearlings were nearly equal for the 5-year period, but variability within years and classes was characteristic. A potential population growth rate of 20% annually was computed from the mean annual increment of 17% cubs.

The mean annual increment of cubs in the park's population was lower than those reported for certain other grizzly populations. Craighead and Craighead (1971) found a mean annual increment of 19% cubs for Yellowstone National Park. The population segment involved was subject to unnatural mortality because the animals concentrated at refuse dumps and frequented campgrounds (Martinka 1970; Cole 1971). A mean annual increment of 22% was described for a hunted grizzly population on Kodiak Island (Troyer and Hensel 1964). Relatively low production of cubs in Glacier National Park possibly reflects low mortality from unnatural causes.

Observation frequency by size and mean sizes for 65 grizzly bear litters are presented by age class in Table 3. Litters containing one or two young were observed most frequently. Percentages of litters with two or three young increased in the yearling class, suggesting that complete yearling litters were more easily observed than cub litters. The trend shown by observation frequency was reflected by a slightly larger size of yearling as compared to cub litters. Small litter sizes in the park's grizzly population probably contributed to the relatively low reproductive rate observed. Troyer and Hensel (1964) reported mean litter sizes of 2.36 and 2.17 for the cub and yearling classes, respectively, of a more productive, hunted grizzly population in Alaska.

TABLE 3. Observation frequency by size and mean sizes for grizzly bear litters in Glacier National Park as determined from yearly observations of individual litters from 1967 through 1971.

Comparable sizes for cub and yearling litters indicated that first-year mortality of cubs was low in the park's population. This was supported by observations of five recognizable family groups where litters accompanied the maternal female for 2 or more years. In each case, the integrity of the litter was maintained for the entire period of observation and no mortalities occurred.

In contrast, field classifications showed a lower percentage of yearlings than cubs in the population (Table 2). This appeared related to observational bias since individual family units with cubs were generally more positively identified than those with yearlings. As a result, a greater proportion of family groups with yearlings would have been excluded from basic data as possible duplicate observations. Some first-year mortality of cubs was also considered a potential contributor to the observed difference. In this case, complete losses of litters may be postulated since mean litter sizes for cubs and yearlings were comparable. A minimum mortality of 12% annually was computed, assuming that cub mortality was entirely responsible for the observed difference and that the yearling class contained no subadults.

Differences in productivity of the grizzly population between years (Table 2) were possibly related to variations in length of the maternal relationship. Hensel et al. (1969) utilized reproductive data from a hunted population to hypothesize that grizzlies normally breed at 3-year intervals. This resulted in a 2-year relationship between the female and young, with dissolution of the bond occurring prior to breeding the third summer. Field observations during this study support the hypothesis with exceptions. In one case, a recognizable family group remained intact into the fourth summer. Several additional family group sightings were recorded where young were sufficiently large to assume that they were older than yearlings.

Extension of the breeding interval by at least some adult females would have contributed to the comparatively low reproductive rate observed in the population. Stokes (1970) considered social intolerance as an important factor which may lead to reproductive responses of this type in a grizzly population. The specific nature of the interaction was not established but inhibition of ovulation, extended lactation, or a combination of both could permit a longer maternal relationship. Erickson and Nellor (1964) reported that lactating female black bears (Ursus americanus) were unreceptive toward males during the normal breeding season. During this study, a relationship between low productivity and high proportions of yearlings was particularly evident in 1968 and 1971.

Most grizzlies in the park's population were observed as single individuals or in family groups (Table 4). Cohesive groups of two or more adults, subadults, or both were observed less frequently. These groups appeared to include adult breeding pairs in June and, more commonly, subadult litter mates which remained together following the dissolution of maternal bonds.

Four reports of unusual groupings of grizzlies were recorded during this study and similar reports were located in park wildlife records. These appeared to result from breeding or maternal behavior. In one case, an adult male was observed with a family group containing a female with three 2-year-old young in May 1969. Dissolution of the maternal bond followed and the female apparently produced two cubs in 1970. Three other sightings of family groups along with detailed descriptions suggest that females with young may occasionally tolerate the presence of subadults or young adults for short periods.

Several reports of large groups of grizzlies appeared to result from seasonal concentrations in the vicinity of preferred food sources. In Glacier National Park, such concentrations have been reported on lowland meadows and snowslides in spring, subalpine areas in late summer, and along a nonnative kokanee salmon (Oncorhynchus nerka) spawning stream in fall. Descriptions of these indicate that distinct spacing among individuals and different social units was characteristic. It seems likely that concentration sites were utilized by those animals with home ranges including or immediately adjacent to the site, as has been described for black bears by Jonkel (1967).

TABLE 4. Observation frequency for grizzly bear singles, groups, and matriarchal family units in Glacier National Park from 1967 through 1971.

Ten grizzly bear deaths were recorded in Glacier National Park from 1967 through 1971 (Table 5). Nine were caused by direct control for management purposes (Martinka 1971) and one by an apparent collision with a motor vehicle. Control deaths generally resulted from attraction of the bears to unnatural food sources and their age distribution suggested a disproportionate involvement of subadult or older adult animals. These were considered as social equivalents and their attraction to unnatural foods may have resulted from their subordinate social rank.

Population losses resulting from control actions averaged two grizzlies per year during the study. This comprised approximately 1% of the estimated population and 6% of the computed mean annual production of 33 cubs. It appears doubtful that this low rate of unnatural loss was of sufficient magnitude to significantly alter population dynamics.

TABLE 5. Summary of known grizzly bear mortality in Glacier National Park from 1967 through 1971.

During the study, 32 grizzly bear deaths were reported for areas immediately adjacent to Glacier National Park (Fig. 1). These included 14 legal kills by hunters and 18 control actions resulting from attraction to unnatural food sources, depredations, or both. Fourteen deaths, which occurred on the Blackfeet Indian Reservation, apparently resulted from bear depredations on livestock. Considerations of habitat requirements and their proximity to the eastern park boundary suggested that some of these bears occupied home ranges which extended across park boundaries.

Four control actions and one hunter kill in the vicinity of West Glacier appeared to result from the presence of an open refuse dump which attracted grizzlies to the area. One additional grizzly was captured and removed from the area. Observations suggested that these were bears that occasionally moved outside the park to use the dump. In contrast, the locations and age classes of 13 hunter kills in the North and Middle Fork Flathead River drainages indicated removals from populations resident to those areas (Jonkel 1967).

Results of this study show that a viable population of 175-230 grizzly bears inhabits Glacier National Park. Numerical stability and relatively low productivity during the study imply that the grizzly population was at or near habitat carrying capacity. Social stress resulting in emigration of subadults may be an important factor contributing to natural regulation of the park's grizzly population. Stokes (1970) considered social intolerance and associated dispersal of subadults as mechanisms which probably contribute to natural control in grizzly populations. Intensive observations of marked black bears by Jonkel (1967) are pertinent to this hypothesis. In that study, home ranges of black bears overlapped broadly among adults of both sexes, but mutually exclusive home ranges were occupied by adults of the same sex. This pattern apparently restricted home-range establishment by subadults, resulting in both dispersal to marginal habitats and higher mortality rates.

Grizzly bears coexist with an increasing number of visitors in Glacier National Park each year (Martinka 1971). However, present levels of visitation do not appear to be sufficient to have adverse effects on the grizzly population. Wild, free-ranging grizzlies commonly frequented areas close to human developments or activity. Control actions which were necessary to protect visitors resulted primarily from improper refuse disposal or human encounters with maternal female grizzlies.

Population removals within the park resulting from direct controls were sufficiently low to have minimal effects on dynamics of the grizzly population. Effects were minimized further because control was mostly limited to socially subordinate individuals. Capture and transplanting of nuisance grizzlies that frequented developed areas also had minor effects on the population. Only two such actions were recorded during 1967-71.

Certain human activities in areas immediately adjacent to Glacier National Park were a potential source of impact to peripheral segments of the park's grizzly population. Deaths resulting from bear depredations on livestock and attraction to unnatural food sources were of particular significance. In cases where adults were removed from established home ranges, temporary effects on the park's population were implied where those ranges extended into the park. These losses were probably replaced by recruitment of subadults to the population. In contrast, extrinsic control actions involving subadults or older adults possibly included emigrants from occupied park habitats which could no longer be considered established members of the park's population. It is doubtful that legal hunting of grizzlies during recent years in areas surrounding the park has appreciably affected the numbers or dynamics of the park's population.

Cole (1972) discussed relationships between grizzly bears and natural area values which are applicable to Glacier National Park. He concluded that the presence of a grizzly population in Yellowstone National Park was essential to (1) have representative natural equilibriums among associated secondary consumers; (2) maintain natural relationships between a variety of primary and secondary consumers; and (3) retain the scientific values of ecological systems with an intact native biota. In Glacier, management procedures which provide for human safety do not appear to have detracted significantly from these values. Effects of a continued increase in visitation cannot be predicted from present data, but it has become apparent that a wild, free-ranging grizzly population offers the least opportunity for conflict with man.

COLE, G. F. 1971. Preservation and management of grizzly bears in Yellowstone National Park. BioScience 21(16):858-864.

______. 1972. Grizzly-elk relationships in Yellowstone National Park. J. Wildl. Manage. 36(2):556-561.

CRAIGHEAD, J. J. and F. C. CRAIGHEAD. 1971. Grizzly bear-man relationships in Yellowstone National Park. BioScience 21(16):845-857.

DIGHTMAN, R. A. 1967. Climate of Glacier National Park, Montana. Glacier Natural History Association Bull. No. 7. 9 p.

ERICKSON, A. W., and J. NELLOR. 1964. Breeding biology of the black bear. Pages 5-45 in A. W. Erickson, J. Nellor, and G. A. Petrides. The black bear in Michigan. Michigan State Univ. Agr. Exp. Sta. Res. Bull. 4. 102 p.

HABECK, J. R. 1970. The vegetation of Glacier National Park, Montana. Spec. Rep. 132 p.

HENSEL, R. J., W. A. TROYER, and A. W. ERICKSON. 1969. Reproduction in the female brown bear. J. Wildl. Manage. 33(2):357-365.

JONKEL, C. J. 1967. Black bear population studies. Ph.D. Thesis. Univ. of British Columbia, Vancouver. 148 p.

MARTINKA, C. J. 1970. Grizzly ecology studies, Glacier National Park, 1969. National Park Service Prog. Rep. 43 p.

______. 1971. Status and management of grizzly bears in Glacier National Park, Montana. Trans. N. Am. Wildl. Conf. 36:312-322.

MUNDY, K. R. D. 1963. Ecology of the grizzly bear (Ursus arctos) in Glacier National Park, British Columbia. MS. Thesis. Univ. of Alberta, Edmonton. 103 p.

______, and W. A. FULLER. 1964. Age determination in the grizzly bear. J. Wildl. Manage. 38(4):863-866.

STOKES, A. W. 1970. An ethologist's views on managing grizzly bears. BioScience 20(21):1154-1157.

TROYER, W. A., and R. J. HENSEL. 1964. Structure and distribution of a Kodiak bear population. J. Wildl. Manage. 28(4):769-772.

The author is indebted to G. F. Cole and D. B. Houston, National Park Service; R. J. Mackie, Montana State University; and S. Herrero, University of Calgary, for critical review of the manuscript. Technical assistance in aging grizzlies was provided by K. R. Greer, Montana Fish and Game Department. Cooperation of park personnel in reporting grizzly bear sightings made the study possible. The paper is a contribution from National Park Service Natural Sciences Project GLAC-N-16.

¹Edited contents of this paper have been published in the Journal of Mammalogy but are included in these proceedings to provide a complete record of the symposium. Please refer to: Martinka, C. J. 1974. Population characteristics of grizzly bears in Glacier National Park, Montana. J. Mamm. 55:(1):21-29.