back to Science Pages

Notes on the Vancouver Island marmot
(Marmota vancouverensis)

Purpose

The purpose of this paper is to serve as a "road-map" to the primary scientific literature concerning Vancouver Island Marmots (Marmota vancouverensis).

What's a Vancouver Island marmot?

Prehistoric distribution Prehistoric bones have been found at 8 locations, all outside of the present core area of distribution (Nagorsen et al. 1996). These bones have been radio-carbon-dated at ages of 800 to 2700 years before present, and show undeniable evidence (tool-marks) that marmots were hunted by humans. It remains impossible to determine over what perod these bones may have been deposited, or to estimate prehistoric population sizes.

Historical distribution Between 1864 and 1969, marmots were recorded from 28 sites on 25 mountains (Bryant and Janz 1996). Unfortunately the records are often vague. It is difficult to know whether they refer to breeding colonies or dispersing individuals. One record simply mentions "swarms of ground hogs" (1893), while another describes a "brace" of marmots shot by a hunter (1922). Given the lack of records, there is little hope of accurately estimating historical population sizes or trends, except to reveal that M. vancouverensis apparently disappeared from 2/3rds of its historic range within the last two or three decades (Bryant and Janz 1996).

Current distribution (1979-present)

In 1979, the B.C. Ministry of Environment, Lands and Parks began sponsoring population counts and searches for M. vancouverensis. Initial counts were conducted by non-government personnel and volunteers. Counts made since 1982 were carried out by government personnel or academic researchers. Bryant (1998) summarized recent results.

Since 1979, marmots or fresh burrows were found at 47 sites on 15 mountains. Reproduction was observed at 35 sites on 14 mountains. With a single exception, all colonies confirmed as active since 1979 were located within 5 adjacent watersheds on south-central Vancouver Island. The exceptional area is on Mount Washington near Comox, an area separated from other known colonies by 100 km.

Figure 1: Active (purple) and inactive (yellow) location records for M. vancouverensis. (Click on the small map to see more detail) Marmots have disappeared from several areas on Vancouver Island, particularly in the northern extremes of their historic range. The detailed map is adapted from Bryant and Janz (1996)

Population trends

Despite differences in count effort from year to year, data are adequate to estimate recent population trends. Marmots expanded during the early 1980s into recently clearcut habitats above 700 m in elevation. At least 10 sites were eventually colonized. Most colonizations occurred within 10 years of logging and within 1 km of existing natural colonies (Bryant and Janz 1996). In recent years marmots declined precipitously, both in numbers and in the incidence of occupied habitats. Populations in natural habitats apparently declined steadily after the early 1980s. Populations in clearcuts "crashed" after 1994 (Bryant 1998). The 1998 population probably comprised fewer than 100 individuals, divided more-or-less equally among natural meadows and clearcuts.

Figure 2: Population trends in M. vancouverensis. Probable marmot numbers were estimated by applying a correction factor to the actual number counted. Updated from Bryant (1998).

Habitat characteristics Vancouver Island marmots need habitats that provide a variety of food plants, deep soil for burrowing and suitable lookout spots. Most marmots live above 1000 m elevation in meadows that face south to west (Bryant and Janz 1996). Some natural meadows may be kept clear of invading trees by snow-creep and periodic avalanches or fire (Milko 1984). Vegetation at natural colonies is diverse (Milko and Bell 1985) and marmots eat over 50 species of grasses and forbs (Martell and Milko 1986). Marmots also inhabit logged habitats where the vegetation is quite different. Few of the preferred plant species found in natural meadows are common in clearcuts. Food habits have not yet been systematically studied. However, we know that marmots grow larger in clearcuts (Bryant 1990) even while annual survival rates are reduced by about 10% (Bryant 1998).

Social structure

Vancouver Island marmots live in colonies comprised of one or more families (Bryant 1990). Families normally contain an adult male, one or more adult females and a variable number of sub-adults, yearlings and young-of-the-year (Heard 1977). The size and number of families varies between colonies and years, and this can produce dramatic fluctuations in population size.

Heard (1977) suggested that M. vancouverensis is monogamous. Milko (1984) predicted some degree of polygyny on the basis of vegetation resources. Bryant (1996a) documented several cases in whish single males apparently sired multiple litters, but that single male:single female groups were much more common.

Reproduction

Mating probably occurs below ground in the first few weeks after emergence from hibernation, as occurs in other marmots (Barash 1989). Gestation is thought to be about 30 days and pups are born hairless in late May or early June (Nagorsen 1987). Pups first emerge from their burrows in very late June or early July. Data for 43 litters from 1987-1998 averaged 3.3 pups. Litters of 3 or 4 were most common, and litters of 2, 5 or 6 pups occurred infrequently. Females may breed in their third year, but most do not breed until age 4 or 5. A few females reproduced in consecutive years. Some females were quite successful, reaching the age of 8 or 9 and producing 10-15 pups over her lifetime (Bryant 1998).

Hibernation

Vancouver Island marmots hibernate from mid-September until late April or early May. The active season is relatively consistent among years, although tracks have been seen as early as April 17th, and animals have been seen aboveground as late as October 18th. Radio-telemetry suggests that marmots hibernate as family groups, and often re-use hibernacula in subsequent years. Hibernacula can be identified either by grass and mud "plugs" found at tunnel entrances in late autumn, or by emergence tunnels through the snowpack in May or early June.

In general, Vancouver Island marmots appear to select hibernacula which are covered during winter by deep snowpack (Bryant 1990).

Dispersal and metapopulation ecology

Pups remain near their natal burrow for their first year and hibernate with the mother (Bryant 1996a). Yearlings generally expand their movements but almost inevitably return to hibernate with their mother a second time. Some "teen-age" marmots leave home in search of a new territory. Dispersal records are scarce, but it appears likely that M. vancouverensis generally disperse when reach the age of 2 or 3 years. Adults that survive to be older than this become quite sedentary and do not disperse (Bryant 1998).

Both males and females can disperse, but only a fraction of animals (perhaps 30%) actually do. Most individuals adopt the alternative strategy, which is to stay put and "take over the family farm". Dispersing marmots can make impressive movements. Ear-tagged marmots moved distances up to 10 km. Records of solitary marmots in unusual locations suggest that even longer movements (>25 km) may sometimes occur. The importance of dispersal and recolonization may be exaggerated for M. vancouverensis because of their small colony sizes (Bryant and Janz 1996).

Figure 3: Dispersal and metapopulations. Metapopulations are sub-divided populations in which the constituent sub-populations occupy discrete habitat patches in an otherwise inhospitable landscape. Individual colonies suffer extinctions so that at any time some patches are unoccupied. Dispersal is the crucial "glue" that allows recolonization of unoccupied habitat patches or "rescue" of colonies that are doing poorly.

Causes of mortality Unsuccessful hibernation and predation are the largest causes of mortality, especially for pups. Only 54% survive their first winter in natural habitats. In clearcuts this figure drops to 43%. Radio-telemetry provides conclusive evidence that predators exact a toll. At least 6 (13%) of the 43 marmots that were surgically equipped with radio-transmitters were killed by predators (Bryant 1998).		Figure 4: Survival of pups in natural meadows and clearcuts. Data are estimates of mean annual survival with 95% confidence. Numbers of individuals in each sample are shown.
Known predators upon marmots include cougars (Felis concolor), wolves (Canis lupus) and golden eagles (Heard 1977, Bryant 1996a). Terrestrial predators may benefit from logging roads that "channel" predators and prey along easy travel routes. Predators may also benefit from reduced visibility in regenerating clearcuts and from increased local density of marmots (Bryant 1998).		Most Vancouver Island marmots apparently die during hibernation, when physiological resources are weakest. This is consistent with what we know about other marmot species (Barash 1989, Armitage 1996).
Disease may also play an important role. Of six marmots experimentally transplanted in 1996, one was killed by a predator, one disappeared without a trace, and four died from a presumed bacterial infection during hibernation (Bryant et al. in press). Catastrophic losses of marmots occurred at several colonies, particularly during the 1992-98 period and especially during the winter of 1994-95. Evidence suggests that the incidence of high-mortality "events" has increased (Bryant 1998).

Conclusions: where do we go from here?

Caughley and Gunn (1996) offered a straightforward model for managing endangered species. First, determine whether populations are declining or whether other evidence suggests that a problem exists. Second, learn about the ecology of the organism and use the accumulated knowledge to test hypotheses about possible causal factors. Finally, use the results to reverse the factors that are causing the problem. It is not yet clear why marmots disappeared from parts of Vancouver Island north of Alberni Inlet. Changes in long-term regional snowfall patterns, hunting by aboriginal peoples, or increased abundance of natural predators may all have played a role (Nagorsen et al. 1996).		It is also possible that landscape connectivity may have changed. For example, tree invasion may have occurred at important "stepping-stone" colonies, such as occurred in other mountain ranges in the Pacific Northwest (Rochefort et al. 1994). Similarly, construction of the Buttle Lake reservoir in Strathcona Provincial Park may have influenced dispersal and metapopulation dynamics (Bryant 1998). South of Alberni Inlet, landscape changes due to forestry appear to be the principal cause of recent marmot population dynamics. Logging reduced the ability of marmots to re-colonize or "rescue" isolated natural colonies by the simple mechanism of creating large amounts of nearby alternative habitat in which dispersers could settle (Bryant 1998).
		...continued at below left...
Because of reduced survival, marmots living in clearcuts produce fewer than half the number of potential dispersers than did their counterparts in natural meadows (Bryant 1998). Probably the most important impact of forestry was to concentrate marmots within a small geographic area, thereby making individuals more vulnerable to predators and disease. The relative importance of various mortality factors remains unclear. In natural habitats, disappearances were uniformly distributed throughout summer, suggesting a constant mortality factor such as predation. In clearcuts, most animals disappeared after late summer, suggesting death during hibernation (Bryant 1998).		Further research is required, particularly in the broad areas of nutrition and hibernation (e.g., Thorp et al. 1994). More work is also needed to identify potential reintroduction habitats on Vancouver Island, and to test for the possibility of systematic environmental change either through tree invasion or more subtle vegetation change in natural sub-alpine meadows (e.g., Walker et al. 1993). With a population numbering fewer that 100 animals, Vancouver Island marmots must be considered as one of North America's most critically endangered mammals. Only by increasing both their numbers and distribution can the future of this engaging rodent be secured. For this reason the Recovery Plan emphasizes captive-breeding combined with marmot reintroductions to formerly occupied sites (Janz et al. in press).

Literature cited