Mountain Goats: Ecology, Behavior, and Conservation of an Alpine Ungulate

book.

CHAPTER 1

Ecological Questions, Conservation Challenges, and Long-Term Research

The conservation of biodiversity and the management of wildlife require an understanding of the basic ecology of animals. That deceptively simple statement conceals a difficult problem, because understanding the basic ecology of species demands years of research. The processes that affect population dynamics of large mammals often develop over many years and cannot be understood without long-term monitoring. Important events (such as forest fires, extreme weather, or the appearance of new predators, competitors, or diseases) may have drastic long-term effects on population dynamics but they may be too rare to be detected, let alone quantified, by a few years of research. In addition, the many factors affecting a species’ abundance seldom act in isolation. Instead, interactions between factors are commonplace: for example, body mass may affect survival only at high population density (Festa-Bianchet et al. 1997), and the impacts of a harsh winter may vary substantially according to a population’s age structure (Coulson et al. 2000). Similar complex relationships affect the consequences of different harvest levels, which can be radically different according to the sex–age composition of the population and of harvested animals.

Consequently, an in-depth understanding of ungulate ecology requires data collected over many years and can best be served by long-term studies that seek to answer fundamental questions: What affects population size? What factors determine age- and sex-specific mortality? How do individuals differ in their ability to contribute to population recruitment, and why do those differences exist?

A long-term approach to the study of the ecology and conservation of large herbivores is particularly appropriate because of their longevity and complex population structure. An individual can experience varying levels of environmental conditions over its lifetime. Consequently, the reproductive strategy of large herbivores likely evolved in response to the range of environmental conditions that an individual may encounter over its lifetime. The complex population structure of large herbivores, sometimes involving a dozen or more coexisting cohorts, means that the population present today is the result of processes and events that took place over the previous decade and will affect population dynamics over the next one. It is therefore essential that management programs to conserve large herbivores, including those that involve some harvest, be mindful of the differences among individuals. In addition, the consequences of conservation actions (or of harvest strategies) can persist over many years. Because of the importance of differences among individuals and of processes occurring over several years, biological knowledge useful for conservation of large herbivores can therefore be best obtained by long-term monitoring of known individuals within a population.

Public finances typically sustain fundamental ecological studies. Our mountain goat study is no exception. In addition to producing novel information, ecological studies have an obligation to clearly communicate the applied implications of their results. The conservation of biodiversity requires long-term research, and long-term research should make a contribution to conservation. We will use our sixteen years of research on mountain goats to show how some aspects of the biology of this species play a fundamental role in its conservation. We will do so by examining the adaptations of mountain goats to their alpine environment, and by underlining differences and similarities between mountain goats and other large herbivores, in particular other mountain-dwelling ungulates.

Why Study Mountain Goats?

Mountain goats provide research challenges and opportunities from both a fundamental and an applied viewpoint. There is much concern for the conservation of this species, which appears highly sensitive to both harvest and disturbance. In addition, its alpine habitat is very sensitive to human activities and is likely at risk from the effects of climate change.

Four factors combined to provide the stimulus to study mountain goats at Caw Ridge. First, an unexplained and rapid decline in mountain goat numbers in Alberta led to the complete closure of goat hunting in the province in 1987 (Smith 1988b). Combined with the lack of information on the ecology of mountain goats, the drop in numbers convinced wildlife managers of the need for a study. Second, earlier research on mountain goats (Chadwick 1977) suggested that they may be good subjects for a study of how social behavior affects individual reproductive success and population dynamics in ungulates, because aggressive interactions seemed to be an important aspect of their ecology. Third, results obtained by studies of mountain goat population dynamics in introduced and native populations were conflicting: goats in native populations are very sensitive to even light levels of harvest (Hamel et al. 2006; Kuck 1977; Smith 1988b), while some introduced populations appear to withstand substantial harvests, similar to those normally associated with deer (Adams and Bailey 1982; Houston and Stevens 1988; Kuck 1977; Swenson 1985; Williams 1999). Finally, by providing new information about the ecology and behavior of a charismatic and economically important mountain ungulate, our work contributes to the conservation of alpine environments. Clearly, one cannot conserve mountain goats without conserving the mountain.

When our research started in 1988, what little was known about mountain goats hinted that they may differ from other ungulates in much of their behavior and ecology. Aggressive behavior was thought to be an important component of their social organization (Geist 1967, 1971), partly because their sharp stiletto-like horns (fig. 1.1) are extremely dangerous in intraspecific interactions. Females appeared to have a protracted period of maternal care, and some yearlings were reported to remain with their mothers and perhaps continue to suckle (Hutchins 1984). There were suggestions of stable associations among specific individuals, possibly indicating long-lasting relationships among female kin (Geist 1971). Therefore, mountain goats appeared to be ideal subjects for a long-term study of the effects of social status on female reproductive success, and of the relationships between behavioral ecology and population dynamics.

We set out to determine what factors affect individual differences in survival and female reproductive success and, by extension, changes in population dynamics of mountain goats. Most recent studies of large herbivores suggest that their populations are limited by a combination of forage availability and weather (Coulson et al. 2000; Gaillard et al. 2000a; Sæther 1997), although predation can also be important for populations that coexist with large predators (Owen-Smith et al. 2005; Sinclair et al. 2003; Wittmer et al. 2005). Because of its key role in population dynamics of other large herbivores (Coulson et al. 2000), we were also interested in the possible effects of changes in population density upon individual reproductive strategies and population dynamics. There was almost no information available on age-specific survival and reproductive success in native populations of mountain goats, and nothing was known about their response to changes in population density or in resource availability. The assumption that mountain goats could be harvested at the same level as deer or bighorn sheep had led to population declines or extirpations: mountain goats may be the only North American ungulate to have suffered local extinctions through sport hunting (Glasgow et al. 2003). Weak density-dependence may explain why harvest mortality appears largely additive in this species, as suggested by a recent review of compensatory and additive hunting mortality (Lebreton 2005).

Figure 1.1. Mountain goats of both sexes have black and sharp stiletto-like horns. Here, an adult female in late August. Photo by S. Côté.

In addition to its possible significance for our understanding of individual reproductive success and population dynamics, differences in reproductive success among females may also be the key to explaining the contrasting impacts of sport hunting on mountain goat populations. Unlike most other North American ungulates, mountain goats are not obviously sexually dimorphic. To an inexperienced hunter looking through binoculars from 200 meters away, males and females do not appear very different unless they stand side by side (fig. 1.2). Most sport hunters prefer to shoot males, but only experienced hunters can identify a mountain goat’s sex at a distance. Male goats are generally more difficult to find than females because males tend to be solitary or in small groups and spend much time in forested areas. Suppose that a hunter has approached a nursery group that includes females, juveniles, and possibly a few two-year-old males. If the selected target is the goat with the longest horns, it almost certainly will be a mature female. Now suppose that most of the recruitment within native herds is provided by mature, dominant females with long horns. The impact of removing a small number of females through sport hunting may then be far greater than that predicted by harvest models which assume that all females in a population have the same probability of contributing to recruitment. Could that be why native mountain goat populations are so sensitive to harvest? And if that is the case, which sex–age groups should be harvested, and in what proportion, if hunting is to be sustainable?

Figure 1.2. Sexual dimorphism is difficult to see from a distance in mountain goats. Here, the goat on the left (37 on left ear) is an adult billy surrounded by adult females. Photo by S. Côté.

Mountain Goat Ecology and the Conservation of Mountain Ungulates

Our research on mountain goats at Caw Ridge demonstrates how long-term studies of marked individuals can both contribute to fundamental ecology and be useful for the conservation and management of ungulates. We ask two main questions: What factors affect population dynamics of mountain goats? and, What selective pressures shape female reproductive strategy? We contend that the answers to these questions are linked by some basic biological characteristics of mountain goats that we will present in coming chapters: high and stable adult female survival, high adult male mortality and dispersal, moderate and variable juvenile survival, slow multiyear physical development, very late age of primiparity, and a strong linear and age-related dominance hierarchy among females.

Long-term studies of marked individuals provided major contributions to our understanding of the ecology of large mammals and therefore our ability to conserve them and their habitat (Gaillard et al. 2003). By monitoring marked, known-age individuals throughout their lifetime, we were able to compare different aspects of their life history, look for correlations between events that occurred in different years, and especially take into account their age. Age has a pervasive influence on almost all aspects of ungulate life history and, by extension, on population dynamics (Coulson et al. 2001; Festa-Bianchet et al. 2003; Gaillard et al. 2001). Mountain goats have a rather long life expectancy: we present here the results of our first sixteen years of work, and at the time of writing there were still two goats on Caw Ridge that had already been there when our research began. Many processes affecting population dynamics exert their effects over long periods of time: ungulate population density does not usually vary much from one year to the next, and for many comparisons (such as the effects of weather, population density, or forage quality on growth and survival) a single data point is collected each year. Consequently, a long-term study was necessary to understand the behavior and ecology of this species.

While focusing on the results of the Caw Ridge study, we will frequently compare it with the results of other long-term studies of ungulates, including other mountain ungulates such as bighorn sheep in Alberta, ibex in the Italian and French Alps, and both Alpine and Pyrenean chamois in France (Gaillard et al. 2000a). Inspired by the pioneering work on red deer on Rum, Scotland (Clutton-Brock et al. 1982), these research programs are all based on long-term monitoring of marked individuals. They have repeatedly demonstrated their value in both advancing our knowledge of population dynamics and evolutionary ecology, and in applying that knowledge to the management and conservation of wild ungulates. Unlike most other long-term studies, however, the Caw Ridge study took place in a relatively pristine environment, on a population that had not been hunted for twenty years. Although the land is scarred by resource exploration (fig. 1.3), accessible by all-terrain vehicle, and the mountain goats are occasionally harassed by helicopters and motorized vehicles, the area maintains all the wildlife species it had before European invasion, particularly the large predators such as wolves, bears, and cougars that are absent from most other long-term study areas of ungulates.

Figure 1.3. A view from the eastern portion of Caw Ridge. Note the roads and trails made in the 1970s for oil, gas, and coal exploration. The trap site and cabin appear as small dots on the top right corner. Photo by J. Mainguy.

Mountain Goat?

The Nisga’a People of northwestern British Columbia call this white mountain-dwelling ungulate Matx. Europeans, however, gave it misleading names. Its scientific name means ram of the mountain, but Oreamnos is not a sheep. English speakers called it mountain goat. It is known as chèvre de montagne (mountain goat) in French, cabra montesa (mountain goat) or cabra blanca de las Rocosas (Rocky Mountain white goat) in Spanish and capra delle nevi (snow goat) in Italian. Indeed Oreamnos lives in the mountains and is often in the snow, so Europeans did not get it completely wrong, but it is not a goat. Oreamnos americanus is classified in the Tribe Rupicaprinae within the subfamily Caprinae, family Bovidae. The subfamily Caprinae includes all true mountain-dwelling ungulates, characterized by highly developed climbing skills, reliance on cliffs or steep terrain to escape predation, presence of horns in both sexes (except for some domesticated forms and one subspecies of mouflon), and for most species a complex seasonal pattern of home-range use (Shackleton 1997).

The systematics of rupicaprins are unclear: they may represent an offshoot from other caprinae (including Capra, the ibex and true goats, and Ovis, the sheep), or they may be derived from a group ancestral to other caprins. Molecular studies disagree about the relationship between Oreamnos and other rupicaprins, often suggesting a phylogenetic relationship between mountain goats and muskox (Groves and Shields 1996; Hartl et al. 1990; Hassanin et al. 1998).

Rupicaprins are found in mountains from the Iberian peninsula through Europe and much of Asia to western North America. The closest contemporary relatives of mountain goats are the Asiatic serows, including the Japanese species and the much larger mainland species, at least two species of goral (Naemorhedus spp.), all found in Asia, and two species of chamois, which inhabit mountains from northwestern Spain to the Caucasus.

Mountain goats do not look much like true goats (fig. 1.4). They are pure white, with sharp recurved horns that resemble those of chamois and serow (fig. 1.5). The horns of males are thicker and more curved than those of females, but there is no difference in length (chapter 6). Males are also much larger than females. Sexual dimorphism increases with age, so that while male and female kids are about the same size, a full-grown adult female weighs about forty percent less than a full-grown male (chapter 6). The mountain goat is an excellent climber (fig. 1.6) and its body appears adapted to a life on the edge: the feet are short and stout, with large hooves that can open very wide, providing a strong grip on rocks and on steep terrain. Its weight is distributed vertically, which presumably helps it maintain its balance on cliff edges. From the side, mountain goats appear to have a very deep chest, but viewed from the front they are surprisingly thin (fig. 1.7).

Once people decided that this animal was a goat, they used domestic names to describe sex–age classes. We used those names in the field and we will sometimes use them here as well. A female is referred to as a nanny, a male as a billy, and a juvenile as a kid.

The ancestors of mountain goats likely originated in central Asia (Geist 1971) and probably entered North America by the Beringia land bridge about forty thousand years ago (Cowan and McCrory 1970; Rideout and Hoffmann 1975). A fossil species, Oreamnos harringtoni, has been found as far south as New Mexico (Jass et al. 2000). The known prehistoric distribution of O. americanus included Vancouver Island (Nagorsen and Keddie 2000) and possibly the Olympic peninsula. Whether or not it reached farther south than the present-day American States of Washington, Idaho, and Montana is a matter of debate, fueled by disparate interests and by preciously little data. There are no recognized subspecies of mountain goat, and little is known about its genetic variability over its geographical range.

Figure 1.4. Mountain goats (here #137 when she was two years old; photo by S. Côté) are not true goats like ibex (an adult male from Grand Paradiso, Italy; photo by M. Festa-Bianchet) or domestic goats (purebred kiko; photo by S. Côté).

Figure 1.5. Mountain goat horns (here a two-year-old female; photo by S. Côté) resemble those of other Rupicaprinae such as chamois (photo by F. Pelletier), and Japanese serow (adult female; photo by K. Ochiai).

Figure 1.6. Mountain goats are excellent climbers. Here are some on a cliff at the west end of Caw Ridge. Photo by S. Hamel.

The distribution of mountain goats includes native, reintroduced, and introduced populations (fig. 1.8). Most mountain goats are in British Columbia and Alaska (table 1.1). Including both native and introduced herds, there are somewhere between 75,000 and 115,000 mountain goats. Because of their vulnerability to hunting, mountain goats were extirpated from parts of their southern range following the arrival of European immigrants. Transplants have been used to reestablish some extirpated populations but also to introduce goats into areas with no clear evidence of their past presence as a native species. One area where mountain goats generate controversy is in the Olympic Mountains National Park in the State of Washington. Goats were introduced there in the 1920s and their numbers and range greatly increased over time. Concern over their exotic status and possible negative effects on alpine vegetation led the U.S. National Park Service to adopt a policy of eradication, which was welcomed by some and denounced by others (Houston 1995; Houston and Stevens 1988; Hutchings 1995; Lyman 1988, 1994, 1995; Pfitsch and Bliss 1985). A similar situation is developing in Yellowstone National Park, where mountain goats are not a native species but are now immigrating following an introduction north of the park in Wyoming (Lemke 2004). Mountain goats have also been introduced to the Black Hills of South Dakota and in several places in Nevada, Colorado, Montana, and Idaho (fig. 1.8), while an attempted introduction to Vancouver Island failed. In Alberta, mountain goats were reintroduced to the southwestern part of the province in 1996–1997, in areas where they disappeared in the early 1960s because of overhunting. Between 1986 and 1988, just before our study began, a few goats from Caw Ridge were captured and transplanted to southern Alberta.

Figure 1.7. Viewed from the side, mountain goats appear to have a very deep chest, but viewed from the front they are surprisingly thin. A two-year-old female, #147. Photos by S. Côté.

Mountain goats occupy a variety of mountainous habitats, from temperate rainforest near sea level in coastal British Columbia and Alaska to xeric tundra at over 4000 meters above sea level in Colorado. Although many goat populations in the southern and western parts of their range use restricted areas with very steep cliffs, farther to the north they are often found in rolling terrain above treeline. Mountain goats are the only caprin in much of the western half of their range, but they share parts of their eastern distribution with bighorn sheep and parts of their northern range with the two subspecies of thinhorn sheep (Dall’s and Stone’s). Little is known about the relationships between mountain goats and mountain sheep (either bighorn or thinhorn) in areas where both species are native, but some introduced populations of mountain goats are suspected to compete for forage or habitat with bighorn sheep and may have a negative effect on bighorn sheep populations, possibly because mountain goats are socially dominant to bighorns (Hobbs et al.