Lessons from a Transboundary Wolf, Elk, Moose and Caribou System
Mark Hebblewhite. 2007. Predator-Prey Management in the National Park Context: Lessons from a Transboundary Wolf, Elk, Moose and Caribou System. Predator-prey Workshop: Predator-prey Management in the National Park Context, Transactions of the 72nd North American Wildlife and Natural Resources Conference.
Full text [here]
Selected excerpts:
Introduction
Wolves (Canis lupus) are recolonizing much of their former range within the lower 48 states through active recovery (Bangs and Fritts 1996) and natural dispersal (Boyd and Pletscher 1999). Wolf recovery is being touted as one of the great conservation successes of the 20th century (Mech 1995; Smith et al. 2003). In addition to being an important single-species conservation success, wolf recovery may also be one of the most important ecological restoration actions ever taken because of the pervasive ecosystem impacts of wolves (Hebblewhite et al. 2005). Wolf predation is now being restored to ecosystems that have been without the presence of major predators for 70 years or more. Whole generations of wildlife managers and biologists have come up through the ranks, trained in an ungulate- management paradigm developed in the absence of the world’s most successful predator of ungulates—the wolf. Many questions are now facing wildlife managers and scientists about the role of wolf recovery in an ecosystem management context. The effects wolves will have on economically important ungulate populations is emerging as a central issue for wildlife managers. But, questions about the important ecosystem effects of wolves are also emerging as a flurry of new studies reveals the dramatic ecosystem impacts of wolves and their implications for the conservation of biodiversity (Smith et al. 2003; Fortin et al. 2005; Hebblewhite et al. 2005; Ripple and Beschta 2006; Hebblewhite and Smith 2007).
In this paper, I provide for wildlife managers and scientists in areas in the lower 48 states (where wolves are recolonizing) a window to their future by reviewing the effects of wolves on montane ecosystems in Banff National Park (BNP), Alberta. Wolves were exterminated in much of southern Alberta, similar to the lower 48 states, but they recovered through natural dispersal populations to the north in the early 1980s, between 10 and 20 years ahead of wolf recovery in the northwestern states (Gunson 1992; Paquet, et al. 1996). Through this review, I aim to answer the following questions: (1) what have the effects of wolves been on population dynamics of large-ungulate prey, including elk (Cervus elaphus), moose (Alces alces) and threatened woodland caribou (Rangifer tarandus tarandus), (2) what other ecosystem effects have wolves had on montane ecosytems, (3) how sensitive are wolf-prey systems to top-down and bottom-up management to achieve certain human objectives, and (4) how is this likely to be constrained in national park settings? Finally, I discuss the implications of this research in the context of ecosystem management and longterm ranges of variation in ungulate abundance. …
Effects of Wolves on Ungulates
Elk
In the Bow Valley, Hebblewhite et al. (2005) compared adult female elk survival and recruitment between the low and high wolf areas during 1997 to 2000. Differences in wolf-caused mortality were tested using chi-square tests. In the high wolf zone, adult survival equaled 0.62 ± 0.06; n equaled 22, where n represents the number of adult female elk. And, calf recruitment equaled 14.6 ± 1.97 percent. The combination of this survival and recruitment led to rapid population decline (Hebblewhite et al. 2005). But, in the low-wolf area, survival equaled 0.89 ± 0.06; n equaled 23. And, recruitment equaled 27.4 ± 1.58 percent, which both are high and the same as before wolf recolonization; it led to a stable or increasing population (Woods 1991; Hebblewhite and Smith 2007). The main survival difference was wolf mortality increasing from about 16 percent to 56 percent; Hebblewhite and Smith 2007) between the low and high wolf area, which was consistent with an increase in wolf-kill rate of elk in the high-wolf area (Hebblewhite et al. 2004). …
In the YHT study area, Hebblewhite et al. (2006) showed that the migratory behavior of elk changed since the 1970s in three ways. First, both the proportion and number of elk migrating into BNP declined. The ratio of migratory to resident elk declined from 13:1, in 1980, to 2.5:1, in 2004; the numbers of migrants declined from 980, in 1984, to 580, in 2004. Second, the spatial distribution of elk shifted to the winter range year round. Third, the duration of migration declined because fall migration occurred almost a month earlier. … Importantly, prescribed fires, competition with horses for winter forage, and human harvest were unrelated to changes in the ratios of migratory to resident elk.
Moose
Hurd (1999) undertook a 4-year study (1993–1997) in BNP of competition between moose and elk to understand causes for moose declines following wolf recolonization. Hurd examined both exploitative competition for forage and apparent competition mediated by predation by wolves. The study revealed, at fine-spatial scales, that elk were exploitatively outcompeting moose because of their greater diet breadth and higher abundance. Yet, at large spatial scales, apparent competition mediated by wolves seemed the most compelling reason for moose declines. Wolves were the leading cause of moose mortality, causing 56 percent. Adult moose (male and female were the same) survival rates were very low (0.71 ± 0.03, n = 45) and were combined with low calf recruitment (23 ± 7.5 percent, most likely a result of predation but unknown). Moose populations were declining at about 8 percent per year because of wolf predation. Moose and elk in the high-wolf area had similar demography evidencing the strong top-down effect of wolf predation. In summary, Hurd found apparent competition mediated by wolves was occurring in combination with exploitative competition in a negatively additive fashion, which caused moose population declines.
Caribou
A similar example of conservation concern is apparent competition between elk and threatened woodland caribou, which have declined during wolf recolonization (Hebblewhite et al. 2007b) in the Canadian Rocky Mountains. Elk and caribou diets differ enough to make exploitative competition an unlikely explanation for caribou declines. Instead, similar to moose, the likely mechanism for caribou declines is competition between elk and caribou mediated by wolf predation, and this hypothesis was supported by modeling work by Hebblewhite et al. (2007b) and Lessard (2005). … Consequences of this for national park management in the Parks Canada system are dramatic; with current densities of wolves and elk in BNP, the Banff caribou subpopulation will almost certainly become extirpated. …
Evaluating Potential Management Scenarios
Relative Sensitivity to Management Changes in Forage
There was essentially no evidence that the extensive prescribed fires (more than 77.22 square miles [200 km2] of burns) actually translated to increased elk populations in BNP. This was despite the higher forage biomass in burns (Sachro et al. 2005) and the higher forage quality for migrants in general (Hebblewhite et al. in press); migrants still declined due to wolf and grizzly predation. Furthermore, time-series modeling in both the Bow Valley and YHT area suggested that burning in areas with high-wolf density can actually reduce elk population growth rates (White et al. 2005, Hebblewhite et al. 2006). Although speculative, these studies suggest a bottom-up effect of fire on wolf numbers instead of elk mediated by rapid numeric responses of wolves. In essence, any increased elk productivity from fires translated to increased wolf productivity through a rapid numeric response. …
Relative Sensitivity of and Management Constraints to Changing Wolf Predation
The typical conclusion of previous studies where wolves limited prey densities to low numbers was usually a recommendation to reduce predation via large-scale wolf control (Hayes et al. 2003). While there is some controversy over the success of wolf controls (Orians et al. 1997), there is some experimental evidence that wolf control—when applied consistently to reduce wolf populations by greater than 80 percent over huge areas (more than 3,861.02 square miles [10,000 km2]) for long terms (5-years) at great financial costs can be partially successful at enhancing ungulate populations (Boertje et al. 1996; Bergerud and Elliot 1998; Hayes et al. 2003; Valkenburg et al. 2004) for short periods of time.
I feel compelled to reiterate, however, that the main conclusions of the authors of perhaps, to date, the best executed wolf-control study in the Yukon (Hayes et al. 2003) pointed out the seeming futility of their wolf-control program as a longterm solution to ungulate population declines. Within 2 years of the end of wolf control, wolf densities and ungulate vital rates returned to precontrol levels. To be successful, wolf control needs to be conducted for long periods of time with greater than 70 percent of the wolf population removed from huge areas (Hayes et al. 2003). While future harvest plans for wolves once delisting occurs will undoubtedly include some wolf harvest, it remains difficult to conceive of states being able to conduct wolf control at the spatial and temporal scales required to even obtain short-term increases in ungulate populations. …
