IV
REDUCING THE RISK OF TRANSMISSION FROM WILDLIFE TO CATTLE

This chapter looks at approaches to reducing the risk of transmission from wildlife to cattle and reviews previous vaccination efforts in state and national parks and national wildlife refuges. Because any control or eradication effort will involve some degree of vaccination, this chapter reviews the difficulties involved in vaccine delivery and the effects of cattle vaccination on control efforts. Eradication efforts necessarily will include a test and slaughter component, and that component is examined for the effects on genetic diversity. And finally, this chapter looks at the prospects of natural regulation and successful brucellosis control.

PREVIOUS BISON-VACCINATION PROGRAMS IN NATIONAL AND STATE PARKS

In the 1960s, government regulations were devised to regulate interstate and intrastate movement of bison, and programs were developed to eradicate brucellosis in bison in some national parks and wildlife refuges. Bison herds managed with a herd plan were generally successful in eliminating brucellosis. Data on those cases are limited to memoranda from state and federal groups concerned with developing programs for brucellosis control and eradication (Gilsdorf 1997).

Wind Cave National Park and Custer State Park in South Dakota are adjacent to each other, and their bison herds had intermingled. High seropositive rates indicated both herds were infected with B. abortus . Brucellosis in adjacent cattle herds was being eradicated and had been eliminated by 1963 through vaccination, testing, and removal of reactor cattle. In Wind Cave, serologic testing for brucellosis in 1945 revealed 85% seropositivity; in 1960,



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 107
IV REDUCING THE RISK OF TRANSMISSION FROM WILDLIFE TO CATTLE This chapter looks at approaches to reducing the risk of transmission from wildlife to cattle and reviews previous vaccination efforts in state and national parks and national wildlife refuges. Because any control or eradication effort will involve some degree of vaccination, this chapter reviews the difficulties involved in vaccine delivery and the effects of cattle vaccination on control efforts. Eradication efforts necessarily will include a test and slaughter component, and that component is examined for the effects on genetic diversity. And finally, this chapter looks at the prospects of natural regulation and successful brucellosis control. PREVIOUS BISON-VACCINATION PROGRAMS IN NATIONAL AND STATE PARKS In the 1960s, government regulations were devised to regulate interstate and intrastate movement of bison, and programs were developed to eradicate brucellosis in bison in some national parks and wildlife refuges. Bison herds managed with a herd plan were generally successful in eliminating brucellosis. Data on those cases are limited to memoranda from state and federal groups concerned with developing programs for brucellosis control and eradication (Gilsdorf 1997). Wind Cave National Park and Custer State Park in South Dakota are adjacent to each other, and their bison herds had intermingled. High seropositive rates indicated both herds were infected with B. abortus . Brucellosis in adjacent cattle herds was being eradicated and had been eliminated by 1963 through vaccination, testing, and removal of reactor cattle. In Wind Cave, serologic testing for brucellosis in 1945 revealed 85% seropositivity; in 1960,

OCR for page 107
a group of bison tested in Wind Cave had a reactor rate of 56%. Custer State Park had a 47% reactor rate. In April 1961, state and federal animal-health officials met with bison managers of both herds, and a herd plan was devised and agreed on. The plan included blood testing of the entire herd of adults and calves, immediate removal of reactors or permanent identification of reactors with later disposal, and continuing calfhood vaccination with S19. The Wind Cave and Custer bison herds were separated by a fence. In Wind Cave, the program followed lines of "natural management," and facilities for active control of the herd of 250 bison were not built. In the first blood testing in 1964, 37% of the bison were seropositive (Table IV-1); by 1985, the herd was seronegative. The combination of vaccination, serologic testing, and management with removal of reactor bison allowed Wind Cave National Park to eliminate brucellosis in 21 years. The program for the bison in Custer State Park followed lines of a commercial ranching operation. Capture facilities were built in 1960-1961. The first herd test, in the winter of 1961, found 119 reactors in 248 bison tested (Table IV-1). Bison were culled annually and sold or sent to abattoirs. All bison calves and yearlings were vaccinated annually. In 1967, the number of bison tested was increased to 2,110; the reactor rate was 5%. In 1973, the herd was seronegative, and in 1974 the park managers reduced the herd size from 1,750 to 1,000. Brucellosis had been eliminated in 10 yr, even though not all bison were tested each year. The U.S. Fish and Wildlife Service's Wichita Mountains Wildlife Refuge located in Comanche County, Oklahoma, had vaccinated its bison and longhorn cattle for brucellosis since the 1940s. Bison tested in October 1963 were seronegative, but in 1964 brucellosis suspects in elk and bison were found in the fall roundup. The origin of the disease is not known. The program developed for this herd included the following steps: blood-test all bison over 1 yr old in the fall of 1972, slaughter all bison that could not be gathered, send all test-positive bison to slaughter and collect tissues for isolation of B. abortus, divide bison herd into isolated groups on different pastures, conduct a complete herd test in the fall of 1973, discontinue vaccination of bison calves in 1973, and test another species for brucellosis. The Refuge also reduced the size of the herd from 781 to 345 in 1973. It took 8 yr to eliminate the disease, and the herd was considered free of brucellosis in May 1974.

OCR for page 107
Table IV-1. Bison seropositivity rates in parks that eliminated brucellosis. (Source: Gilsdorf 1997)   Reactors/No. Tested   Year Custer State Park Wind Cave National Park 1961 119/248 (48%)   1962 20/141 (14%)   1963 ?   1964 2/84 (2%) 81/220 (37%) 1965 0/16 (0%) 41/175 (23%) 1966 20/905 (2.2%) 16/173 (9.2%) 1967 113/2,110 (5%) 12/185 (6.5%) 1968 53/2,493 (2%) 7/194 (3.6%) 1969 3/1,335 (0.2%) 7/282 (2.5%) 1970 7/1,439 (0.5%) 1/75 (1.3%) 1971 1/1,142 (0.09%) 1/146 (0.7%) 1972 12/1,379 (0.9%) 1/146 (0/7%) 1973 0/108 (0%) ? 1974 ? 2/120 (1.7%) 1975 0/172 (0%)   1977 0/237 (0%)   1979   12/185 (6.5%) 1982   3/128 (2.3%) 1983   15/264 (5.7%) 1984   7/337 (2.1%) 1985   0/225 (0%) 1986   0/217 (0%) 1987   0/205 (0%) 1989   0/191 (0%) APPROACHES TO CONTROLLING OR ELIMINATING BRUCELLOSIS IN YNP Numerous approaches to controlling or eliminating brucellosis from the Greater Yellowstone Area (GYA) have been identified. Some are theoretical,

OCR for page 107
some are experimental, and others are technically possible. All should be considered for short- and long-term solutions. Vaccination alone with a vaccine that is protective will reduce but not eliminate B. abortus from the GYA (Peterson et al. 1991a). Development and use of an efficacious vaccine could greatly reduce the prevalence of brucellosis in the GYA. To be successful in bison, vaccination must be accompanied by prevention of contact with infected elk, and reduction of brucellosis in elk by reducing feeding-ground concentrations. Artificially controlling population growth in bison would make administering programs to eliminate brucellosis easier. The disparity of seroprevalence between feeding-ground elk and the northern herd suggests that exposure to infected material on the feeding ground is the driving force maintaining infection in elk. Management strategies to disperse elk from the feeding grounds for the 3 months before calving combined with an intensive vaccination program might eliminate the disease from elk. Discontinuing winter feeding of elk would eliminate the problem of elk congregating but have the consequence of drastically reducing the number of elk in the Jackson area. Vaccinating a high-enough proportion of elk is problematic because they are widely distributed, and foci of infection are numerous in Wyoming and the National Elk Refuge (NER). Vaccination and gradual removal of feeding grounds as elk foci would probably allow the gradual natural extinction of the disease in elk (K. Aune, Mont. Dept. Fish, Wildlife and Parks, pers. Commun., 1997). Vaccinating cattle and bison would make the risk of transmission from bison extremely low under current conditions. Spatial and temporal separation of cattle and bison would be a good first step toward risk reduction. Regional surveillance and monitoring of surrounding cattle in Montana, Wyoming, and Idaho might be required for early phases of any program. Vaccination combined with herd management—including culling and test-and-slaughter procedures, close surveillance and monitoring of disease prevalence, and spatial and temporal management of wildlife—could be used to eliminate the disease in bison. Vaccination alone would have to be continued indefinitely, but if it were combined with a test-and-slaughter program, brucellosis potentially could be eradicated in the GYA over time. The time required for eradication would be contingent on vaccine effectiveness, the slaughter rate, and efforts to reduce the population. Case studies of Custer State Park have shown that eradication would take 10-20 yr (Gilsdorf 1997).

OCR for page 107
"In theory and in practice, vaccination combined with test and slaughter is effective, second only to depopulation, in eradicating brucellosis from cattle" (T. Kreeger, WGFD, pers. commun., 1997). One program suggested for control and eventual eradication is vaccination and test and slaughter of bison and elk with restriction of perimeter cattle herds to steers and monitoring of peripheral cow herds. Key elements of the program are listed in below in order of importance; the first three were considered essential (S. Amosson, Texas A&M, pers. Commun., 1997): Test and slaughter of all segments. Perimeter control through use of steers (or full vaccination of cow herds). Vaccination of bison herds. Vaccination of elk herds. Opinions differ as to the likelihood of successful outcomes of the various programs. The likelihood of reduction of abortions in bison and elk and reduction in transmission to cattle seems high. The likelihood of eradication of B. abortus in the short term is low but would increase with appropriate levels of funding and an adequate vaccination program (E. Williams, Univ. Wyom., pers. commun., 1997). R. Mead (State vet., Wash., pers. commun., 1997) compared different approaches: Pros and Cons of Approaches to Control Vaccination only Vaccination with test-and-cull program slow faster expensive more traumatic to the animal more acceptable to the public less acceptable to the public requires more research greater risk of human injury adverse impact on GYA management Cost is a major problem. For the past decade, the annual federal budget for the U.S. Department of Agriculture program for brucellosis eradication in cattle has been about $65 million. Added to that are the indirect costs of

OCR for page 107
state programs and producer vaccination efforts, and production losses must be considered. Brucellosis eradication in cattle is a costly process. In combination with vaccination, alternative methods might be acceptable and effective. Testing and neutering of seropositive animals is an option that has not been considered. Neutered animals are not likely to spread brucellosis. However, surgical neutering might not be publically acceptable and might have undesirable effects on herd behavior. Immunocontraception to suppress persistence of B. abortus in bison should be examined, but current technical barriers, especially with delivery systems, and the potential for introducing genetic selection makes it unacceptable at present. Pregnancy reduction has been achieved in captive deer (Turner et al. 1996a) and free-roaming feral equids (Turner et al. 1996b), and new techniques also provide promise (Miller et al. 1997). Whatever methods are used, results will depend on the intertwined effects of bison and elk populations, predator numbers, food supply, and weather. FIELD DELIVERY OF A VACCINATION PROGRAM FOR YNP BISON Given the high seropositivity rate in YNP bison (about 50%), several people have noted that a test-and-slaughter program to eliminate brucellosis would differ little from a depopulation program. Neither depopulation nor a test-and-slaughter program alone is likely to be publicly acceptable in YNP. More realistic is the implementation first, of vaccination to reduce the seropositivity rate to a low level, and then, when the numbers that have to be removed are small, a test-and-slaughter program. That strategy could be conducted within the framework of an adaptive management approach. A program of vaccination for bison in the field in YNP in all likelihood will have to be conducted under several constraints. Most bison populations have been managed for many years by rounding up in specially designed corrals, and any incorrigible individuals that could not be managed were shot. Administering a brucellosis-elimination program similar to that used for domestic livestock (vaccination in conjunction with test and slaughter) is feasible in those cases. But rounding up has the consequence of some artificial selection for domestication because wildness and intractability, salient traits of wild bison, are disfavored. Those are important traits to retain in YNP bison, one of the few herds where it is feasible to maintain natural behavior, so rounding up is not likely to be acceptable. In addition, the construction of facilities necessary to handle bison would detract from the natural aura of the park and might have detrimental effects on the park ecosystem.

OCR for page 107
Consequently, whatever vaccine is developed probably will have to be applied to free-roaming bison. As noted above, brucellosis was eliminated in one herd (Wind Cave National Park) with vaccination and test and slaughter. Most research veterinarians think that the vaccine should be developed first and a method of application found later. However, in view of the constraints likely to pertain to application in YNP, it seems prudent to keep the field-delivery problems of YNP in mind, for they might influence the characteristics required of the vaccine. Vaccine Delivery in Food or via Injection Vaccine is likely to be delivered as bait, in food, or via remote injection. All have serious problems in the context of YNP bison. Putting vaccine in bait or artificial food has the drawback of not allowing control of doses. Dominant animals are likely to get multiple doses; other animals might get none. Bulls would be treated with cows. Perhaps low doses given ad lib would result in all animals receiving a common dose per body mass, but that seems optimistic. An alternative strategy would be to give the vaccine over a short time with feed spread to allow consumption by all individuals. Nevertheless, control of dose by feeding strategy would be difficult. Furthermore, it would be difficult to prevent other species from eating treated feed. Imaginative approaches using genetic engineering to put the vaccine in native plants have been proposed (D. Sands, Mont. State Univ., pers. commun., 1997), but that technology is probably many years away and will be subject to the same problems of control of dose as the artificial feed route or worse. In addition, serious issues of ecologic and evolutionary consequences probably are best not worked out in YNP, the crown jewel of the U.S. national park system. Remote injection is probably a more realistic approach. However, bison are not fed in the winter and are not as approachable as elk that are currently vaccinated on winter feeding grounds with biobullets. The range of biobullets is very short, and this is not a good technique for wary animals. Syringe darts have a greater range, and they can be fired from the ground or by helicopter. Disturbance is an issue. A vaccination program probably would best be conducted in winter when visitors are fewer, and perhaps it could be carried out away from the roads to which snow machines are confined. Whether unrecovered syringes are an environmental hazard would have to be addressed. A major problem will be to distinguish treated animals from untreated animals. Temporary marking as is done with elk will be possible with biobullets or syringes; even then, the dark coat of the bison will make such marks

OCR for page 107
more difficult to see. In a milling herd, the identification of treated individuals will pose a problem. Directing the program at recognizable age or sex classes would reduce the number of animals that need to be vaccinated each year. Because of they are small and easy to recognize, calves are a favorable target group, except that it can be difficult to distinguish male from female calves (females are the likely sex to be vaccinated) in the field. Yearling cows might have greater potential in being present in small numbers, distinguishable from bulls, and recognizable by experienced field personnel. Venereal Immunization The vaccination of bulls to immunize cows has been suggested by some. Venereal immunization of cows by purposeful infection of dominant breeding bulls at the right time so that the infection is at its peak in the breeding season theoretically would immunize most of the adult females in the population by vaccinating only a small subset of the population. Old dominant bulls are most recognizable by size and often recognizable individual marks. According to the study of Berger and Cunningham (1994; J. Berger, U. Nev., pers. commun., 1997), 6-yr-old and older bulls do most of the breeding, and they constitute about only 6-7% of the population. The low number of vaccinations would be more feasible to deliver in the field and would intrude on the segment of the population that would manifest the least demographic consequences of disturbance. There is no evidence to suggest that venereal immunization would be effective. Vaginal epithelium is a strong barrier and lacks the macrophages that make the uterus susceptible to infection. Failure of venereal transmission of B. abortus in cattle and other species is based on that difference and underlies the failure of experimental intravaginal inoculation to transmit Brucella spp. The number of males that become infected, the percent of infected males that excrete B. abortus in semen, and the failure of bulls with testicular lesions and pain to breed all make venereal transmission unlikely. In addition, new vaccine strains of B. abortus have the attribute of reduction of reproductive tract tropism. Selecting the vaccine strain of B. abortus for appropriate characteristics (such as higher rates of shedding of B. abortus in the semen), timing of injection in the bulls, or similar refinements of the technique might overcome these problems.

OCR for page 107
VACCINATION OF CATTLE Given the difficulties of vaccinating bison, the most workable method of reducing the risk of transmission of brucellosis from bison and elk to cattle in the GYA is vaccination of cattle. Cattle are already rounded up and handled, so the major impediment to uniform vaccination against brucellosis is the associated cost. Most cattle in the region already are being vaccinated for brucellosis, and this program is the most cost-effective way of reducing potential transmission from wildlife in the short term. Vaccination is required in Idaho and strongly recommended in Montana and Wyoming. Until a program of elimination is in the implementation stage, cattle vaccination should be universal in the area surrounding the GYA. LIMITING CATTLE NEAR PARK BORDERS TO STEERS The presence of geographic barriers that reduce the spread of brucellosis by limiting contact of infected bison and elk in the GYA with susceptible cattle clearly is important. One approach based on this principle is to reduce contact by making the first line of contact a population of cattle that has a reduced likelihood of maintaining B. abortus in the herd. Limiting cattle near YNP borders to steers or spayed heifers could lower the risk of transmission in the treated animals. Castrated males and spayed heifers are unlikely to transmit brucellosis. Removal of the testes and ovaries deletes the source of gonadal hormones that initiate reproductive growth at puberty and maintain the reproductive system in adulthood. Although the animals might become infected, they will not transmit the disease when living and do not develop tissue titers of bacteria that can sustain the disease in nature. Certainly steers and spayed heifers do not transmit brucellosis through abortion or its byproducts, but the requirement for sexual maturity and the presence of gonads in transmission has not been clearly established. Opposition has been voiced to limiting cattle production around the GYA to steers. Limiting cattle production will not eliminate brucellosis; elk will remain throughout a large landscape. Some also believe that this method wrongly places responsibility on agricultural segments of society. A related plan that has merit and could be carried out immediately is to establish perimeter zones in which animal populations are monitored in progressively vigorous ways. Zones established nearest the GYA would have increased disease surveillance, vigilant monitoring, vaccination, and contact-reporting

OCR for page 107
programs. Implementation of such a perimeter-zone strategy should include collection of serologic data in cattle vaccinated with RB51. This would more clearly establish whether transmission of B. abortus actually is occurring in the GYA. EFFECTS OF TEST-AND-SLAUGHTER PROGRAMS ON GENETIC DIVERSITY Reduction of wild populations to low numbers for any reason raises concerns over loss of genetic diversity (Denniston 1977; Frankel and Soulé 1981). Bison in YNP contain lineages that go back without interruption to the aboriginal stocks in the area (Meagher 1973). In fact, the YNP population is the only extant bison population that has not been derived solely from stocks held in captivity at some point in their history. Plains bison (Bison bison bison) stock was introduced to YNP (Meagher 1973), but if any current population is likely to contain unique alleles from the original bison (Bison bison athabaska, which occupied the valleys in the Rocky Mountains), it is the YNP herd. Consideration of minimal numbers must include genetically effective population size, which is influenced by sex ratio, breeding behavior, the number of nonbreeding individuals, and other factors. It can be substantially less than actual population size in a polygynous species like bison. Berger and Cunningham (1994) calculated effective population size in bison to be 21-46% of actual size, depending on the variables included in the formula applied. For example, if the goal were to maintain an effective population of at least 500 bison for gene conservation, an actual population of 1,087-2,381 would be required. Protein-electrophoresis data suggest that the YNP herd and the Wind Cave National Park herd have the highest heterozygosity (a measure of genetic diversity) among the 12 public herds of bison in the United States (Stormont 1993). However, DNA studies of the YNP bison, using both mitochondrial DNA (inherited only through the female lineage) and nuclear DNA (microsatellites—a sensitive measure of genetic changes over time), revealed no unique alleles in that population (J. Derr, Texas A&M, pers. commun., 1997). The lack of unique alleles in YNP might indicate that mountain bison were not very different from plains bison or that genetic diversity was lost because of the bottlenecks and long periods at small population sizes characteristic of the population (see Figure II-2). Considering the influence of effective population size, the number of bison was small in the early years. Alternatively, it might be that because bison stocks have been mixed frequently, including the

OCR for page 107
movement of bison from YNP to other populations, their genes are represented in other herds. Whatever the explanation for the apparent absence of unique alleles, the DNA evidence suggests that conservation of genetic diversity is not a major issue in the management of the YNP bison. Technical decisions can be based on mainly demographic criteria. However, only a small part of the genome has been analyzed, and prudence dictates that minimal effective size be considered in any program of brucellosis eradication. NATURAL REGULATION AND BRUCELLOSIS CONTROL The analysis of movements of bison and elk outside YNP highlights the importance of the park's policy of ''natural regulation" of ungulates in relation to the possibility of transmission of B. abortus to domestic livestock. Given that bison and elk populations are large, they will continue to move out of the park; that is especially true of bison in years with hard winters. It cannot be determined with precision what the transmission risk is, because with current knowledge, it is too small to measure with accuracy. However, whatever the risk, it will be increased by more frequent movement of greater numbers of bison and elk beyond park boundaries. Whether the increase in risk is trivial or important depends on how the epidemiologic evidence is interpreted. In any event, the YNP policy of natural regulation influences the probability of transmission of B. abortus from wildlife to cattle and therefore must be considered in this study, although specific recommendations regarding the policy are beyond the study charge. Natural-regulation policy, particularly as it pertains to the northern YNP elk herd, has been controversial (Houston 1982; Chase 1986; Kay 1990). As with brucellosis, the science is insufficient to settle arguments over whether it is wise. Critical tests are difficult because the issues are linked with larger patterns of nature that are not readily reduced to a research hypothesis. The "experiment" is conducted by nature, lacks controls and replications, and yields only one set of data points per year. Each side in the debate interprets the results with reference to its own position. Natural regulation is a useful label for the controversy, but the real issue is human intervention: should it occur? If so, to what degree and when? The mandate of the National Park Service in large wild parks is to maintain wildlife in as natural a state as possible. If that mandate cannot be carried out in a park as large as Yellowstone, it has little prospect in any of the other national parks in the lower 48 states. Given the ubiquitous alteration of landscapes by human pressures outside the parks, YNP and a few other large

OCR for page 107
parks are the main remaining baseline areas where the course of nature can be observed (Sinclair 1983; Arcese and Sinclair 1997). They are the controls for the national and global human experiment. No one would seriously argue that YNP—with its infrastructure of roads, accommodations, and millions of visitors—is in a natural state. It is not immune to edge effects along its borders or to regional or global phenomena. It is not a complete ecologic entity, as indicated by larger designations, such as the "Greater Yellowstone Area" and the "Greater Yellowstone Ecosystem." Nevertheless, retaining the ecologic integrity of the park requires setting limits and not intruding where nature can manage without human intervention. In fact, many historical human intrusions, such as husbandry of ungulates and attempted elimination of large predators, are lamented by most conservation scientists and require restoration efforts, such as the recent reintroduction of the wolf. Those interventions were deemed advisable at the time, and they should stand as stark reminders of the limits of knowledge and understanding—then and now. Decisions to intervene should be supported by clear and compelling evidence and a consensus of experts that they are necessary. On the other side of the issue is a long history and practice of managing ungulate populations to meet prescribed goals. Virtually every bison and elk population in the country outside YNP is managed to some degree, and they were managed inside YNP at an earlier time—bison until 1967 (Meagher 1973) and elk until 1969 (Houston 1982). Neither has been managed in YNP since the implementation of the natural-regulation policy in 1969. Elk that migrate out of YNP in the fall and winter are hunted in the surrounding states, and elk are still managed in the Grand Teton National Park area (Boyce 1989; Toman et al. 1997). Management goals for ungulates usually are to stabilize the population within some range deemed to be in the best interest of the health of the population and to maintain some state of vegetation that is judged to be desirable. To some critics of YNP policies, the need to control ungulates to prevent irruptive population behavior and its consequent detriment to vegetation is a guiding principle in ecology. Obviously, the knowledge and technical capability are available to manage bison and elk to stabilize their numbers inside YNP at some upper limit. The important question, therefore, is not whether we can, but whether we should do so. The debate over bison and elk management in YNP has taken place before a larger backdrop of a shift in paradigms in the field of ecology. Stated simply, the earlier paradigm was based on equilibrium states—the "balance of nature"—whereas recent emphasis has been on nonequilibrium fluctuations over time and space (Simberloff 1982; Pickett et al. 1992). According to the latter paradigm, fluctuations are the norm, and any apparent balance

OCR for page 107
is an artifact of averaging over space or long periods. The shift was prompted by the growing body of evidence that variation, rather than similarity, across landscapes is the predominant characteristic of nature and that analyses of longer time series were showing the same to be true over time (for example, Sousa 1984; Pickett and White 1985; Whitlock 1993; Reice 1994; Russell 1994). At the same time, little hard evidence could be marshaled to support the equilibrium view. Traditionally, irruptive population behavior in ungulates (exceeding average carrying capacity and suffering declines due to starvation) was viewed as exceptional and attributable to the actions of humans: confinement by fencing or development, introductions to new areas, removal of predators, and so on. Such irruptions showed strong associations in time with human effects on ecosystems, and certainly the actions of humans contributed to the degree of irruption, even if not they were not the cause. But close examination of the evidence supporting those cases shows that it is often deficient (Caughley 1987; McCullough 1997). In fact, there is little solid evidence to support the traditional view. The population estimates in most cases were largely guesses, and contributing variables were not measured. Even if the traditional view is accepted on logical grounds, it cannot be assumed that because human intervention can produce irruptions, nature does not. Actually, the earlier irruptions can be characterized better as comparisons between unhunted populations and hunting-controlled populations than as human intrusion versus the natural order. Human intervention was ubiquitous; it was, and still is, hard to find places where natural behavior could be observed as a "control." That ungulate populations in undisturbed nature tended toward an equilibrium based on interactions with predation and resources was inferred from the assumption that natural predators caused effects equivalent to human hunting in the cases in which hunting had stabilized populations (Leopold 1933, 1940). Early results in the moose and wolf populations in Isle Royale National Park gave initial credence to the equilibrium view (Mech 1966, 1970; Allen 1979), but later results dispelled any semblance of equilibrium (Peterson and Page 1988; McLaren and Peterson 1994). Despite restoration efforts in parks and reserves, finding natural areas from which to obtain time-series data to address the behavior of ungulate-predator-resource-climate interactions continues to be difficult. Control and replicate areas on a scale necessary for ungulates and their predators are not available, and treatments are not possible if natural management is pursued. Consequently, long series of years are necessary to observe the effects of "natural experiments" that occur by chance. Interpretations are based on

OCR for page 107
correlations, and, because nature seldom conducts clean experiments, cause and effect often are complicated by covariance of variables. The difficulties are similar to those of determining whether global warming is occurring and, if so, whether it is due to natural processes or anthropogenic carbon dioxide emissions. The answers will come, but they are not in the immediate offing. Despite the difficulties, a growing body evidence from largely intact natural areas seems to indicate that ungulate populations commonly fluctuate over considerable numbers. Abundant evidence from YNP indicates that neither bison nor elk conform to a stable equilibrium model. Both species have shown a persistent tendency to increase to the limits of the environment. That the northern elk herd shows a dynamic equilibrium should not obscure the fact that the amplitude of its population fluctuation over time is considerable. The elk herd shows a dynamic equilibrium in that the rate of increase has declined in a density-dependent manner at high population numbers (Coughenour and Singer 1996; M. Taper, Mont. State Univ., and P. Gogan, USGS, pers. commun., 1997). The dynamic equilibrium mean carrying capacity (that is, where elk population growth rate equals 0) appears to be about 14,000-18,000 elk (Houston 1982; Merrill and Boyce 1991; M. Taper, Mont. State Univ., and P. Gogan, USGS, pers. commun., 1997). Bison, in contrast, have not yet shown evidence of natural regulation over the range of numbers recorded, and their geographic expansion has already exceeded the boundaries of YNP. Natural regulation of bison in YNP appears to be unlikely. Control of bison numbers presents difficult choices that had to be addressed in the recent past and probably will have to be addressed again, independently of the brucellosis issue. Although brucellosis has catalyzed the recent controversy, the fundamental issue is the need to respond to burgeoning bison numbers that are overflowing park boundaries. Growth of bison and elk populations has been expressed in the presence of native predators other than the wolf. Whether the recently reintroduced wolves, whose population has grown quickly to about 100, will have an appreciable effect on bison and elk population growth remains to be seen. A computer model using a wolf population of 76 predicted an elk population reduction over 100 yr of 15-25% but no evidence of imposing an equilibrium (Boyce 1993). The same model predicted a bison population reduction of less than 10%. But judging by the experiences with moose on Isle Royale (Peterson and Page 1988; McLaren and Peterson 1994) and other ungulates elsewhere (for example, Ballard et al. 1987; Gasaway et al. 1983, 1992; NRC 1997), it is questionable whether wolves will impose a stable equilibrium on either bison or elk in YNP. No stable equilibrium is apparent on Isle Royale,

OCR for page 107
with almost 40 yr of data probably the best documented case of interaction between populations of a natural, large ungulate and wolves. Mech et al. (1987) noted that winter snow is the most important variable controlling the number of moose on Isle Royale and that wolves had a relatively minor effect. A concept in ecology that well could be applied to the YNP elk and bison population issue is that of source-and-sink dynamics (Pulliam 1988). It is based on the common observation that some habitat patches favorable to a species result in production of new individuals greater than can be supported in the habitat patch; subsequent population pressure results in the dispersal of part of the population into surrounding poor habitats where reproduction and survival are low. Thus, a source population is a consistent exporter of individuals, whereas a sink population cannot maintain itself except for the continuous influx of individuals from the source. The net outcome for both areas is a dynamic equilibrium, with the majority of reproduction happening in source habitats and the majority of mortality in sink habitats. The concept would appear to apply well to the YNP elk and bison situation. As long as the natural-regulation policy is followed, increasing elk and bison populations will stretch the winter capacity of YNP, and, at least in harder winters, animals will be forced out of the park. The incompatibility of bison with developed areas and private lands will require either culling or relocation; both have the demographic consequence of removing the animals from the system. Because it is a source habitat, YNP can continue to be managed according to the natural-regulation policy. The sink habitat outside the park can be the area where adjustment is applied by the combination of relocation and mortality that is compatible with brucellosis containment and public acceptance. (The sink habitat in the GYA is a sink not because of inadequate habitat quality. It is a sink because of high mortality, in this case, mortality induced by humans.) The populations of bison and elk can be stabilized over the combined areas (the ecosystem) in a manner that duplicates or mimics a common system in nature without violating the mandates under which these lands are managed. In many ways, source-and-sink dynamics already applies to the management of YNP elk and bison. At minimum, source-and-sink dynamics has heuristic value. For example, it puts various proposals about management of federal lands (mainly by the Forest Service) into a different perspective. It has often been suggested that bison and elk should be favored over other uses, particularly livestock grazing, on these lands. That proposal is attractive for its inherent appeal of contributing to the conservation of bison and elk, and the lands are already in public ownership. The perspective of source-and-sink dynamics, however, reveals two drawbacks of this approach.

OCR for page 107
First, it assumes that the additional area will contribute to natural regulation. That might be true for northern-range elk, whose population shows evidence of regulation. However, Coughenour and Singer (1996) and M. Taper (Mont. State Univ.) and P. Gogan (USGS, pers. commun., 1997) have noted that elk populations in the northern range previously have expanded in response to habitat increase. It is possible that they might do so again, although Coughenour and Singer (1996) note that additional suitable habitat might not be available. Bison, however, have shown no evidence of regulation, but only range expansion. The likely consequence of shifting the boundary of protection from YNP to surrounding public lands is that bison, and perhaps elk, populations will simply increase further, shifting the boundary to a new point—private lands—where even greater numbers of bison will have to be dealt with. Those limits need to be confronted unless our nation is ready to make a substantial commitment to acquire private lands for bison conservation. If such a commitment is to be made, it needs to be determined whether it should be made in the GYA where bison conservation is already near the potential of the ecosystem. Bison conservation might be better served if, for example, the commitment were directed to the Great Plains, the heartland of the aboriginal bison range. Second, the sharp juxtaposition of source-and-sink areas maximizes the conflicts because the policies of one jurisdiction or the other will have to be compromised to some extent by the lack of a transition area. Establishment of buffer zones between parks or reserves and the surrounding lands used for agriculture or other purposes is a well-accepted approach in land planning (for example, Harris 1984; Western and Wright 1994). The buffer zone is an area in which management can facilitate the transition between goals of two contrasting land uses. In a source-and-sink model, then, the shift from favorable to unfavorable habitat (because of conflict with human land uses) is accommodated along a gradient in the buffer zone between protected and unprotected areas. Federal lands outside YNP could, and to some degree already do, serve that function. The buffer zones also could be linked with perimeter zones for brucellosis control discussed earlier. ADAPTIVE MANAGEMENT Because neither sufficient information nor technical capability is available to implement a brucellosis-eradication program in the GYA at present, eradication as a goal is more a statement of principle than a workable program. The best that will be possible in the near future will be reduction of the risk of transmission of B. abortus from wildlife to cattle.

OCR for page 107
Biologically sound wildlife policy can be developed most efficiently using adaptive management (Walters 1986; Lancia 1996, NRC 1997). An adaptive management approach that had research designed to provide data to reduce areas of current uncertainty should eventually give a more realistic assessment of the feasibility of eradication of B. abortus in the GYA. Adaptive management means conducting management activities as hypothesis tests, the outcome of which will direct the subsequent efforts to achieve the ultimate goal. Adaptive management is not just modifying management in light of experience; it is designing management intervention to maximize what can be learned from the experiments (NRC 1997).