UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Ecology of disease in bighorn sheep in Hells Canyon, USA Cassirer, Elizabeth Frances 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2005-99452x.pdf [ 7.24MB ]
Metadata
JSON: 831-1.0092258.json
JSON-LD: 831-1.0092258-ld.json
RDF/XML (Pretty): 831-1.0092258-rdf.xml
RDF/JSON: 831-1.0092258-rdf.json
Turtle: 831-1.0092258-turtle.txt
N-Triples: 831-1.0092258-rdf-ntriples.txt
Original Record: 831-1.0092258-source.json
Full Text
831-1.0092258-fulltext.txt
Citation
831-1.0092258.ris

Full Text

E C O L O G Y O F D I S E A S E I N B I G H O R N S H E E P I N H E L L S C A N Y O N , U S A by E L I Z A B E T H F R A N C E S C A S S I R E R B.Sc. , University o f Montana, 1981 M . S c , University of Idaho, 1990 A THESIS S U B M I T T E D PN P A R T I A L F U L F I L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in I : '1- D E P A R T M E N T O F ( Z O O L O G Y ) T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A January 2005 © Elizabeth Frances Cassirer, 2005 11 A B S T R A C T I investigated the dynamics of eight populations of a bighorn sheep (Ovis canadensis) metapopulation in Hells Canyon, U S A from 1997 - 2003. Pneumonia was the most common cause (46%) o f adult mortality and the primary factor limiting population growth. Cougar (Puma concolor) predation was the second most frequent source (30%) o f adult mortality but did not reduce the rate o f population growth significantly. Average annual survival o f adult males (0.84) was lower than females (0.91). Pneumonia was the most common known cause of lamb mortality (86%) and pneumonia-related mortality was detected whenever summer lamb survival was less than 50%. Summer pneumonia epizootics in lambs were independent o f epizootics in adults. Survival o f transplanted sheep was initially high, but was lower than resident sheep during two of six years and at two o f three release locations, due to pneumonia-related mortality. Survival and pneumonia outbreaks were not related to population size, growth rates, climate, or nutrition. Sex and source (resident or transplant) were the best predictors of adult survival and disease-related mortality. Individuals that died from pneumonia were not closer to domestic sheep (O. aries) or goats (Capra hircus) than those that did not, but the potential for transfer of pathogens between wi ld sheep and domestic and goats existed for all populations. Over 70 biovariants o f Pasteurella spp. and Mannheimia spp. bacteria were cultured from live sheep during health sampling at capture and from dead sheep at necropsy. Only two biovariants o f P. trehalosi were isolated from live sheep in all populations at capture. Both biovariants were also prevalent in sheep that died from pneumonia Pasteurella multocida was detected more commonly in the lungs of dead sheep than in resident sheep at capture and was not found in transplanted sheep. Lungworms (Protostrongylus spp.) did not appear to play a role in pneumonia-related mortality. Mic ro - and macroparasites differed among populations, but prevalence o f potentially pathogenic organisms in pneumonic sheep was lower or did not differ from that sheep at capture or sheep that died from causes other than pneumonia. Clinical results suggested multiple causes o f pneumonia outbreaks within this population, presence of undetermined factors affecting immunocompetence, or an inability to detect or correctly classify virulent organisms. iv T A B L E O F C O N T E N T S Abstract i i List o f Tables • v List o f Figures v i i Acknowledgments ix Chapter 1 Introduction 1 1.1 Research Hypothesis 6 1.2 Literature cited 7 Chapter 2 Population dynamics of bighorn sheep in Hells Canyon 14 2.1 Methods 14 2.2 Results 21 2.3 Discussion 34 2.4 Literature cited 36 Chapter 3 Ecological and individual characteristics associated with bighorn sheep mortality in Hells Canyon, 1997 - 2003 42 3.1 Introduction 42 3.2 Methods 44 3.3 Results 50 3.4 Discussion 64 3.5 Literature Cited 67 Chapter 4 Pathogens associated with bronchopneumonia in bighorn sheep in Hells Canyon 76 4.1 Introduction 76 4.2 Materials and methods 78 4.3 Results 82 4.4 Discussion 103 4.5 Literature cited 109 Chapter 5 Conclusions 117 5.1 Literature cited 125 V L I S T OF T A B L E S Table 2.1. Annual adult survival rates in 8 Hells Canyon bighorn sheep populations, 1997 -2003 24 Table 2.2. Candidate models of adult bighorn sheep survival in Hells Canyon, 1997 - 2003.. 25 Table 2.3. Parameter estimates derived from the best model of adult bighorn sheep survival in Hells Canyon, 1997 - 2003 25 Table 2.4. Observed productivity of radiocollared bighorn ewes in 8 Hells Canyon bighorn populations, 1997 - 2003 26 Table 2.5. Summer survival and recruitment of lambs in 8 Hells Canyon populations, 1997 -2003 27 Table 2.6. Occurrence of epizootics in adults and lambs 1997 - 2003 30 Table 2.7. Bighorn population size and average annual rate o f increase (r) in Hells Canyon, 1997-2003 32 Table 3.1. Bighorn demographics in 8 Hells Canyon populations, 1997 - 2003 51 Table 3.2. Whole blood and liver selenium values (ppm) in 8 bighorn populations in Hells Canyon, 1997-2003 58 Table 3.3. Min imum observed distance between radio-collared bighorn sheep and domestic sheep and goats in 8 Hells Canyon populations, 1997 - 2003 59 Table 3.4. Candidate models of survival of adult bighorn sheep in 8 populations in Hells Canyon, 1997-2003 60 Table 3.5. Candidate models of disease-related mortality o f adult bighorn sheep in 8 populations in Hells Canyon, 1997 - 2003 61 Table 3.6. Parameter estimates derived from the best model of adult survival in Hells Canyon, 1997-2003 61 Table 3.7. Parameter estimates derived from the best model of disease-related mortality in Hells Canyon, 1997 - 2003 : '. 61 Table 4.1. Pasteurella trehalosi and P. multocida biovariants in bighorn sheep Hells Canyon and in source populations of sheep transplanted to Hells Canyon, 1997 - 2004 84 L I S T O F T A B L E S , cont'd. v i Table 4.2. Mannheimia spp. biovariants in bighorn sheep Hells Canyon and in source populations of sheep transplanted to Hells Canyon, 1997 - 2004 85 Table 4.3. Pasteurella and Mannheimia bacteria cultured from 23 adult bighorn sheep and 25 lambs that died from bronchopneumonia in Hells Canyon, 1997 - 2004 89 Table 4.4. Prevalence of Pasteurella and Mannheimia species and beta-hemolysis in 6 bighorn sheep populations in Hells Canyon, 1997 - 2004 94 Table 4.5. Average titers to Pasteurella in adult bighorn sheep in 6 Hells Canyon populations and source populations of sheep transplanted to Hells Canyon, 1997 - 2004 94 Table 4.6. Prevalence o f titers to respiratory viruses and Anaplasma spp. in adult bighorn sheep in 6 Hells Canyon populations at capture, 1997 - 2004 97 Table 4.7. Prevalence o f titers to respiratory viruses and Anaplasma spp. in adult bighorn sheep in 6 Hells Canyon populations and source populations of sheep transplanted to Hells Canyon, 1997 - 2004 98 Table 4.8. Fecal prevalence and mean intensity of gastrointestinal parasites and lungworms in bighorn sheep Hells Canyon and in source populations o f sheep transplanted to Hells Canyon, 1997 - 2004 98 Table 4.9. Fecal lungworm (Protostrongylus spp.) larvae infection prevalence and intensity (larvae per gram) in 175 bighorn sheep sampled at capture in 6 populations in Hells Canyon, 1997-2004 99 Table 4.10. Gastrointestinal parasite prevalence and abundance in feces o f 132 adult bighorn sheep captured in Hells Canyon, 1997 - 2004 101 Vl l L I S T O F F I G U R E S Figure 2.1. Hells Canyon bighorn sheep metapopulation, 2003 17 Figure 2.2 Causes o f mortality of adult radio-collared bighorn sheep in 8 Hells Canyon populations, 1997-2003 22 Figure 2.3 Average survival rates and causes of mortality of adult radio-collared bighorn sheep by month in 8 Hells Canyon populations, 1997 - 2003 23 Figure 2.4. Lamb survival distribution functions in relation to occurrence o f pneumonia-caused summer mortality in Hells Canyon bighorn sheep, 1997 - 2002 28 Figure 2.5. Hazard functions estimated at two-week intervals from birth to 133 days o f age and 3-week intervals from 22 A p r i l - 15 September in 9 population-years where pneumonia epizootics occurred and 27 population-years where no pneumonia was detected 29 Figure 2.6. Least-squares regression between survival o f lambs from birth through October (weaning) and lamb:ewe ratios in March (recruitment) 31 Figure 2.7. Change in Hells Canyon bighorn sheep population sizes 1997 - 2003 in relation to pneumonia-caused adult mortality 32 Figure 2.8. Annual bighorn ewe survival and exponential population growth rate (r) in Hells Canyon 1997-2003 33 Figure 3.1. Population size, adult survival, and juvenile survival in 8 bighorn sheep populations in Hells Canyon, 1997 - 2003 50 Figure 3.2. Normal monthly average temperature and total precipitation at 5 weather stations in the Hells Canyon study area, 1974 - 2003 53 Figure 3.3. Annual variation in precipitation, temperature, and adult survival in Hells Canyon, 1997-2003 54 Figure 3.4. Average seasonal pattern o f F N values in three climate zones within the Hells Canyon project area, 1999 - 2003 56 Figure 3.5. Average monthly fecal nitrogen values in 8 bighorn populations in Hells Canyon, 1999-2003 57 Figure 3.6. Median lambing dates in 8 bighorn populations 1997 - 2002. 62 Figure 4.1. Prevalence o f biochemical types ofPasteurella and Manneheimia spp. isolated from bighorn sheep in Hells Canyon at capture, source populations of bighorn sheep transplanted to Hells Canyon, and mortalities in Hells Canyon, 1997 - 2004 88 L I S T O F F I G U R E S , cont'd. v i i i Figure 4.2. Prevalence of Pasteurella trehalosi, Mannheimia spp., and P. multocida isolated from resident and transplanted bighorn sheep at capture and mortalities in Hells Canyon, 1997-2004 91 Figure 4.3. Proportion of pneumonic adults and lambs where only nonhemolytic P. trehalosi was isolated from the lungs and those where beta-hemoloytic P. trehalosi or Mannheimia spp., P. multocida were isolated from lungs either separately or together in Hells Canyon 1997 - 2004 92 Figure 4.4. Prevalence of beta-hemolytic P, trehalosi and Mannheimia spp. isolates from bighorn sheep in Hells Canyon and in source populations o f bighorn sheep transplanted to Hells Canyon 93 Figure 4.5. Least-squares regression between bacteria and virus data collected at capture and demographic characteristics of bighorn sheep in 6 Hells Canyon populations, 1997 -2004 103 ix A C K N O W L E D G M E N T S This research would not have been started or finished without the support of Pete Zager at the Idaho Department of Fish and Game. M y advisor, Tony Sinclair, was instrumental in designing the project and I appreciate his support throughout. M y committee, at various times Jamie Smith, Peter Arcese, Martin Adamson, Charley Krebs, Carl Walters, and David Shackleton provided guidance, support, and constructive criticism. External examiner John Gross also provided helpful suggestions as did examining committee members John Richardson and Judy Myers. Members o f the Hells Canyon Bighorn Sheep Restoration Committee assisted the project from the outset and sustained support for the duration. L loyd Oldenburg, T i m Schommer, V i c Coggins, and John Beecham were among the first of many individuals instrumental in obtaining and maintaining support for the Hells Canyon Initiative. Thanks to many people for assistance in the field, including Wendy Lammers, Hol ly Akenson, Jamie Nelson, Crystal Strobl, Corey Kallstrom, Regan Berkley, Mar ia Bennett, Marc Hammond, Bryce Krueger, Glen Landrus, Ray Vinkey, Robert Adair, Gene Majors, R ick Ward, Joe Myers, Doug Gadwa, Mark H i l l , R ick Cooper, and Roy Lombardo. Appreciation is also extended to the British Columbia Ministry of Water, Land, and A i r Protection, Environmental Stewardship Divis ion; Alberta Ministry of Sustainable Resource Development, Fish and Wildlife Divis ion; and the Montana Dept. o f Fish, Wildlife, and Parks for support of bighorn sheep restoration in Hells Canyon. This study was funded under the Hells Canyon Initiative by the Idaho Dept. of Fish and Game, Oregon Dept. o f Fish and Wildlife, Washington Dept. o f Fish and Wildlife, Bureau o f Land Management, Foundation for North American W i l d Sheep, Oregon Hunters Association, Turner Foundation, and Wildlife Forever. Ultimately, this project was the result of indispensable and much appreciated contributions and collaboration by many individuals and organizations. 1 Chapter 1 I N T R O D U C T I O N Historically, both species of North American wi ld sheep, bighorn (Ovis canadensis), and thinhorn (O. dalli) were relatively abundant and widely distributed throughout rugged mountain and desert terrain in the west. However, by the early 20th century, bighorn sheep had been extirpated from much o f their range in the United States. A s with other wi ld ungulates in North America, over hunting and competition with livestock were factors in this historical decline, however, diseases transferred from livestock also played an especially important role (Buechner 1960). Despite restrictive hunting regulations and an active program o f transplants and habitat conservation, bighorn sheep in the United States have increased from 15,000 - 18,000 in 1960 (Buechner 1960) to approximately 45,000 today (Shackleton 1997) an average annual increase of less than 4%. Cougar (Puma concolor) predation (Hass 1989, Wehausen 1996, Ross et al. 1997, Hayes et al. 2000, Kamler et al. 2002), habitat (Risenhoover et al. 1988), and disease (Hobbs and Mi l l e r 1992, Singer et al. 2000) have been proposed as limiting factors for native and restored bighorn sheep populations. Disease continues to be an important conservation and management issue for bighorn sheep. Pneumonia epizootics have occurred in Desert bighorn sheep (O. c. nelsoni) in Arizona, north through every state to Rocky Mountain bighorn sheep (O. c. canadensis) in Alberta and California bighorn sheep in British Columbia (O. c. California). On average, at least one bighorn population in North America experiences an all-age pneumonia epizootic every year (Monello et al. 2001). Often, 20 - 80% of the affected population dies from bronchopneumonia. Pneumonia epizootics also occur frequently in lambs and are typically described subsequent to all-age epizootics. These post all-age epizootics in lambs are 2 thought to be caused by transmission of pathogens to lambs from surviving ewes that have become temporarily resistant carriers. Lamb epizootics have been documented for 3 -10 years following all-age epizootics (Spraker et al. 1984, Festa-Bianchet 1988, Coggins 1992, Ryder et al. 1992). Pneumonia epizootics are not reported in thinhorn sheep populations, (Heimer 1992) or in the northern Rocky Mountain bighorn range in west-central Alberta. Epizootics occur when a new pathogen is introduced into a naive population, or when an existing pathogen becomes virulent due to a change in host susceptibility, or a change in pathogen virulence or abundance. Epizootics continue as long as transmission of the lethal pathogen occurs and stop when the conditions that caused initiation dissipate (Heesterbeek and Roberts 1995, Wills 1996). Spread of directly-transmitted microorganisms is usually considered density-dependent: the higher host density, the greater the opportunity for transmission of the pathogen (Anderson and May 1978). Bighorn sheep are affected by many interacting pathogens, most transferred from livestock, some endemic, and some where the source is unknown (Becklund and Senger 1967, Spraker and Adrian 1990, Boyce and Zarnke 1996). Contagious ecthyma, Parainfluenza-3, sinusitis, epizootic hemorrhagic disease, and Chlamydia have all been documented in bighorn sheep. Common macroparasites include lungworm (usually Protostrongylous rushi and P. stilesi and occasionally Muellarius capillaris) and scabies (Psorptes ovis). Scabies is likely introduced (Buechner 1960, Boyce and Zarnke 1996), (although apparently bighorn and domestic sheep scabies are currently host-specific; Boyce and Zarnke 1996) whereas Protostrongylus spp. are probably endemic (Becklund and Senger 1967). Intestinal parasites include Coccidia and nematodes including Trichuris spp., 3 Moniezia spp., Ostertagia spp, and Skrjabinema spp. (Becklund and Senger 1967, Worley and Seesee 1992). ! Historically, scabies and lungworm, particularly P. stilesi (Buechner 1960) were thought to be precipitating factors in most bighorn sheep pneumonia epizootics. However, anthelmintic treatments, although effective at reducing parasite loads (Miller et al. 1988, Foreyt et al. 1990, Worley and Seesee 1990), have not been shown to increase population growth (Miller et al. 1996), infecting lambs with lungworms does not reduce survival (Samson et al. 1987), and fecal lungworm intensity is not necessarily a good predictor of pneumonia epizootics (Festa-Bianchet 1991). Pneumonia epizootics also occur in lambs in the absence o f any macroparasite infection (Marsh 1938, Spraker et al. 1984). Bacteria in the genera Pasteurella and Mannheimia (formerly Pasteurella haemolytica), (Angen et al. 1999) are currently considered important proximate causal agents o f pneumonia in bighorn sheep either in conjunction with or independently o f other factors (Miller 2001). Pasteurella trehalosi, P. multocida, and Mannheimia spp. bacteria are common, diverse, and normally commensal organisms colonizing the upper respiratory tract of ruminants (Boyce 2004) with multiple serotypes, biovariants, and strains of differing virulence (Frank 1989, Davies et al 1997a, 1997b, Saadati et al. 1997, Brogden et al. 1998). Certain strains are apparently endemic to bighorn sheep whereas others are not (Heimer et al. 1992, Ward etal . 1997). Unlike lungworms, which infect bighorn sheep through an intermediate host (several species of land snails; Soulsby 1968), Pasteurella bacteria are spread through mucous and direct contact. Transmission occurs when hosts make physical contact (nose to nose) or 4 through mucous transfer via sneezing or coughing. There is currently no effective vaccine for preventing pasteurellosis in bighorn sheep or livestock. Little is known about why bighorn pneumonia epizootics start, what determines transmission rates or the intensity of mortality, or why die-offs stop. However, several hypotheses have been proposed for initiation. Pneumonia epizootics are known to occur when bighorn sheep come into contact with domestic sheep. Captive studies have shown that healthy domestic sheep can carry Pasteurella spp. ox Mannheimia spp. bacteria that cause lethal pneumonia in bighorn sheep (Foreyt and Jessup 1982, Onderka and Wishart 1988, Foreyt 1989, Callan et al. 1991). One possible reason for the differential susceptibility to pathogens in domestic and bighorn sheep is that although domestic and wi ld sheep share a common Asian ancestor (O. orientalis musimon), ovids probably colonized North America via the Bering Sea Land Bridge during the Pleistocene (Cowan 1940, Geist 1971) and were isolated from contact with their ancestors. Asiatic caprids may be resistant to diseases carried by domestic sheep (Valdez 1982), but American wi ld sheep are apparently completely vulnerable to certain pathogens carried by domestic and Eurasian sheep (Silflow et al. 1989, Foreyt 1994, Foreyt et al. 1996). Specifically, bighorn and Dal l ' s sheep (O. dalli dalli) neutrophils have been demonstrated to be more susceptible to leukotoxins produced by Pasteurella than domestic sheep or cattle neutrophils (Sillflow et al. 1994). The transmission o f lethal pathogens from domestic to wi ld sheep is well documented. However, bighorn pneumonia epizootics can also occur in the absence of known contact with domestic sheep. Theories as to the initiating factors in these epizootics include suboptimal habitat quality (Cook et al. 1989), loss o f traditional movement patterns, which concentrates bighorn sheep rather than distributing use over available habitat 5 (Risenhoover et al. 1988), mixing source populations of transplanted bighorn sheep with potentially differential vulnerability to pathogens (Sandoval et al. 1987), chronic stress due to human disturbance, or as a result of predisposition through initial infection with viral or verminous pathogens as occurs in "shipping fever" in livestock (Spraker et al. 1984, Belden et al. 1988, Kraabel and Mi l l e r 1997). Bighorn die-offs may also be correlated with increasing population densities or numbers o f susceptible individuals in the absence o f overt precipitating environmental or pathological factors (Hobbs and Mi l l e r 1992, Dunbar 1992). Although catastrophic die-offs have been occurring in bighorn sheep since the 1800's, there is little empirical evidence for recent evolution of resistance to livestock-associated pathogens. Certain alleles in the major histocompatibility complex ( M H C ) , which plays an important role in initiating the immune response to pathogens, are associated with parasite resistance and juvenile survival in Soay sheep (O. aries) (Paterson et al. 1998). However, there is no evidence o f selection in the bighorn M H C (Boyce et al. 1997) which might suggest increased resistance, and there is no indication of reduced variation in the M H C (Gutierrez-Espeleta et al. 2001) that might explain historic susceptibility to infection. The rapidness and relative unpredictability o f epizootics and the logistic difficulties associated with studying them have hindered progress in understanding etiology and epidemiology. However, as more information has been gathered, it has also generated new theories on the mechanism, rather than eliminating old ones. The lack o f consensus on causes of epizootics suggests that different and multiple mechanisms may come into play during various die-offs, and there may not be one unifying cause. 6 1.1 R E S E A R C H H Y P O T H E S I S In this research I tested the hypothesis that the Hells Canyon bighorn sheep population is regulated by disease through density-dependent transmission o f introduced pathogens into large populations of sheep independently o f environmental factors such as nutrition and climate. Under this hypothesis I predicted the following: 1. Disease w i l l be the major source o f mortality. 2. Disease-related mortality and survival rates w i l l not be correlated with nutrition or climatic conditions. 3. Disease-related mortality w i l l increase with increasing population size and there w i l l be a threshold population size for the establishment o f infection. 4. Populations experiencing disease-related mortality w i l l be more likely to contact domestic sheep. 5. Populations experiencing disease-related mortality w i l l carry pathogens not endemic to bighorn sheep. 6. Antibody titers to or prevalence of Pasteurella or other pathogens w i l l be highest in populations experiencing pneumonia-related mortality and lowest in healthy populations. In chapter two I evaluate whether disease was a major source of mortality (prediction one) by analyzing causes and patterns o f adult and lamb mortality and effects on rate of population growth. In chapter three I describe the effects o f extrinsic and intrinsic factors at the metapopulation, population, and individual level on survival, population growth, and 7 occurrence o f pneumonia in adults and lambs (predictions two - four). In chapter four I summarize health testing and necropsy data to assess whether introduced pathogens were associated with disease-related mortality (predictions five and six). 1.2 L I T E R A T U R E C I T E D Anderson, R. M . and R. M . May . 1978. Regulation and stability of host-parasite population interactions. I. Regulatory processes. Journal of Animal Ecology 47:219 - 247. Angen, O., R. Mutters, D . A . Caugant, J. E . Olsen, and M . Bisgaard. 1999. Taxonomic relationships of the [Pasteurella] haemolytica complex as evaluated by D N A - D N A hybridizations and 16S r R N A sequencing with proposal of Mannheimia haemolytica gen nov.. comb, nov., Mannheimia granulomatis comb, nov, Mannheimia glucosidal sp. nov., Mannheimia ruminalis sp. nov. and Mannheimia varigena sp. nov. International Journal o f Systematic Bacteriology 49:67 - 86. Becklund, W . W . and C . M . Senger. Parasites of Ovis canadensis canadensis in Montana, with a checklist of the internal and external parasites of the Rocky Mountain bighorn sheep in North America. Journal o f Parasitology 53:157-165. Belden, E . L . , E . S. Will iams, E . T. Thorne, H . J. Harlow, K . White, and S. Anderson. 1988. Effect o f chronic stress on immune system function of Rocky Mountain bighorn sheep. Proceedings of the biennial symposium of the Northern W i l d Sheep and Goat Council 7:76-91. Boyce, W . M . . P. W . Hedrick. N . E . Muggli-Cockett, S. Kalinowski , M . C . T. Penedo, and R. R. Ramey II. 1997. Genetic variaton of major histocompatibility complex and microsatellite loci: a comparison in bighorn sheep. Genetics 145:421-433. 8 and R. L . Zarnke. 1996. Antibody responses to Psorptes sp. mites in Dal l sheep (Ovis dalli). Journal of Wildlife Diseases 32:711-713. Boyce, J. D . , R. Y . C. L o , I. Wilkie , A n d B . Adler. 2004. Pasteurella and Mannheimia. Pp. 273 - 294 in C . L . Gyles, J. F . Prescott, J. G . Songer, and C. O. Thoen eds. Pathogenesis of bacterial infections in animals, 3 r d ed. Blackwell Publ., Ames, IA . Brogden, K . A . , Lehmkuhl, H . D . and R. C . Cutlip. 1998. Pasteurella haemolytica complicated respiratory infections in sheep and goats. Veterinary Research 29:233-254. Buechner, H . K . 1960. The bighorn sheep in the United States, its past, present, and future. Wildlife Monographs 4:174 pp. Callan, R. J., T. D . Bunch, G . W . Workman, and R. E . Mock . 1991. Development of pneumonia in desert bighorn sheep after exposure to a flock o f exotic wi ld and domestic sheep. Journal of the American Veteterinary Medicine Association 198:1052-1056. Cook, J. G . , E . B . Arnett, L . L . Irwin, and F . G . Lindzey. 1989. Ecology and population dynamics of two transplanted bighorn sheep populations in southcentral Wyoming. Univ . o f Wyoming, Laramie. 234 pp. Cowan, I. M . 1940. Distribution and variation in the native sheep o f North America. A m . M i d i . Nat. 24:505-580. Davies, R. L . , S. Arkinsaw, and R. K . Selander. 1997a. Evolutionary genetics of Pasteurella haemolytica isolates recovered from cattle and sheep. Infection and Immunity, 65:3585-3593. 9 , , and . 1997b. Evolutionary genetics of Pasteurella haemolytica isolates recovered from cattle and sheep. Infection and immunity 65:3585-3593. Dunbar, M . R. 1992. Theoretical concepts of disease versus nutrition as primary factors in population regulation o f wi ld sheep. Proceedings of the biennial symposium of the Northern Wi ld Sheep and Goat Council 8:174-192. Festa-Bianchet,M. 1988. A pneumonia epizootic in bighorn sheep, with comments on preventive management. Proceedings of the biennial symposium o f the Northern W i l d Sheep and Goat Council 6:66-76. . 1991. Numbers of lungworm larvae in feces of bighorn sheep: yearly changes, influence of host sex, and effects on host survival. Canadian Journal of Zoology 69:547-554. Foreyt, W . J. A n d Jessup D . A . 1982. Fatal pneumonia o f bighorn sheep following association with domestic sheep. Journal o f Wildlife Diseases 18:163-168. 1989. Fatal Pasteurella haemolytica pneumonia in bighorn sheep following direct contact with clinically normal domestic sheep. American Journal of Veterinary Research 50:341-344. , V . L . Coggins, and P. Fowler. 1990. Psoroptic scabies in bighorn sheep in Washington and Oregon. Proceedings of the biennial symposium of the Northern W i l d Sheep and Goat Council 7:135-142. , T. Parker, and V . Coggins. 1990. Safety and efficacy o f fenbendazole against Protosrongylous infections in Rocky Mountain bighorn sheep (Ovis canadensis canadensis). Proceedings of the biennial symposium of the Northern W i l d Sheep and Goat Council 7:118-124. 10 . 1994. Effects of controlled contact exposure between healthy bighorn sheep and llamas, domestic goats, mountain goats, cattle, domestic sheep, or mouffon sheep. Proceedings of the biennial symposium of the Northern W i l d Sheep and Goat Council 9:7-14. , R. M . Silflow, and J. E . Lagerquist. 1996. Susceptibility of Dal l sheep (Ovis dalli dalli) to pneumonia caused by Pasteurella haemolytica. Journal of Wildlife Diseases 32:586-593. Frank, G . H . 1989. Pasteurellosis of cattle. Pp. 197-222 in C . F . Adlam and J. M . Rutter eds. Pasteurella and pasteurellosis. Academic Press, London, United Kingdom. Geist, V . 1971. Mountain sheep: a study in behavior and evolution. Univ . of Chicago Press. 383 pp. Gutierrez-Espeleta, G . A . , P. Hedrick, S. T. Kalinowski , D . Garrigan, and W . M . Boyce. 2001. Is the decline o f desert bighorn sheep from infectious disease the result of low M H C variation? Heredity 86:439-450. Hass, C . C . 1989. Bighorn lamb mortality: Predation, inbreeding, and population effects. Canadian Journal o f Zoology 67:699-705. Hayes, C . L . , E . S. Rubin, M . C . Jorgensen, R. A . Botta, and W . M . Boyce. 2000. Mountain lion predation o f bighorn sheep in the peninsular ranges, California. Journal o f Wildlife Management. 64:954-959. Heesterbeek, J. A . P. and M . G . Roberts. 1995. Mathematical models for microparasites of wildlife. Pp. 9 0 - 122 in B . T. Grenfell and A . P. Dobson eds. Ecology of infectious diseases in natural populations. Cambridge University Press, Cambridge, U K . 11 Heimer, W . , R . L . Zarnke, and F. J. Mauer. 1992. Pharyngeal microflora o f Dal l and domestic sheep in Alaska: management implications. Proceedings of the biennial symposium o f the Northern W i l d Sheep and Goat Council 8:193-201. Hobbs, N . T. and M . W . Mi l le r . 1992. Interactions between pathogens and hosts: simulation of pasteurellosis epizootics in bighorn sheep populations. Pp. 997-1007 in D . R. McCul lough and R . H . Barrett eds. Wildlife 2001. Populations. Elsevier Appl ied Science, Essex, England. 1163 pp. Kamler, J. F. , R. M . Lee, J. C . deVos, W . B . Ballard, and H . A . Whitlaw. 2002. Survival and cougar predation of translocated bighorn sheep in Arizona. Journal of Wildlife Management 66:1267-1272. Kraabel, B . J. and M . W . Mi l le r . 1997. Effect of simulated stress on susceptibility of bighorn sheep neutrophils to Pasteurella haemolytica leukotoxin. Journal of Wildlife Diseases 33:558 - 566. Marsh, H . 1938. Pneumonia in Rocky Mountain bighorn sheep. Journal of Mammalogy 19:214-219. Mil le r , M . W . 2001. Pasteurellosis. Pp. 330 - 339 in . E . S. Williams and I. K . Barker eds. Infectious diseases o f wi ld mammals 3 r d ed. Iowa State University Press, Ames. 558 PP-and N . T. Hobbs. 1988. Ivermectin for treating lungworms in bighorn sheep: an appropriate technology with a specific utility. Wildlife Society Bulletin 16:236-238. , Vayhinger, J . , Rousch, S., Verry, T., Torres, A . A n d V . Jurgens. 1996. Experimental evaluation o f mountain sheep transplanting and disease treatment. Colorado Division o f Wildlife. Final Report. Monello, R. J., D . L . Murray, and E . F . Cassirer. 2001. Ecological correlates of pneumonia epizootics in bighorn sheep populations. Canadian Journal of Zoology 79:1423-1432. Onderka, D . K . and W. D . Wishart. 1988. Experimental contact transmission of Pasteurella haemolytica from clinically normal domestic sheep causing pneumonia in Rocky Mountain bighorn sheep. Journal of Wildlife Diseases 24:663-667. Paterson, S., K . Wilson, and J. M . Pemberton. 1998. Major histocompatibility complex variation associated with juvenile survival and parasite resistance in a large unmanaged ungulate population (Ovis aries L . ) . Proceedings of the National Academy of Sciences, U S A 95:3714-1719. Risenhoover, K . L . , J. A . Bailey, and K . A . Wakelyn. 1988. Assessing the Rocky Mountain bighorn sheep management problem. Wildlife Society Bulletin 16:346-352. Ross, P. I., M . G . Jalkotzy, and M . Festa-Bianchet. 1997. Cougar predation on bighorn sheep in southwestern Alberta during winter. Canadian Journal of Zoology 74:771-775. Ryder, T.J., E.S.Will iams, K . W . M i l l s , K . H . Bowles, and E . T Thorne. 1992. Effect of pneumonia on population size and lamb recruitment in Whiskey Mountain bighorn sheep. Proceedings o f the biennial symposium o f the Northern W i l d Sheep and Goat Council 8:136-146 Samson, J., Holmes J. C , Jorgenson J. T., and W . D . Wishart. 1987. Experimental infections of free-ranging Rocky Mountain bighorn sheep with lungworms (Protostrongylous spp.; Nematoda: Protostrongylidae). Journal o f Wildlife Diseases 23:396-403. Shackleton, D . M . (ed.) and the I U C N / S S C Caprinae Specialist Group. 1997. W i l d sheep and goats and their relatives. Status survey and conservation action plan for caprinae. I U C N , Gland, Cambridge, U K . 390pp. Silflow, R. M . , Foreyt, W.J . , Lagerquist, J. E . 1994. Evaluation of the cytotoxicity of various isolates of Pasteurella haemolytica from bighorn sheep and other ungulate populations. Proceedings of the biennial symposium of the Northern W i l d Sheep and Goat Council 9:1-6. , Foreyt,W. J., Taylor, S. M . , Laegreid, W . W., Liggitt, H . D . , and R. W . Leid . 1989. Comparison of pulmonary defense mechanisms in Rocky Mountain bighorn (Ovis canadensis canadensis) and domestic sheep. Journal of Wildlife Diseases 25: 514-520. Singer, F . J., E . S. Williams, M . W . Mil ler , and L . C . Zeigenfuss. 2000. Population growth, fecundity, and survivorship in recovering populations o f bighorn sheep. Restoration Ecology 8:75-84. Soulsby, E . J. L . 1968. Helminths, arthropods, and protozoa of domesticated animals. Will iams and Wilkins Co . 824 pp. Spraker, T. R . , C . P. Hibler, G . G . Schoonyeld, and W. S. Adney. 1984. Pathologic changes and microorganisms found in bighorn sheep during a stress-related die-off. Journal o f Wildlife Diseases 20:319-327. and W . J. Adrian. 1990. Problems with "multiple land use" dealing with bighorn sheep and domestic livestock. Proceedings of the biennial symposium of the Northern W i l d Sheep and Goat Council 7:67-75. Ward, A . C . S., Hunter, D . L . Jaworski, M . D . , Benolkin, P. J., Dobel, M . P., Jeffress, J .B. and G . A . Tanner. 1997. Pasteurella spp. in sympatric bighorn and domestic sheep. Journal o f Wildlife Diseases 33:544-557 Wehausen, J. D . 1996. Effects o f mountain lion predation on bighorn sheep in the Sierra Nevada and Granite Mountains o f California. Wildlife Society Bulletin 24:471-479. Wil ls , C . 1996. Ye l low fever, black goddess: the coevolution o f people and plagues. Addison-Wesley Publ. Co. , Inc. 324 pp. Worley, D . E . and M . E . Seesee. 1992. Gastrointestinal parasites of bighorn sheep in western Montana and their relationship to herd health. Proceedings of the biennial symposium o f the Northern W i l d Sheep and Goat Council 8:202-212. 15 Chapter 2 P O P U L A T I O N D Y N A M I C S O F B I G H O R N S H E E P I N H E L L S C A N Y O N , 1997-2003 Historically, bighorn sheep were abundant and widely distributed in western North America. However, by the early 20th century, bighorn sheep had been extirpated from much o f their range in the United States. Over hunting and competition with livestock were factors in this historical decline, however, diseases transferred from livestock, particularly domestic sheep, also played an important role (Buechner 1960). Despite restrictive hunting regulations and an active program o f transplants and habitat conservation, the United States bighorn sheep population remain at 5 - 10% of estimated historical numbers (Buechner 1960, Valdez and Krausman 1999). Cougar predation (Wehausen 1996, Hayes et al. 2000, Kamler et al. 2002), habitat loss (Risenhoover et al. 1988), and disease (Hobbs and Mi l l e r 1992, Singer et al. 2000) have been proposed as limiting factors for native and restored bighorn sheep populations. I evaluated patterns o f survival and population growth in a restored bighorn sheep metapopulation in Hells Canyon of the Snake River in Oregon, Idaho, and Washington 1997 - 2 0 0 3 . 2.1 M E T H O D S Study Area The study area encompassed 22,732 k m 2 along the Snake, Salmon, and Grande Ronde Rivers in Idaho, Oregon, and Washington (117°52 '30"W, 46°30"N to 116 °15 'N, 16 44°45 'W) (Figure 2.1). Elevations ranged from 245 m in river canyons to over 2740 m in the Seven Devils, ID and Wallowa Mountains, OR. Climate is generally continental and dry with light precipitation (25 cm to 127 cm), low relative humidity, and wide ranges in temperature (-20 °C to >40 °C). Columbia River basalts are the dominant geologic formation. Plant associations include primarily perennial bunchgrass (Agropyron spicatum and Festuca idahoensis) communities, with deciduous riparian shrub stringers and upland shrub-fields. Douglas-fir (Pseudotsuga menziesii) and ponderosa pine (Pinus ponderosa) stands occur on northerly aspects (Johnson and Simon 1987). Over 50% of the area was publicly owned and managed by federal and state agencies. Habitat improvements have included termination o f most domestic sheep grazing allotments, the most recent in 1999. Several domestic sheep operations still remain on the edges o f and peripheral to the bighorn metapopulation on public and private land. Bighorn sheep were extirpated from Hells Canyon and the surrounding area by 1945 (Smith 1954, Johnson 1980, Oregon Department o f Fish and Wildlife 1992). From 1971 to 2002, 492 bighorn sheep (O. c. canadensis) from 11 source populations were released into the project area and 124 bighorn sheep were relocated within the Hells Canyon area. Fifteen bighorn populations were established during this period and at least 6 pneumonia epizootics occurred between 1972 and 1996 (Hells Canyon Bighorn Sheep Restoration Committee 2004). The most recent epizootic prior to this study occurred in 1995-96, when about one third o f the population died. Most mortality was concentrated in populations in Oregon and Washington (Cassirer et al. 1996). In 2003, 756 bighorn sheep were counted during spring aerial and ground surveys and the metapopulation was estimated at approximately 900 bighorn sheep. Legend I I Bighorn sheep study population I I Other bighorn sheep population Figure 2.1. Hells Canyon bighorn sheep metapopulation, 2003 18 Data collection and analysis Between 1997 and 2003, 151 radio-collared adult (>1 yr old) sheep (112 F, 39 M ) were monitored in 8 populations (Figure 2.1), for a total o f 39 population-years (ewes 316 radio-years and rams 80 radio-years). These included 98 radio-collared resident sheep captured in 5 populations by netgunning from a helicopter or in a corral trap, and 53 radio-collared sheep captured under drop nets and translocated from Spences Bridge, British Columbia (33) in December 1997, and Cadomin, Alberta (20) in February 2000. The relocated sheep started the B i g Canyon and M u i r Creek populations and supplemented the Asotin population (Figure 2.1) Radio-collared bighorn sheep were located from the ground or from a fixed-wing aircraft at least bi-weekly, and often several times per week during the spring and summer for 3 - 6 years. Radio collars were equipped with a 4-hour delay mortality switch. When the mortality sensor was activated, the site was visited, and where possible, the sheep was collected for evaluation at the Washington Animal Disease and Diagnostic Laboratory ( W A D D L ) at the Washington State University Veterinary School in Pullman, W A . When this was not possible, a field necropsy was conducted and tissue and organs were collected for gross and histological investigation at W A D D L . In order to exclude capture and relocation-related mortality, data from transplanted animals were only included after the sheep had been in Hells Canyon for one year, and data from resident sheep were only included > 60 days post-capture. A l l resident sheep were judged to be healthy when handled in 1997 and all transplanted sheep were certified healthy by a provincial veterinarian. In 2000, one resident 19 ewe was diagnosed with chronic pneumonia and 2 ewes were diagnosed with mastitis. Two o f these sheep subsequently died during this study., A l l other sheep appeared healthy at capture. L o w to moderate levels of Psoroptes (scabies) infection were found in all resident populations except the Lostine population and in none of the transplanted sheep. The productivity o f radio-collared ewes and survival to weaning (October) o f their lambs (n = 283) were monitored through visual observation from 3 - 6 years per population. One year in the B i g Canyon population (2000) was omitted from analysis because monitoring was not sufficiently intensive to accurately evaluate timing and causes o f extensive lamb mortality. Ewes were observed weekly and usually several times per week during this period. A ewe was determined to have a lamb when it was observed alone with a lamb or nursing. Lambs were assumed dead when the ewe was no longer observes nursing or associating with a lamb. Mortality date was estimated as the midpoint between when the ewe was last observed with a lamb and when it was first observed without the lamb. Dead lambs were located through visual observation, often assisted by the behavior o f the ewe (Akenson 1998), and recovered where possible. The dam often stayed with the dead lamb for days and sometimes up to a week after the death o f the lamb. Recruitment was estimated from lamb:ewe ratios observed in February and March ground and aerial surveys (lamb age 10 months). Population size and survival of adults in populations and years where data on radio-collared animals were not available were also estimated from annual ground and aerial surveys. Observability of bighorn sheep in Hells Canyon is very high (approximately 87% from a helicopter, Idaho Dept. of Fish and Game, unpubl. data). 20 Monthly and annual survival rates of radio-collared sheep and summer lamb survival and hazard functions were calculated using staggered entry Kaplan-Meier analysis (Kaplan and Meier 1958, Pollock et al. 1989). I used the log-rank test to evaluate statistical equivalence among K - M curves for lambs. In this metapopulation, most lambs are born in May, and so annual survival and exponential rate of population increase (r = ln [N t/N t.i]) were calculated on a biological year (June - May). Annual survival and population growth rates were calculated for each population and are reported population-years. Proportional hazard survival models for discrete time intervals were used to evaluate the effects of sex, year, population, and the occurrence of predation and disease-caused mortality in a population on adult survival. The analysis over discrete time intervals (in this case, the 6 study years) allows inclusion of time-dependent covariates and variable hazard ratios among years. I used known-fate analysis in program SURPH (Skalski 2003) with a log-log or hazard link function: h(t,X) = ho(t)exp £/JX, where ho(t) is the baseline hazard function and X denotes a collection of p covariates. The hazard link assumes an exponential effect of hazards on survival S(t,,X)= [So(r)] where So(r) is the baseline survival function (Kleinbaum 1996). Model selection was based on Akaike's Information Critereon adjusted for small samples (AICC). An AIC C , difference from the best (smallest) value (AAICC) and Akaike weight (w) were calculated for each model (Burnham and Anderson 2002). T-tests were used to compare population growth rates in population-years with and without mortality due to pneumonia and predation and least-squares linear regression was used for analysis of survival rates and population growth. I used a general linear model and Tukey's studentized range test to test for differences in productivity among populations, and 21 a two-factor general linear model to test for an interactive effect o f predation and disease on exponential growth rates. Seasonal patterns and differences between sexes in cause-specific mortality were compared with chi-square tests. These analyses were conducted in S A S V8.2 (SAS 2001). Elastic6 was used for proportional sensitivities (elasticity) analysis (Wisdom and M i l l s 1997). 2.2 R E S U L T S Adult survival Sixty-one radio-collared animals (39 ewes and 22 rams) died during the study and cause o f mortality could be determined for 49. The most frequent known source o f adult mortality was disease (n = 21, 43%), followed by cougar predation (n = 13, 27%), falls or injuries (n = 11, 22%), and human-caused death (harvest, poaching, or vehicle collisions) (n = 4, 8%) (Figure 2.3). A l l populations were either closed to hunting or managed with limited entry harvest with 2 - 6 tags per year. Ninety percent o f disease-related mortality was due to pneumonia. Two ewes (10%) died as a consequence of severe scabies infections. Human-caused mortality accounted for a higher proportion o f known ram mortality (rams 20%, n = 4, ewes 0%). When human-caused and injury mortalities were combined there were no differences in causes of mortality in ewes and rams (%2 = 0.79; 2 df; P = 0.67). Most adult mortality occurred September - May . Causes of mortality differed seasonally (%2 = 15.77; 4 df; P = 0.003). Disease accounted for 70% o f mortality where a cause could be determined during fall and winter (October - January) and 67% of disease-22 • Disease • Predation • Fall/Injury ED Human-caused SUnknown Figure 2.2 Causes o f mortality o f adult radio-collared bighorn sheep (n = 61) in 8 Hells Canyon populations, 1997 - 2003. related adult mortality occurred during this period. Cougar predation accounted for 50% of mortality from known causes during spring (February - May) and 77% of predation occurred during this period. The combined category of injuries and human-caused mortalities did not vary seasonally (Figure 2.3). Annual survival o f radio-collared ewes in the 8 study populations averaged 0.91 (95% CI 0.87 - 0.95) and was higher than average annual survival o f radio-collared rams (0.84, 95% CI 0.75 - 0.92). Adult survival was highly variable among populations and among years (Table 2.1). Disease-related adult mortality was detected in 1999, 2000, 2001, and 2002 in 6 of 8 populations for 1 to 3 years per population (9 population-years) and not detected in 30 population-years (Table 2.1). 23 1.00 0.99 A 0 Unknown 0.98 B Human-caused 0 Fall/Injury • Disease 0.97 • Predation • Monthly survival 0.96 0.95 J Figure 2.3 Average survival rates and causes of mortality by month of adult radio-collared bighorn sheep (n = 151) in 8 Hells Canyon populations, 1997 - 2003. Ewe and ram survival were lower in populations and years in which pneumonia was detected (SeWe = 0.81; S r a m = 0.70) than in those where no disease was detected (S e w e = 0.94; Sram = Cougar predation of radiocollared animals occurred every year during the study except 1998 in at least 1 of 6 of the 8 populations (1 to 3 years per population). Ewe and ram survival were lower during years and populations in which cougar predation occurred in the absence of disease (n - 7; S e w e = 0.87, S^m - 0.78), than in populations and years with no predation or where both disease and predation occurred (n = 32; S e w e = 0.91, S r a m = 0.85). Although there were differences in survival between sexes, among populations, among years, and in the presence of predation and disease, the best model of adult survival included only a reduction in survival in populations and years when pneumonia was detected and differential survival between sexes (Table 2.2). 0.91). 24 Adult bighorn sheep were nearly 4 times as likely to die when pneumonia was present in the population than at other times, and rams were nearly twice as likely to die as ewes (inverse o f hazard ratio, Table 2.3). Table 2.1. Annual adult survival rates in 8 Hells Canyon bighorn sheep populations, 1997 -2003. E W E S 1997-98 1998- 99 1999-2000 2000-01 2001-02 2002-03 X Asotin 0.88 0.86 1 0.83 1 0.91 Big Canyon - 1 0.93 0.602 1 0.91 0.87 Black Butte 0.92 1 0.58 0.71 0.80 1 0.84 Imnaha 0.08 0.74 0.78 0.85 0.95 1 0.95 Lostine 0.81 0.71 0.98 1 1 0.94 0.96 Muir Creek - 1 0.87 0.69 1 0.50 0.82 Redbird 1 1 1 1 0.85 0.91 0.96 Wenaha 0.83 1 1 0.73 1 1 0.93 Average 0.87 0.92 0.88 0.81 0.93 0.91 0.91 R A M S 1997-98 1998- 99 1999-2000 2000-01 2001-02 2002- 03 X Asotin - - - 0.78 1.00 0.61 0.81 Big Canyon - 1 1 0.80 0.80 0.50 0.77 Black Butte 0.46 0.99 0.73 1 0.80 0.30 0.72 Imnaha 0.68 0.80 1 0.71 1 1 0.92 Lostine 0.72 0.77 0.87 0.80 1 0.75 0.94 Muir Creek - 1 0.83 0.50 1 - 0.83 Redbird 1 0.66 0.97 1 0.80 0.75 0.92 Wenaha 1 1 1 0.67 1 1 0.87 Average 0.77 0.89 0.91 0.78 0.91 0.72 0.84 Survival estimated from annual counts in italics. All other survival estimates from radio-collared animals. Population-years with pneumonia-related mortality are in bold. 25 Table 2.2. Candidate models o f adult bighorn sheep survival in 8 populations in Hells Canyon, 1997 - 2003. The best fitting model included accounting for the occurrence of pneumonia epizootics and differences in survival between ewes and rams. Model name N o . parameters L n Likelihood AICc AAIC C Wi N u l l 1 -195 93 393 89 27.43 0 00 Sex 2 -193 56 391 20 24.74 0 00 Year 6 -188 18 388 94 22.48 0 00 Population 8 -184 41 385 84 19.39 0 00 Population and Sex 9 -181 82 382 91 16.46 0 00 Year and Sex 7 -186 40 387 58 21.13 0 00 Population and Year 48 -149 37 440 85 74.40 0 00 Population, Year, and Sex 49 -147 70 441 90 75.45 0 00 Predation 2 -194 89 393 86 27.41 0 00 Predation and Sex 3 -192 57 391 31 24.86 0 00 Pneumonia 2 -182 11 368 30 1.84 0 28 Pneumonia and Sex 3 -180 15 366 45 0.00 0 72 Table 2.3. Parameter estimates derived from the best model of adult bighorn sheep survival in Hells Canyon, 1997 - 2003. The hazard ratio is defined as the hazard for one individual divided by the hazard for another individual, controlling for other covariates. Baseline survival was 0.91 (standard error 0.02). Variable P Standard error Hazard ratio (e13) Presence o f pneumonia (present = 1, absent = 0) 1.34 0.26 3.82 Sex (female = 1, male = 0) -0.55 0.27 0.57 Productivity, lamb survival, and recruitment Observed productivity (proportion of radio-collared ewes observed with lambs) averaged 0.80 over all populations and years, and ranged from 0.45 to 1. Productivity was not significantly different among populations except for the populations with the highest 26 (Redbird x = 0.92) and lowest (Wenaha x = 0.63) productivity (F7,29 = 2.85; P = 0.02) (Table 2.4). Lamb survival to weaning, (birth through October) and recruitment were highly variable among populations and among years (Table 2.5). Survival to weaning of lambs born to radio-collared ewes ranged from 0 - 100%. Forty-three recently-dead lambs were found during the study. The cause of death could be determined in 25 cases, and 88% (22) of these were due to pneumonia. The other 3 were cougar predation, probable contagious ecthyma, trauma, and starvation due to mortality o f the ewe (one each). Lambs as young as 2 days were diagnosed with pneumonia and pneumonia-caused lamb mortality was detected as early as 30 M a y and as late as 26 December. Pneumonia-related lamb mortality was detected in at least one population every summer except 2000, and occurred from 1-4 summers in 5 o f the 8 populations (9 population-years, Table 2.5). In all populations, during years where lamb survival to weaning was less than 50%, dead lambs were recovered and diagnosed with pneumonia. Table 2.4. Observed productivity of radiocollared bighorn ewes in 8 Hells Canyon bighorn populations, 1997 - 2003. Population N o . years No . ewe-years Average productivity SD Asotin 6 40 0.80 0.11 B i g Canyon 5 60 0.84 0.11 Black Butte 7 62 0.84 0.16 Imnaha 4 52 0.72 0.21 Lostine 3 47 0.79 0.04 M u i r 5 52 0.84 0.18 Redbird 7 88 0.92 0.06 Wenaha 7 68 0.63 0.11 Average (total) 5 52 (469) 0.80 27 Table 2.5. Summer survival o f lambs born to radio-collared bighorn ewes and recruitment based on March lamb:ewe ratios in 8 Hells Canyon populations, 1997 - 2003. Summer Survival Recruitment 1997 1998 1999 2000 2001 2002 1998~ 1999 2000 2001 2002 2003 X X Asotin - 1 0.83 0.80 0.60 0.75 0.80 - 0.64 0.53 0.13 0.5 0.48 0.46 Big Canyon - 0.92 0.86 0.50 0.78 0.171 0.65 - 0.47 0.71 0.06 0.21 0 0.29 Black Butte 0.63 0.65 0.50 0,8 ' 0.25 1 0.64 0.33 0.32 0.39 0.30 0.23 0.53 0.35 Imnaha - - ' - 0.80 0.56 0.56 0.64 - - 0.28 0.42 0.35 0.60 0.43 Lostine - - - 1 0.85 0.72 0.86 0.24 0.35 0.27 0.50 0.55 0.36 0.57 Muir - - - 0.91 0.78 0.14 0.61 - 0.29 0.57 0.79 0.35 0.13 0.43 Redbird 1 0.25 0.55 0.72 0.83 0.66 0.67 0.32 0.18 0.24 0.55 0.16 0.59 0.34 Wenaha 0.22 0.14 0.27 1 0.25 0.80 0.45 0.08 0.20 0.19 0.29 0.17 0.41 0.23 Average 0.62 0.59 0.60 0.82 0.61 0.60 0.67 0.24 0.35 0.40 0.38 0.32 0.39 0.39 1 Population-years summer pneumonia-related mortality was diagnosed in lambs in bold. Survival distribution functions differed significantly during 9 population-years (71 lambs) when lambs were diagnosed with pneumonia M a y - August (X2 = 53.506, 1 df, pO.OOOl , (Figure 2.1, Figure 2.4). Mortality rates o f lambs were significantly higher in populations affected by pneumonia between 24 June - 5 August and from 4 2 - 9 8 days o f age (X2 > 54.193, 1 df, p < 0.0001) (Figure 2.4). N o relationship was apparent between the occurrence of summer epizootics in lambs and pneumonia-related mortality in adults. Pneumonia in lambs was detected in 1 population where no pneumonia was detected in adults and no lambs were diagnosed with pneumonia in 2 populations where pneumonia-related mortality occurred in at least 1 adult during the study (Table 2.6). Two pneumonia epizootics in lambs occurred the summer after pneumonia-related mortality was detected in adults and 5 occurred following a year without adult pneumonia-related mortality. One o f 9 epizootics in lambs occurred the summer prior to an adult epizootic in the population. 28 Figure 2.4. Survival distribution functions in relation to occurrence of pneumonia-caused summer mortality in lambs in Hells Canyon, 1997 - 2002. Lamb survival was similar from birth through the first week in June, and up to 24 days o f age, and then declined rapidly in populations affected by pneumonia. 29 0.025 0.02 0.015 0.01 H 0.005 49 63 77 Age (days) 91 105 119 0.025 i +J E 0.02 -ortal 0.015 -E dai o . o i -E O.OOS -> < O -3-May 24-May 14-Jun 5-Jul Date 26-Jul 16-Aug 6-Sep Figure 2.5. Hazard functions estimated at two-week intervals from birth to 133 days of age and 3-week intervals from 22 A p r i l - 15 September in 9 population-years where pneumonia epizootics occurred (solid line) and 27 population-years where no pneumonia was detected (dashed line). Table 2.6. Occurrence of epizootics in adults and lambs in Hells Canyon, 1997 - 2003. L indicates summer lamb pneumonia mortalities, A indicates adult pneumonia mortalities. Shaded areas indicate no data. ADULT YEAR 1997-98 1998-99 1999-00 2000-01 2001-02 2002-03 LAMB YEAR (SUMMER) 1997 1998 1999 2000 2001 2002 Redbird L Black Butte1 L L A Wenaha1 L I. L A L Big Canyon AL(?)2 A L Muir A A AL Imnaha A Lostine A Populations experiencing high rates of adult mortality in 1995-96 epizootic. 2 Possible pneumonia outbreak in Big Canyon lambs in summer 2000. Monitoring was not sufficiently intensive to detect source of lamb mortality. In population-years when summer pneumonia epizootics did not occur, lambs experienced significant fall and winter mortality. When summer lamb survival was greater than 50%, there was no correlation between summer lamb survival and recruitment (Figure 2.6). Lamb recruitment was not significantly lower in population-years with disease-related adult mortality (t = 0.87; 36 df; P = 0.39) or in populations the year following those with disease-related adult mortality (t = 0.42; 35 df; P = 0.68). Lamb survival to weaning Figure 2.6. Least-squares regression between survival of lambs from birth through October (weaning) and lamb:ewe ratios in March (recruitment). Recruitment was related to summer lamb survival (y = 0.4003x + 0.0875, R2 = 0.288, solid line). However, due to variable fall and winter mortality, survival to weaning and recruitment were not related when summer survival was greater than 50% (y = 0.1013x + 0.3574, R 2 = 0.0119, dotted line). Population growth Between 1997 - 2003, the Hells Canyon metapopulation increased from approximately 640 - 895 sheep (r = 0.06). This included releases of 115 sheep into the metapopulation. In the absence of releases, annual population growth rate was estimated at r = 0.04. The 8 study populations increased from 5 1 0 - 6 1 0 sheep, r = 0.03 (Table 2.7). Annual growth rates of individual study populations were highly variable ranging from r = -0.01 to r = 0.16. Populations declined (r = -0.08) during population-years when disease-related adult mortality was detected and did not decline in its absence (r = 0 .11 ; / = 3.83; 37 df; P = 0.0005). N o disease-related adult mortality was detected in the 2 populations (Redbird and Asotin) that doubled in size during the study. Populations incurring pneumonia-related adult mortality increased by less than 30% (Table 2.7, Figure 2.7). 32 Table 2.7. Bighorn population size and average annual rate of increase (r) in Hells Canyon, 1997-2003. Population N \ N R 1997-98 2002-2003 Asotin 21 42 0.13 Big Canyon' 23 30 0.01 Black Butte 78 89 0.02 Imnaha 141 152 0.02 Lostine2 86 73 0.06 Muir Creek3 26 17 -0.05 Redbird 73 148 0.16 Wenaha 62 59 0.01 Total 510 610 0.03 1 Big Canyon population started with releases in 1997-98. N 1997-98 is the number of sheep released into population in 1997-98 (16) and 1999-00 (7). 2 30 sheep transplanted out of Lostine population and released at other sites in Hells Canyon 1999-00(15) and 2001-02(15). 3 Muir Creek population started with releases in 1997-98. N 1997-98 is the number of sheep released into population in 1997-98 (13) and 1999-00 (13). z 4 at D) O -1 3.5 -Redbird -Asotin B 2.5 H -Black Butte -Imnaha -Wenaha -Big Canyon -Muir 1997-98 1998-99 1999- 2008-01 2001-02 2002-03 2000 1997-98 1998-99 1999- 2000-01 2001-02 2002-03 2000 Figure 2.7. Change in Hells Canyon bighorn sheep population sizes 1997 - 2003 in relation to pneumonia-caused adult mortality. Populations where no pneumonia-caused adult mortality was detcted (A) and those where pneumonia-caused adult mortality occurred (B). Average annual exponential rate of increase (r) was 0.13 in the Asotin and 0.16 in the Redbird populations where no adults died from pneumonia during the study. Average annual rate of increase ranged from -0.05 to 0.02 in other populations in which pneumonia-caused adult mortality was detected. 33 Population growth in 7 population-years where predation of radio-collared adults by cougars occurred (r = 0.02) was not significantly lower than in 32 population-years where no cougar predation was detected or where cougar predation was accompanied by disease (r = 0.08; t = 0.92; 37df; P= 0.36). Disease and predation-related mortality occurred in the same year within a population in only 4 of 30 combined occurrences, and there was no interaction between the effects of predation and disease-related mortality on population growth (Fijg = 0.18; P = 0.67). 0.4 n Ewe survival Figure 2.8. Annual bighorn ewe survival and exponential population growth rate (r) in Hells Canyon 1997-2003. Y = 0.6849x - 0.559, R 2 = 0.57, p< 0.0001. Population growth was related to ewe survival and recruitment (R2 = 0.57; F= 25.37; 41df; P< 0.001) (Figure 2.8), but not to ram survival (R2 = 0.08; F= 3.38; 38 df; P= 0.07). Rate of population growth was most sensitive to changes in adult ewe survival (elasticity = 0.566, SE 0.027) and secondly to recruitment (elasticity = 0.145, SE 0.009). 34 2.3 D I S C U S S I O N Adult and juvenile survival were highly variable and unpredictable among years and among Hells Canyon bighorn populations, due to periodic pneumonia-caused mortality in adults and in lambs. Adult mortality was seasonal, with most disease-related mortality occurring in the fall and early winter, and most predation occurring in the spring. Little adult mortality occurred during summer. These mortality patterns are similar to those found in other bighorn populations where both disease and predation occur (Enk et al. 2001). Preponderance of cougar predation on bighorn sheep in winter and early spring has also been observed elsewhere (Hayes et al. 2000). Seasonal increase in the incidence o f disease during fall and winter, such as influenza and pneumonia in humans, has been attributed to variation in immunocompetence caused by energy availability, stressors, or to seasonal behavior patterns that may facilitate pathogen transmission. In some animals, these predictable seasonal stressors are compensated for by an increase in immune function possibly regulated by photoperiod (Nelson and Dumas 1996, Nelson 2004), however seasonality o f immune response in bighorn sheep has not been evaluated. Cougar predation was the second most frequent source of adult mortality in Hells Canyon. However, contrary to observations in other areas (Wehausen 1996, Hayes et al. 2000, Kamler et al. 2002), predation did not have a significant effect on population growth rates and only the presence o f disease-related adult mortality caused significant population declines. Although parasite infections can predispose animals to predation (Hudson et. al.1992, Murray et al. 1997), the lack of interaction between predation and disease-related mortality suggested that vulnerability to predation was not higher in conjunction with disease 35 outbreaks. In the few instances where lung tissue was present for analysis, pneumonia was not detected in predator-killed animals. Periodic summer pneumonia epizootics resulted in high rates of lamb mortality June -August. The timing of peak lamb mortality (24 - 91 days of age) during epizootics differed from that observed in ungulate populations where pneumonia is not an important source o f mortality. In ungulates, most pre-weaning mortality typically occurs within one month o f birth (Gaillard et al. 2000). The relative vigor of neonatal lambs and subsequent infection suggests that morbidity may coincide with the waning o f passive immunity acquired through the colostrum (Miller et al. 1997). Previous reports of summer lamb epizootics have described consistently high rates of juvenile mortality subsequent to all-age pneumonia dieoffs (Spraker et al. 1984, Coggins and Matthews 1992, Ryder et al. 1992). In this study, sporadic lamb pneumonia epizootics were interspersed with years of high lamb survival both following and in the absence of pneumonia epizootics in adults. Although lamb epizootics occurred in at least one o f the 8 populations in 5 of the 6 years of the study, each year, summer lamb epizootics were confined to only a few populations and these varied from year to year. A s with disease-related mortality in adults, lamb epizootics were not synchronized through space or time. While we were only able to establish causes of summer lamb mortality, post-weaning survival of lambs was also important in determining recruitment. Growth observed in this bighorn sheep metapopulation (r = 0.04) was far below the intrinsic rate o f increase (r m , Caughley 1977) calculated for bighorn sheep at approximately 0.26 by Buechner (1960) and below the observed average rate of increase o f 0.13 for successfully introduced Rocky Mountain bighorn populations in Colorado (McCarty and 36 Mi l l e r 1998). Overall, disease played a chronic although spatially and temporally heterogeneous role in limiting the rate of population growth by affecting both adult survival and recruitment. M A N A G E M E N T I M P L I C A T I O N S Disease-related mortality was largely responsible for the extirpation of bighorn sheep throughout much o f their range in the western U.S. , presumably as a result o f pathogens associated with domestic sheep (Buechner 1960, Mi l l e r 2001). Despite intensive efforts by wildlife managers to eliminate contact between wi ld and domestic sheep, disease continues to play a major role in depressing the rate of growth of bighorn sheep populations and contributing to local extirpations. A recent survey of wildlife managers indicated that epizootics had occurred in Rocky Mountain or California (O. c. California) sheep populations in 14 (88%) o f 16 responding states and provinces between 1974 and 1999. Eleven o f 16 (67%) state and provincial managers listed disease as a significant limiting factor to the wi ld sheep population in their state or province (Thomas and Thomas 2000). In addition to precipitating large-scale die offs, recurring disease-related mortality in adults and juveniles can chronically limit bighorn populations. 2.4 L I T E R A T U R E C I T E D Akenson, H . A . 1998. Predicting summer lamb mortality in free ranging bighorn sheep. Proceedings of the Biennial Meeting of the Northern W i l d Sheep and Goat Council . 11:70-76. 37 Buechner, H . K . 1960. The bighorn sheep in the United States, its past, present, and future. Wildlife Monographs 4:174pp. Burnham, K . P. and D . R. Anderson. 2002. Model selction and multmodel inference: a practical information-theoretic approach. Springer-Verlag New York, Inc. 488pp. Cassirer, E . F. , L . E . Oldenburg, V . L . Coggins, P. Fowler, K . Rudolph, D . L . Hunter, and W . J. Foreyt. 1996. Overview and preliminary analysis o f Hells Canyon bighorn sheep die-off, 1995-96. Proceedings of the Biennial Meeting of the Northern W i l d Sheep and Goat Council 10:78-86. Caughley, G . 1977. Analysis o f vertebrate populations. John Wiley & Sons. 234 pp. Coggins, V . C. and P. E . Matthews. 1992. Lamb survival and herd status of the Lostine bighorn herd following a Pasteurella die-off. Proceedings of the Biennial Meeting o f the Northern W i l d Sheep and Goat Council 8:147-154. Enk, T. A . , H . D . Picton, and J. S. Will iams. 2001. Factors limiting a bighorn sheep population in Montana following a dieoff. Northwest Science. 75(3):280-291. Gaillard, J . - M . , M . Festa-Bianchet, N . G . Yoccoz, A . Loison, and C . Toigo. 2000. Temporal variation in fitness components and population dynamics of large herbivores. Annual Review of Ecology and Systematics. 31:367-93. Hayes, C . L . , E . S. Rubin, M . C . Jorgensen, R . A . Botta, and W . M . Boyce. 2000. Mountain lion predation of bighorn sheep in the peninsular ranges, California. Journal of Wildlife Management. 64:954-959. 38 Hobbs, N . T. and M . W. Mi l le r . 1992. Interactions between pathogens and hosts: simulation o f pasteurellosis epidemics in bighorn sheep populations. Pp. 997-1007 in D . R. McCul lough and R. H . Barrett eds. Wildlife 2001. Population. Elsevier Applied Science, Essex, England. 1163pp. Hudson, P. J., A . P. Dobson, and D . Newborn. 1992. Do parasites make prey more vulnerable to predation? Red grouse and parasites. Journal o f Animal Ecology 61: 681 - 6 9 2 . Johnson, R . L . 1980. Re-introduction of bighorn sheep in Washington. Proceedings o f the Biennial Meeting o f the Northern W i l d Sheep and Goat Council 2:106-112. Johnson, C . G . , and S. A . Simon. 1987. Plant associations of the Wallowa-Snake Province. U S D A Forest Service, Pacific Northwest Region, R6-ECOL-TP-255B-86 . 399 pp. + app. Jorgenson, J. T., M . Festa-Bianchet, J. M . Gaillard, and W . D . Wishart. 1997. Effects of age, sex, disease, and density on survival o f bighorn sheep. Ecology 78:1019-1032. Kamler, J. F. , R. M . Lee, J. C . deVos, W. B . Ballard, and H . A . Whitlaw. 2002. Survival and cougar predation o f translocated bighorn sheep in Arizona. Journal o f Wildlife Management 66:1267-1272. Kaplan, E . L . and P. Meier. 1958. Nonparametric estimation from incomplete observations. Journal of the American Statistical Association 53:457-481. Kle inbaum, D . G . 1996. Survival analysis: a self-learning text. Springer-Verlag New York, Inc. 324 pp. 39 Mil le r , M . W. , J. A . Conlon, H . J. M c N e i l , J. M . Bulgin, and A . C . S. Ward. 1997. Evaluation of a multivalent Pasteurella vaccine in bighorn sheep: Safety and serologic responses. Journal of Wildlife Diseases 33:738-748. , 2001. Pasteurellosis. Pp. 330 - 339 in_ E . S. Williams and I. K . Barker eds. Infectious diseases of wi ld mammals 3 r d ed. Iowa State University Press, Ames. 558 pp. McCarty, C . W . and M . W . Mil le r . 1998. Modeling the population dynamics of bighorn sheep. Colorado Division of Wildlife Special Report 73:35 pp. Murray, D . L . , John R. Cary, and L l o y d B . Keith. 1997. Interactive effects o f sublethal menatodes and nutritional status on snowshoe hare vulnerability to predation. Journal o f Animal Ecology 66:250-264. Nelson, R. J. and G . E . Dumas. 1996. Seasonal changes in immune function. Quarterly Review of Biology 71:511-548. . 2004. Seasonal immune function and sickness responses. Trends in immunology. 25: 1 8 7 - 192. Oregon Department of Fish and Wildlife. 1992. Oregon's bighorn sheep management plan, 1992 - 1997. Portland, OR. 30pp. Pollock, K . H . , S. R. Winterstein, C . M . Bunck, and P. D . Curtis. 1989. Survival analysis in telemetry studies: the staggered entry design. Journal o f Wildlife Management 53:7-15. Risenhoover, K . L . , J. A . Bailey, and L . A . Wakelyn. 1988. The Rocky Mountain bighorn sheep management problem. Wildlife Society Bulletin 16:346-352. 40 Rubin, E . S., W. M . Boyce, and E. P. Caswell-Chen. 2002. Modeling demographic processes in an endangered population of bighorn sheep. Journal of Wildlife Management 66:796 - 810. , E . S., W. M . Boyce, M . C. Jorgenson, S.G. Torres, C. L . Hayes, C. S. O'Brien, and D. A. Jessup. 1998. Distribution and abundance of bighorn sheep in the Peninsular Ranges, California. Wildlife Society Bulletin 26:539-551. Ryder, T. J., E . S. Williams, K. M . Mills, K. H. Bowles, and E. T. Thorne. 1992. Effect of • pneumonia on population size and lamb recruitment in Whiskey Mountain bighorn sheep. Proceedings of the Biennial Meeting of the Northern Wild Sheep and Goat Council 8:136-146. Singer, F. J., E . S. Williams, M . W. Miller, and L. C. Zeigenfuss. 2000. Population growth, fecundity, and survivorship in recovering populations of bighorn sheep. Restoration Ecology 8:75-84. Smith, D. R. 1954. The bighorn sheep in Idaho: it's status, life history, and management. Wildlife Bulletin No. 1. Idaho Dep. of Fish and Game, Boise, ID. 154 pp. Spraker, T. R., C. P. Hibler, G. G. Schoonveld, and W. S. Adney. 1984. Pathological changes and microorganisms found in bighorn sheep during a stress-related die-off. J. Wildlife Diseases 20:319-327. Valdez, R. and P. R. Krausman. 1999. Description, distribution, and abundance of mountain sheep in North America. Pages 3-22 in R. Valdez and P. R. Krausman, editors. Mountain Sheep of North America. The University of Arizona Press, Tucson, Arizona, USA. Thomas, A . E . and H . L . Thomas eds. 2000. Transactions of the 2 n North American W i l d Sheep Conference. A p r i l 6-9, 1999, Reno, N V . 470pp. Wehausen, J. D. 1996. Effects o f mountain lion predation on bighorn sheep in the Sierra Nevada and Granite Mountains o f California. Wildlife Society Bulletin 24:471-479. Wisdom, M . J. and L . S. M i l l s . 1997. Sensitivity analysis to guide population recovery: prairie chickens as an example. Journal o f Wildlife Management 61:302-312. 42 Chapter 3 E C O L O G I C A L A N D I N D I V I D U A L C H A R A C T E R I S T I C S A S S O C I A T E D W I T H B I G H O R N S H E E P M O R T A L I T Y IN H E L L S C A N Y O N , 1997 - 2003 3.1 Introduction Ungulate populations are typically regulated by a combination o f biotic factors such as nutrition, predation, and disease and influenced by abiotic variables such as climate (Krebs 2003). Disease often interacts with other variables. For example, nematode-related population crashes in Soay sheep (Ovis aries) have been described as a consequence of high population density and severe winter weather conditions (Gulland 1992). Nutritional condition, nematode infection, and predation interact to affect snowshoe hare survival (Murray et al. 1997) and nematode infection can predispose red grouse to predation (Hudson et al. 1992). However, endemic nematode infection can regulate red grouse (Lagopus lagopus scotius) populations independently o f nutrition or climate (Hudson et al. 1998). Introduced pathogens can also have significant independent limiting effects on wildlife populations as observed by population declines of wildebeest (Connochaetes taurinus) and other ungulate species after the introduction of the rinderpest virus in East Afr ica and their recovery subsequent to rinderpest eradication (Tompkins et al. 2002). Density, climate, nutrition, and introduction of diseases by domestic sheep have all been proposed as causes o f epizootics in wi ld sheep. Numerous authors have suggested that pneumonia epizootics in bighorn sheep are a consequence o f a combination of climatic, nutritional, or other stressors in conjunction with high population densities (Stelfox 1974, Dunbar 1992, Bunch et al.1999). Selenium deficiency has been suggested as a potential nutritional factor associated with disease-related mortality in domestic and wi ld lambs (Rock 43 et al. 2001, Hnilicka et al. 2002). Contact between bighorn and domestic sheep has also been implicated as a cause of bighorn pneumonia epizootics. Captive studies have demonstrated that healthy domestic sheep can carry pathogens that cause lethal pneumonia in bighorn sheep (Onderka and Wishart 1988, Foreyt 1989, Callan et al. 1991) and epizootics in free-ranging bighorn sheep have been linked to contact with domestic sheep (Goodson 1982, Martin et al. 1996). Self-limiting recurring epizootics in bighorn sheep following initial pathogen introduction have also been simulated in the absence of extrinsic factors (Hobbs and Miller 1992). In that simulation, animals that recover from an initial pathogen introduction remained mildly infective and subsequent epizootics were precipitated by recruitment of susceptible animals into the population through time. Population-level effects of disease were observed in a reintroduced bighorn sheep (O. canadensis) metapopulation in Hells Canyon, USA, affected by recurrent pneumonia-caused mortality in adults and juveniles between 1997 and 2003. Pneumonia-caused mortality in adults occurred at least once in 6 of 8 populations studied in 4 of the 6 study years and pneumonia epizootics occurred in lambs nearly every summer in at least one population (Chapter 2). Here I evaluate individual, population, and metapopulation-level factors to test hypotheses about the roles of nutrition, population density, weather, and introduction of novel pathogens in precipitating disease outbreaks and limiting population growth of bighorn sheep in Hells Canyon. M y hypotheses were that climate (a metapopulation-level factor) and nutrition (a population-level factor) would have no effects on the occurrence of disease but that distance to domestic sheep and goats would (a population or individual-level factor). 44 3.2 Methods Study area The project area encompassed 22,732 k m 2 along the Snake, Salmon, and Grande Ronde Rivers in Idaho, Oregon, and Washington (117°52 '30"W, 46°30"N to 116 °15 'N, 44°45 'W). The area included portions o f the Blue Mountain and Columbia Plateau ecoregions. Elevations ranged from 245m in river canyons to over 2740 m in the Seven Devils, Idaho (ID) and Wallowa Mountains, Oregon (OR). Climate is generally continental and dry with light precipitation (25 cm to 127 cm), low relative humidity, and wide ranges in temperature (-20 °C to >40 °C). The low elevation Snake River canyon is typically warm and dry with temperatures averaging 11 °C at Lewiston, ID. Average annual precipitation o f 49 cm occurs fairly evenly year-round except during the months of July and August. The adjacent uplands including the Blue Mountains in Washington ( W A ) and Oregon are cooler and wetter with average temperatures of 10 °C in Pomeroy, W A and average annual precipitation of 61 cm at Asotin, W A and 66 cm in Pomeroy. The upper elevations in the Wallowa and Seven Devils mountains receive annual precipitation of up to 205cm, over two-thirds o f which occurs as snow during winter (Johnson and Simon 1987). Temperature averages 7 °C at the base o f the Wallowa Mountains in Enterprise, O R and annual precipitation averages 76 cm. Over fifty percent of the study area was publicly owned and managed by federal and state agencies. Habitat improvements have included termination o f most domestic sheep grazing allotments, the most recent in 1999. Several domestic sheep and goat operations still operate on public and private land adjacent to or within bighorn population areas. 45 Data collection Between 1997 and 2003, 151 adult (>1 yr old) sheep (112 F , 39 M ) were radio-collared and monitored in 8 study populations for a total of 39 population-years (ewes 316 radio-years and rams 80 radio-years). Adult and juvenile survival and rates of population growth were determined as described in Chapter 2. Weather In order to reflect geographic differences in climate within the study area, temperature and precipitation data collected by the National Oceanic and Atmospheric Administration, National Climatic Data Center (Western Regional Climate Center 2004) at 5 weather stations in the study area were used in analyses of weather effects. Weather data were obtained from: Lewiston/Nez Perce County Airport, ID (Coop Sta ID# 105241, elev 437.7m, 46°227117°,01'W), Asotin, W A (Coop Sta ID#450294, elev 1066.8m, 46 0 12 'N /117°15 'W) , Enterprise, O R (Coop Sta ID#352678, elev 999.7m, 4 5 0 4 3 ' N / l 17 °09 'W), andPomeroy, W A (Coop Sta ID#456610, elev 579.1m, 4 6 ° 2 8 ' N / 1 1 7 ° 3 5 ' W ) . We used data from the Wallowa, O R weather station (Coop St ID#, 358997, elev 890.9m, 4 5 0 3 4 ' N / l 17 °32 'W) to substitute for 3 missing values in the Enterprise data. Total monthly precipitation and average monthly temperature were selected to represent climate. Analysis was conducted on reported data and both variables were standardized by subtracting the 30-year mean, 1974 - 2003, (except for precipitation at 46 Asotin where data were only available from July 1976 - 2003) and dividing by the standard deviation to obtain z-scores (Zar 1999). Temperature and precipitation data collected at the Lewiston weather station were used to represent weather conditions along the Snake River corridor in the Redbird, Black Butte, Imnaha, B i g Canyon, and M u i r Creek populations (Snake River climate zone). Data from the Enterprise weather station was used for the Lostine population (Wallowa climate zone). Asotin precipitation and Lewiston temperatures were used in analyses o f the Asotin population and Pomeroy data were used to characterize weather in the Wenaha population (Blue Mountains climate zone). Nutrition Fecal pellets were collected monthly from ewe groups for 4 years (June 1999 - M a y 2003). Although the relationship between fecal nitrogen and dietary nitrogen can vary (Hobbs 1987) fecal nitrogen is correlated with diet quality and with population performance in bighorn sheep (Irwin et al. 1993, Blanchard et al. 2003). A n average o f 21 pellet groups were collected from adult animals in each population per month. Lamb pellets were collected primarily during summer. Samples were frozen after collection and oven dried prior to analysis. Samples were composited to an average o f 5 adult samples per population per month (n = 1,722) and analyzed for nitrogen using the Kjeldahl method (Drew 1970). Analyses were conducted on the log e o f fecal nitrogen (FN) to linearize the relationship with apparent digestibility (Wehausen 1995). I interpolated missing monthly F N values in 7 o f the populations by averaging the difference between the preceeding and succeeding months 47 within the same population in years where data were available (Redbird n = 1, Asotin n = 2, Wenaha n = 2, Imnaha n = 1, Lostine n = 5, B i g Canyon n = 4, M u i r n = 3). This correction was applied to the data collected in the year of the missing month. Year-round data were available for 3 years (June 2000 - M a y 2003) in the B i g Canyon population and for 2 years (June 2001 - M a y 2003) in the M u i r population. Four years of data (June 1999 - M a y 2003) were available for all other populations. Fecal nitrogen values were standardized by subtracting the population mean and dividing by the standard deviation to obtain z-scores (Zar 1999). Blood samples for selenium analysis were collected at capture and immediately injected into E D T A blood tubes. Whole blood selenium was analyzed by inductively coupled plasma spectrophotometer atomic emission using hydride generation (Tracy and Moller 1990) at the Ho lm Research Center, University of Idaho, Moscow, and values were reported on a wet weight basis. Liver samples were collected from dead sheep at necropsy. Liver and blood data were combined for analysis in those populations where they did not differ significantly (p > 0.05). Analysis Univariate t-tests and correlation analyses were conducted on population and environmental variables to evaluate relationships prior to inclusion in multivariate models. Correlations between the total number of sheep in the current year (time t) and previous year (time t -1) and survival and population growth were evaluated to assess density dependence. Bighorn sheep are sexually segregated during much of the year (Geist 1971) and males and 48 females may differ in their role in disease dynamics (Skorping and Jensen 2004), therefore the relationship of male and female numbers and male:female ratios at time t and time t - 1 on demographics and occurrence of pneumonia outbreaks were also evaluated. General linear models were used to compare nutrition, weather, and population variables to adult and lamb survival rates by population. General linear model analysis of ranked lambing dates and Tukey's studentized range test were used to test for difference in lambing dates among populations. The relationship between fecal nitrogen and climate was examined with stepwise multiple regression between monthly F N and climate with up to two months lag in climate variables. Logistic regression, with the occurrence o f population pneumonia outbreaks as the response variable, was used to test for the significance of ecological variables as predictors o f adult and lamb epizootics within each population and across populations. Seventy-four percent of adult pneumonia-related mortality occurred October - January and no adult pneumonia mortalities occurred June - August. Therefore, actual monthly precipitation and average temperatures and z-scores and fecal nitrogen values during the months of Jun - Dec were evaluated as potential predictors of pneumonia-related mortalities. Bighorn lamb recruitment has been positively correlated with autumn, spring, and summer temperature and precipitation (Douglas and Leslie, 1986, Portier et al. 1998, Enk et al. 2001). Summer nutrition during the year prior to birth may also affect juvenile survival (Cook et al. 2004). Therefore, monthy temperature and precipitation values and z-scores and F N values from the previous July through July of the birth year were included as potential predictors o f summer pneumonia epizootics. I also evaluated the correlation of these 49 variables with summer lamb survival and recruitment. The above analyses were conducted in S A S (SAS Institute 2001). Proportional hazard survival models for discrete time intervals were used to evaluate effects, and the occurrence o f predation and disease-caused mortality in a population on adult survival. The analysis over discrete time intervals allows inclusion of time-dependent covariates and variable hazard ratios among years. I used known-fate analysis in program S U R P H (Skalski 2003) with a log-log or hazard link function. In order to evaluate effects o f covariates specifically on pneumonia-related mortality I censored all other sources of mortality (All ison 2001). I conducted a similar analysis on the second most important cause o f mortality, cougar predation. Model selection was based on Akaike 's Information Critereon adjusted for small samples (AIC C ) . A n A I C C , difference from the best (smallest) value ( A A I C C ) and Akaike weight (w) were calculated for each model (Burnham and Anderson 2002). Individual characteristics included in survival analysis were sex, age at death or censoring, closest location o f each individual to domestic sheep or goats during the study, transplanted sheep or born in study area, and selenium level in blood or liver. Sheep were aged by tooth replacement (< 4 years) (Wishart 1978) and horn annuli (>4 years) at capture (Geist 1966). Cementum analysis (Turner 1977) o f incisors (Matson's Laboratory, Mi l l town, M T ) was used to age dead sheep. Where selenium measurements were not available for individuals I used population means. Monthly total precipitation and monthly average temperatures as well as z-scores and fecal nitrogen values during the months o f Jun - Dec were included as group (population) covariates. 50 3.3 Results Survival and population growth were highly variable and spatially and temporally asynchronous within the Hells Canyon metapopulation (Figure 3.1). None of the demographic parameters (population growth, ewe or ram survival, or lamb survival and recruitment) were correlated with total population size or composition within or across populations (Figure 3.1, Table 3.1, p >0.05). Size of populations experiencing adult epizootics (3c = 53) was actually smaller but not significantly different than populations that did not (x =76). Pneumonia outbreaks in lambs occurred in populations that were smaller (x = 60) and had fewer ewes (x = 36) and rams (x = 17) than populations where no epizootics occurred (N x - 80, ewes x = 41, rams x = 22, p < 0.0001). 160 1997-88 1998-89 1999-00 2000-01 2001-02 2002-03 1997-98 1998-99 1999-00 20004)1 2001-02 2002-03 Figure 3.1. Population size, adult survival, and juvenile survival in 8 bighorn sheep populations in Hells Canyon, 1997 - 2003. Survival and population growth was highly variable and not synchronized among populations. 51 Table 3.1. Bighorn sheep demographics in 8 Hells Canyon populations, 1997 - 2003. Bo ld denotes pneumonia in adults and/or lambs. Population Year N N Ram: ewe R S 1 >J ewe s 1 *-> ram recruitment S • 5 0 weaning Asotin 1998 -99 24 11 0.55 0.13 0.88 0.64 1.00 1999-2000 29 15 0.40 0.1.9 0.86 0.53 0.83 2000 - 01 31 23 0.35 0.07 1.00 0.13 0.80 2001 -02 36 18 0.50 0.1.5 1.00 0.50 0.60 2002 - 03 42 23 0.35 0.15 1.00 0.48 0.75 Big Canyon3 1998-99 27 15 0.33 0.33 1.00 1.00 0.47 0.92 1999-2000 37 17 0.47 0.32 0.93 1.00 0.71 0.86 2000 - 01 25 16 0.5 -0.39 0.60 0.80 0.06 0.50 2001 - 02 32 19 0.47 0.25 1.00 0.80 0.21 0.78 2002 - 03 30 20 0.60 0 0.91 0.50 0 0 Black Butte 1997-98 78 45 0.29 -0.07 0.92 0.46 0.33 0.63 1998-99 72 44 0.32 -0.01 1.00 0.99 0.32 0.65 1999- 82 46 0.39 0.1.3 0.58 0.73 0.39 0.51 2000 2000 - 01 87 44 0.68 0.06 0.71 1.00 0.30 0.80 2001 -02 75 35 0.91 -0.15 0.80 0.80 0.23 0.25 2002 - 03 89 43 0.53 0.17 1.00 0.30 0.53 1.00 Imnaha 1997-98 141 69 0.42 0.17 0.62 1998 - 99 142 71 0.42 0.01 0.58 1999- 125 61 0.77 -0.13 0.28 2000 2000-01 123 69 0.36 . -0.02 0.85 0.71 0.42 0.80 2001 - 02 144 78 0.50 0.16 ' 0.95 1.00 0.35 0.56 2002 - 03 152 77 0.38 0.05 1.00 1.00 0.60 0.56 Lostine4 1997-98 86 44 0.48 0.12 0.24 1998-99 77 50 0.58 -0.11 0.35 1999- 82 47 0.53 0.06 0.27 2000 2000-01 74 38 0.45 0.10 1.00 0.80 0.50 1.00 2001 -02 89 42 0.57 0.18- 1.00 1.00 0.55 0.85 2002-03 73 39 0.51 -0.01 0.94 0.80 0.36 0.72 Muir Creek 1998-99 25 17 0.18 0.04 1.00 1.00 0.29 1999-2000 29 14 0.50 0.15 0.93 1.00 0.57 0.91 2000-01 29 14 0.36 0 0.71 0.50 0.79 0.78 2001 -02 37 17 0.59 0.24 1.00 1.00 0.35 0.14 2002 - 03 17 16 0.38 -0.43 0.56 0.13 0 Redbird 1997-98 73 38 0.61 0.28 1.00 0.32 1.00 1998-99 80 44 0.64 0.09 1.00 0.18 0.25 1999- 86 46 0.63 0.07 1.00 0.24 0.55 2000 2000-01 121 55 0.65 0.34 1.00 1.00 0.55 0.72 2001 -02 130 61 0.82 0.07 0.92 0.80 0.16 0.83 2002 - 03 148 66 0.64 0.13 0.91 0.75 0.59 0.66 Wenaha 1997-98 62 50 0.16 0.24 0.83 0.08 0.22 1998-99 64 41 0.37 0.03 1.00 0.20 0.14 1999- 60 36 0.47 -0.06 1.00 0.19 0.27 2000 2000-01 50 28 0.50 -0.1.8 0.73 0.67 0.29 1.00 2001 -02 60 35 0.54 0.18 1.00 1.00 0.17 0.25 2002-03 59 29 0.62 -0.02 1.00 1.00 0.41 0.25 1. Survival of radio-collared ewes and rams. 2. Survival of lambs born to radio-collared ewes from birth - October. 52 Climate Seasonal temperatures at all stations were similar, with highs in July and August and lows in December and January (Figure 3.2). In the Snake River Canyon climate zone (Lewiston station), the driest portion o f the study area, precipitation was lowest during summer, increased slightly during fall and winter, and peaked in May . Snow occurred rarely and briefly in the canyon. Annual precipitation was higher in the Blue Mountains (Asotin and Pomeroy) and Wallowa Mountains (Enterprise and Wallowa), and much of this occurred as snow between November and February, with a secondary maximum in M a y and June (Figure 3.2). Precipitation was higher than average during 1997 - 99 and 2002 - 03 and lower than average during 2000 - 2002. Mean temperature was higher than average 1997 -2000 and 2003 - 03, lower than average 2000 - 01, and average in 2002 - 03 (Figure 3.3). Actual climate values and deviations from average were correlated among weather stations. Temperature was more highly correlated among stations than precipitation (precipitation z-score n = 72, 0.64 < r < 0.74, p < 0.0001; temperature z-score n = 72, 0.69 < r < 0.89, p < 0.0001). 53 25 _5 l 1 1 1 1 1 1 1 Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May I— 0.2 0.1 0 I 1 1 1 1 ' ' ' ' 1 ' Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Figure 3.2. Normal monthly average temperature and total precipitation at 5 weather stations in the Hells Canyon study area, 1974 - 2003. 54 1.5 c o 1997 98 1998 99 1999 00 2000_01 2001_02 2002_03 Figure 3.3. Annual variation in precipitation, temperature, and adult survival in Hells Canyon, 1997-2003 . 55 Fecal nitrogen Percent fecal nitrogen (FN) in lamb pellets averaged 17% higher than in adult pellets collected simultaneously in the same population (t - 2.38, 7 df, p = 0.03). Lamb pellets were not available for all sampling periods, so only adult pellets were used for analysis. FN was lowest in all populations during the winter months of November, December, and January. FN values in the Snake River Canyon populations (Imnaha, Big Canyon, Muir, Redbird, and Black Butte) peaked during "green-up" in March, April, and May and were moderate to low the remainder of the year. FN in the Blue Mountains suppopulations (Asotin and Wenaha) also peaked during spring "green-up", but remained moderately high throughout the summer. FN in the Wallowas population (Lostine) was the lowest during winter and remained low through March. These bighorn sheep migrated to alpine summer range and FN peaked on the summer range during June and July (Figure 3.4). Spring temperatures and fall precipitation explained significant proportion of the variation in fecal nitrogen values across the project area. May - July FN values were negatively related to average May temperatures (May FN n = 31, R 2 = 0.40, p = 0.002; June FN n = 29, R 2 = 0.50, p = 0.002; July FN n = 28, R 2 = 0.23, p = 0.01). October and November FN were positively related to actual precipitation and deviations from normal in September (October FN n = 30, R 2 = 0.17, p = 0.02; November FN n = 30, R 2 = 0.24, p = 0.006). 56 Figure 3.4. Average seasonal pattern of F N values in three climate zones within the Hells Canyon project area, 1999 - 2003. Average monthly F N varied among populations ( F 7 > 2 i = 5.53, p = 0.0010) and was lowest in the B i g Canyon, Black Butte, Imnaha, and M u i r Snake River Canyon populations and highest in the Blue Mountain and Redbird populations (Figure 3.5). Selenium Selenium values differed among populations (Table 3.2) but were comparable to those observed regionally (Hein et al. 1994). In general, selenium levels were adequate to slightly deficient for domestic sheep. Differences in selenium values among populations were not significantly related to adult survival, summer lamb survival, recruitment, population growth or occurrence of epizootics. Selenium values were lowest in the population (Asotin) where no pneumonia was detected in adults or lambs. 57 Ui Asotin Wenaha Redbird Figure 3.5. Average monthly fecal nitrogen values in 8 bighorn populations in Hells Canyon, 1999 - 2003. Means connected by horizontal lines are not significantly different (p > 0.05). a M u i r Creek F N data collected 2001 - 2003. b B i g Canyon F N data collected 2000 - 2003. Commercially available salt and mineral blocks containing selenium were available to all 4 populations in Oregon (Imnaha, Lostine, Muir , and Wenaha) throughout the study, however availablility of supplemental selenium apparently only increased blood values in the Lostine population. Selenium concentration in liver tissue (n = 2) in the Lostine population were similar to concentrations in other blood and liver in other populations (Table 3.2). Table 3.2. Whole blood and liver selenium values (ppm) in 8 bighorn populations in Hells Canyon, 1997 - 2003. Selenium concentrations in blood and liver did not differ within populations (p > 0.502) except in the Lostine population (p = 0.002). Population means with different superscript letters are significantly different (p < 0.05). Asotin Imnaha Wenaha Black Butte Muir B i g Canyon Redbird Lostine Blood 3c (n) 0.09(13) 0.11 (27) 0.11 (25) 0.19(16) N A N A 0.21 (17) 0.40 c (29) s.d. 0.06 0.07 0.08 0.09 0.07 0.15 range 0.03 - 0.22 0.02 - 0.26 0.02 - 0.33 0.06-0.38 0.11-0.33 0.12-0.65 Liver 0 .14 A B (2) x (n) N A 0.11 (7) 0.09 (4) 0.16(4) 0.17(9) 0.19(6) 0.31 (6) s.d. 0.05 0.03 0.08 0.03 0.03 0.14 0.14 range 0.03-0.17 0.066-0.12 0.04 - 0.22 0.14-0.24 0.14-0.21 0.13-0.47 0.05 - 0.24 Mean 0.092A 0 . 1 1 A B 0 . 1 1 A B 0.18 A B 0 .17 A B 0.19 A B 0.24B L /1 59 Introduction of pathogens To assess the potential for introduction of pathogens, the minimum straight-line distance observed to domestic sheep and goats was calculated for each radio-collared bighorn. Distance to domestic sheep and goats was also calculated for each mortality location. Minimum observed distance between bighorn sheep that died during the study and domestic sheep and goats averaged 6.34 km and was not significantly different from that of bighorn sheep that survived (t = -0.30, 110 df, p = 6.768). Average minimum distance to domestic sheep and goats was slightly greater for bighorn sheep that died from disease-related causes than for bighorn sheep that died due to other known causes (t = 2.5, 47 df, p = 0.037; Table 3.3). Table 3.3. Minimum observed distance between radio-collared bighorn sheep and domestic sheep and goats in 8 Hells Canyon populations. , 1997 -2003. N Mean (km) SD Min Max Median Survivors 90 6.34 5.01 0.50 17.00 6.85 A l l mortalities 61 6.59 6.28 0.50 19.60 2.90 Disease-related 21 9.37 6.22 1.50 19.50 7.70 Other known-cause 28 5.88 5.81 0.50 19.00 2.55 Adult survival Several covariates selected for analysis of adult survival were correlated. Average size of transplanted populations (28) was significantly smaller than resident populations (90) (t = 7.80, 46 df,;? <0.0001) as was number of ewes and rams. Marked males were younger than females entering the study (males 3 years-old, females 5 years-old) due to the effect of transplanting young males (average age at entry into the study was 2 years-old for 60 transplanted males and 4 years-old for resident males) (F i , 147 = 4.27, p = 0.0063). Males were also younger than females at death (males 6 years-old, females 8 years-old; F i ; 147 = 130, p < 0.008) irrespective of place of birth. Age of adults dying from pneumonia (6 years) did not differ from those dying from other causes.(F5 > 5 5 = 1.44, p = 0.225). The best model of overall adult survival and o f adult disease-related mortality included sex, and whether the sheep were in the B i g Canyon and M u i r Creek populations in 2000-2001 and 2002-2003 (Table 3.4 and Table 3.5). The hazard-ratio for sex indicated that overall, males were nearly twice as likely to die from any cause than females and over 3 times more likely to die from disease-related mortality. Sheep released in the B i g Canyon and M u i r Creek populations were over 4 times more likely to die than resident sheep or transplanted sheep released at Asotin Creek during 2000-2001 and 2002-2003 and nearly 13 times more likely to die from disease-related mortality (Table 3.6 and Table 3.7). The high rate o f mortality o f adults released in these populations was reflected in population declines during 2001-2002 and 2002-2003 (Figure 3.1). When sex and the transplant effect were included in the models, none of the other population, climate or nutrition covariates were predictive o f survival nor was distance to domestic sheep and goats. Table 3.4. Candidate models o f adult survival of bighorn sheep in 8 populations in Hells Canyon, 1997-2003 . Model name N o . L o g A I C C A A I C c W parameters likelihood S E X , S O U R C E 2000 2002 1 1 -195.93 393.89 22.75 >0.99 S O U R C E 2000 2002 2 -193.56 391.20 20.06 <0.01 S O U R C E (0 = Resident, 1 = Transplant) 2 -192.36 388.79 17.66 O . 0 1 S E X (0 = Male, 1 - Female) 2 -185.14 384.91 13.77 <0.01 N o covariates ( N U L L ) 3 -182.48 371.13 0.00 <0.01 1 SOURCE 2000 2002 = Transplant effect in Muir and Big Canyon release sites in 2000-2001 and 2002-2003. 61 Table 3.5. Candidate models of disease-related mortality of adult bighorn sheep in 8 populations in Hells Canyon, 1997 - 2003. Model name ' No. Ln r AIC C AAICC w parameters Likelihood SEX , SOURCE 2000 2002 1 -88.25 178.53 29.86 >0.01 SOURCE 2000 2002 2 -84.8 173.68 25.02 >0.01 SOURCE 0 = Resident, 1 = Transplant 2 -81 165.98 17.32 >0.01 SEX 0 = Male, 1 = Female 2 -73.17 150.43 1.77 0.29 No covariates (NULL) 3 -69.38 148.66 0.00 0.71 Table 3.6. Parameter estimates derived from the best model of adult survival in Hells Canyon, 1997-2003. Variable 3 SE Hazard ratio (e13)1 Source (0 = resident, 1= transplant) 1.46 0.28 4.30 Sex (0 = m, 1 = f) -0.65 0.27 0.52 1 The hazard ratio is defined as the hazard for one individual divided by the hazard for another individual. Table 3.7. Parameter estimates derived from the best model of disease-related mortality in Hells Canyon, 1997-2003. Variable 3 SE Risk ratio (e15) Source (0 = resident, 1= transplant) 2.61 0.44 13.59 Sex (0 = m, 1 = f) -1.26 0.44 0.28 Summer lamb survival Lambing occurred from April - July, median lambing date for the metapopulation was 15 May, and 87% of lambs were born before 1 June. Median lambing dates differed among populations and ranged from 8 May to 21 May (Figure 3.6). Median lambing date differed among years in the Asotin, Big Canyon, Muir, Imnaha, and Wenaha, populations 62 but these differences were not correlated. Summer lamb survival decreased by 3.7% (hazard ratio) for each day born after the population median for that year (A*= 0.846, 1 df,p = 0.0046). Median population lambing dates did not differ significantly in years with pneumonia epizootics {X2< 1.148, 1 df,p > 0.282) except in the M u i r Creek population, where median lambing date was 12 days earlier during the year of the summer lamb epizootic (A*= 4.429, \df,p = 0.035). LOSTINE ^ • • • • • • • • • • • ^ • • • • • • • • • • • i REDBIRD ••••••••MaHBBBBBBB^BMBHHHHHBBBl ASOTIN l a a a a a a H B B a a a a a a a f l H B B B IMNAHA • • • • • • • • • • • • • • • • r a WENAHA • • • • • • • • • • • • • • • • • I BLACK.BUTTE • • • • • • • • • • • • • • • • • B MUIR • • • • • • • j j a j B H J B J I BIG CANYON • • • • • • 5/5 5/7 5/9 5/11 5/13 5/15 5/17 5/19 5/21 5/23 Median lambing date Figure 3.6. Median lambing dates in 8 bighorn populations 1997 - 2002. Dates in populations connected by lines were not significantly different p > 0.05 Summer survival did not differ between lambs born to resident and transplanted ewes (X2= 0.846, 1 df, p = 0.3576). Ewe age class did not have a significant effect on lamb survival, probably because most adults were prime age. Survival senescence in bighorn sheep, especially males, has been observed to commence at 8 years (Jorgenson et al. 1997) and reproductive senescence in ewes has been observed starting at 13 years of age (Berube et al. 1999). Only 4 individuals were over 12 and most sheep were between the ages o f 3 and 10. Two-thirds (66%) of lambs (n = 273) born to ewes age' 2 - 1 1 survived through the 63 summer and only 42% of lambs (n = 7) born to ewes age 12-17 survived (J^= 1.605, 1 df, p = 0.205). In the absence of disease-caused lamb mortality, a ewe's success at raising a lamb to weaning was positively related to success the previous year (^=4.721, ldf,/? = 0.03), however, during epizootic years summer lamb survival was independent of ewe history (J^= 0.007, 1 df,p = 0.935). Climate and FN were not predictors of summer lamb epizootics. Summer lamb survival and recruitment were also not correlated to weather variables except for a correlation between summer lamb survival and August precipitation the previous year (r = 0.321, p = 0.046, n = 39). Although there was no evidence that the occurrence of pneumonia-related mortality was related to climate, there were some correlations between lamb survival, precipitation and temperature. During population-years with epizootics (n = 9), summer lamb survival was correlated to September precipitation the previous year (r — 0.795, p = 0.015) and negatively correlated to July temperature (r = -0.646, p = 0.060). Recruitment was negatively correlated to October temperatures (r = -0.704, p =0.0025). During years without epizootics (n = 29), summer lamb survival was also negatively correlated to October temperatures the previous year (r = - 0.409, p = 0.027) and May precipitation (r = 0.372, p = 0.047). 3.4 Discussion The patterns of mortality observed in this study suggest that factors associated with disease in this metapopulation operated at the individual and population level. Bighorn sheep in some populations were apparently periodically exposed to virulent pathogens, and 64 mortality occurred in susceptible individuals, particularly transplanted sheep and lambs. Pneumonia outbreaks occurred in lambs born both to resident and transplanted sheep, suggesting some acquired immunity in adults and transmission o f pneumonia-causing pathogens to lambs. This could result in intrinsic cycling o f disease (Hobbs and Mi l l e r 1992) or could be caused by movements of bighorn sheep among populations. However, despite intensive management to maintain separation between domestic sheep and goats and wi ld sheep, average minimum distance o f marked individuals to domestic sheep and goats in this population was less than 6 km. Buffers between domestic and bighorn sheep that appear to be effective at preventing disease outbreaks have been identified as 20 km (Singer et al. 2000), 23 km (Zeigenfuss et al. 2000) and 40 km (Monello et al. 2001). Therefore, presumably, all populations had potential for contact with domestic sheep and goats and periodic introduction o f novel pathogens may have been a primary external factor precipitating pneumonia outbreaks. N o relationship was observed between the occurrence of pneumonia-related mortality and climate, or fecal nitrogen values, which were to some extent correlated to climate. Bighorn populations occurring within the same climate zones experienced similar weather conditions and fecal nitrogen regimes, and although climate regimes differed geographically, climate was highly correlated among zones. When environmental variation has a strong effect on survival, this synchronizes dynamics among populations (Hanski and Woiwood 1993, Grenfell et al. 1998) contrary to the pattern observed in this study. Asynchronous dynamics o f other bighorn sheep populations has also been attributed to the relative importance of local factors over climate (Rubin et al . 1998, Bleich et al. 1990). 65 In the absence of pneumonia outbreaks, heterogeneity in individual fitness was apparent in the positive correlation o f reproductive success in adult ewes through time. This positive association o f offspring survival with survival the previous year is contrary to the pattern observed in desert bighorn sheep (Rubin et al. 2000) and could be a consequence of a relative abundance of forage in Hells Canyon (Festa-Bianchet and Jorgenson 1998). Immunbcompetence may be positively correlated with genetic heterogeneity (Coltman et al. 1999), and transplanted populations may exhibit genetic drift and low genetic heterozygosity due to small founder size (Fitzsimmons et al. 1997). We did not evaluate genetic characteristics in this study, but it may be significant that transplanted sheep had reduced resistance to disease relative to resident sheep (descendants o f previous transplants). In particular, the sheep transplanted from Cadomin, A B in this study came from a large native and presumably outbred population, yet exhibited high vulnerability to pneumonia. The link between host density and rate o f spread of infection is a common assumption for directly transmitted pathogens (Anderson and M a y 1978, Swinton et al. 2002). However, although pneumonia outbreaks have previously been attributed to high sheep densities, and coincidentally high population size, no evidence for density-dependence in bighorn sheep pneumonia outbreaks has been demonstrated (Jorgenson et al.1997, Aune et al. 1998). Monello et al. (2001) suggested that density played a role in epizootics, but no density-related differences, such as population size or growth rates, were found between bighorn populations that suffered pneumonia epizootics and those that did not. In this study, although I did not directly measure density, I also found no relationship between population size or growth rates and pneumonia epizootics. Previous epizootics in this population have affected populations as a large as 220 (Cassirer et al. 1996), but in this study pneumonia outbreaks 66 occurred in populations as small as 29 bighorn sheep and with growth rates (r) as high as 0.32 the year prior to the epizootic. Conversely, no pneumonia-caused mortality in adults was observed in a population of 150. The occurrence of epizootics in small populations and the lack o f a relationship between population size and pneumonia-caused mortality suggests a frequency-dependent model of pathogen transmission (Swinton et al. 2002). The gregarious behavior of sheep maintaining relatively high rates of contact, even at low numbers, could allow rapid transmission of bacteria regardless of population size (Altizer et al. 2003). Similar transmission dynamics have been used to describe chronic wasting disease in mule deer (Odocoileus hemonius) (Gross and Mi l l e r 2001) and phocine distemper virus in harbour seals (Phoca vitulind) (Swinton et al. 2002). I f transmission rates are independent of population size, no threshold population size is required for establishment and spread of infection and a pathogen could exterminate the host. This could provide one mechanism for the high rate o f extinction observed in small bighorn populations (Berger 1990, Krausman et al. 1993, Goodson, 1994, Wehausen 1999). 3.5 Literature Cited Altizer, S., C . L . Nunn, P. H . Thrall, A . A . Cunningham, A . P. Dobson, V . Ezenwa, K . E . Jones, A . B . Pederson, M . Poss, and J. R. C . Pulliam. 2003. Social organization and parasite risk in mammals: integrating theory and empirical studies. Annual Review of Ecology and Evolutionary Systematics 34:517-547. Al l i son , P. D . 2001. Survival analysis using the S A S system: a practical guide. S A S Institute, Inc. Cary, N C . 67 Anderson, R. M . and R. M . May . 1978. Regulation and stability o f host-parasite population interactions. I. Regulatory processes. Journal of Animal Ecology 47:219 - 247. Aune, K . , N . Anderson, D . Worley, L . Stackhouse, J. Henderson, and J. Daniel. 1998. A comparison of population and health histories among seven Montana bighorn sheep populations. Proceedings Northern W i l d Sheep and Goat Council 11:46-69. Berger, J. 1990. Persistence of different-sized populations: an empirical assessment o f rapid extinctions in bighorn sheep. Conservation Biology 4:91 - 98. Berube, C . H . , M . Festa-Bianchet, and J. T. Jorgenson. 1999. Individual differences, longevity, and reproductive senescence in bighorn ewes. Ecology 80:2555-2565. Blanchard, P., M . Festa-Bianchet, J - M Gaillard, and J. T. Jorgenson. 2003. A test o f long-term fecal nitrogen monitoring to evaluate nutritional status in bighorn sheep. Journal o f Wildlife Management 67:477-484. Bleich, V . C , J. D . Wehausen, R. R. Ramey II, and J. L . Rechel. 1990. Metapopulation theory and mountain sheep: implications for conservation. Pp. 353-373 in D . R. McCul lough ed. Metapopulations and Wildlife Conservation. Island Press. 429 pp. Bunch, T. D . , W. M . Boyce, C . P. Hibler, W . R. Lance, T. R. Spraker, and E . S. Will iams. 1999. Diseases of North American W i l d Sheep. Pp. 209 - 237 in R. Valdez and P. R. Krausman eds. Mountain Sheep of North America, University o f Arizona Press, Tuscon. 353 pp. Callan, R. J., T. D . Bunch, and G . W . Workman. 1991. Development of pneumonia in desert bighorn sheep after exposure to to a flock o f exotic wi ld and domestic sheep. Journal of the American Veterinary Medical Association 198:1052 - 1056. 68 Cassirer, E. F., L. E. Oldenburg, V. L. Coggins, P. Fowler, K. Rudolph, D. L. Hunter, and W. J. Foreyt. 1996. Overview and preliminary analysis of a bighorn sheep dieoff, Hells Canyon 1995-96. Proceedings Northern Wild Sheep and Goat Council 10:78-86. , W. M . Lammers, and A. R. E. Sinclair. 2002. Population density and mortality of bighorn sheep in Hells Canyon. Proceedings Northern Wild Sheep and Goat Council 13:27-38. Coltman, D. W., J. G. Pilkington, J. A. Smith, and J. M . Pemberton. 1999. Parasite-mediated selection against inbred Soay sheep in a free-living island population. Evolution 53:1259-1267. Cook, J. G, B. K. Johnson, R. C. Cook, R. A. Riggs, T. Delcurto, L. D. Bryant, and L. L. Irwin. 2004. Effects of summer-autumn nutrition and parturition date on reproduction and survival of elk. Wildlife Monographs 155. Douglas, C. L. and D. M . Leslie. Influence of weather and density on lamb survival in desert bighorn sheep. Journal of Wildlife Management 50:153-156. Drew, H. D. 1970. Determination of total nitrogen. Pages 1-27 in C. A. Streuli and P. R. Averell, editors. The analytical chemistry of nitrogen and its compounds. Wiley-Interscience, New York, New York, USA. Dunbar, M . R. 1992. Theoretical concepts of disease versus nutrition as primary factors in population regulation of wild sheep. Northern Wild Sheep and Goat Council 8:174 -192. Enk, T. A., H. D. Picton, and J. S. Williams. 2001. Factors limiting a bighorn sheep population in Montana following a dieoff. Northwest Science. 75(3):280-291. 69 Festa-Bianchet, M . and J. Jorgenson. 1988. Selfish mothers: reproductive expenditure and resource availability in bighorn ewes. Behavioral ecology 9:144 - 150. Fitzsimmons, N . N . , S. W . Buskirk, and M . H . Smith. 1997. Genetic changes in reintroduced Rocky Mountain bighorn sheep populations. Journal of Wildlife Management 61:863 - 872. Foreyt, W . J. 1989. Fatal Pasteurella haemolytica pneumonia in bighorn sheep after direct contact with clinically normal domestic sheep. American Journal o f Veterinary Research 50:341 - 3 4 4 . Gaillard, J . - M . , Festa-Bianchet, M . , Yoccoz, N . G . 1998. Population dynamics of large herbivores: variable recruitment with constant adult survival. Trends in ecology and evolution. 13:58-63. Geist, V . 1966. Validi ty of horn segment counts in aging bighorn sheep. Journal of Wildlife Management 30:634-635. . 1971. Mountain sheep: a study in behavior and evolution. University of Chicago Press, Chicago, 111. 383pp. Goodson, N . J. 1982. Effects o f domestic sheep grazing on bighorn ranges: an overview. Biennial Symposium Northern W i l d Sheep and Goat Council 3:287 - 313. . 1994. Persistence and population size in mountain sheep: why different interpretations? Conservation Biology 3:617-618. Grenfell, B . T., K . Wilson, B . F . Finkenstadt, T. N . Coulson, S. Murray, S. D . Albon , J. M . Pemberton, T. H . Clutton-Brock and M . J. Crawley. 1998. Noise and determinism in synchronized sheep dynamics. Nature 394:674-677. 70 Gross, J. E . and M . W . Mil le r . 2001. Chronic wasting disease in mule deer: disease dynamics and control. Journal o f Wildlife Management 65:205-215. Gulland, F . M . D . 1992. The role o f nematode parasites in Soay sheep (Ovis aries L) mortality during a population crash. Parasitology 105:493 - 503. Hanski, Ilkka and I. P. Woiwood. 1993. Spatial synchrony in the dynamics of moth and aphid populations. The Journal of Animal Ecology 62:656-668. Hein, R. G . , P. A . Talcott, J. L . Smith, and W . L . Myers. 1994. Blood selenium values of selected wildlife populations in Washington. Northwest Science 68:185-188. Hnlicka, P. A . , J. Mionczynski , B . J. Mincher, J. States, M . Hinschberger, S. Oberlie, C . Thompson, B . Yates, and D . D . Siemer. 2002. Bighorn sheep lamb survival, trace minerals, rainfall, and air pollution: are there any connections? Northern W i l d Sheep and Goat Council 13:69-94. Hobbs, N . T. and M . W . Mi l le r . 1992. Interactions between pathogens and hosts: simulation o f pasteurellosis epizootics in bighorn sheep populations. Pp. 997 - 1007 in D.R. McCul lough and R. H . Barret eds. Wildlife 2001: Populations. Elsevier Applied Science, Essex, England. 1163 pp. . 1987. Fecal indices to dietary quality: a critique. Journal o f Wildlife Management 51:317-320. Hudson, P. J., A . P. Dobson, and D . Newborn. 1992. Do parasites make prey more vulnerable to predation? Red grouse and parasites. Journal of Animal Ecology 61: 681 - 6 9 2 . , , and_ . 1998. Prevention of population cycles by parasite removal. Science 282 :2256-2258 . Irwin, L . L . , J. G . Cook, D . E . McWhirter, S. G . Smith, and E . B . Arnett. 1993. Assessing winter dietary quality in bighorn sheep via fecal nitrogen. Journal o f Wildlife Management 57:413-421. Jorgenson, J. T., M . Festa-Bianchet, J . - M . Gaillard, and W. D . Wishart. 1997. Effects o f age, sex, disease, and density on survival o f bighorn sheep. Ecology 78:1019-1032. Krausman, P. R., R. C . Etchberger, and R . M . Lee. 1993. Persistence of mountain sheep. Conservation Biology 7:219. Krebs, C . J. 2003. Two complementary paradigms for analyzing population dynamics. In Sibly, R. M . , J. Hone, and T. H . Clutton-Brock. Wildlife population growth rates. Cambridge University Press, New York. Martin, K . D . , T. Schommer, and V . L . Coggins. 1996. Literature review regarding the compatibility between bighorn and domestic sheep. Proceedings Northern W i l d Sheep and Goat Council 10:72-77. Monello, R. J., D . L . Murray, and E . F . Cassirer. 2001. Ecological correlates of pneumonia epizootics in bighorn sheep populations. Canadian Journal of Zoology 79:1423-1432. Onderka, D . K . and W. D . Wishart. 1988. Experimental contact transmission of Pasteurella haemolytica from clinically normal domestic sheep causing pneumonia in Rocky Mountain bighorn sheep. Journal of Wildlife Diseases 24:663 - 667. Portier, C , M . Festa-Bianchet, J . - M . Gaillard, J. T. Jorgenson, and N . G . Yoccoz. 1998. Effects o f density and weather on survival o f bighorn sheep lambs (Ovis candensis). Journal of Zoology London 245:271-278. 72 Rock, M . J., R. L . Kincaid, and G . E . Carstens. 2001. Effects of prenatal source and level o f dietary selenium on passive immunity and thermometabolism o f newborn lambs. Small Ruminant Research 40:129 - 138. Rubin, E . S., W . M . Boyce, M . C . Jorgenson, S. G . Torres, C . L . Hayes, C . S. O'Brien, and D . A . Jessup. 1998. Distribution and abundance o f bighorn sheep in the Peninsular ranges, California. Wildlife Society Bulletin 26:539^551. , W . M . Boyce, and V . C . Bleich. 2000. Reproductive strategies o f desert bighorn sheep. Journal of Mammalogy 81:769 - 786. Samson, J., J. T. Jorgenson, and W . D . Wishart. 1989. Glutathione peroxidase activity and selenium levels in Rocky Mountain bighorn sheep and mountain goats. Canadian Journal o f Zoology 67: 2493-2496. SAS Institute. 2001. S A S version 8.2. Cary, North Carolina, U S A . Singer, F . J., C . M . Papouchis, and K . M . Symonds. 2000. Translocations as a tool for restoring populations of bighorn sheep. Restoration Ecology 8:7-13. Skalski, J. R. 2003. S U R P H Survival under proportional hazards v. 2.2. University of Washington, Seattle. http://www.cbr.washington.edu/paramEst/SURPH/ Skorping, A . and K . H . Jensen. 2004. Disease dynamics: all caused by males? Trends in Ecology and Evolution 19:219-220. Stelfox, J. G . 1974. Range ecology o f bighorn sheep in relation to self-regulation theories. Proc. Biennial Symposium of the Northern W i l d Sheep Council 3:67-76. Swinton, J., M . , E . J. Woolhouse, M . E . Begon, A . P. Dobson, E . Ferroglio, B . T. Grenfell, V . Guberti, R. S. Hails, J. A . P. Heesterbeek, A . Lavazza, M . G . Roberts, P. J. White, and K . W i l s o n . 2002. Microparasite transmission and persistence. Pp. 83 - 101 in P. 73 J. Hudson, A . Rizzo l i , B . T. Grenfell, H . Heesterbok, and A . P. Dobson eds. The ecology of wildlife diseases. Oxford University Press, Oxford, U K . 197 pp. Tompkins, D . M . , A . P. Dobson, P. Arneberg, M . E . Begon, I. M . Cattadori, J. V . Greenman, J. A . P. Heesterbok, P. J. Hudson, D . newborn, A . Pugliese, A . P. Rizzo l i , R. Rosa, F . Rosso, and K . Wilson. 2002. Parasites and host population dynamics. Pp. 45 - 62 in P. J. Hudson, A . Rizzo l i , B . T. Grenfell, H . Heesterbok, and A . P. Dobson eds. The ecology of wildlife diseases. Oxford University Press, Oxford, U K 197 pp. Tracy, M . L . and G . Moller . 1990. Continuous flow vapor generation for inductively coupled argon plasma spectrometric analysis. Part I: Selenium. Journal of the Association of Official Analytical Chemists 73:404-410. Turner, J. C . 1977. Cemental annulations as an age criterion in North American sheep. Journal Wildlife Management 41:211-217. Wehausen, J. D . 1995. Fecal measures of diet quality in wi ld and domestic ruminants. Journal of Wildlife Management. 59:816-823. . 1999. Rapid extinction o f mountain sheep populations revisited. Conservation Biology 13:378-384. Western Regional Climate Center. 2004. Historical Climate Information. Desert Research Institute, Reno, N V http://www.wrcc.dri.edu/ Wishart, W . 1978. Bighorn sheep. Pp. 161 - 171 in J. L . Schmidt amd D . L . Gilbert eds. B i g Game o f North America. Stackpole Books, Harrisburg, P A , U S A . Zar, J. H . 1999. Biostatistical analysis, 4 t h ed. Prentice-Hall, Inc., New Jersey, U S A . 663 pp. 74 Zeigenfuss L . C , F . J. Singer, and M . Gudorf. 2000. Test o f a modified habitat suitability model for bighorn sheep. Restoration ecology 8:38 - 46. 75 Chapter 4 P A T H O G E N S A S S O C I A T E D W I T H B R O N C H O P N E U M O N I A I N R E S I D E N T A N D T R A N S P L A N T E D B I G H O R N S H E E P I N H E L L S C A N Y O N 4.1 I N T R O D U C T I O N Pneumonia has serious acute and chronic effects on bighorn sheep (Ovis canadensis) populations, and is an important factor limiting abundance (Miller, 2001). However, the etiology of pneumonia outbreaks in bighorn sheep has proven elusive. Most investigators have concluded that multiple factors are involved in precipitating outbreaks and that domestic sheep (O. canadensis) and.goats (Capra hircus) can carry pathogens that are lethal to bighorn sheep. However, attempts to correlate disease outbreaks with population demographics, weather, nutrition, and even with known contact with domestic sheep and goats have met with limited success (Mil ler et al., 1991; Ward et al., 1997; Aune et al., 1998; Monello et al., 2001; Rudolph et al., 2003). This has seriously hampered the ability to predict or prevent pneumonia epizootics in bighorn sheep populations. Bacteria in the genera Pasteurella and Mannheimia (Angen et al., 1999) are considered important proximate causal agents o f pneumonia in bighorn sheep either in conjunction with or independently o f other factors (Miller, 2001). Pasteurella trehalosi, P. multocida, and Mannheimia spp. bacteria are common, diverse, and normally commensal organisms colonizing the upper respiratory tract o f ruminants (Boyce 2004). Numerous biovariants of Pasteurella have been isolated from domestic and wi ld animals (Bisgaard et al. 1986, Jaworski et al., 1998). Virulence varies among strains, and specific strains linked to pneumonia in bighorn sheep include M. haemolytica (formerly Pasteurella haemolytica) 76 serotype 2 (Foreyt et al., 1994), P. trehalosi serotype 10 (Kraabel et al., 1998), and P. multocida multocida a (Rudoph et al., 2003; Weiser et al., 2003; Rudolph et al., in review). Beta-hemolytic activity of P. trehalosi and Mannheimia spp. on blood agar is also an indicator o f leukotoxin production (Burrows et al., 1993; Fisher et al., 1999) and has been suggested to be a predictor of virulence in bighorn sheep (Onderka and Wishart, 1998; Mi l l e r et al., 1991; Silflow and Foreyt, 1994; Sweeney et al., 1994; Jaworski et al., 1998). In domestic ruminants, environmental factors and/or intercurrent infections are widely regarded as prerequisites to pneumonic pasteurellosis (Boyce et al., 2004). Lungworms (Forrester and Senger, 1964), viruses (Parks et al., 1972; Spraker and Collins, 1986), other bacteria (Woolf et al. 1970), and various environmental stressors (Spraker et al. 1984) have been implicated as potential single or multiple contributing agents in all-age and lamb pneumonia epizootics in free-ranging and captive bighorn sheep. However, pneumonic pasteurellosis also occurs in bighorn sheep in the absence o f known predisposing factors (Miller, 2001; Cassirer and Sinclair, in prep). The Rocky Mountain bighorn sheep (O. c. canadensis) metapopulation in the Hells Canyon area of Idaho, Oregon, and Washington is composed of 14 reintroduced populations that totalled about 900 bighorn sheep over an area o f 23,000 k m 2 during this study. Between 1971 and 1996, the metapopulation experienced at least seven pneumonia epizootics in one or more populations (Hells Canyon Bighorn Sheep Restoration Committee, 2004). During the period 1997 - 2004, population growth was limited by recurrent pneumonia epizootics in adults and lambs. Bronchopneumonia in lambs was detected in at least one population every year from 1997 - 2004, and bronchopneumonia in adults was detected in one or more populations from 2000 - 2004 (Chapter 2). 77 In this study, field data were used to test the hypothesis that specific biotypes of Pasteurella and Mannheimia are associated with the occurrence of bronchopneumonia in Hells Canyon. Results of routine testing of healthy sheep captured in Hells Canyon and healthy sheep transplanted to Hells Canyon during this period were compared with similar data collected from sheep that died in Hells Canyon of pneumonia and from other causes. I f certain Pasteurella and Mannheimia spp. biovariants were important in causing pneumonia, prevalence o f those biovariants should be higher in pneumonic sheep than in healthy sheep at capture or in sheep that died from causes other than pneumonia. In comparison to resident sheep, sheep transplanted in to Hells Canyon during this period were from populations with no known history of pneumonia. Therefore, it was also expected that pneumonia-causing organisms might be more prevalent in resident sheep than in transplanted sheep. Evidence of verminous or viral involvement in pneumonia-related mortalities and interaction with Pasteurella and Mannheimia spp. strains was also examined. Finally, pathogen data collected at capture were evaluated as predictors of adult or juvenile survival and population growth. 4.2 M A T E R I A L S A N D M E T H O D S Pharyngeal swabs, blood, and fecal samples were collected from resident adult sheep captured, radio-collared, and released in 6 of 14 Hells Canyon bighorn populations in Idaho, Oregon, and Washington, (117°52 '30"W, 46°30"N to 116 °15 'N, 44°45 'W), Dec - Feb 1997 and 1999 - 2004. Similar health survey data were collected from sheep relocated into Hells Canyon Dec - Feb 1997, 1999, 2000, and 2002. Source populations for transplanted sheep were located near Cadomin, Alberta (117°35 '0"W 53°24 '0" N) , Spences Bridge, British 78 Columbia (121 0 20 '59"W, 50° 25 '0" N) and in the Missouri Breaks, Montana (109° 52'26"W, 4 7 0 43 '30"N). Dead adults and lambs were necropsied at the Washington Animal Disease and Diagnostic Laboratory, Pullman, W A ( W A D D L ) (n = 108) or at the Idaho Department of Fish and Game Wildlife Health Laboratory, Caldwell , ID ( IWHL) and University o f Idaho Caine Veterinary Teaching Center, Caldwell , ID ( C V T C ) (n = 12). Culture and bacteriological analysis o f swabs collected at capture and respiratory tissue from necropsies at the I W H L and C V T C was conducted at C V T C by inoculation onto Columbia blood agar (Difco Laboratories, Detroit, Michigan, U S A ) with 5% sheep blood ( C B A ) and a selective Columbia blood agar which contained 5% bovine blood plus antibiotics selective for Pasteurellaceae ( C B A A ) (Jaworski et al., 1993). A l l cultures were incubated at 35-37 C with 10% added C O 2 . A t C V T C , culture media were evaluated after 24 and 48 hrs incubation and representative colonies with characteristics o f Pasteurella were plated on fresh C B A agar for further testing. Biochemical utilization tests were conducted on all selected colonies for biovariant identification (Jaworski et al., 1998). Respiratory tissue and swab samples collected during necropsies at W A D D L were cultured by standard procedures (Carter and Cole, 1990) and classified to genus or genus and species. Some of the Pasteurella organisms isolated at W A D D L were biochemically typed at C V T C . A l l fecal samples, except those from the Missouri Breaks sheep population were evaluated at W A D D L , using the sugar fecal flotation technique (Foreyt, 2001). Presence of lungworm larvae in feces was evaluated with a modified Baermann technique (Beane and Hobbs, 1983). Fecal samples from the Missouri Breaks population were evaluated at Montana Department o f Fish, Wildlife, and Parks Wildlife Laboratory using the modified 79 Baermann technique and the modified Lane fecal flotation procedure (Dewhirst and Hansen, 1961). Serologic tests on samples from all sheep except the Missouri Breaks population were conducted for bluetongue virus ( B T V ) (Agar Gel Immunodiffusion [AGID] , V M R D , Inc. Pullman, W A , U S A ) , epizootic hemorrhagic disease virus ( E H D V ) ( A G I D : Veterinary Diagnostic Technology, Inc. Wheat Ridge, C O , U S A 1991), bovine respiratory syncytial virus ( B R S V ) (Serum neutralization [SN]: neg@l:4, National Veterinary Services Laboratories [NVSL]Testing Protocol 1998), parainfluenza-3 virus (PI3) (SN: neg@l:4, N V S L Testing Protocol 1998), infectious bovine rhinotracheitis virus ( IBRV) S N : neg @ 1:4, N V S L Testing Protocol BPPR02104.02, 1998), bovine viral diarrhea virus ( B V D V ) (SN: neg@l:4, N V S L Testing Protocol 1998), Brucella ovis (Elisa:Walker et al., 1985), serovars o f Leptospira interrogans (Microaggultination: neg @ 1:50), and Anaplasma spp. (Complement fixation: neg @ 1:5) at the Idaho State Bureau of Animal Health Laboratories, Boise, Idaho. A complement fixation test was conducted for titers to Anaplasma spp. Serum samples from the Missouri Breaks population were analyzed for exposure to B. abortus, B. ovis, B R S V , E H D V , I B R V , B V D V , and P B at the Montana Dept. o f Livestock Laboratory, Bozeman, M T . Levels of serum antibodies against M. haemolytica A l , A 2 , and P. trehalosi T10 serotype-specific surface antigens were measured using direct microagglutination assay (Reggiardo, 1981) as described by Mi l l e r et al. (1997). Antibodies to serotype A l and A 2 antigens generally suggest exposure to M. haemolytica biogroup 1, and antibodies to serotype T10 generally suggest exposure to P. trehalosi biogroups 2 and 4 C D (Ward et al., 1997; Jaworski et al., 1998); in all cases, some cross-reaction with other intraspecific serotypes may 80 occur. Levels of leukotoxin neutralizing (LN) antibodies in bighorn sera were measured using a modified in vitro leukotoxin neutralization assay (Greer and Shewen, 1985; Shewen and Wilke, 1988; Miller et al., 1997). Al l serological tests for Pasteurella and Mannheimia spp. antibody titers were conducted at the Department of Pathobiology, University of Guelph, Ontario, Canada. Titers were reported as reciprocal log2 of endpoint dilutions for agglutination assays and as reciprocal log2 dilution that yielded 50% neutralization of toxicity for leukotoxin neutralization assays. Seroprevalence of titers to potential pathogens was compared among groups of sheep using contingency tables and a log-likelihood test statistic (G). Average titers were compared among groups using general linear models and Tukey's studentized range test for pairwise comparisons (Zar, 1999). Prevalence and/or presence of P. multocida and beta-hemolytic isolates, as well as seroprevalence of titers to PB and median positive titer to PB at capture of sheep in Hells Canyon were compared to population dynamics with least-squares linear regression and general linear models incorporating two-way interactions. Demographic parameters were estimated on an annual basis from June - May by population (Cassirer and Sinclair, in prep). Pathogen data were compared to population growth rate (r = ln [Nt/Nt-i]), recruitment (March lamb:ewe ratio), pre-weaning survival (to October) of lambs born to radio-collared ewes the year of sampling and the following year, and to radio-collared ewe and ram survival during the year of sampling. Al l analyses were conducted using SAS statistical software v8.2 (SAS, 2001). 81 4.3 R E S U L T S Pharyngeal swabs, blood, and fecal samples were collected from 196 resident adult sheep in Hells Canyon at capture and from 152 sheep in source populations for transplants into Hells Canyon Dec - Feb 1997, 1999, 2000, and 2002. Thirty-three resident sheep were recaptured at least once for a total o f 229 resident samples. Data on transplanted sheep included 89 sheep sampled in Cadomin, Alberta (58 of which were relocated into Hells Canyon), 43 sheep sampled in Spences Bridge, British Columbia (37 of which were relocated into Hells Canyon) and 20 sampled and relocated from the Missouri Breaks, Montana. A l l sheep transplanted into Hells Canyon were from increasing populations and were certified healthy at capture by a provincial, federal, and/or state veterinarian. One resident ewe captured in the resident Wenaha, Oregon population in 2000 was diagnosed with chronic pneumonia and two resident ewes captured in the resident Black Butte, Washington population in 2000 were diagnosed with mastitis. The Wenaha ewe and one of the Black Butte ewes subsequently died during this study. A l l other resident sheep were considered by veterinarians and wildlife biologists to be healthy at capture. Necropsies were conducted on 40 adults (including 4 yearlings) and 37 lambs that died from bronchopneumonia and 39 adults (including 4 yearlings) and 4 lambs that died from causes other than bronchopneumonia, 1997 - 2004. Causes o f mortality other than pneumonia included hunter-kill (8), cougar predation (7), trauma (6), capture mortality or intentional removal (5), roadkill (4), infection from injury (4), meningitis (2), hypothermia from scabies infection (2), endometritis (1), drowning (1), and 3 unknown (no evidence of pneumonia). 82 Moderate to severe; acute, subacute, to chronic fibrinosuppurative, necrotizing bronchopneumonia was observed at necropsy of adults and iambs diagnosed with pneumonia. Multifocal or diffuse pneumonia was observed in the cranioventral lobes, occasionally in the caudal lobes, and was sometimes accompanied by pleuritis or tracheitis. Lesions were consistent with those caused by Mannheimia or Pasteurella. M i l d verminous involvement was noted in the lungs o f only two adults with bronchopneumonia and no evidence of lungworms was detected in pneumonic lambs. Twenty-five (62%) o f the adults sampled that died from bronchopneumonia had been transplanted to Hells Canyon and 38% were resident sheep. This included 10 o f 38 (26%) sheep transplanted from Spences Bridge, 12 o f 58 (20%) transplanted from Cadomin, 1 of 20 (5%) transplanted from the Missouri Breaks, 8 of 175 radio-collared resident sheep (4%), and 7 uncollared resident sheep. One sheep transplanted from Cadomin in 1995 that died from pneumonia during this period was also included. Among transplanted animals that died from pneumonia, average time to death after release in Hells Canyon was 2.8 years. Seven (18%) of the adults that died from causes other than pneumonia were transplanted and 82% were resident. Average time to death after release for transplanted sheep from causes other than pneumonia was 2 years. Although lambs o f both resident and transplanted sheep died from pneumonia, most (89%) o f the pneumonic lambs recovered were the offspring o f resident sheep. Bacteria Nineteen biovariants of P. trehalosi, 59 biovariants of Mannheimia spp., and 2 subspecies and a non-speciated biotype of P. multocida were isolated from pharyngeal swabs 83 of 163 resident sheep (232 isolates) and 126 transplanted sheep (230 isolates); and from the lungs, throat, nose, tracheobronchial lymph nodes, and/or marrow from 23 pneumonic adults, 25 pneumonic lambs, and 25 adults and 3 lambs that died from causes other than pneumonia. Seventy percent of the P. trehalosi (12) and Mannheimia (42) biovariants have not previously been reported in Rocky Mountain bighorn sheep (Jaworski et al. 1998). Thirty-six percent of the P. trehalosi (6) and M. haemolytica (23) biovariants were isolated only once during the study (Tables 4.1 and 4.2). Only 2 isolates (in 2 sheep) were recovered from 20 sheep sampled in the Missouri Breaks source population, a nonhemolytic biovariant 2B and a hemolytic biotype 1. The low recovery rate in these samples was probably due to problems with sampling, handling, or culturing, (Dunbar, 1990; W i l d and Mil le r , 1994; Foreyt and Lagerquist, 1994). Therefore, this population was excluded from further bacterial analyses. More biovariants were detected in Hells Canyon sheep at capture than in Spences Bridge or Cadomin, and Hells Canyon had the highest proportion o f unique P. trehalosi biovariants not found in the other populations (P. trehalosi: Hells Canyon, 11 biovariants (7 unique); Spences Bridge 6 biovariants (1 unique); Cadomin 6 biovariants (2 unique). Mannheimia spp.: Hells Canyon, 36 biovariants (23 unique); Spences Bridge, 17 biovariants (6 unique); Cadomin, 22 biovariants (11 unique). The average biovariant diversity per sheep sampled was highest in Spences Bridge (23 biovariants/36 sheep sampled = 0.63) and similar in Cadomin (28 biovariants/88 sheep sampled = 0.31) and Hells Canyon (47 biovariants/163 sheep sampled = 0.29). Only 2 biovariants occurred in all three populations: beta-hemolytic biovariant 2 and nonhemolytic biovariant 2B (Tables 4.1 and 4.2). Hells Canyon and 84 Spences Bridge shared more biovariants (9/70 = 0.13) than Spences Bridge and Cadomin (5/51 = 0.10) or Hells Canyon and Cadomin (5/75 = 0.07). Table 4.1. Pasteurella trehalosi and P. multocida biovariants in bighorn sheep Hells Canyon and in source populations of sheep transplanted to Hel ls Canyon , 1997 - 2004. All isolates are nonhemolytic unless noted (j3). N refers to number of sheep represented. H E L L S C A N Y O N SPENCES CADOM1 Biovariant Adult Lamb Other Capture BRIDGE pneumonia Pneumonia mortality N = 11 N = 18 N = 12 N = 163 N = 36 N = 88 P. trehalosi 2 3 3/?, 4 i / U 8 0, 29 2 p, 10 ifi 2 A 0 0 0 1 0 0 2 B 7 5 2 0,2 2 P, 64 \p,21 70 2BCD 0 0 1 1 0 0 2 B C E 1 0 0 0 0 0 2 B D 0 0 0 1 0 0 2 B E 1 0 0 0 0 0 2 B G 0 1 0 0 0 ip,3 2 B S 2 0 0 12 0 0 2 B G 1 0 0 0 0 0 2 C D S 0 0 1 0 0 0 2 E D G 1 0 0 0 0 0 2 G 0 0 0 0 1/9,1 2 G S 0 0 0 1 1 0 4 1 0 0 1 0 0 4 B 1 1 0 4 3 0 ^BLS 0 0 0 2 0 0 4 B S 0 0 4 0 0 ^CDES 1 0 0 0 0 0 P. multocida multocida a 3 3 1 5 0 0 multocida b 3 3 2 4 0 0 biotype U6 0 1 0 1 0 0 Table 4.2. Mannheimia spp. biovariants in bighorn sheep Hells Canyon and in source populations of sheep transplanted to Hells Canyon, 1997 - 2004. Al l isolates are nonhemolytic unless noted (J3). N refers to number of sheep sampled. H E L L S C A N Y O N SPENCES C A D O M I N Biovariant Adult pneumonia Lamb Pneumonia Other mortality Capture BRIDGE N = 23 N = 25 N = 31 N = 163 N = 36 N = 88 1 0 1/9 2B 2B 0 0 1° 0 0 0 1/5 3/3 0 0 0 0 0 0 1 jaE 0 0 0 2B 0 0 1E 0 0 0 0 0 1/? 1° 0 0 0 0 3/? 0 3 a 0 2 28 \p, 5 1 5 3 a B 0 0 0 1 0 0 0 0 0 1 0 0 yxBE 0 0 0 1 0 0 ^aBEX 0 0 0 0 2 4 3 a B S 0 0 0 0 1. 0 3 a E 0 0 0 0 1 3<xEG 0 0 0 1 0 0 3 a E X 0 0 0 0 0 3 B 0 0 0 0. 0 4 3 B E 0 0 0 1 0 1 3BEX 0 0 0 0 0 1 p, 6 3 E 0 0 0 1 0 0 6 a R 0 0 0 0 2/? 2p 6R 0 0 0 0 3/? 0 7 0 1 0 0 0 0 7 B 1 0 0 0 0 jBX 0 0 0 1 1 0 8 0 0 0 2p 0 17/?, 1 8B 0 0 0 2 0 0 8 s 0 0 0 1 0 0 9<x3B 0 1 0 ifi.s 0 0 0 0 0 0 3 / U 0 9ABL 0 0 0 1 0 0 10 0 0 1 0 0 1 10a 0 0 0 28 0 0 1 0 a B 0 0 1 2p 0 0 i n a | 3 B 0 0 1 0 0 0 1 0 a P S 0 0 0 0 1 0 10B 0 0 0 3 1 0 86 Table 4.2, cont'd. Mannheimia spp. biovariants in bighorn sheep Hells Canyon and in source populations of sheep transplanted to Hells Canyon, 1997 - 2004. All isolates are nonhemolytic unless noted. HELLS CANYON SPENCES CADOMIN Biovariant Adult Lamb Other Capture BRIDGE pneumonia Pneumonia mortality N = 2 3 N = 2 5 N = 3 1 N= 1 6 3 N = 3 6 'N = 8 8 10WJK 0 0 0 0 0 1 j 0 B G 0 0 0 0 0 1 1 Q B X 0 0 0 0 0 2 1 0 S 0 0 0 0 1 0 j j a G X 0 0 0 1 0 0 1 6 A 0 0 1 0 2 0 • 1 6 A E 0 0 1 2 0 0 1 6 B 0 0 0 0 1 0 y a p 0 0 0 1 / 3 1 0 j j a P B 0 0 0 2 0 1 1 y A B P 0 0 0 1 0 0 I j a P B C 0 0 0 2 0 0 j j a P E R 0 0 0 1 0 0 J J A E L 0 0 0 1 / ? 0 0 j j a E R 0 0 0 IP 1 0 Ifi 3 / ? 0 1 0 0 UPB 0 0 0 1 0 0 J J P B E X 0 0 0 0 0 2 y j P B X 0 0 0 0 1 0 T j P G L 0 0 0 0 0 1 U 3 L 0 0 0 0 0 1 j j B L X 0 0 0 1 0 0 U L 0 0 0 2 0 0 Pasteurella trehalosi biotype 2 was the most common biotype overall, and was the only trehalosi biotype isolated in the Cadomin source population. Mannheimia spp. biotypes 3 and 8 made up a greater proportion of the isolates from the Cadomin population, and biotype 6 was only detected in the Cadomin and Spences Bridge source populations. Biotype 11 was only detected once and was in the Hells Canyon population (Figure 4.1). 87 A s found in the sheep at capture, a diversity of Pasteurella biovariants was cultured from pneumonic sheep. Nine biovariants of Pasteurella and Mannheimia spp. were cultured from the lungs of 11 pneumonic lambs and 7 biovariants were cultured from the lungs of 8 pneumonic adults. Nonhemolytic P. trehalosi 2B was cultured from the lungs of 5 sheep, P. multocida multocida a and P. m. multocida b were cultured from the lungs of 4 sheep each, other biovariants occurred in either 1 or 2 pneumonic sheep (Table 3). The two biovariants shared by all three populations at capture (P. trehalosi 2 and 2B) were the only biovariants cultured from the lungs of 26% (5) of pneumonic sheep. Two biovariants of Mannheimia spp. and four biovariants o f P. trehalosi were isolated from pneumonic sheep but not from any sheep at capture. A M. haemolytica biovariant 7 was isolated from the lungs of a pneumonic resident lamb from the Redbird population. This biovariant has been previously described in Rocky Mountain bighorn sheep and in domestic sheep (Jaworski et al. 1998). A beta-hemolytic Mannheimia spp. U p biovariant was isolated from pharyngeal swabs from a resident adult in the M u i r Creek population that died of bronchopneumonia in 2001 and pharyngeal and nasal swabs from 3 lambs that died o f pneumonia in 2003 and 2004 in the Redbird (2) and Mui r Creek (1) populations. A single nonhemolytic strain o f this biovariant was cultured from an adult male in the Black Butte population in 2000 and from bighorn sheep and two feral domestic goats in Hells Canyon in 1995 (Rudolph et al. 2003). Five biovariants (2BCE, 2 B E , 2 C D S , 2 E D G , 4 C D E S ) of P. trehalosi were cultured from the lungs, marrow, or throat of 4 sheep that died from bronchopneumonia but not from any sheep at capture (Table 4.1 and Table 4.2) 88 70 <o 60 Q) ra 50 o ~ 40 ~ 30 e 20 -i °- 10 0 10 11 16 U6 70 60 in S 50 8 40 ° 30 | 20 <v ° - 10 B • Cadomin • Spences Bridge — T l X L 8 9 10 11 16 U a U6 70 60 V) 1 50 » 40 | 30 ^ | 20 ° - 10 i L a, Mi-• lamb pneumonia • adult pneumonia a other adult mortality 10 11 16 U U6 Biotype or s u b s p e c i e s Figure 4.1. Prevalence of biochemical types of Pasteurella and Manneheimia spp. isolated from bighorn sheep in (A) Hells Canyon at capture, (B) source populations of bighorn sheep transplanted to Hells Canyon, and (C) mortalities in Hells Canyon, 1997 - 2004. Pasteurella trehalosi is represented by biotypes 2 and 4. Mannheimia spp. are represented by biotypes 1,3,6-11, 16, and U. P. multocida is represented by subspecies a (multocida a), b (multocida b), and U6 (untypable biotype U6). 89 Table 4.3. Pasteurella and Mannheimia bacteria cultured from 23 adult bighorn sheep and 25 Iambs that died from bronchopneumonia in Hells Canyon, 1997 -2004. Biovariant 3 Hemolysis Sample type N Adults N lambs P. trehalosi 21 19 2 B Lung 0 2 2 B Pharyngeal 0 1 2 Nh Nasal 1 0 2 Nh Pharyngeal 1 2 2 Nh Lung. 1 2 2B '. Nh Nasal 1 0 2B Nh Pharyngeal 1 3 2B Nh Marrow 1 0 2B Nh TB Lymph node c 1 0 2B Nh Lung 2 2BCE Nh Lung 1 0 2BE Nh Lung 1 0 2BS Nh Marrow 1 0 2BS Nh Pharyngeal 1 0 2DCS Nh Lung 1 0 2EDG Nh Marrow 1 0 2BG Nh lung 1 2G B Pharyngeal 1 4 Nh Nasal 1 0 4B Nh Nasal 1 0 4B Nh Pharyngeal 0 1 4CDES Nh Pharyngeal 1 0 untyped b TB Lymph node 2 1 untyped Pharyngeal 2 0 untyped Lung 9 5 M. haemolytica 2 10 1 B lung 0 1 3a Nh pharyngeal 0 2 Ub B pharyngeal 1 2 7 Nh lung 0 1 7B Nh pharyngeal 1 0 9abQ Nh nasal 0 1 Untyped pharyngeal 0 1 P. multocida 11 9 Multocida a lung 3 1 Multocida a pharyngeal 0 2 Multocida b lung 2 2 Multocida b pharyngeal 0 1 biotype U6 lung 0 1 Untyped lung 3 2 a More than one biovariant was cultured from some individuals. b Not all isolates classified to species were biochemically typed. 0 TB = tracheobronchial. 90 These strains have not previously been described in bighorn sheep (Jaworski et al., 1998) , however biotype 4 C D E S was detected in a captive bison population (Ward et al., 1999) . Three of these biovariants were isolated from sheep in the Sheep Mountain population in Hells Canyon, and one was from a sheep that had been transplanted from Cadomin, A B and died in the Wenaha population. A l l other biovariants found in sheep that died from pneumonia were also detected in sheep at capture (Tables 4.1 and 4.2). Pasteurella trehalosi was cultured from the greatest proportion of individuals in all groups, followed by P. multocida, and Mannheimia spp. (Table 4.4). N o P. multocida was isolated from source populations of bighorn sheep transplanted in to Hells Canyon (Figure 4.1). Prevalence of P. multocida (28 - 35%) cultured from 71 dead sheep (all tissue) was significantly higher than in pharyngeal swabs collected from 163 resident (6%) or 124 transplanted sheep at capture (0) (G = 37.050, 4 D F , p < 0.001) (Figure 4.1). However, the prevalence of P. trehalosi (84 - 73%), Mannheimia spp. (8 - 27%), and P. multocida (40 -41%) in 25 pneumonic adults and 22 pneumonic lambs did not differ from that cultured from 27 sheep that died from other causes (P. trehalosi 52%, Mannheimia spp. 33%, P. multocida 41%) (G = 0.272, 2 D F , p = 0.256) (Figure 4.2). P. trehalosi was the most frequently cultured Pasteurella or Mannheimia spp. from the lungs o f pneumonic lambs (11 o f 15) and adults (15 of 21). In 62% of adult cases (13 of 21), P. trehalosi was the only Pasteurella or Mannheimia spp. cultured from the lungs and in all cases (n = 7) where the P. trehalosi 91 • He l l s C a n y o n • C a d o m i n 0 S p e n c e s B r i dge P. trehalosi Mannheimia spp. P. multocida B • lamb pneumonia • adult pneumonia H other mortality P. trehalosi Mannheimia spp. P. multocida Figure 4.2. Prevalence of Pasteurella trehalosi, Mannheimia spp., and P. multocida isolated from (A) resident and transplanted bighorn sheep at capture and (B) mortalities in Hells Canyon, 1997 - 2004. Isolates from live sheep cultured from pharyngeal swabs. Isolates from dead sheep cultured from pharyngeal and nasal swabs, tracheobronchial lymph nodes, and lungs 92 cultured from the lungs of adult sheep were typed biochemically, they were nonhemolytic biotype 2 (Figure 4.3). P. multocida was isolated from the lungs of 5 lambs (0.33) and and 8 adults (0.37) and M. haemolytica (biotypes 1 and 7) were isolated from the lungs o f 1 lamb each (0.13) (Table 4.3). Prevalence of beta-hemolytic strains was lower in adults that died from pneumonia (1/20 isolates), the Cadomin source population (23/163), and resident sheep at capture (39/242), than in the Spences Bridge source population (19/67), lambs that died from pneumonia (7/22), and sheep that died from causes other than pneumonia (7/13) (G = 24.10, 5 D F , p = 0.0002) (Figure 4.4). • Adults • Lambs nonhemolytic beta-hemolytic multocida multocida and only beta-hemolytic Figure 4.3. Propor t ion of pneumonic adults and lambs where only nonhemolytic P. trehalosi was isolated from the lungs and those where beta-hemoloytic P. trehalosi or Mannheimia spp., P. multocida were isolated from lungs either separately or together in Hells Canyon 1997-2004 . 93 60% HC adult Cadomin HC capture Spences HCIamb HC other pneumonia Bridge pneumonia mortality Figure 4.4. Prevalence of beta-hemolytic P. trehalosi and Mannheimia spp. isolates from bighorn sheep in Hel ls Canyon (HC) and in source populations of bighorn sheep transplanted to Hel ls Canyon. Sample size (number of isolates) above each bar. Leukotoxin neutralizing titers differed significantly among all populations (F3 i i 5 2 = 150.92, p < 0.0001) and were lowest in the Cadomin source population and highest in Hells Canyon (Table 4.5). Agglutinating titers to M. haemolytica serotype A l were significantly lower in the Cadomin and the Missouri Breaks source populations than in Hells Canyon and were not analyzed for the Spences Bridge population (F2,162 = 47.56,/? O.0001). Agglutinating titers to P. trehalosi serotype T10 were also lower in the Cadomin population than in any other population (F = 110.15, 3 DF,/? < 0.0001). Titers to M. haemolytica serotype A2 showed less divergence among populations and were highest in the Spences Bridge source population and lower in the other three populations (F3241 = 5.95, 3 DF,/? = 0.0006) (Table 4.5). 94 Table 4.4. Prevalence (percent of bighorn sheep) of Pasteurella and Mannheimia species and beta-hemolysis in 6 bighorn sheep populations in Hel ls Canyon , 1997 - 2004. Population Date N sheep n isolates P. trehalosi Mannheimia spp. P. multocida beta-hemolytic Wenaha Mar 1997 12 15 1.00 0.08 0.08 0.50 Wenaha Jan 2000 10 12 1.00 0.10 0 0.10 Wenaha Mar 2003 5 7 1.00 0.20 0 0.60 Redbird Mar 1997 9 14 0.67 0.56 0.11 0.33 Redbird Jan 2000 7 9 0.86 0.29 0 0.14 Redbird Mar 2003 3 3 1.00 0 0 0 Black Butte Mar 1997 7 10 0.71 0.29 0.29 0.29 Black Butte Jan 2000 7 10 0.57 0.43 0.43 0.14 Black Butte Mar 2003 2 2 1.00 0 0 0 Asotin Mar 2003 13 15 1.00 0.08 0 0 Imnaha Jan 2000 21 38 0.95 0.38 0 0.38 Imnaha Mar 2003 5 9 0.80 0.80 0.20 0.80 Lostine Jan 1999 15 28 0.93 0.40 0 0 Lostine Jan 2000 10 20 0.80 0.80 0 0.20 Lostine Jan 2001 6 10 0.66 0.50 0 0 Lostine Feb 2002 10 13 0.90 0.10 0 0 Lostine Jan 2003 8 9 0.75 0 0.25 0.25 Lostine Feb 2004 13 18 0.92 0.31 0 0.23 Total (prevalence) 163 242 143 (0.88) 50 (0.30) 10 (0.06) 36 (0.22) Table 4.5. Average titers (standard deviation) to Pasteurella in adult bighorn sheep in 6 Hells Canyon populations and source populations of sheep transplanted to Hells Canyon, 1997-2004 . Population Year 3 L K T A A l A 2 T10 Hells Canyon" 1997-2000 7.34 (2.18) 5.17(1.25) 6.60(1.49) 9.93 (1.86) Spences Bridge 1997 6.74(1.62) N D C 7.88 (0.85) 11.68 (0.77) Cadomin 1999, 2000 0.47(1.28) 3.61 (0.90) 7.09 (2.05) 6.20(1.16) Missouri Breaks 2002 3.00 (0.82) 3.05(1.16) 6.60(1.78) 10.30(1.16) all samples collected Dec. - Mar. b Sample sizes: Hells Canyon L K T A = 46, A l = 63, A2 = 103, T10 = 104; Spences Bridge all tests = 40; Cadomin L K T A = 50; Cadomin A l , A2 , T10 = 82; Missouri Breaks all tests = 10. ° N D = N o d a t a 95 Other bacteria isolated from the lungs of pneumonic sheep included Arcanobacterium pyogenes (5 lambs and 6 adults), Streptococcus spp. (4 adults), Moraxella spp. (2 lambs), Histophilus somnus (1 adult and 1 lamb) and a Histophilus somnus-like organism (1 lamb). Acinetobacter sp., Clostridium sp., Peptostreptococcus sp., Prevotella sp., and Eschericia coli were isolated from the lungs of one lamb each. Klebsiella oxytoca, Agrobacterium sp., Pseudomonas sp., Staphylococcus sp., Clostridum sordelli, and Fusobacterium necrophorum were each isolated from a single adult. None of these organisms were isolated at capture. Other bacteria isolated from the lungs o f adult sheep that did not die from pneumonia included Arcanobacterium pyogenes (3), Streptococcus spp. (4), Enterococcus spp. (3) , Eschericia coli (2), Bacillus spp. (2). Pseudomonas sp., and Citrobacter sp., were isolated from one individual each. Viruses and Rickettsia Immunohistochemistry of pneumonic lamb lung tissue (n = 4) was negative for PI3 and B R S V . One o f two pneumonic adults tested prior to death had a positive titer o f 8 to PI3 and immunohistochemistry of lung tissue of 3 o f 4 pneumonic adults tested was positive for PI3. Serology o f 4 of 5 adults that did not die of pneumonia was also positive for PI3 (titer range 32 - 64). One sheep that did not die from pneumonia tested positive on an El isa to B R S V . N o viruses were cultured from dead sheep. Antibodies to PI3 were commonly detected in resident sheep at capture (57%, Table 5). Positive titers to PI3 occurred in every population every year sampled except for the Lostine population in 2002, and prevalence ranged from 9% -100%. Positive titers ranged from 4 to 128, with median positive titers of 4 in the Lostine population in 2003 (90% prevalence) to 64 in the Black Butte population in 1997 (25% prevalence). Antibodies to Anaplasma were detected in 45 of 229 individuals sampled in 5 o f 6 resident populations. Titers to other viral pathogens occurred with low prevalence in resident sheep at capture, including B T V (5 individuals in 5 populations), B R S V (2 individuals in 1 population), B V D (7 individuals in 3 populations), E H D (5 individuals in 4 populations), and OPP (3 individuals in 2 populations) (Table 6). N o titers were detected to Brucella ovis or IBR. In transplanted sheep, antibodies to PI3 (positive titer range 1 6 - 3 2 ) were only detected in the Spences Bridge source population, and prevalence (23%) was significantly lower than in Hells Canyon (57%) (G = 20.24, 1 df, p < 0.001). Seroprevalence of positive titers to E H D V (85%) and B R S V (100%) were significantly higher in the Missouri Breaks than in Hells Canyon (G > 83.86, 1 df, p < 0.001) as was prevalence of positive titers to I B R (9%) in the Spences Bridge population (G = 15.11, 1. df, p < 0.001) (Table 6). N o positive titers were detected to Anaplasma spp., B V D V , or B T V in source populations for transplanted sheep. Lungworm Fecal lungworm larvae prevalence (G= 56.57, 5 df, p < 0.0001) and intensity (F 5 > 3 46 = 25.07, p < 0.0001) were higher in sheep transplanted to Hells Canyon from Cadomin and the Missouri Breaks than in resident sheep at capture or adults that died from pneumonia (Table 4.8). Table 4.6. Prevalence of titers to respiratory viruses and Anaplasma spp. in adult bighorn sheep in 6 Hells Canyon populations at capture, 1997 - 2004. Number positive in parentheses. Subpopn Year 3 N A n a b B T V BRSV B V D V E H D V OPP PI3 PI3 median +c PI3 + range Asotin 2002- 03 13 0 0.08(1) 0 0 0.08 0.15(2) 0.38 (5) 8 Black Butte 1997 - 98 12 0 0 0 0 0 0 0.25 (3) 64 Black Butte 1999 - 00 6 0.16(1) 0 0.16(1) 0 0 N D 1.00 (6) 32 8 -64 Black Butte 2002- 03 10 0.50 (5) 0 0 0 2 (0.20) 0 0.9 (9) 12 Imnaha 1999 - 00 21 0.38 (8) 0 0 0.14(3) 0 0 0.86(18) 24 8 -64 Imnaha 2002- 03 6 0 0 0 0 1 (0.17) 0 1.00 (6) 8 8 -16 Lostine 1998- 99 15 0.47 (7) 0.07(1) 0 0 0 1 (0.07) 0.87(13) 16 8 -16 Lostine. 1999 - 00 21 0.38 (8) 0 0 0.05 (1) 1 (0.05) 0 0.67 (14) 32 ' 4 - 6 4 Lostine 2000- 01 11 0 0 0 0 0 0 0.09(1) 16 16 Lostine 2001 - 02 25 0 0 0 0 0 0 0 0 0 Lostine 2002- 03 10 0 0 0 0 0 0 0.90 (9) 4 4 Lostine 2003- 04 15 0 0 0 0 0 0 0.47(7) 12 8 - 3 2 Redbird 1997- 98 12 0 0 0 0.08(1) 0 0 0.50 (6) 24 ' 16-32 Redbird 1999 - 00 6 0.17(1) 0 0 0.33 (2) 0 0 0.83 (5) 16 8 - 3 2 Redbird 2002- 03 12 0 0.17(1) 0 0 0 0 1.00(12) 12 I-"" 4 - 1 2 8 Wenaha 1997- 98 10 0.33 (3) 0 0 0 0 0 0.10(1) 16 16 Wenaha 1999- 00 16 0.75 (12) 0 2(0.13) 0 0 0 0.81 (13) 32 8 -64 Wenaha 2002- 03 8 0 0.13 (1) 0 0 0 0 0.38 (3) 16 16 , TOTAL 229 0.20 (45) 0.02 (4) 0.01 (3) 0.03 (7) 0.02 (5) 0.01 (3) 0.57(132) a Year = all samples collected Dec. - Mar. b Pathogens: Ana, Anaplasma sp.; B T V , bluetongue virus; BRSV, bovine respiratory syncitial virus; E H D V , epizootic hemorrhagic disease virus; IBR infectious bovine rhinotracheitis; OPP ovine progressive pneumonia; PI3, parainfluenza-3 virus 0 Median value of positive titers to parainfluenza-3 virus vo Table 4.7. Prevalence of titers to respiratory viruses and Anaplasma spp. in adult bighorn sheep in 6 Hells Canyon populations and source populations of sheep transplanted to Hells Canyon, 1997 - 2004. Population Year" N A n a b B T V BRSV B V D V E H D V IBR OPP PI3 Hells Canyon 1997-2004 229 0.20 0.02 0.01 0.03 0.02 0 0.01 0.57 Spences Bridge 1997 43 0 0 0 0 0 0.09 0 0.24 Cadomin 1999, 2000 89 0 0 0 0 0 0 0.01 0 Missouri Breaks 2002 20 N D N D 1.0 0 0.85 0.05 0 0 a all samples collected Dec. - Mar. b Pathogens: Ana, Anaplasma sp.; B T V , bluetongue virus; BRSV, bovine respiratory syncitial virus; E H D V , epizootic hemorrhagic disease virus; IBR infectious bovine rhinotracheitis; OPP ovine progressive pneumonia; PI3, parainfluenza-3 virus Table 4.8. Fecal prevalence and mean intensity (larvae, eggs, or oocysts per gram in parentheses) of gastrointestinal parasites and lungworms in bighorn sheep Hells Canyon and in source populations of sheep transplanted to Hells Canyon, 1997 - 2004. Parasite H E L L S C A N Y O N S P E N C E S C A D O M I N M I S S O U R I Adult Lamb Other Capture B R I D G E B R E A K S pneumonia pneumonia mortality n = 14 n = 16 n = 18 n = 117 n = 41 n = 71 n = 14 Coccidia (Eimeria spp.) 0.27(15) 0 0.26(17) 0.58 (184) 0.17(5) 0.97 (410) 0 Strongyles 0.36 (35) 0 0.63 (19) 0.12(3) 0.09 (0) 0 0.71 (18) Nematodirus spp. 0.43 (21) 0 0.56 (5) 0.65 (5) 0.56 (3) 0.59 (7) 0.14(0) Trichuris spp. 0.07 (2) 0 0.17(1) 0.20(1) 0.29 (2) 0.45 (5) 0 Protostrongylus spp. 1 0.27(1) 0 0.61 (7) 0.42 (10) 0.61 (3) 0.83 (47) 0.85 (25) Sample sizes for Protostrongylus - Adult pneumonia 16; Capture 175; Cadomin 83, Missouri Breaks 20. so 00 99 N o lungworms or gastrointestinal parasites were detected in lambs at necropsy (n = 41) or in their feces (n = 16). Prevalence of fecal lungworm larvae was also lower in adults that died from bronchopneumonia (27%) than in adult mortalities due to causes other than bronchopneumonia (61%) (p = 0.046). Average intensity of infection was also lower in adults that died from pneumonia than from other causes (Table 4.8) Protostrongylus spp. larvae were detected in feces of 4 o f 6 resident populations sampled in Hells Canyon. Prevalence in the populations where lungworm was detected averaged 75% (77/102) with a mean intensity of 12 larvae per gram of feces (lpg) (95% CI = 8 - 16 lpg). Lungworm prevalence was higher in the Black Butte (24/29, 83%) and Wenaha (21/27, 78%) populations than in the Imnaha (10/21, 48%) and Redbird (22/52, 42%) populations = 17.88, 3 df, p = 0.005), but mean intensity ( 9 - 1 6 lpg) was low. Intensity greater than 48 lpg was only detected in 1 animal (Table 4.9). Table 4.9. Fecal lungworm {Protostrongylus spp.) larvae infection prevalence and intensity (larvae per gram) in 175 bighorn sheep sampled at capture in 6 populations in Hells Canyon, 1997 - 2004. Population Year Prevalence Mean intensity Range Asotin 2003 0/13 0 0 Black Butte 1997 10/13 20.17 6 - 4 8 Black Butte 2000 6/6 24.5 6 - 4 0 Black Butte 2003 8/10 5.9 2 - 1 6 Imnaha 2000 10/21 12 1 --176 Lostine 2000 0/15 0 0 Lostine 2001 0/14 0 0 Lostine 2002 0/13 0 0 Lostine 2003 0/7 0 0 Lostine 2004 0/10 0 0 Redbird 1997 8/11 13.64 5 - 3 3 Redbird 2000 2/3 10.7 4 - 2 8 Redbird 2003 12/12 9.3 2 - 2 2 Wenaha 1997 6/12 8 8 - 3 0 Wenaha 2000 15/15 9.8 3 - 2 2 A V E R A G E 1 0.42 8 0--176 Mean of 6 populations tested. 100 Gastrointestinal parasites Gastrointestinal parasite eggs, oocysts, or larvae detected in feces of resident Hells Canyon sheep included Eimeria spp. 58%, the thread-necked strongyle (Nematodirus spp.) 65%, unidentified strongyles 12%, and the whipworm (Trichuris spp).19% (Table 4.8). The tapeworm Wyominia was detected in 1 of 5 samples from 1 population (Asotin), and the pinworm Skjrabinema was detected in 20 sheep in all populations except the Lostine (prevalence 15%). The abomasal nematode Marshallagia was reported only in the Missouri Breaks source population (prevalence 43%, mean intensity 10 epg). N o Eimeria spp. or Trichuris spp. were detected in the Missouri Breaks population and no Strongyles (other than Nematodirus spp.) were detected in the Cadomin population. The highest prevalence and intensity of Eimeria spp. and Trichuris spp. occurred in the Cadomin source population and the highest prevalence o f strongyles (other than Nematodirus spp.) occurred in the Missouri Breaks source population. Mean intensity of strongyles was highest in the sheep that died from all causes and the Missouri Breaks source population (F 5 236 = 8.59, 5 D F , / j < 0.0001) (Table 4.8). Ectoparasites L o w to moderate levels of Psoroptes (scabies) infection were found in all resident populations in Hells Canyon except the Lostine population. N o scabies infection was detected in source Table 4.10. Gastrointestinal parasite prevalence and abundance in feces of 132 adult bighorn sheep captured in Hells Canyon, 1997 - 2004. Population Year Coccidia Mean (sd) Strongyles Mean (sd) Nematodirus spp. Mean (sd) Trichuris spp. Mean (sd) Asotin 2003 0/5 0 0/5 0 4/5 2 (2.58) 0/5 0 Black Butte 1997 2/7 34 (86.48) 5/7 1.71 (3.73) 111 6 (4.04) '1/7 0.57(1.51) Black Butte 2000 1/6 0.5(1.22) 4/6 18.67 22.90) 6/6 4(3.16) 1/6 0.50 (4.90) Black Butte 2003 0/1 0 0/1 0 1/1 3 0/1 0 Imnaha 2000 3/21 5 (18.62) 0/21 0 7/21 1.19(1.99) 1/21 0.19(0.87) Lostine 2000 14/15 91.67 (93.01) 0/21 0 13/15 7.47 (5.88) 5/15 1.4(2.61) Lostine 20001 15/15 169.87 (188.07) 0/15 0 12/15 6.13(5.95) 3/15 1.07 (2.71) Lostine 2001 8/9 42.7 (54.25) 0/9 0 6/9 7.7(13.25) 2/9 2.2 (5.33) Lostine 2002 8/10 45.1 (79.86) 0/10 0 7/10 9.2(10.51) 6/10 4.2 (5.37) Lostine 2003 6/7 112(114.20) 0/7 0 6/7 11.86(9.53) 1/7 0.29 (0.76) Lostine 2004 4/4 88 (98.05) 0/4 0 0/4 0 0/4 0 Redbird 1997 2/6 20.83 (42.00) 3/6 4.17(5.85) 2/6 0.33 (0.52) 1/6 0.67(1.63) Redbird 2000 0/3 0 3/3 20 (8) 2/3 2.67 (3.06) 1/3 2.33 (4.04) Redbird 2003 3/3 188.67(177.79) 3/3 44.33 (52.69) 3/3 4.67 (5.51) 2/3 0.67 (0.58) Wenaha 1997 4/5 1243.8(1235.31) 0/5 0 5/5 9.2 (6.72) 0/5 0 Wenaha 2000 7/15 65.93 (103.32) 0/15 0 9/15 5.3 (6.84) 3/15 0.40 (0.82 A V E R A G E 1 77/132 183.64 (417.04) 16/132 2.59 (16.07) 86/132 5.33 (7.02) 26/132 0.99 (2.69) Mean of 6 populations tested 102 populations of transplanted sheep. The ixodid tick Dermacentor albipictus and the ear tick Otobius megnini were also documented in resident sheep. Demographic effects of pathogens The relationship between demographic parameters and PI3 and characteristics of Pasteurella and Mannheimia isolates were compared for 6 populations in Hells Canyon for 1 - 6 years (17 population-years, Table 4.4 and Table 4.6). Ewe survival was negatively related to PI3 seroprevalence ( r = 0.253, p = 0.047, n = 16) and population growth was negatively related to median positive titer to PI3 (r 2 = 0.249,/? = 0.041, n = 17). Lamb survival the following summer was negatively related to the percent of individuals carrying beta-hemolytic isolates at capture, however this was not significant (r 2 = 0.205, p = 0.090, n = 15) (Figure 4.5). There were no significant interactions between pathogens (p > 0.10). 4.4 D I S C U S S I O N Three subspecies or biovariants of Pasteurella bacteria were most common in the lungs of pneumonic sheep: P. m. multocida a, P. m. multocida b, and P. trehalosi 2B. P. multocida was more common in dead sheep (all causes) than in resident sheep at capture and was not detected in source populations o f transplanted sheep. Pasteurella multocida (primarily P. m. multocida a) was an important pathogen in an epizootic that occurred in Hells Canyon sheep moved to captivity in 1995 - 96, just prior to this study (Weiser et al., 2003) and it was also detected in a previous pneumonia epizootic in captive sheep in y = -0.4786X + 0.8078 r2 = 0.205 p = 0.090 0.2 0.3 0.4 0.5 0.6 0.7 Prevalence of beta-hemolytic isolates 0.8 0.9 Figure 4.5. Least-squares regression between bacteria and virus data collected at capture and demographic characteristics of bighorn sheep in 6 Hells Canyon populations, 1997 - 2004. 104 Colorado (Demartini and Davies, 1977) and in wi ld sheep epizootics in Montana (Aune et al., 1998). However, the prevalence of P. multocida in apparently healthy Hells Canyon sheep that died from other causes is inconsistent with the hypothesis that P. multocida was the cause of pneumonia. Prevalence o f P. multocida at capture was also not correlated with adult or lamb survival or population growth. Nonhemolytic biovariants of P. trehalosi (especially biovariant 2B) were as prevalent as P. multocida in the lungs of pneumonic sheep. This is similar to observations o f animals that died in the field in the 1995 - 96 epizootic in Hells Canyon (Rudolph et al., in review) and in other epizootics (Aune et al., 1998). P. trehalosi 2B was also the most common biovariant in all populations of resident and transplanted sheep at capture. Although P. multocida and P. trehalosi 2B were most common, 11 biovariants of Pasteurella and Manneheimia bacteria were isolated from the lungs o f 19 pneumonic sheep. This diversity o f organisms could be a result of several factors. Since these data were collected over seven years in different populations, it may be that each outbreak had a distinct etiology. The isolates in pneumonic sheep may be representative o f various strains of virulent Pasteurella and Mannheimia circulating within and among populations causing pneumonia in naive resident and transplanted sheep, or strains that are periodically introduced into populations from domestic sheep and goats in the study area. Two biovariants of M. haemolytica and 5 biovariants o f P. trehalosi were cultured from the lungs, nose, throat, or marrow o f pneumonic sheep but not any sheep at capture. These organisms may have been introduced from a domestic or wi ld source not sampled in this study. Ten of the Mannheimia and 2 of the P. trehalosi biovariants detected at capture in Hells Canyon that 105 were not previously recorded in Rocky Mountain bighorn sheep have been recorded in domestic sheep and/or domestic goats (Jaworski et al., 1998; Ward et al., 2002). However, since only Pasteurella biovariants considered nonpathogenic were detected in a high proportion of sheep with pneumonia, and some of these biovariants occurred widely in both resident and transplanted sheep, it may be that epizootics were initiated by a combination of environmental and/or other pathological contributing factors. A number o f opportunistically pathogenic bacteria, such as Histophilus somnus and Arcanobacterium pyogenes were identified in dead sheep and not in samples at capture. These organisms may have been contaminants or ancillary findings, however they have been detected in other bighorn sheep epizootics and could also play a role in the pathology o f pneumonia. Some of these potential pathogens may be difficult to detect on nonselective media (Ward et al., 1986) and were likely overlooked using the sampling methods employed in this study. Parainfluenza-3 was the only virus with persistent, widespread exposure detected in Hells Canyon bighorn sheep. This virus was detected in the 1995-96 die-off (Cassirer et al. 1996, Rudolph et al. submitted), and may play a role in causing pneumonia in Hells Canyon. However, exposure to PI3 was not observed in pneumonic lambs and was not consistently detected in pneumonic adults. Exposure to PI3 was also observed in sheep that died from causes other than pneumonia and no PI3 virus was isolated. Serologic titers to PI3 are also frequently observed in healthy bighorn sheep populations (Foreyt et al., 1996; Aune et al., 1998) and pneumonia epizootics in bighorn sheep have occurred in the absence o f PI3 (Miller et al., 1991). Widespread titers to Anaplasma, as observed in this study, have been documented in other bighorn sheep populations and have not been linked to pneumonia outbreaks (Foreyt et al., 1996; Crosbie et al., 1997). 106 Finally, while the biochemical typing classification of Pasteurella and Mannheimia used in this study resolved previous problems with cross-reactions and untypeable isolates using serotyping (Jarworski et al., 1998), this classification method may not be appropriate for epidemiological analysis. Since biochemical typing is based on characteristics unrelated to pathology (Jaworski et al., 1998), it is possible that virulent and avirulent isolates may be classified as the same biovariant, or that strains with the same virulence factors may be classified as different biovariants biochemically. Beta-hemolytic strains of Mannheimia and Pasteurella, including biotype 2 corresponding to known virulent strain serotype 10 (Kraabel et al. 1998, Ward et al. 1997), occurred in all populations. Beta-hemolytic isolates were isolated from pneumonic lambs but were nearly absent from pneumonic adults. Neutralizing titers to leukotoxin in Hells Canyon adults were higher than those in transplanted sheep and in captive sheep at and after challenge with beta-hemolytic P. trehalosi serotype 10 (Kraabel et al., 1998) suggesting exposure to cytotoxic and presumably beta-hemolytic strains of Mannheimia and Pasteurella in Hells Canyon. Overall however, beta-hemolytic P. trehalosi or Mannheimia spp. isolates were not commonly detected in the lungs of animals that died from pneumonia and prevalence was not a consistent predictor of pneumonia outbreaks in adults or lambs. M. haemolytica biovariant 1 (serotype 2), is known to be pathogenic in bighorn sheep (Foreyt et al., 1994). It was uncommon and detected in only a single pneumonic lamb, two sheep that died from other causes, and two sheep at capture. This biovariant was also detected in one sheep transplanted into Hells Canyon from the Missouri Breaks source population. This strain did not appear to be an important disease agent in this study or in the 1995 - 96 epizootic (Rudolph et al., in review). 107 Macroparasite infection in Hells Canyon sheep was similar or lower than that described for healthy Rocky Mountain bighorn populations in the region (Forrester and Senger 1964, Foreyt et al., 1996; Hoar et al. 1996). Gastrointestinal parasites differed among resident sheep and source populations for sheep transplanted to Hells Canyon. Intensity of lungworm larvae in feces was low in Hells Canyon relative to transplanted sheep, and there was no evidence o f significant verminous involvement bronchopneumonia in adults or lambs in Hells Canyon. Some potentially pathogenic organisms were identified in this study, such as P. multocida, Mannheimia beta-hemolytic biovariant U 1 5 , several beta-hemolytic and nonhemolytic biovariants o f P. trehalosi biotypes 2 and 4, and parainfluenza-3 virus. However, with a few exceptions, prevalence of these organisms in sheep that died from pneumonia did not differ from that in apparently healthy sheep and/or sheep that died from causes other than pneumonia. A s a result, pathogen data collected from bighorn sheep at capture using techniques similar to those employed in this study may not reliably provide information that can be used to predict whether individuals are healthy or likely to either experience or cause pneumonia i f mixed with other sheep. Currently, demographics (rate of increase, recruitment rates, and cause-specific mortality) may be more useful for evaluating population health than health sampling data. In the future, detection and recognition of pathogens might be improved by use of molecular techniques to identify and classify microorganisms and evaluate virulence factors and gene expression, especially i f these were accompanied by experimental challenge tests. In particular, better identification of virulence factors associated with P. trehalosi and P. multocida in bighorn sheep would be useful (McNei l et al. 2002, Weiser et al. 2003). A broader sampling strategy targeting a wider 108 range of potential pathogens, including bacteria other than Pasteurella and Mannheimia spp., might also more accurately identify causes o f epizootics and risk factors for bighorn sheep. 4.5 L I T E R A T U R E C I T E D A N G E N , O., R. M U T T E R S , D . A . C A U G A N T , J. E . O L S E N , A N D M . B I S G A A R D . 1999. Taxonomic relationships of the [Pasteurella] haemolytica complex as evaluated by D N A - D N A hybridizations and 16S r R N A sequencing with proposal of Mannheimia haemolytica gen nov.. comb, nov., Mannheimia granulomatis comb, nov, Mannheimia glucosidal sp. nov., Mannheimia ruminalis sp. nov. and Mannheimia varigena sp. nov. International Journal of Systematic Bacteriology 49:67 - 86. A U N E , K . A . , N . A N D E R S O N , D . W O R L E Y , L . S T A C K H O U S E , J. H E N D E R S O N , A N D J. D A N I E L . 1998. A comparison o f population and health histories among seven Montana bighorn sheep populations. Biennial Symposium o f the Northern W i l d Sheep and Goat Council 11:46 - 69. B E A N E , R. D . A N D N . T. H O B B S . 1983. The Baermann technique for estimating Protostrongylous infection in bighorn sheep: effect o f laboratory procedures. Journal of Wildlife Diseases 19:7 - 9. B I S G A A R D , M . , J. R. P H I L L I P S , A N D W . M A N N H E I M . Characterization and identification of bovine and ovine Pasteurellaceae isolated from the oral cavity and rumen of apparently normal cattle and sheep. Acta Pathologica et Microbiologica Immnunologica Scandanavica Section B Microbiology. 94:9 - 17. B O Y C E , J. D . , R. Y . C . L O , I. W I L K I E , A N D B . A D L E R . 2004. Pasteurella and Mannheimia. Pp. 273 - 294 in C . L . Gyles, J. F . Prescott, J. G . Songer, and C . O. Thoen eds. Pathogenesis of bacterial infections in animals, 3 r d ed. Blackwell Publ., Ames, IA . B U E C H N E R , H . K . 1960. The bighorn sheep in the United States, its past, present, and future. Wildlife Monographs 4:174 pp. B U R R O W S , L . L . , E . O L A H - W T N F I E L D , A N D R . Y . C . L O . 1993. Molecular analysis of the leukotoxin determinants from Pasteurella haemolytica serotypes 1 to 16. Infection and Immunity 61:5001-5007. C A R T E R , G.R. , J.R. COLE.(eds) . 1990. Diagnostic procedures in veterinary bacteriology and mycology, 5 t h ed., Pasteurella and Francisella, pp. 129-142. Academic Press, New York, N Y . C A S S I R E R , E . F . and A . R. E . S I N C L A I R , in prep. The dynamics of disease in a reintroduced Rocky Mountain bighorn sheep metapopulation in Hells Canyon, U S A . , L . E . O L D E N B U R G , V . L . C O G G I N S , P. F O W L E R , K . R U D O L P H , D . L . H U N T E R , A N D W . J. F O R E Y T . 1996. Overview and preliminary analysis of Hells Canyon bighorn sheep die-off, 1995-96. Biennial Symposium of the Northern W i l d Sheep and Goat Council 10:78-86. C R O S B I E , P. R., W. G . G O F F , D . S T I L L E R , D . A . JESSUP, A N D W . M . B O Y C E . 1997. The distribution o f Dermacentor hunteri and Anaplasma sp. In desert bighorn sheep (Ovis Canadensis). Journal of Parasitology 83:31 - 37. 110 D E M A R T I N I , J. C . A N D R. B . D A V I E S . 1977. A n epizootic .of pneumonia in captive bighorn sheep infected with Muellerius sp. Journal of Wildlife Diseases 13:117 — 124. D E W H I R S T , L . W . A N D M . F . H A N S E N . 1961. Methods to differentiate and estimate worm burdens in cattle. Veterinary Medicine 56:84 - 89. D U N B A R , M . R., A . C . S. W A R D , A N D G . P O W E R . 1990. Isolation of Pasteurella haemolytica from tonsillar biopsies o f Rocky Mountain bighorn sheep. Journal o f Wildlife Diseases 26:210 - 213. F I S H E R , M . A . , G . C . W E I S E R , and D . A . H U N T E R . 1999. Use of a polymerase chain reaction method to detect the leukotoxin gene lk tA in biogroup and biovariant isolates of Pasteurella haemolytica and P. trehalosi. American Journal o f Veterinary Research 60 :1402- 1406. F O R E Y T , W . J. 2001. Veterinary parasitology reference manual, 5 t h ed. Blackwell Publishing Professional, Ames IA, U S A . ., S. Z E N D E R , A N D R. J O H N S O N . 1996. A 20-year health evaluation study of a healthy bighorn sheep population in northeastern Washington. Biennial Symposium of the Northern Wi ld Sheep and Goat Council 10:66 - 71. . K . P. SNIPES, A N D R. W. K A S T E N . 1994. Fatal pneumonia following inoculation o f healthy bighorn sheep with Pasteurella haemolytica from healthy domestic sheep. Journal of Wildlife Diseases 30: 137 - 145. A N D J. L A G E R Q U I S T . 1994. A reliable transport method for isolating Pasteurella haemolytica from bighorn sheep. Journal of Wildlife Diseases 30:263-266. F O R R E S T E R , D . J. A N D C . M . S E N G E R . 1964. A survey of lungworm infection in bighorn sheep in Montana. Journal of Wildlife Management 28:481 - 491. G R E E R , C. N . , A N D P. E . S H E W E N . 1986. Automated colorimetric assay for the detection of Pasteurella haemolytica leukotoxin. Veterinary Microbiology 12: 33-42. H O A R , K . L . , D . E . W O R L E Y , K . A U N E . 1996. Parasite loads and their relationship to herd health in the Highlands bighorn sheep herds in southwestern Montana. Biennial Symposium o f the Northern Wi ld Sheep and Goat Council 10:57 - 65. H E L L S C A N Y O N B I G H O R N S H E E P R E S T O R A T I O N C O M M I T T E E . 2004. The Hells Canyon Initiative: Hells Canyon Bighorn Sheep Restoration Plan. Unpubl. Rept. Idaho Dept. o f Fish and Game, Lewiston, ID. 58 pp. J A W O R S K I , M . D . , D . L . H U N T E R , A N D A . C . S. W A R D . 1998. Biovariants o f isolates of Pasteurella from domestic and wi ld ruminants. Journal of Veteterinary Diagnostic Investigation 10 :49-55 . , A . C. S. W A R D , D . L . H U N T E R , A N D I. V . W E S L E Y . 1993. Use of D N A analysis o f Pasteurella haemolytica biotype T isolates to monitor transmission in bighorn sheep (Ovis canadensis canadensis). Journal of Cl inical Microbiology 31:831 - 835. K R A A B E L , B . J., M . W . M I L L E R , J. A . C O N L O N , A N D H . J. M C N E I L . 1998. Evaluation of a multivalent Pasteurella haemolytica vaccine in bighorn sheep: protection from experimental challenge. Journal of Wildlife Diseases 34:325-333. M C N E I L , H . J., P. E . S H E W E N , R. Y . C . L O , J. A . C O N L O N , A N D M . W . M I L L E R . 2002. Mannheimia haemolytica serotype 1 and Pasteurella trehalosi serotype 10 culture supernatants contain fivrinogen binding proteins. Veterinary Immunology and Immunopathology 90:107 - 110. 112 M I L L E R , M . W . 2001. Pasteurellosis. Pp. 330 - 339 in E . S. Williams and I. K . Barker eds. Infectious diseases o f wi ld mammals 3 r d ed. Iowa State University Press, Ames, IA . , J. A . C O N L O N , H . J. M C N E I L , J. M . B U L G I N , A N D A . C . S. W A R D . 1997. Evaluation of a multivalent Pasteurella haemolytica vaccine in bighorn sheep: Safety and serologic responses. Journal o f Wildlife Diseases 33: 738-748. , N . T. H O B B S , A N D E.S. W I L L I A M S . 1991. Spontaneous pasteurellosis in captive Rocky Mountain bighorn sheep (Ovis Canadensis Canadensis): clinical, laboratory, and epizootiological observations. Journal o f Wildlife Diseases 27:534-541. M O N E L L O , R. J., D . L . M U R R A Y , A N D E . F . C A S S I R E R . 2001. Ecological correlates of pneumonia epizootics in bighorn sheep herds. Canadian Journal o f Zoology 79:1423-1432. O N D E R K A , D . K . A N D W. D . W I S H A R T . 1988. Experimental contact transmission o f Pasteurella haemolytica from clinically normal domestic sheep causing pneumonia in Rocky Mountain bighorn sheep. Journal of Wildlife Diseases 24:663-667. P A R K S , J. B . , POST, G. , T H O R N E , T., A N D P. N A S H . 1972. Parainfluenza-3 virus infection in Rocky Mountain bighorn sheep. Journal o f the American Veterinary Association 6 6 9 - 6 7 2 . R E G G I A R D O , C . 1981. Serological tests in bovine bacterial pneumonias. Epidemiological application and potential as a research tool. Advances in Experimental Medicine and Biology 137: 818-822. 113 R U D O L P H , K . M . , D . L . H U N T E R , W . J. F O R E Y T , E . F . C A S S I R E R , R. B . R I M L E R , A N D A . C . S. W A R D . 2003. Sharing o f Pasteurella spp. between free-ranging bighorn sheep and feral goats. . Journal o f Wildlife Diseases 39:897-903. , D . L . H U N T E R , R. B . R I M L E R , E . F . C A S S I R E R , W . J. F O R E Y T , W . J. D E L O N G , G . C . W E I S E R , A N D A . C . S. W A R D , submitted. Micro-organisms associated with a bighorn sheep pneumonia epizootic in Hells Canyon. Journal of Wildlife Diseases. S A S I N S T I T U T E . 2001. S A S version 8.2. Cary, North Carolina, U S A . S H E W E N , P. E . , A N D B . N . W I L K I E . 1988. Vaccination of calves with leukotoxic culture supernatant from Pasteurella haemolytica. Canadian Journal of Veterinary Research 52: 30-36. S I L F L O W , R. M . A N D W . J. F O R E Y T . 1994. Susceptibility o f phagocytes from elk, deer, bighorn sheep, and domestic sheep to Pasteurella haemolytica cytotoxins. Journal of Wildlife Diseases 30 :529-535 . S P R A K E R , T. R., H I B L E R , C. P., G . G . S C H O O N V E L D , A N D W . S. A D N E Y . 1984. Pathological changes and microorganisms found in bighorn sheep during a stress-related die-off. Journal o f Wildlife Diseases 20 :319-327 . , A N D J. K . C O L L I N S . 1986. Isolation and serologic evidence of a respiratory syncytial virus in bighorn sheep from Colorado. Journal of Wildlife Diseases 22: 418 - 420. S W E E N E Y , S. J., R. M . S I L L F L O W , A N D W . J. F O R E Y T . 1994. Comparative leukotoxicities of Pasteurella haemolytica isolates from domestic sheep and free-ranging bighorn sheep (Ovis canadensis). Journal of Wildlife Diseases 30:523-528. 114 W A L K E R , R. L . , B . R. L E A M A S T E R , J. N . S T E L L F L U G , A N D E . L . B I B E R S T E I N . 1985. Use of enzyme-linked immunosorbent assay for detection of antibodies to Brucella ovis in sheep: Field trial. American Journal of Veterinary Reseach 46: 1642-1646. W A R D , A . C . S., W E I S E R , G . C. , W . J. D E L O N G , A N D G . H . F R A N K . 2002. Characterization of Pasteurella spp. isolated from healthy domestic pack goats and evaluation o f the effects of a commercial Pasteurella vaccine. American Journal of Veterinary Research 63: 119 - 123. , D Y E R , N . W., A N D B . W . F E N W I C K . 1999. Pasteurellaceae isolated from tonsillar samples of commercially-reared American Bison {Bison bison). Canadian Journal of Veterinary Research 63:161-165. , D . L . H U N T E R , M . D . J A W O R S K I , P. J. B E N O L K I N , M . P. D O B E L , J. B . v J E F R E S S , A N D G . A . T A N N E R . 1997. Pasteurella spp. in sympatric bighorn and domestic sheep. Journal of Wildlife Diseases 33:544 - 557. , L . R. S T E P H E N S , B . J. W I N S L O W , R. P. G O G L E W S K I , D . C . S C H A E F E R , S. K . W A S S O N , A N D B . L . W I L L I A M S . 1986. Isolation of Haemophilus somnus: a comparative study of selective media. American Association of Veterinary Laboratory Diagnosticians 29:479 - 486. W E I S E R , G . C , W. J. D E L O N G , J. L . P A Z , B . SHAFI I , W . J. P R I C E , A N D A . C . S. W A R D . 2003. Characterization o f Pasteurella multocida associated with pneumonia in bighorn sheep. Journal of Wildlife Diseases 39:536-544. 115 W I L D , M . A . A N D M . W. M I L L E R . 1994. Effects o f modified Cary and Blair medium on recovery of nonemolyic Pasteurella haemolytica from Rocky Mountain bighorn sheep (Ovis candensis Canadensis) pharyngeal swabs. Journal of Wildlife Diseases 30: 16 - 1 9 . W O O L F , A . , D . C . K R A D E L , A N D G . R. B U B A S H . 1970. Mycoplasma isolates from pneumonia in captive Rocky Mountain bighorn sheep. Journal o f Wildlife Diseases 6 :169- 170. Z A R , J H . 1999. Biostatistical Analysis, 4 t h ed. Prentice-Hall, Inc. Upper Saddle River, N J 663 p 116 Chapter 5 C O N C L U S I O N S This research tested the hypothesis that the Hells Canyon bighorn sheep metapopulation was regulated by disease through transmission of introduced pathogens at a rate proportional to the number of sheep, independently o f environmental factors such as nutrition and climate. Under this hypothesis it was predicted that disease would be the major source o f mortality, and this prediction was confirmed. Adult and juvenile survival were highly variable among years and among Hells Canyon bighorn populations, due to periodic outbreaks o f pneumonia. The most frequent known source o f adult mortality was disease (n -21, 43%), followed by cougar predation (n = 13, 27%), falls or injuries (n = 11, 22%), and human-caused death (harvest, poaching, or vehicle collisions) (n = 4, 8%). Populations declined (r = -0.08) during population-years when disease-related adult mortality was detected and did not decline in its absence (r = 0.11; t = 2.99; 9.94 df; P = 0.01). Two populations doubled in size during the study in the absence o f pneumonia-caused adult mortality. The other 6 populations incurred pneumonia-related adult mortality and increased by less than 30%. Cougar predation was the second most frequent source o f adult mortality in Hells Canyon, however, contrary to observations in other areas (Wehausen 1996, Hayes et al. 2000, Kamler et al. 2002), predation did not have a significant effect on population growth rates and only the presence of disease-related adult mortality caused significant declines in population growth rates. Although parasite infections can predispose animals to predation, particularly when nutritional constraints are present (Murray et al. 1997), a lack of interactive 117 effect between predation and disease-related mortality on population growth indicated that vulnerability to predation was not higher in conjunction with disease outbreaks. Periodic summer pneumonia epizootics resulted in high rates of lamb mortality June -August. The timing of peak lamb mortality (24-91 days of age) during epizootics differed from that observed in ungulate populations where pneumonia is not an important source of mortality. In ungulates, most pre-weaning mortality usually occurs within one month of birth (Gaillard et al. 2000). The relative vigor of neonatal lambs and subsequent infection suggests that morbidity may coincide with the waning of passive immunity acquired through the colostrum (Miller et al. 1997). Previous reports of summer lamb epizootics have described consistently high rates of juvenile mortality subsequent to all-age pneumonia die-offs (Spraker et al. 1984, Coggins and Matthews 1992, Ryder et al. 1992). I observed sporadic lamb pneumonia epizootics interspersed with years of high lamb survival both following and in the absence of pneumonia epizootics in adults. Although lamb epizootics occurred in at least one of the 8 populations in 5 of the 6 years of the study, each year, summer lamb epizootics were confined to only a few populations and these changed from year to year. As with disease-related mortality in adults, lamb epizootics were not synchronized through space or time. While only causes of summer lamb mortality could be determined, fall and early winter survival were also important in determining recruitment. Growth observed in this bighorn sheep metapopulation (r = 0.04) was far below the intrinsic rate of increase (rm, Caughley 1977) calculated for bighorn sheep at approximately 0.26 by Buechner (1960) and below the observed average rate of increase of 0.13 for successfully introduced Rocky Mountain bighorn populations in Colorado (McCarty and 118 Mi l l e r 1998). Overall, disease played a chronic although spatially and temporally heterogeneous role in limiting the rate of population growth by affecting both adult survival and recruitment. The second prediction of the hypothesis tested in this research was that disease-related mortality and survival rates would not be correlated with nutrition or climatic conditions. The data supported this prediction. N o relationship was observed between the occurrence o f pneumonia-related mortality and climate, selenium, or fecal nitrogen values. Bighorn populations occurring within the same climate zones experienced similar weather conditions and fecal nitrogen regimes, and, although climate regimes differed geographically, climate was highly correlated among zones. When correlated environmental variation plays an important role in demographics, this synchronizes dynamics among populations (Grenfell et al. 1998) contrary to the pattern observed in this study. Asynchronous dynamics o f other bighorn sheep populations has also been attributed to the relative importance of local factors over climate (Rubin et al. 1998, Bleich et al. 1990). Factors associated with disease in this population appeared to operate at the individual and population rather than landscape level. Bighorn sheep in some populations apparently periodically succumbed to virulent pathogens, and mortality occurred in susceptible individuals, particularly transplanted sheep and lambs. Pneumonia outbreaks occurred in lambs born both to resident and transplanted sheep, suggesting some acquired immunity in adults and transmission of pneumonia-causing pathogens to lambs. This could result in intrinsic cycling o f disease (Hobbs and Mi l l e r 1992). However, despite intensive management to maintain separation between domestic sheep and goats and wi ld sheep, average minimum distance o f marked individuals to domestic sheep and goats in this 119 population was less than 6 km. This may be why the third prediction, that populations experiencing epizootics w i l l be more likely to contact domestic sheep was not supported. Buffers between domestic and bighorn sheep that appear to be effective at preventing disease outbreaks have been identified as 20 km (Singer et al. 2000), 23 km (Zeigenfuss et al. 2000) and 40 km (Monello et al. 2001), therefore, presumably, all populations had potential for contact with domestic sheep and goats and periodic introduction of novel pathogens may have been a primary external factor precipitating pneumonia outbreaks. In the absence of pneumonia outbreaks, heterogeneity in individual fitness was apparent in the positive correlation of reproductive success (poor or good) in adult ewes through time. This positive association o f offspring survival with survival the previous year is contrary to the pattern observed in desert bighorn sheep (Rubin et al. 2000) and could be a consequence of a relative abundance o f forage in Hells Canyon (Festa-Bianchet and Jorgenson 1998). Immunocompetence may be positively correlated with genetic heterogeneity (Coltman et al. 1999), and transplanted populations may exhibit genetic drift and low genetic heterozygosity due to small founder size (Fitzsimmons et al. 1997). We did not evaluate genetic characteristics in this study, but it may be significant that transplanted sheep had reduced resistance to disease relative to resident sheep (descendants of previous transplants). In particular, the sheep transplanted from Cadomin, A B in this study came from a large native and presumably outbred population, yet exhibited high vulnerability to pneumonia. The third prediction of the research hypothesis tested was that disease-related mortality w i l l increase when populations are at high densities (and presumably at higher numbers). This prediction was not supported by the data. The link between host density and 120 parasite abundance is a basic principle o f epidemiological theory (Anderson and M a y 1978). However, although pneumonia outbreaks have previously been attributed to high sheep densities, and coincidentally high population size, no evidence for density-dependence in bighorn sheep pneumonia outbreaks has previously been demonstrated (Jorgenson et al.1997, Aune et al. 1998). Monello et al. (2001) suggested that density played a role in epizootics, but no density-related differences, such as population size or growth rates, were found between bighorn populations that suffered pneumonia epizootics and those that did not. In this study, I also found no relationship between population size or growth rates and pneumonia epizootics. Previous epizootics in this population have affected populations as a large as 220 (Cassirer et al. 1996), but in this study pneumonia outbreaks occurred in populations as small as 29 bighorn sheep and with growth rates (r) as high as 0.32 the year prior to the epizootic. Conversely, no outbreaks (in adults) were observed in a population o f 150. This suggests a mass-action or frequency-dependent model of pathogen transmission (Swinton et al. 2002) in which there is no threshold population size for establishment of disease. Virulent pathogens and the gregarious behavior of sheep maintaining relatively high rates of contact, even at low numbers, could allow rapid transmission and high mortality rates regardless o f population size (Altizer et al. 2003). The resulting independence o f disease-related mortality from population size could provide one mechanism for the high rate o f extinction observed in small bighorn populations (Berger 1990, Krausman et al. 1993, Goodson 1994, Wehausen 1999). Three subspecies or biovariants of Pasteurella bacteria were most common in the lungs o f pneumonic sheep: P. m. multocida a, P. m. multocida b, and P. trehalosi biovariant 2B. P. multocida was more common in dead sheep (all causes) than in resident sheep at 121 capture and was not detected in source populations of transplanted sheep. However, the prevalence of P. multocida in Hells Canyon sheep that died from other causes is inconsistent with the hypothesis that P. multocida causes pneumonia. Prevalence o f P. multocida at capture was also not correlated with adult or lamb survival or population growth. Nonhemolytic biovariants of P. trehalosi (especially biovariant 2B) were as prevalent as P. multocida in the lungs of pneumonic sheep. This is similar to observations o f animals that died in the field in the 1995 - 96 epizootic in Hells Canyon (Rudolph et al., in review) and in other epizootics (Aune et al., 1998). P. trehalosi 2B was also the most common biovariant in all populations of resident and transplanted sheep at capture. Although P. multocida and P. trehalosi 2B were most common, 11 biovariants o f Pasteurella and Manneheimia bacteria were isolated from the lungs o f 19 pneumonic sheep. This diversity.of organisms could be a result of several factors. Since these data were collected over seven years in different populations, it may be that each outbreak had a distinct etiology. The isolates in pneumonic sheep may be representative of various strains o f virulent Pasteurella and Mannheimia circulating within and among populations causing pneumonia in naive resident and transplanted sheep, or strains that are periodically introduced into populations from domestic sheep and goats in the study area. However, since only Pasteurella biovariants considered nonpathogenic were detected in a high proportion of sheep with pneumonia, and some of these biovariants occurred widely in both resident and transplanted sheep, it may be that epizootics were initiated by a combination of environmental and/or other pathological contributing factors. A number o f opportunistically pathogenic bacteria, such as Histophilus somnus and Arcanobacterium pyogenes were identified in dead sheep and not in samples at capture. These organisms may 122 have been contaminants or ancillary findings, however they have been detected in other bighorn sheep epizootics and could also play a role in the pathology o f pneumonia. Some o f these potential pathogens may be difficult to detect on nonselective media (Ward et al. 1986) and were likely overlooked using the sampling methods employed in this study. Parainfluenza-3 was the only virus with persistent, widespread exposure detected in Hells Canyon bighorn sheep. This virus was detected in the 1995-96 die-off (Cassirer et al. 1996, Rudolph et al. submitted), and may play a role in causing pneumonia in Hells Canyon. However, exposure to PI3 was not observed in pneumonic lambs and was not consistently detected in pneumonic adults. Exposure to PI3 was also observed in sheep that died from causes other than pneumonia and no PI3 virus was isolated. Serologic titers to PI3 are also frequently observed in healthy bighorn sheep populations (Foreyt et al. 1996; Aune et al., 1998) and pneumonia epizootics in bighorn sheep have occurred in the absence o f PI3 (Mil ler etal . 1991). Finally, while the biochemical typing classification of Pasteurella and Mannheimia used in this study resolved previous problems with cross-reactions and untypeable isolates using serotyping (Jarworski et al. 1998), this classification method may not be appropriate for epidemiological analysis. Since biochemical typing is based on characteristics unrelated to pathology (Jaworski et al. 1998), it is possible that virulent and avirulent isolates may be classified as the same biovariant, or that strains with the same virulence factors may be classified as different biovariants biochemically. Beta-hemolytic strains of Mannheimia and Pasteurella, including biotype 2 corresponding to known virulent strain serotype 10 (Kraabel et al. 1998, Ward et al. 1997), occurred in all populations. Overall however, beta-hemolytic P. trehalosi or Mannheimia 123 spp. isolates were not commonly detected in the lungs of animals that died from pneumonia and prevalence was not a consistent predictor of pneumonia outbreaks in adults or lambs. M. haemolytica biovariant 1 (serotype 2), also known to be pathogenic in bighorn sheep (Foreyt et al. 1994) was uncommon and did not appear to be an important disease agent in this study. Macroparasite infection in Hells Canyon sheep was similar or lower than that described for healthy Rocky Mountain bighorn populations in the region (Forrester and Senger 1964, Foreyt et al., 1996; Hoar et al. 1996). Intensity o f lungworm larvae in feces was low in Hells Canyon relative to transplanted sheep, and there was no evidence of significant verminous involvement bronchopneumonia in adults or lambs in Hells Canyon. The final predictions, that populations experiencing epizootics w i l l carry pathogens not endemic to bighorn sheep and that antibody titers to Pasteurella or other pathogens w i l l be highest in sick populations and lowest in healthy populations were not supported. Some potentially pathogenic organisms were identified in this study. However, with a few exceptions, prevalence of these organisms in sheep that died from pneumonia did not differ from that in apparently healthy sheep and/or sheep that died from causes other than pneumonia. Detection and recognition of pathogens might be improved by use of molecular techniques to identify and classify microorganisms and evaluate virulence factors and gene expression, especially i f these were accompanied by experimental challenge tests. In particular, better identification of virulence factors associated with P. trehalosi and P. multocida in bighorn sheep would be useful (McNe i l et al. 2002, Weiser et al. 2003). A broader sampling strategy targeting a wider range o f potential pathogens, including bacteria other than Pasteurella and Mannheimia spp., might also more accurately identify agents responsible epizootics and allow better assessment of risk factors for bighorn sheep. 5.1 L I T E R A T U R E C I T E D Altizer, S., C . L . Nunn, P. H . Thrall, A . A . Cunningham, A . P. Dobson, V . Ezenwa, K . E . Jones, A . B . Pederson, M . Poss, and J. R. C . Pulliam. 2003. Social organization and parasite risk in mammals: integrating theory and empirical studies. Annual Review of Ecology and Evolutionary Systems 34:517-547. Anderson, R. M . and R. M . May . 1978. Regulation and stability o f host-parasite population interactions. I. Regulatory processes. Journal of Animal Ecology 47:219 - 247. Aune, K . A . , N . Anderson, D . Worley, L . Stackhouse, J. Henderson, A n d J. Daniel. 1998. A comparison of population and health histories among seven Montana bighorn sheep populations. Biennial Symposium of the Northern W i l d Sheep and Goat Council 11 :46 -69 . Berger, J. 1990. Persistence of different-sized populations: an empirical assessment of rapid extinctions in bighorn sheep. Conservation Biology 4:91 - 98. Bleich, V . C , J. D . Wehausen, R. R. Ramey II, and J. L . Rechel. 1990. Metapopulation theory and mountain sheep: implications for conservation. Pp. 353-373 in D . R. McCul lough ed. Metapopulations and Wildlife Conservation. Island Press. 429pp. Cassirer, E . F. , L . E . Oldenburg, V . L . Coggins, P. Fowler, K . Rudolph, D . L . Hunter, and W. J. Foreyt. 1996. Overview and preliminary analysis of a bighorn sheep dieoff, Hells Canyon 1995-96. Proceedings Northern W i l d Sheep and Goat Council 10:78-86. Caughley, G . 1977. Analysis of vertebrate populations. John Wiley & Sons. 234 pp. Coggins, V . C . and P. E . Matthews. 1992. Lamb survival and herd status of the Lostine bighorn herd following a Pasteurella die-off. Proceedings o f the Biennial Meeting of the Northern W i l d Sheep and Goat Council 8:147-154. 125 Coltman, D . W. , J. G . Pilkington, J. A . Smith, and J. M . Pemberton. 1999. Parasite-mediated selection against inbred Soay sheep in a free-living island population. Evolution 53:1259-1267. Enk, T. A . , H . D . Picton, and J. S. Will iams. 2001. Factors limiting a bighorn sheep population in Montana following a dieoffi Northwest Science. 75(3):280-291. Festa-Bianchet, M . and J. Jorgenson. 1988. Selfish mothers: reproductive expenditure and resource availability in bighorn ewes. Behavioral ecology 9:144 - 150. Fitzsimmons, N . N . , S. W . Buskirk, and M . H . Smith. 1997. Genetic changes in reintroduced Rocky Mountain bighorn sheep populations. Journal of Wildlife Management 61:863 - 872. Forrester, D . J. A n d C. M . Senger. 1964. A survey of lungworm infection in bighorn sheep in Montana. Journal of Wildlife Management 28:481 - 4 9 1 . Foreyt, W . J. 2001. Veterinary parasitology reference manual, 5 t h ed. Blackwell Publishing Professional, Ames IA, U S A . . K . P. Snipes, and R. W . Kasten. 1994. Fatal pneumonia following inoculation of healthy bighorn sheep with Pasteurella haemolytica from healthy domestic sheep. Journal of Wildlife Diseases 30: 1 3 7 - 145. Gaillard, J . - M . , M . Festa-Bianchet, N . G . Yoccoz, A . Loison, and C . Toigo. 2000. Temporal variation in fitness components and population dynamics of large herbivores. Annual Review o f Ecology and Systematics. 31:367-93. Goodson, N . J. 1994. Persistence and population size in mountain sheep: why different interpretations? Conservation Biology 3:617-618. Grenfell, B . T., K . Wilson, B . F . Finkenstadt, T. N . Coulson, S. Murray, S. D . A lbon ; J. M . Pemberton, T. H . Clutton-Brock and M . J. Crawley. 1998. Noise and determinism in synchronized sheep dynamics. Nature 394:674-677. Hayes, C . L . , E . S. Rubin, M . C . Jorgensen, R . A . Botta, and W . M . Boyce. 2000. Mountain lion predation of bighorn sheep in the peninsular ranges, California. Journal o f Wildlife Management. 64:954-959. Hoar, K . L . , D . E . Worley, K . Aune. 1996. Parasite loads and their relationship to herd health in the Highlands bighorn sheep herds in southwestern Montana. Biennial Symposium o f the Northern W i l d Sheep and Goat Council 10:57-65. Hobbs, N . T. and M . W . Mi l l e r . 1992. Interactions between pathogens and hosts: simulation o f pasteurellosis epidemics in bighorn sheep populations. Pp. 997-1007 in D . R. McCul lough and R. H . Barrett eds. Wildlife 2001. Population. Elsevier Applied Science, Essex, England. 1163pp. Jaworski, M . D . , D . L . Hunter, and A . C . S. Ward. 1998. Biovariants o f isolates of Pasteurella from domestic and wi ld ruminants. Journal of Veteterinary Diagnostic Investigation 10 :49-55 . Jorgenson, J. T., M . Festa-Bianchet, J. M . Gaillard, and W . D . Wishart. 1997. Effects o f age, sex, disease, and density on survival o f bighorn sheep. Ecology 78:1019-1032. Kamler, J. F. , R. M . Lee, J. C . deVos, W . B . Ballard, and H . A . Whitlaw. 2002. Survival and cougar predation of translocated bighorn sheep in Arizona. Journal of Wildlife Management 66:1267-1272. Kraabel, B . J., M . W . Mil le r , J. A . Conlon, and H . J. Mcnei l . 1998. Evaluation o f a multivalent Pasteurella haemolytica vaccine in bighorn sheep: protection from experimental challenge. Journal o f Wildlife Diseases 34:325-333. Krausman, P. R., R. C . Etchberger, and R . M . Lee. 1993. Persistence of mountain sheep. Conservation Biology 7:219. McCarty, C . W . and M . W . Mi l le r . 1998. Modeling the population dynamics of bighorn sheep. Colorado Division o f Wildlife Special Report 73:35 pp. M c N e i l , H . J., P. E . Shewen, R. Y . C . L o , J. A . Conlon, and M . W. Mi l le r . 2002. Mannheimia haemolytica serotype 1 and Pasteurella trehalosi serotype 10 culture supernatants contain fibrinogen binding proteins. Veterinary Immunology and Immunopathology 90:107 - 110. Mi l le r , M . W. , J. A . Conlon, H . J. Mcnei l , J. M . Bulgin, and A . C. S. Ward. 1997. Evaluation of a multivalent Pasteurella haemolytica vaccine in bighorn sheep: Safety and serologic responses. Journal o f Wildlife Diseases 33: 738-748. Mil le r , M . W., N . T. Hobbs, and E.S. Will iams. 1991. Spontaneous pasteurellosis in captive Rocky Mountain bighorn sheep (Ovis Canadensis Canadensis): clinical, laboratory, and epizootiological observations. Journal o f Wildlife Diseases 27:534-541. Monello, R. J., D . L . Murray, and E . F . Cassirer. 2001. Ecological correlates of pneumonia epizootics in bighorn sheep herds. Canadian Journal of Zoology 79:1423-1432. Murray, D . L . , John R . Cary, and L l o y d B . Keith. 1997. Interactive effects o f sublethal menatodes and nutritional status on snowshoe hare vulnerability to predation. Journal o f Animal Ecology 66:250-264. Nelson, R. J. and G . E . Dumas. 1996. Seasonal changes in immune function. Quarterly Review o f Biology 71:511 -548. 128 . 2004. Seasonal immune function and sickness responses. Trends in immunology. 25: 1 8 7 - 192. Rubin, E . S., W . M . Boyce, M . C . Jorgenson, S. G . Torres, C . L . Hayes, C . S. O'Brien, and D . A . Jessup. 1998. Distribution and abundance of bighorn sheep in the Peninsular ranges, California. Wildlife Society Bulletin 26:539-551. , W . M . Boyce, and V . C . Bleich. 2000. Reproductive strategies o f desert bighorn sheep. Journal of Mammalogy 81:769 - 786. Rudolph, K . M . , D . L . Hunter, R. B . Rimler, E . F . Cassirer, W . J. Foreyt, W . J. Delong, G . C . Weiser, and A . C. S. Ward, submitted. Micro-organisms associated with a bighorn sheep pneumonia epizootic in Hells Canyon. Journal of Wildlife Diseases. Ryder, T. J., E . S. Williams, K . M . M i l l s , K . H . Bowles, and E . T. Thorne. 1992. Effect o f pneumonia on population size and lamb recruitment in Whiskey Mountain bighorn sheep. Proceedings of the Biennial Meeting of the Northern W i l d Sheep and Goat Council 8:136-146. Singer, F . J., C . M . Papouchis, and K . M . Symonds. 2000. Translocations as a tool for restoring populations o f bighorn sheep. Restoration Ecology 8:7-13. Spraker, T. R., Hibler, C . P., G . G . Schoonveld, A n d W. S. Adney. 1984. Pathological changes and microorganisms found in bighorn sheep during a stress-related die-off. Journal of Wildlife Diseases 20:319 - 327. Swinton, J., M . , E . J. Woolhouse, M . E . Begon, A . P. Dobson, E . Ferroglio, B . T. Grenflee, V . Guberti, R. S. Hails, J. A . P. Heesterbeek, A . Lavazza, M . G . Roberts, P. J. White, and K . Wilson. 2002. Microparasite transmission and persistence. Pp. 83 - 101 in P. J. Hudson, A . Rizzo l i , B . T. Grenfell, H . Heesterbok, and A . P. Dobson eds. The ecology of wildlife diseases. Oxford University Press, Oxford, U K . 197 pp. Ward, A . L . , L . R. Stephens, B . J. Winslow, R. P. Goglewski, D . C . Schaefer, S. K . Wasson, A n d B . L . Will iams. 1986. Isolation of Haemophilus somnus: a comparative study o f selective media. American Association o f Veterinary Laboratory Diagnosticians 29 :479-486 . Ward, A . L . , D . L . Hunter, M . D . Jaworski, P. J. Benolkin, M . P. Dobel, J. B . Jefress, and G . A . Tanner. 1997. Pasteurella spp. in sympatric bighorn and domestic sheep. Journal of Wildlife Diseases 33:544 - 557. Wehausen, J. D . 1996. Effects o f mountain lion predation on bighorn sheep in the Sierra Nevada and Granite Mountains o f California. Wildlife Society Bulletin 24:471-479. . 1999. Rapid extinction of mountain sheep populations revisited. Conservation Biology 13:378-384. Weiser, G . C , W . J. Delong, J. L . Paz, B . Shafii, W. J. Price, A n d A . C . S. Ward. 2003. Characterization of Pasteurella multocida associated with pneumonia in bighorn sheep. Journal of Wildlife Diseases 39:536-544. Zeigenfuss L . C , F . J. Singer, and M . Gudorf. 2000. Test o f a modified habitat suitability model for bighorn sheep. Restoration ecology 8:38 - 46. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092258/manifest

Comment

Related Items