A Comparative Health Assessment of Urban and Non-Urban Mule Deer (Odocoileus hemionus) in the Kootenay Region, British Columbia, Canada
Thesis
Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in the Graduate School of The Ohio State University
By Amélie Mathieu, DMV Graduate Program in Comparative and Veterinary Medicine
The Ohio State University 2018
Thesis Committee Dr. Thomas E. Wittum, Advisor Dr. Mark S. Flint, Co-Advisor Dr. Barbara A. Wolfe Dr. Randall E. Junge
1
Copyright by Amélie Mathieu 2018
2
Abstract
The provincial wildlife management agency, British Columbia Ministry of Forests, Lands,
Natural Resource Operations and Rural Development, performed a translocation trial from 2015 to 2017 to control the urban mule deer (Odocoileus hemionus; uMD) overpopulation and supplement the declining non-urban mule deer (nuMD) population in the Kootenay region,
British Columbia, Canada. Several local communities are now considering using uMD translocations as a long-term wildlife management method. The aim of this study was to characterize the health risks associated with the translocation initiative by comparing pathogen exposure, body condition scores (BCS) and pregnancy rates of urban and non-urban mule deer
(nuMD) and to develop predictive disease models to inform management decisions related to urban deer translocations. Blood samples collected from 200 free-ranging mule deer captured in urban and non-urban environments in the Kootenay region from 2014 to 2017 were tested for exposure to selected pathogens and pregnancy status. Body condition scoring (BCS) and morphometric examinations were performed for each deer. BCS averaged 3.4 on a five-point scale, was greater in nuMD, and significantly differed between years. Antibodies were detected for adenovirus hemorrhagic disease virus (AHDV) (38.4% (uMD 43.7%, nuMD 33.3%)), bluetongue virus (BTV) (0.6% (uMD 1.2%, nuMD 0%)), bovine respiratory syncytial virus
(BRSV) (8.4% (uMD 4.6%, nuMD 12.1%)), bovine viral diarrhea virus (BVDV) (1.1% (uMD
0%, nuMD 2.2%)), bovine parainfluenza-3 virus (PI3) (27.0% (uMD 27.6%, nuMD 26.4%)),
Neospora caninum (22.1% (uMD 24.4%, nuMD 19.7%)) and Toxoplasma gondii (8.2% (uMD
12.3%, nuMD 3.9%)). No antibodies against epizootic hemorrhagic disease virus were detected. ii
Pregnancy rates did not differ between the two deer populations (90.7% (uMD 90.6%, nuMD
90.9%)). Exposure to N. caninum was associated with a reduction in pregnancy rates. uMD were more likely to be exposed to T. gondii than nuMD. Comparison of body condition scores, pregnancy rates and pathogen exposure of uMD and nuMD showed that the health of the two populations did not significantly differ, suggesting that these particular pathogens do not factor in the decline of deer populations nor do they pose a risk of uMD translocations. However, predictive models indicated that exposure to AHDV, BRSV and T. gondii currently differs between deer populations and that uMD translocations would result in an increased risk of
AHDV, BRSV and T. gondii transmission. The risk of BRSV and T. gondii transmission under a range of pathogen prevalence conditions was established to be of current concern and the risk of
EHDV, BTV, BRSV, BVDV and T. gondii transmission was established to be of potential future concern for either populations. The number of uMD to translocate under the current conditions should be limited to 344 individuals as to minimize the risk of ADHV transmission and its potential negative effects on the nuMD population. Targeted continuous pathogen monitoring was shown to necessitate the sampling of a total of at least 133 individuals to detect realistic outbreaks in all pathogens of concern, including EHDV, BTV, BRSV, BVDV and T. gondii. These results should be considered as part of a formal risk assessment for future uMD translocations in southeastern British Columbia.
iii
Dedication
To Patrick and his undying love for mulligans.
iv
Acknowledgments
I would first like to thank my advisors Dr. Thomas E. Wittum and Dr. Mark Flint for their continuous support of my master’s research, and for their patience, kindness, and extensive knowledge, as well as my committee members, Dr. Barbara A. Wolfe and Dr. Randall E. Junge, for helping me juggle my master’s research and my clinical responsibilities. I would also like to express my gratitude to Dr. Helen Schwantje and the British Columbia Ministry of Forests,
Lands, Natural Resource Operations and Rural Development for entrusting me with this project. I thank Patrick Stent and Cait Nelson for their assistance with sample organization, and Dr. Aruna
Ambagala, Dr. Robert Bildfell, Josh Branen, Dr. Ling Jin, Dr. Emily Jenkins, Dr. Tomy Joseph,
Rajnish Sharma, Dr. Erin Zabek for coordinating and performing the laboratory assays. I also thank the following people for their assistance with the deer captures: Patrick Stent, Ian Adams,
Holger Bohm, Dr. Nigel and Joan Caulkett, Dr. Adam Hering, Larry Ingham, David Lewis, Dr.
Bryan MacBeth, Becky Phillips, Dr. Kylie Pon, Dr. Pat Rice, Irene Teske, Cliff Wilson and the many volunteers. Finally, I express profound gratitude to my parents and to my partner for providing me with unfailing support and continuous encouragement throughout my years of study. This accomplishment would not have been possible without them.
v
Vita
2005 ...... D.E.S. Collège Mont-Saint-Louis
2007 ...... D.E.C. Sciences, lettres et arts, Collège de Bois-
de-Boulogne
2012 ...... D.M.V. Faculté de Médecine Vétérinaire,
Université de Montréal
2015 to present ...... Residency in Wildlife and Ecosystem Health,
The Ohio State University, The Columbus Zoo
and Aquarium, The Wilds
Publications
Mathieu A, Caulkett NA, Stent PM, Schwantje HM. 2017. Capture of free-ranging mule deer (Odocoileus hemionus) with a combination of medetomidine, azaperone and alfaxalone. J
Wildl Dis 53:296-303.
Mathieu A, Pastor AR, Berkvens CN, Gara-Boivin C, Hébert M, Léveillé AN, Barta JR,
Smith DA. 2018. Babesia odocoilei as a cause of mortality in captive cervids in Canada. Can Vet
J 59:52-58.
Mathieu A, Flint M, Stent P, Schwantje H, and Wittum T. 2018. Health assessment of urban and non-urban free-ranging mule deer (Odocoileus hemionus) in the southeastern British
Columbia, Canada. Accepted for publication by PeerJ. vi
Fields of Study
Major Field: Comparative and Veterinary Medicine
Specialty: Conservation Medicine
vii
Table of Contents
Abstract ...... ii
Dedication...... iv
Acknowledgments ...... v
Vita ...... vi
Table of Contents ...... viii
List of Tables ...... xii
List of Figures ...... xiii
Chapter 1. Literature Review ...... 1
1.1 Suburban and Urban Deer Overabundance ...... 1
1.2 Wildlife Translocation as an Urban Deer Population Management Tool ...... 4
1.3 Risk of Disease Transmission Associated with Wildlife Translocations...... 5
1.4 Health Risk Assessment Approaches for Wildlife Translocations ...... 8
1.5 Health Assessment ...... 12
1.5.1 Nutritional Condition Assessment ...... 13
1.5.2 Pregnancy Status ...... 13
1.5.3 Pathogen Exposure ...... 14
1.6 Management of Mule Deer in the Kootenay Region, British Columbia...... 24
Chapter 2. Rational and Hypothesis ...... 27
viii
Chapter 3. Health Assessment of Urban and Non-Urban Free-Ranging Mule Deer (Odocoileus hemionus) in the Kootenay Region, British Columbia, Canada ...... 29
3.1 Introduction ...... 29
3.2 Materials & Methods ...... 31
3.2.1 Study Area and Animals ...... 31
3.2.2 Capture Techniques ...... 32
3.2.3 Sample Collection and Testing ...... 32
3.2.4 Statistical Analysis ...... 34
3.3 Results ...... 35
3.3.1 Animals Sampled ...... 35
3.3.2 Biometrics...... 35
3.3.3 Pregnancy-Specific Protein B Quantification ...... 36
3.3.4 Pathogen Exposure ...... 36
3.4 Discussion ...... 38
3.4.1 Biometrics...... 39
3.4.2 Pathogen Exposure ...... 40
3.4.3 Study Limitations ...... 44
3.5 Conclusion ...... 45
Chapter 4. Predictive Models for Pathogen Transmission Risk Assessment Associated with Urban
Deer Translocations in Southeastern British Columbia, Canada ...... 46
4.1 Introduction ...... 46
4.2 Materials and Methods ...... 47
4.2.1 Model 1: Stochastic Risk of Pathogen Transmission Associated with uMD
Translocations (Table 4)...... 48
ix
4.2.2 Model 2: Dynamic Risk of Pathogen Transmission Associated with uMD
Translocations (Table 5)...... 49
4.2.3 Model 3: Continuous Pathogen Monitoring Guidelines (Table 6) ...... 50
4.3 Results ...... 51
4.3.1 Model 1: Stochastic Risk of Pathogen Transmission Associated with uMD
Translocations (Table 4)...... 51
4.3.2 Model 2: Dynamic Risk of Pathogen Transmission Associated with uMD
Translocations (Table 5)...... 52
4.3.3 Model 3: Continuous Pathogen Monitoring Guidelines (Table 6) ...... 52
4.4 Discussion ...... 53
4.4.1 Urban Mule Deer Translocation Guidelines ...... 54
4.4.2 Continuous Pathogen Monitoring Guidelines ...... 56
4.4.3 Model Limitations ...... 57
4.5 Conclusion ...... 57
Chapter 5. General Discussion ...... 59
5.1 Introduction ...... 59
5.2 General Hypothesis ...... 59
5.3 Aims ...... 60
5.3.1 Aim 1: To Characterize the Pathogen Exposure, Body Condition and Pregnancy
Rates of Urban and Non-Urban Mule Deer ...... 60
5.3.2 Aim 2: To Determine if the Health of the Urban Deer Population Differs from the
Health of the Non-Urban Deer Population and Characterize the Health Risk Associated with
the Translocation Project ...... 61
x
5.3.3 Aim 3: To Investigate the Role of Infectious Diseases in the Decline of the Non-
Urban Mule Deer Population ...... 62
5.3.4 Aim 4: To Develop Predictive Models to Inform Management Decisions Related
to Urban Deer Translocations ...... 63
5.4 Future Directions ...... 64
5.5 Recommendations ...... 65
Appendix: Tables and Figures ...... 67
References ...... 79
xi
List of Tables
Table 1. Characteristics of urban and non-urban free-ranging mule (Odocoileus hemionus) deer sampled in southeastern British Columbia, Canada...... 68
Table 2. Prevalence of pregnancy based on Pregnancy-Specific Protein B levels in free-ranging mule deer (Odocoileus hemionus) in southeastern British Columbia, Canada...... 69
Table 3. Pathogen exposure prevalence in free-ranging mule deer (Odocoileus hemionus) in southeastern British Columbia, Canada...... 70
Table 4. Predicted minimal sample size (PMSS) required to detect statistically significant differences in pathogen exposure between urban and non-urban mule deer (Odocoileus hemionus) populations...... 73
Table 5. Predicted minimal sample size for a range of pathogen prevalence conditions in the urban (A) and non-urban (B) mule deer (Odocoileus hemionus) populations...... 74
Table 6. Predicted minimum sample size required to detect realistic pathogen outbreaks the urban
(A) and non-urban (B) mule deer (Odocoileus hemionus) populations...... 76
xii
List of Figures
Figure 1. Location of capture sites of urban and non-urban of free-ranging mule deer (Odocoileus hemionus) in southeastern British Columbia, Canada...... 77
Figure 2. An urban mule deer (Odocoileus hemionus) forages on tree branches in April 2017 in
Marysville, British Columbia, Canada...... 78
xiii
Chapter 1. Literature Review
1.1 Suburban and Urban Deer Overabundance
In the early 20th century, rapid change in landscape, market hunting, weak enforcement of game laws, and habitat loss drove deer populations to dangerously low numbers, with white- tailed deer (Odocoileus virginianus) becoming endangered or even locally extinct in several
American states (Whitney 1994). Deer population numbers have since recovered as the result of tenacious conservation efforts undertaken by state wildlife agencies and deer hunters. Those actions included changes in hunting practices and regulations, predator control, creation of game refuges, and deer reintroduction projects (McCabe & McCabe 1984). The increase in interspersion of forest and cropland secondary to farm abandonment and succession on cutover areas also contributed to the recovery of deer populations (Swihart et al. 1995). Since the conservation efforts were implemented, deer population have steadily increased, and are now far beyond their historic size in many areas (Mumford & Whitaker 1982).
More recently, wildlife managers were faced with the opposite deer management challenge - urban deer overpopulation. Indeed, urbanization has resulted in habitat encroachment and spill-over of deer within urban and suburban environments. Deer are highly adaptable animals, and thrive in urbanized environments for a variety of reasons (Swihart et al. 1995).
Urbanized areas provide deer with protection from human predation during the hunting season and year-round protection from predators and, oftentimes, with richer and more diverse sources of food. Deer are opportunistic browsers and adapt well to a diet that includes plants grown on fertilized lawns and gardens, especially in areas of limited foraging opportunities. Together, these 1 factors supersede the risks of living in close proximity to humans and have led to the steady growth of urban deer populations.
Caughley (1981) rationalizes that overabundance of a species ensues when its population size negatively affects human well-being and the fitness of the species itself, reduces the ecosystem richness and diversity, and causes dysfunctions in the ecosystem. As so, deer overabundance within urban and suburban communities is not without negative consequences.
Deer eat and damage private and commercially-grown vegetation. Bucks can also damage trees and shrubs by rubbing the bark with their antler disrupting the flow of phloem (“ring barking”). It has been estimated that residential landscape damage in the U.S. may exceed $250 million per year (Conover 2002). Moreover, urban deer may become a public safety threat. Human- habituated deer may become aggressive and pose a danger to residents (bluff-charging and attacking people, chasing joggers and postal workers, and killing small pets). Five to ten people are killed annually in the U.S. by aggressive bucks (Conover 2002). There also is an increased potential for deer-vehicle collisions. It is estimated that, in the U.S., deer-vehicle collisions result in approximately 29,000 injuries, 100-200 human fatalities annually and more than $2 billion in damage (Conover 1997; Conover et al. 1995; Winter 1999). An estimated 1.5 million deer are killed annually (Conover 2002). Deer populations also attract predators such as cougars (Puma concolor) to urban areas, creating a possible hazard for local residents and their pets as indirect prey targets. Lastly, deer may represent a risk to human health. Zoonotic diseases directly or indirectly associated with deer include Q fever (Coxiella burnetii), leptospirosis (Leptospira interrogans), Lyme disease (Borrelia burgdorferi), brucellosis (Brucella abortus), bovine tuberculosis (Mycobacterium bovis), human babesiosis (Babesia microti), cryptosporidiosis
(Cryptosporidium sp.), and giardiasis (Giardia sp.) (Kruse et al. 2004; Medrano et al. 2012;
Piesman et al. 1979; Rickard et al. 1999; Rijks et al. 2011; Roth 1970). These pathogens may
2 infect humans that come in direct contact with infected deer carcasses, deer ticks, or deer bodily fluids or feces. Although deer were proven incompetent as a reservoir for B. burgdorferi, they remain instrumental in the maintenance of tick abundance and of endemic levels of Lyme disease
(Kilpatrick et al. 2014; Telford et al. 1988).
The public opinion regarding urban deer is greatly polarized and varies as a result of the assortment of stakeholders (government, farmers, local residents, businesses, animal rights activist groups and conservationists) involved in the issue. While some enjoy the presence of deer in urban communities and wish for a peaceful cohabitation, others perceive them as a hazard and nuisance and want their population numbers reduced. Public perception of urban deer appears to vary based on the socioeconomic characteristics of communities surveyed. For example, a survey conducted in large American metropolitan areas in the 1990s revealed urban deer were perceived favorably by the vast majority of urban residents surveyed despite the economic losses, nuisance problems, and threats to human health and safety they incur (Conover et al. 1995). By contrast, a different survey performed in the same decade in the small rural town of Chincoteague, Virginia, showed most residents perceived urban deer negatively (Green et al. 1997). Given the dichotomy of opinions and the overall complexity of the human dimension in the context of the urban deer issue, making wildlife management decision to minimize the negative effects of urban deer overabundance can be challenging. In fact, deer control methods that are perceived has socially acceptable in one situation may not be in another. Because the public opinion is often unpredictable, the degree of satisfaction with a management outcome is likely to be improved at the local community level when stakeholders and their values are identified and collaborative relationships with the community and the media are established (Decker et al. 1992; Triezenberg et al. 2012).
3
The impact of urban deer overabundance may be mitigated through the reduction of the damage they incur or through the reduction of their population. Damage control methods include fencing, use of unpalatable plants and repellants, road signs and harassment techniques such as noise, startling lights or motion-activated sprinklers. Population control methods can be categorized as lethal and non-lethal methods. Lethal methods include hunting, predator reintroduction, and controlled culling whereas non-lethal methods include live-capture and translocation as well as immunocontraception and surgical sterilization. While lethal methods are often the most cost and logistic efficient means of urban deer population control, they are frequently disregarded by public stakeholders. However, it was shown that residents may be more supportive of lethal methods when faced directly with urban deer-related problems and when their tolerance level for such issues has been exceeded (Loker 1996). As such, urban deer management methods need to be evaluated for social acceptance as well as effectiveness, cost efficiency and animal welfare considerations (Clark 1995).
1.2 Wildlife Translocation as an Urban Deer Population Management Tool
Wildlife translocations are defined by the IUCN as the “intentional movement of species from one geographic area to another” (IUCN 1987). Such projects are developed to (i) either reinforce or reintroduce a species within its indigenous range, (ii) introduce a species outside of its indigenous range, or (iii) reduce local human-wildlife conflicts by removing the animals from the area of conflict. As such, translocated animals may have been captive-reared or wild-caught, and may be moved into areas in the core, periphery or outside of historical ranges (Griffith 1993).
Translocations must aim to profit the population, species or ecosystem, and not only the translocated animals (IUCN/SSC 2013). Translocation projects are intricate endeavors and must be carefully planned and executed. When employed as a population management tool,
4 translocation projects should be designed as to maximize the benefits to the source population
(reduction of the population density, resolution of human-wildlife conflicts or improvement of population and habitat health) and to limit the potential for undesirable outcomes on the translocated individuals (increased mortality rates, decreased breeding success or expression of wandering behavior), the recipient populations and their habitats. Evaluation of the fate of the translocated animals necessitates long-term monitoring and is rarely easily achieved. Nonetheless, wildlife managers should consider the potential impacts on the translocated individuals as animal welfare issues may surpass the benefits to the source population and render the translocation effort unjustifiable (Whisson et al. 2012).
1.3 Risk of Disease Transmission Associated with Wildlife Translocations
Anthropogenic movement of wildlife, whether purposeful or not, carries an inherent risk of disease transmission, and is recognized as one of two main drivers of wildlife disease emergence (Daszak et al. 2001). It has been long established that animal translocations may alter host-pathogen interactions and can negatively affect the health of the translocated individuals and that of the resident populations (livestock, other domestic animals, and wildlife), and have important economic and/or public health repercussions (Woodford 2000; Woodford & Rossiter
1993). As explained by Nettles (1988), “a translocated animal is not the representative of a single species but is rather a biological package containing a selection of viruses, bacteria, [fungi], protozoa, helminths and arthropods”. As so, an animal cannot be dissociated from its pathogens, and translocating it implies translocating its pathogens. Many pathogens have a localized geographic distribution as they require specific ecological conditions to thrive. Hence, translocated individuals can carry pathogens that are not found at the release site or can be exposed to pathogens at the release site to which they lack immunity (Kock et al. 2010).
5
Expression of disease may be the result of a lack of acquired immunity to a novel pathogen, or be precipitated by immunosuppression caused by the stress experienced during the translocation and prior to becoming habituated to the new environment or by co-morbidities (Kock et al. 2010;
Sainsbury & Vaughan-Higgins 2012). The risk of disease transmission exists regardless of the origin and destination of the translocated animals but is greatest when animals are born and raised under artificial conditions (zoological facilities, farms and ranches, captive breeding centers).
These animals may become asymptomatically infected with local endemic pathogens or pathogens carried by allopatric species with which they commingle in captivity, and their translocation may result in the transmission of exotic pathogens to the resident fauna. Animals translocated in situ generally carry a lesser risk of disease transmission, as coevolution of pathogens and hosts and natural selection pressures decrease the likelihood of pathogen persistence and disease expression in healthy animals (Kock et al. 2010; Walker et al. 2008).
The risk of disease transmission is reciprocal between wild and domestic species and between translocated and resident individuals. As such, the risk posed by translocated wildlife to resident domestic species and wildlife and the risk posed by resident domestic species and wildlife to translocated wildlife should be considered when planning a translocation. Several disease transmission scenarios exist between translocated wildlife and resident fauna.
Translocated individuals may encounter pathogens at the release site to which they are immunologically naïve. These pathogens may be carried by wild or domestic species. For example, several naïve translocated caribou (Rangifer tarandus) and moose (Alces alces) died after exposure to the meningeal worm Parelaphostrongylus tenuis carried by white-tailed deer present in eastern Canada and U.S (Lankester 2001). The pathogens encountered at the release site may also be carried by domestic species, as seen with enzootic pneumonia-associated die-offs of bighorn sheep (Ovis canadensis) secondary to comingling with healthy domestic sheep (Ovis
6 aries) (Onderka et al. 1988). Translocated individuals may introduce pathogens to the release site which can cause disease among co-existing, immunologically naïve, wild and domestic animals.
Brucellosis was introduced to the Wood Buffalo National Park in Canada and its herd of wood bison (Bison bison athabascae) after Plains bison (B. bison) were translocation from Montana
(Carbyn & Watson 2001). Myxobolus cerebralis, the protozoan responsible for whirling disease in salmonids, was co-introduced to North America along with the brown trout (Salmo trutta)
(Hoffman 1970). The African horse sickness virus was introduced into Spain by two zebras
(Equus burchelli) translocated from Namibia in 1987 (Rodriguez et al. 1992). Resident pathogens may be spread or amplified following the introduction of naïve hosts into the release site. This scenario was seen with Varroa jacobsoni, a blood-sucking mite of bee pupae and adults, after trans-Siberian railroad traffic led to the introduction of the naïve western honeybees (Apis mellifera) to enzootic areas of Asia, allowing comingling with the infected resident Apis cerana honeybees (Oldroyd 1999). Common brushtail possums (Trichosaurus vulpecula) became a new reservoir host for bovine tuberculosis (Mycobacterium bovis) following their translocation from
Australia to New Zealand (Viggers et al. 1993).
Several epizootics caused by wildlife translocations have been documented over the years, some of which having disastrous repercussions. However, the extent of the impact of disease introduction is likely largely underestimated given how difficult detection of disease outbreaks in wildlife is (Wobeser 2006). Disease introduction can result in direct or indirect mortality of translocated or resident individuals, and may have profound repercussions on the ecosystem (Ewen et al. 2012). For example, the introduction of the rinderpest virus in Africa in the 1890s caused a massive wildebeest (Connochaetes spp.) die-off in the Serengeti ecosystem, which resulted in the reduction of lion (Panthera leo) and hyena (Hyaenidae) abundance, and the epidemic of sylvatic plague in prairie dogs (Cynomys leucurtus) resulted in population reduction
7 of the black-footed ferret (Mustela nigripes) (Dobson & Hudson 1995; Williams et al. 1994).
Moreover, wildlife disease introduction can have disastrous socioeconomic and human health implications. The rinderpest outbreak mentioned above also caused mass mortality of cattle, sheep, and goats, which led to starvation and death of a large portion of the human population of
Ethiopia and of the Maasai people of Tanzania (Phoofolo 1993). Disease introduction can have more subtle effects on population health by leading to the reduction of genetic diversity, species richness, recruitment and herd fitness as well as the alteration of age structure (Miller & Thorne
1993). Interestingly, Keesing et al. (2010) found that biodiversity loss frequently increases disease transmission, which suggest that translocations that increase species richness at the release site may decrease the local risk of disease transmission and disease outbreaks. The risk of disease transmission may surpass the wildlife conservation benefits of a translocation project and should be critically evaluated as to prevent far-reaching, long-term, and unpredictable effects on the ecosystem.
1.4 Health Risk Assessment Approaches for Wildlife Translocations
Health risk assessment is the systematic and evidence-based evaluation of the health- related hazards inherent to an activity (Jakob-Hoff et al. 2014; Leighton 2002; MacDiarmid
1997). In the context of wildlife translocations, health hazards may include exposure novel infectious agents, toxins or any other ecological hazards, and can affect both the translocated and resident animal populations. Disease risk analysis is a subset of the health risk assessment and is the process through which the potential consequences of the introduction of selected diseases are evaluated. The probability of such events occurring is then assessed (Murray et al. 2004).
Analysis findings are ultimately used by wildlife managers to modify translocation plans in order to mitigate the risk of disease transmission. The health risk assessment should be transparent and
8 performed independently of those involved in the decision-making process and of those with a vested interest in a particular outcome (Leighton 2002). This must be done in order to ensure fairness, scientific rigor, rationality and credibility in the risk assessment report (Dalziel et al.
2017).
For many years, wildlife translocations were planned and executed with little, if any, consideration for the risk of disease transmission (Griffith 1993). This omission to conduct a systematic evaluation of the risk of disease transmission prior to conducting wildlife translocations resulted in several disastrous disease outbreaks and has had negative impacts on both translocated and resident animal populations. Wild or domestic animal movements and overabundance of wildlife, along with open air farming and expansion or introduction of vectors, have since been identified as the most common risk factors for wildlife diseases introduction
(Gortázar et al. 2007; Kock et al. 2010), and the role of infectious diseases on the health and sustainability of wildlife populations and on the outcome of wildlife translocations has been better defined. Risk of disease transmission has become a pivotal aspect of wildlife management decisions involving translocations. Given the extreme difficulty of controlling diseases once introduced to the wild and the potential disastrous effect of disease translocation, it is crucial that the risk of disease introduction be critically investigated prior to performing a wildlife translocation, and that the translocation be planned in light of the disease risk analysis findings
(Leighton 2002). In some cases, risk assessment findings may reveal an unacceptable risk and preclude execution of the translocation project. The health risk assessment evaluates the general health of the translocated and the sympatric wild and domestic resident animal populations and incorporates evaluation of any pathogens of importance to the either populations. Ideally, the health status of the resident animals of the same or related species to the one being translocated should be evaluated (Cunningham 1996). Emphasis should be placed on the identity, number,
9 distribution, mode of transmission and pathogenicity of said pathogens (Sainsbury & Vaughan-
Higgins 2012). Information required to perform a disease assessment may not exist or may not be easily accessible. The feasibility of a disease risk assessment should be determined early in the translocation planning process, and translocation should not be performed if a valid disease risk analysis cannot be generated (Leighton 2002). Leighton (2002), Davidson & Nettles (1992),
Armstrong et al. (2003) and Miller (2007) established qualitative methods to assess the health risks inherent to wildlife translocations that can be broadly summarized as follow:
i. Redaction of the translocation plan
The translocation plan must be clearly detailed in order for a risk assessment to be
performed. It must include information pertaining to the goals of the project, the animals
to be translocated, the source and destination ecosystems, the methods of capture and
release, the veterinary protocols to be employed as well as a general outline of the
potential health, ecological and economic risks associated with the proposed
translocation.
ii. Identification of health hazards
All potential health hazards associated with the proposed translocation are listed. Health
hazards include all infectious agents the translocated animals and resident animals may
carry or be susceptible to as well as any environmental hazards that may be present in the
destination ecosystem (Murray et al. 2004). A detailed risk assessment often cannot
reasonably be produced for each of the many potential health hazards (Sainsbury &
Vaughan-Higgins 2012). Thus, it is more practicable to perform an in-depth risk
assessment for a few selected hazards. Priority should be placed on health hazards that (i)
may be carried by the translocated animals, (ii) may cause significant harm in resident
animals and/or humans, and (iii) may have substantial ecological or economic
10
repercussions if translocated. Prioritization of health hazards can be difficult given the
epidemiology of wildlife diseases is oftentimes poorly understood; however, novelty of
an infectious agent to the host may be a sufficient reason for it to be classified as
hazardous (Hartley & Sainsbury 2017; Sainsbury & Vaughan-Higgins 2012). iii. Assessment of health risks
The probability of occurrence of each selected health hazard and their potential to cause
damage is assessed. Risk may be categorized as the (i) probability of a pathogen carried
by the translocated animals to cause harm to resident animals or as the (ii) probability of
translocated animals to be harmed by a health hazard present at the release site. For each
pathogen that may be carried by a translocated animal, one must estimate the probability
of the pathogen to be transferred to the release site and the probability that the resident
animals will be exposed to the pathogen. The extent of the damage caused to the
susceptible species, the ecosystem and the economy is then assessed. For each health
hazard that may be present at the release site, one must estimate the probability of
translocated animals of being exposed. The extent of the ecological and economic
damage caused to the translocated animals and the impact on the outcome of the
translocation are then assessed. iv. Assessment of the overall health risk
The overall risk assessment is synthesized based on the evaluation of the objective
information available and on the individual health risk assessments. The overall risk may
be largely driven by a health hazard of particular concern or may be the combination of
several health hazards of comparable concern. The overall health risk is summarized into
a statement that details how the assessment was formulated and comments on the validity
of the information used.
11
v. Additional hazards and risks
Any additional hazards, either directly health-related or indirectly, that were identified
during the health assessment process should be included in the risk assessment report and
further evaluated.
vi. Reduction of risk
In many instances, the health risks identified during the risk assessment can be reduced
through modification of the translocation plan. For examples, risk may be mitigated by
altering details pertaining to the methods of capture and transportation, the veterinary
protocols and the release site.
1.5 Health Assessment
Wildlife health was once interpreted as the simple absence of disease within an animal population. It is now broadly recognized that the health status of a wild animal population reflects its ability to thrive in a changing environment, and is impacted by intrinsic (genetics and physiological capacity) and extrinsic factors (climate, social dynamics, exposure to pathogens and environmental threats, interactions with humans) (Stephen 2014). Parameters commonly evaluated for wildlife health status determination include body condition, herd/population age structure, genetics, physiological health, habitat modifications, environmental contaminants and pathogen diversity and abundance. Knowledge of the health status of a wild animal population and of its determinants and comparison of health data over time and with other wildlife populations can assist wildlife managers in planning conservation management actions.
12
1.5.1 Nutritional Condition Assessment
Nutritional condition can provide insights into individual and herd health and habitat quality. Harder & Kirkpatrick (1994) defined nutritional condition as “the state of body components (e.g., fat, protein) which in turn influence an animal’s future fitness”. As such, it reflects the quantity and quality of forage available and the ability of animals to metabolize its nutrients. Nutritional condition has been linked to reproductive success, overwinter and neonatal survival probability, susceptibility to predation and maternal lactation yields in cervids (Bender et al. 2002; Cook et al. 2004; Loudon et al. 1983; Verme 1969). Several methods have been developed to assess nutritional condition in both live and dead cervids. Live animal methods include a body condition score (BCS), body mass and ultrasonographic assessment of subcutaneous fat and muscles thickness while dead animal methods include mandible, femoral and metatarsal marrow fat indexes, morphometric measurements (weight, size, circumferences) and a variety of kidney fat indices and carcass scoring methods (Cook et al. 2007; Jakob et al.
1996; Peig & Green 2009). Subcutaneous fat thickness, a rump BCS, and an arithmetic combination of subcutaneous fat thickness and the rump BCS labelled rLIVINDEX were shown to be better correlated indices to nutritional condition in live mule deer (Odocoileus hemionus) whereas kidney fat and the Kistner score were most representative of fat and gross energy in dead mule deer (Cook et al. 2007). Unfortunately, ultrasonographic assessment of body condition is not always practical, especially when capturing live deer via helicopter or working in remote areas.
1.5.2 Pregnancy Status
Since physiologically or environmentally stressed animals tend to not reproduce well, pregnancy status is valuable indicator of population health. Several serologic methods for
13 pregnancy diagnosis are used for wildlife species, including the quantification of the pregnancy- specific protein B (PSPB). PSPB is one of several pregnancy-associated glycoproteins produced by the binucleate trophoblastic cells of the ruminant placenta. This placenta-specific protein can be found in the maternal circulation, and its quantification can be used to detect pregnancy
(Butler et al. 1982; Sasser & Ruder 1987; Sasser et al. 1986; Wooding et al. 2005). The
BioPRYNwild, an enzyme-linked immunosorbent assay developed by BioTracking, Inc. that was initially validated for cattle, moose and elk (Cervus canadensis), was shown to efficiently quantify PSPB levels in mule deer sera (Huang et al. 1999; Huang et al. 2000), and may help determine the number of fetuses (Andelt et al. 2004). Its sensitivity and specificity are, respectively, 95% and 99.9%. Timing of sampling is important when trying to detect pregnancy using a PSPB radioimmunoassay. The BioPRYNwild may detect PSPB in sera from 40 days of gestation (BioTracking). False negative results may occur if sampling is performed too early after conception, and false positives can arise if sampling is done shortly following the loss of the fetus. In cattle, PSPB has a half-life of over 7 days (Kiracofe et al. 1993). Enough PSPB may remain in circulation to lead to a positive test result despite the lack of a placenta.
1.5.3 Pathogen Exposure
Here being reviewed are diseases of potential importance to mule deer health as well as to other wild and domestic ruminants’ health. These select pathogens include three hemorrhagic disease viruses (epizootic hemorrhagic disease virus, bluetongue virus, adenovirus hemorrhagic disease virus), three viruses of the bovine respiratory disease complex (BVDV, BRSV, PI3), and two protozoan parasites (Neospora caninum and Toxoplasma gondii).
14
1.5.3.1 Hemorrhagic Diseases
Epizootic hemorrhagic disease (EHD), bluetongue (BT) and adenovirus hemorrhagic disease (AHD) are grouped under the umbrella term “hemorrhagic diseases” (HD) as they are clinically indistinguishable and associated with similar post-mortem lesions. HD has been documented in a variety of domestic and wild ruminants, including white-tailed deer, mule deer, bighorn sheep and pronghorn antelope (Antilocapra americana) (Gibbs & Greiner 1989; Myers et al. 2015; Roug et al. 2012; Thorne 1982).
1.5.3.1.1 Epizootic Hemorrhagic Disease Virus (EHDV)
Epizootic hemorrhagic disease (EHD) is a seasonal arthropod-borne disease caused by an orbivirus. It relies on hematophagous midges of the genus Culicoides for transmission (Foster et al. 1977). The clinical signs associated with EHD are most severe in white-tailed deer (Nettles et al. 1992; Nettles & Stallknecht 1992). Mule deer and pronghorn antelope are also sensitive to the disease, but to a lesser extent (Barker et al. 1992). EHD in white-tailed deer may result in rapid death or death after 5-7 days of illness during which weakness, dehydration, anorexia and salivation may be observed (Fletch & Karstad 1971). Extensive internal hemorrhage lead to vascular shock, collapse and death. Prominent post-mortem lesions in affected animals are hemorrhage and edema of varying extent (Thorne 1982).
EHD is currently epidemic in Canada, and outbreaks have been documented in southern portions of British Columbia, Alberta and Saskatchewan (Ditchfield et al. 1964; Dulac et al.
1989; Dulac et al. 1992). The latest documented EHDV outbreak in Canada took place in southern Alberta in 2013, and resulted in the death of at least 50 deer (primarily white-tailed deer) and three pronghorns (Pybus et al. 2014). This outbreak was the first to be recorded since 1962
(Ditchfield et al. 1964). Outbreaks occur during late summer and early fall during hot, dry
15 weather when animals congregate near water sources (Couvillion et al. 1981). Culicoides sp. are not known to overwinter in Canada, and the temporal pattern reflects the vector’s presence in the western provinces only during the months of August and October. It has been proposed that seasonal wind patterns move the infected midges northwards from endemic northern American areas into Southwestern Canada, or that the virus’ range is expanding northwards as a result of climate change (Purse et al. 2005; Sellers & Maarouf 1991; Wittmann & Baylis 2000).
1.5.3.1.2 Bluetongue Virus (BTV)
Analogously to EHD, bluetongue (BT) is a seasonal disease caused by an orbivirus transmitted by the hematophagous midges of the genus Culicoides (Tabachnick 1996). The two viruses are closely related, but certainly distinct (Moore & Lee 1972). Of the 26 reported serotypes, five are endemic to the U.S. In Canada, those are immediately notifiable under the
Health of Animals Regulations while all exotic serotype to the U.S. are federally reportable under the Health of Animals Act (CFIA 2016c).
BT is usually considered a disease of sheep, but clinical disease following experimental or natural infection has also been reported in a variety of domestic and wild ruminants, including cattle, pronghorn, elk, yak, white-tailed deer and desert bighorn sheep (Ovis canadensis nelsoni)
(Drolet et al. 2013; Mauroy et al. 2008; Murray & Trainer 1970; Noon et al. 2002; Thorne et al.
1988). Clinical signs seen in sheep include anorexia, dehydration, depression, fever, nasal lesions and discharge, cyanotic tongue, excessive salivation, coronitis, facial edema, conjunctivitis, and secondary pneumonia, with a mortality rate of up to 50% (Parsonson 1990). White-tailed deer and pronghorn antelopes are the most commonly affected wildlife species, with severe hemorrhagic disease leading to sudden death (Hoff & Trainer 1978; Kocan et al. 1987; Stallknecht & Howerth
2004; Thorne et al. 1988). Clinical signs in white-tailed deer are similar to those seen in sheep
16 and may include anorexia, excessive salivation, nasal discharge, facial edema, conjunctival erythema, abnormal gait, loss of balance and respiratory distress (Drolet et al. 2013).
Hemorrhages and edema are the most common gross pathologic changes observed at necropsy in white-tailed deer, and are undistinguishable from lesions of epizootic hemorrhagic disease without molecular analysis (Thorne et al. 1988). The extent of the disease in mule deer is not as well documented, but exposure to the virus has been documented and recent bluetongue outbreak news reports indicate that mule deer can fatally contract the disease (Myers et al. 2015; Roug et al. 2012; The Spokesman Review 2015). Cattle and other wild ruminants rarely show overt clinical disease and are generally considered reservoir hosts for the viral agent (Hoff & Trainer
1972; Hoff & Trainer 1978).
Over the past 40 years, BT has sporadically been reported in Canada, and all cases, with one recent exception, have taken place in the Okanagan Valley, British Columbia (Clavijo et al.
2000; Dulac et al. 1989; Pasick et al. 2001; Thomas et al. 1982). The most recent occurrence of bluetongue, and first occurrence of serotype 13, was detected in three southwestern Ontario cattle
(OIE 2015). The detection of bluetongue cases in cattle herds has repercussions on livestock trade with bluetongue-free countries, and is a concern for the health of farmed sheep and wild ruminants.
1.5.3.1.3 Adenovirus Hemorrhagic Disease Virus (AHDV)
Adenovirus hemorrhagic disease (AHD) is caused by an adenovirus closely related to bovine adenovirus-3, and has been described in mule deer and white-tailed deer (Lapointe et al.
1999; Sorden et al. 2000; Woods et al. 1996). The infection presents itself as a systemic or as a localized disease. The systemic form is associated with clinical signs similar to those described in deer infected with BTV or EHDV (i.e. open mouth breathing, foaming at the mouth, diarrhea,
17 weakness), and with necropsy findings that include pulmonary edema, hemorrhagic enteropathy and panvasculitis (Woods et al. 2001). Death may occur within 3 to 5 days of exposure to the virus. Animals affected by the localized form show abscessation and deep ulceration of the mouth and the forestomachs, become emaciated and die of starvation (Woods et al. 1996). Transmission occurs through direct contact between animals, contact with infected bodily secretions and potentially via aerosols.
The disease was responsible for sporadic deaths in captive white-tailed deer in 1997-1998 in Iowa, for the death of over a thousand mule deer in northern California in 1999 and for the death of more than 400 mule deer in Oregon in 2002 (ODFW 2003). There are no reports of cases occurring in Canada.
1.5.3.2 Bovine Respiratory Disease Complex
The bovine respiratory disease complex (BRDC) is a major cause of bronchopneumonia in cattle and is of great economic importance to the farming industry worldwide. The disease is multifactorial, and results from intricate interactions between a variety of infectious agents, host factors and environmental stressors (Guzman & Taylor 2015). Both viral (bovine viral diarrhea virus-1 (BVDv), bovine herpesvirus-1, bovine respiratory syncytial virus (BRSV), parainfluenza-
3 virus (PI3), and bovine coronavirus, adenoviruses and enteroviruses) and bacterial agents
(Pasteurella multocida, Mannheimia haemolytica, Histophilus somni, Mycoplasma bovis) are involved in the BRDC. Stressors, such as abrupt weaning, castration, as well as adverse climatic or transport conditions, have been linked to alteration of the immune function and expression of the disease in cattle (Blecha et al. 1984; Griebel et al. 2014; Murata & Hirose 1991). It is generally accepted that primary impairment of the host’s immune system by viral agents allows for secondary bacterial infections of the lower respiratory system (Caswell 2014; Guzman &
18
Taylor 2015). In cattle, the BRDC is defined into three distinct clinical entities: 1) “Shipping fever” complex, 2) Enzootic pneumonia of calves, and 3) Atypical interstitial pneumonia (Lillie
1974). Such clinical categorization has not been established for deer as clinical significance of
BRDC in wildlife is not well documented, but serological evidence of exposure to various agents of the BRDC is abundant.
1.5.3.2.1 Bovine Respiratory Syncytial Virus (BRSV)
Respiratory syncytial viruses (RSV) (genus Pneumovirus, family Paramyxoviridae) have been isolated from cattle (BRSV), goats (caprine RSV), sheep (ovine RSV) and humans (human
RSV). These enveloped RNA viruses are cytopathic and have the ability to cause lower respiratory epithelial cells to fuse into syncytia. Like other viral agents of the BRDC, BRSV predisposes the host to secondary bacterial infections. BRSV is indigenous to cattle worldwide, and predominantly causes respiratory disease in calves. Disease is most severe upon first exposure to the virus, and progression of clinical signs is often rapid. Signs of disease in cattle may include fever, depression, decrease feed intake, tachypnea, ptyalism, cough, nasal and lacrimal discharge, reduction of milk production in dairy cattle (Bryson et al. 1983). Common post-mortem findings are emphysematous bulla, bronchiolitis and interstitial pneumonia. BRSV transmission mainly occurs via direct contact with respiratory secretions (Van der Poel et al.
1993), but can also occur indirectly via inhalation of aerosolized viral particles (Mars et al. 1999).
Morbidity is high during outbreaks in cattle herds, and mortality can reach 20%. Interspecies transmission of RSVs is not well characterized, but ovine RSV was experimentally shown to induce respiratory disease in calves and deer (Bryson et al. 1988). Serological evidence of exposure has been documented various wild ruminants, including mule deer, bighorn sheep,
19
European bison (Bison bonasus) and chamois (Rupicapra rupricapra rupicapra) (Gaffuri et al.
2006; Myers et al. 2015; Spraker et al. 1986; Urban-Chmiel et al. 2017).
1.5.3.2.2 Parainfluenza-3 Virus (PI3)
Parainfluenza-3 virus (PI3) is an enveloped single-stranded RNA virus within the
Respirovirus genus of the Paramyxovirus family. In cattle, infections are common and clinical signs are mild unless complicated by a secondary bacterial infection. Clinical signs may include tachypnea, increased breath sounds, cough, serous nasal and lacrimal discharge and fever. PI3 also predisposes animals to secondary bacterial pneumonia and to co-infections with other viruses such as the BRSV and BVDV (Van Campen & Early 2001). Mortality rate associated with PI3 infection is low, and most fatality are associated with concurrent bacterial bronchopneumonia.
Virus transmission occurs through inhalation of aerosolized viral particles or contact with nasal secretions (Mars et al. 1999). The epidemiology and clinical significance of PI3 infection in deer and is not well documented, but evidence of exposure is plentiful (Dubay et al. 2015;
Ingebrigtsen et al. 1986; Myers et al. 2015).
1.5.3.2.3 Pestiviruses
Bovine Viral Diarrhea Virus (BVDV) and Border Disease Virus (BDV) are enveloped single-stranded RNA viruses that belong to the Pestivirus genus of the family Flaviviridae.
Whereas BDV mainly affects sheep, BVDV primarily cause disease in young domestic cattle, but can also infect sheep, goats and wild ungulates. Transmission may occur vertically via direct or indirect contact within contaminated bodily fluids or horizontally. The clinical presentation ranges from unapparent infection to acute and severe enteric disease to the highly fatal mucosal disease complex characterized by profuse enteritis in association with mucosal lesions. BVDV
20 infection can also result in reproductive failure, persistently infected calves or congenital abnormalities, depending on the age of the fetus at the time of infection. Clinical presentation varies depending on the age of the host at the moment of infection, on its immune status and on the presence of stressors. BVDV contributes to the bovine respiratory disease complex by causing immunosuppression and predisposing animals to infection by other respiratory pathogens.
Although the significance of pestiviruses in deer is uncertain, antibodies have been reported in mule deer, white-tailed deer, moose and pronghorn antelopes (Barrett & Chalmers 1975; Karstad
1981; Roug et al. 2012; Thorsen & Henderson 1971).
1.5.3.3 Neospora caninum
Neospora caninum, the etiologic agent of neosporosis, is a protozoal parasite with a wide host range, a global distribution and a heteroxenous life cycle that involves a canid definitive host and a ruminant intermediate host (Gondim 2006). The parasite is transmitted either horizontally when the intermediate host ingests N. caninum oocysts shed in the feces of infected canids, or vertically via transplacental migration (Dubey et al. 2013; Dubey et al. 1996; Trees & Williams
2005). In North America, a sylvatic cycle exists between domestic and wild ruminants, and wild and domestic canids (Dubey et al. 2007; Gondim 2006).
Neosporosis causes reproductive failure (abortion, stillbirth and perinatal death) in cattle and other ruminants, and neuromuscular disease in domestic dogs. The role of wildlife in the epidemiology of N. caninum and the impact of the pathogen on wildlife health remain poorly understood, with most reports on N. caninum infection in wildlife consisting of seroprevalence studies in asymptomatic animals. Evidence of N. caninum exposure is well documented in wild cervids, and has been reported in mule deer, white-tailed deer, black-tailed deer (Odocoileus hemionus columbianus), pampas deer (Ozotoceros bezoarticus), roe deer (Capreolus capreolus),
21 red deer (Cervus elaphus), elk, moose and caribou (Anderson et al. 2007; Dubey et al. 1999;
Dubey et al. 2009; Dubey et al. 2008; Dubey & Thulliez 2005; Ferroglio & Rossi 2001;
Gutiérrez-Expósito et al. 2012; Lindsay et al. 2002; Malmsten et al. 2011; Myers et al. 2015;
Tavernier et al. 2015; Tiemann et al. 2005; Woods et al. 1994).
Only four reports of neosporosis in cervids have been published. Similar to cattle, congenitally infected cervids may be born dead, weak or with neurologic deficits or evidence of systemic disease, with the most common lesion consisting of a nonsuppurative encephalitis. The first reported case of neosporosis in a wildlife occurred in a female black-tailed fawn in 1993 in
California, USA. Histological lesions consisted of a diffuse, moderate interstitial pneumonia, hepatic architecture disruption and tubular necrosis, and N. caninum tachyzoites were identified by histopathology and immunohistochemistry in the kidneys and lungs (Woods et al. 1994). It was first proposed that deer are an intermediate host for N. caninum and that they may act as a wildlife reservoir (Woods et al. 1994). Fatal neosporosis was subsequently diagnosed in a fallow deer fawn with a 2-day history of hind limb paresis and a multifocal necrotizing and granulomatous meningoencephalomyelitis in Switzerland (Soldati et al. 2004); in an axis deer fawn with anal sphincter incontinence, weakness, ataxia, a non-suppurative encephalitis, a suppurative bronchopneumonia, a fibrin necrotic enteritis and degenerative changes in the liver in
Argentina (Basso et al. 2014); and in a Eld’s deer stillborn with a nonsuppurative encephalitis in
France (Dubey et al. 1996).
In Canada, N. caninum infection has been reported in beef and dairy cattle in most provinces (Haddad et al. 2005). Red foxes (Vulpes vulpes) and coyotes (Canis latrans) have been shown to be asymptomatic carriers in Prince Edward Island (Haddad et al. 2005; Wapenaar et al.
2007; Wapenaar et al. 2006). Exposure to N. caninum was reported in mountain and boreal caribou (Rangifer tarandus caribou) in British Columbia and in woodland caribou in the
22
Northwest Territories (Mathieu et al. 2015; Schwantje et al. 2014; Schwantje et al. 2016; Sifton
2005). There are currently no published reports of neoporosis in cervids in Canada.
1.5.3.4 Toxoplasma gondii
Toxoplasma gondii, the etiologic agent of toxoplasmosis, is a protozoal parasite widespread in most vertebrates worldwide. Its heteroxenous life cycle involves a felid definitive host and a variety of warm-blooded intermediate hosts, including wild ungulates (Dubey et al.
2014). The parasite is transmitted to the intermediate host either horizontally when the intermediate host ingests T. gondii oocysts shed in the feces of infected felids, or vertically via transplacental migration. Oocysts mature into motile tachyzoites after ingestion, which disseminate through the body via lymphatics and blood vessels, and later encyst in tissues. Cysts containing bradyzoites are most prevalent in neural and muscular tissues, but may also be found in visceral organs (lungs, liver, kidney) (Dubey & Odening 2001).
Post-natal infection may result in clinical disease or remain asymptomatic depending on the parasite strain and the susceptibility of the host. Clinical disease varies in severity from mild to fatal, and may present as acute systemic disease, retinochoroiditis, anterior uveitis or encephalitis. Congenital infection may result in fetal resorption, abortion, stillbirth, neonatal death and blindness. Rats, cattle, horses, pronghorn and Old World monkeys appear resistant to clinical toxoplamosis (Wolfe 2003). Immunosuppression secondary to stress or co-morbidity can result in recrudescence of bradyzoites and subsequent active disease. Serological evidence of T. gondii exposure has been documented in mule deer and other wild game species in North
America, and muscle tissue cysts were found in muscles of road-killed mule deer (Dubey et al.
2009; Dubey et al. 1982; Lindsay et al. 2005). Experimental intraruminal inoculation of T. gondii oocysts induced fatal systemic toxoplasmosis in mule deer, but no disease in elk (Chomel et al.
23
1994; Dubey & Odening 2001; Dubey et al. 1980; Dubey et al. 1982). The clinical importance of
T. gondii in naturally-infected deer is poorly characterized.
T. gondii is a major public health concern as humans can be infected via ingestion of any of the parasite life stages or via in utero transmission. Infection is often asymptomatic or results in mild-flu-like symptoms, but may cause severe disease in pregnant women and immunocompromised individuals. Cases of clinical toxoplasmosis have been documented in hunters who had consumed raw or under-cooked venison (Ross et al. 2001). Those individuals exhibited flu-like symptoms and visual loss due to a focal necrotizing retinitis.
1.6 Management of Mule Deer in the Kootenay Region, British Columbia
Over the past decade, a dramatic increase in suburban and urban mule deer population density has been observed in communities in the Kootenay. In parallel, the non-urban mule deer population has been declining, experiencing more than a 20% decline from 2011 to 2014 after a prolonged period of relative stability (Mule Deer Working Group 2016). While the urban deer population growth is likely linked to the abundance of forage and water sources, the restricted hunting due to urban bylaws and the scarcity of predators, the factors driving the non-urban mule deer population decline remain unknown. It has been proposed that forage of limited quality and quantity, variations in predator-prey dynamics, and the harsh winter conditions are involved
(Mule Deer Working Group 2016).
Wildlife management agencies in the Kootenay region have yet to perfect an urban deer management technique that is time and cost-effective and that accommodates the public opinion.
Most Kootenay communities, not without generating negative reactions from animal rights and conservation organizations, have instituted culls in an effort to control the urban mule deer population overabundance. While some residents are simply opposed to the act of killing animals,
24 others challenge the morality of destroying seemingly healthy urban animals while their non- urban counterparts are struggling to maintain their numbers. These stakeholder groups would prefer to see a population management approach that would simultaneously allow for the reduction of the overabundant urban deer population and for the supplementation of the declining non-urban deer population. For those reasons, local wildlife management agencies elected to perform live captures and translocations (Adams 2018). This translocation project has health implications for the translocated urban mule deer, the resident non-urban mule deer, as well as other sympatric wildlife species and the domestic livestock (cattle, sheep and goats) present at or near the release sites.
The British Columbia government had been reluctant to translocate mule deer. This hesitation to engage in deer translocation stemmed from translocation projects having a reputation of incurring high mortality during capture and transport, having a poor post-release survival, being expensive and labor intensive, and being accompanied with a risk of disease transmission.
Several authors have characterized capture and translocation as an “ineffective and costly” urban deer management method, with low survival of translocated deer (Creacy 2006; Jones & Witham
1990). The high mortality rate has been attributed to various decimating factors such as capture- related injuries, capture myopathy, nutritional status of the animal, unfamiliarity with the release site, human activities (hunting, vehicle traffic), predation, habitat quality, competition for resources with resident deer and encounters with new infectious agents (Beringer et al. 1996;
Jones & Witham 1990). Earlier studies have shown that as many as 85% of translocated deer may die within the first three months of translocation, and that 45-85% of deer may not survive longer than a year post-release (Diehl 1988; McCullough et al. 1997; O’Bryan & McCullough 1985).
These morose statistics, along with a poor cost-effectiveness compared to lethal techniques, have given translocation as an urban deer management technique a poor reputation amongst wildlife
25 managers. However, recent studies demonstrate more encouraging numbers, and have given the
British Columbia government the incentive for performing a translocation locally. Mule deer translocation efforts conducted in 2014 in Utah and New Mexico resulted in fewer capture and translocation related mortalities than previously reported, and in one-year post-release survival rates of approximately 84% and 61-68%, respectively. Both groups also identified that the mortality rate during the second-year post-release is similar to what native deer experienced
(Howard 2015; Mule Deer Working Group 2015). A study on the survival and movements of translocated white-tailed deer in South Texas showed an overall survival rate that compared to the adult resident deer’s (64-80%) (Foley et al. 2008).
Little is known about the driving factors of mule deer health in the Kootenay region and the diseases that afflict them. In order to create a basis for the health risk assessment to be performed as part of plan for the urban deer translocation project, health data must be gathered from the urban and non-urban populations. This health assessment will be integrated to the health risk assessment analysis and ultimately will (i) be used to characterize the disease risk, (ii) assist wildlife health managers in determining whether this risk is acceptable and (iii) provide insight as to how the health risks associated with the translocation project may be mitigated.
26
Chapter 2. Rational and Hypothesis
It has been postulated that urban dwelling mule deer have high disease burdens and pose a disease risk to non-urban mule deer and other sympatric species and a zoonotic risk to people if they are translocated back to the wild. Based on this postulate, control of urban deer populations has been limited to controlled culling. However, to assess the true disease transmission risk associated with the translocation of urban mule deer to non-urban areas, a thorough evaluation of the health of both the urban and non-urban deer populations is needed.
Little is known about the health status of mule deer populations in the Kootenay region, with the majority of information gleaned from game harvest reports and opportunistic post- mortem examinations of road-killed deer. Gaining a better understanding of the determinants of their health in the Kootenay region and identifying health hazards is necessary to quantify the health risk associated with wildlife translocation, and constituted the goal of the present study
(Leighton 2002).
Health surveillance was achieved through the determination of seroprevalence to various bovid and cervid disease antigens and pregnancy rates. These components were used to help determine if the health of urban deer differs from the health of non-urban deer, and characterize the risk for disease transmission when translocating urban deer to non-urban regions to reduce the urban deer overabundance. The results of this research project were ultimately used to help guide the management of free-ranging mule deer in the Kootenay region, and potentially benefit their conservation in the long-term.
27
The null hypotheses of this study were:
- Hypothesis 1: The health of the urban population is superior (lower pathogen exposure
rate, higher BCS, higher pregnancy rate) to the health of the non-urban population.
a. Abundance of forage and protection from predator harassment leads to lower
stress and better health, and consequently to a higher pregnancy rate and better
nutritional condition in the urban population.
b. Non-urban deer are at an increased risk of disease exposure as they are likely to
come in contact with sympatric wildlife and livestock through their migratory
behavior. Urban deer tend to stay within the city limits and to minimally
commingle with other animal populations.
- Hypothesis 2: Introducing urban deer to a non-urban environment through translocation
may negatively affect the health and survivorship of the urban deer.
The null hypothesis was tested by accomplishing the following specific objectives:
- Aim 1: To characterize the pathogen exposure, body condition and pregnancy rates of
urban and non-urban mule deer.
- Aim 2: To determine if the health of the urban deer population differs from the health of
the non-urban deer population and characterize the health risk associated with the
translocation project.
- Aim 3: To investigate the role of infectious diseases in the decline of the non-urban mule
deer population.
- Aim 4: To develop predictive models to inform management decisions related to urban
deer translocations
28
Chapter 3. Health Assessment of Urban and Non-Urban Free-Ranging Mule Deer
(Odocoileus hemionus) in the Kootenay Region, British Columbia, Canada
(Manuscript accepted in January 2018 for publication by PeerJ)
3.1 Introduction
Wildlife translocations carry inherent health and disease transmission risks, and are recognized as an important driver of wildlife disease emergence (Cunningham 1996; Daszak et al.
2001). By altering host-pathogen interactions, they can negatively affect the health of the translocated individuals and that of the resident animal populations as well as have negative conservation, economic and ecological repercussions (Carbyn & Watson 2001; Lankester 2001;
Woodford 2000; Woodford & Rossiter 1993). To ensure a translocation project is beneficial, it is crucial that the health risk be critically evaluated prior to performing a wildlife translocation, and that the translocation be planned in light of the health risk analysis findings (Cunningham 1996).
Over the past decade, a dramatic increase in urban mule deer (uMD) population density was observed in communities in the Kootenay region of southeastern British Columbia, Canada.
In parallel, the non-urban mule deer (nuMD) population declined from 2011 to 2014 (Mule Deer
Working Group 2016). While the uMD population growth is likely linked to the abundance of forage, the restricted hunting due to urban bylaws and the scarcity of predators, the factors driving the nuMD population decline remain unknown. It has been proposed that forage of limited quality and quantity, variations in predator-prey dynamics, and winter weather conditions are involved (Mule Deer Working Group 2016). To simultaneously reduce the overabundant
29 urban population and supplement the declining non-urban population, local wildlife management agencies performed an uMD translocation of 85 animals over a 2-year period. This translocation project has potential health implications for the translocated uMD, the resident nuMD, as well as other sympatric wildlife species and domestic livestock present at or near the release sites.
Although the health risk was acknowledged, the decision to conduct the health assessment concurrent with the translocation was made by the provincial government.
Little is currently known of the driving factors of mule deer health in the Kootenay region.
We designed a cross-sectional study to characterize mule deer health based on selected health indicators including pathogen exposure, nutritional status and pregnancy rates. Pathogens were chosen based on their capacity to negatively affect deer health (epizootic hemorrhagic disease virus (EHDV), bluetongue virus (BTV), adenovirus hemorrhagic disease virus (AHDV),
Neospora caninum) (Basso et al. 2014; Ditchfield et al. 1964; Dubey et al. 1996; Dulac et al.
1989; Dulac et al. 1992; Lapointe et al. 1999; ODFW 2003; Pybus et al. 2014; Soldati et al. 2004;
Sorden et al. 2000; WDFW 2017; Woods et al. 1994; Woods et al. 1996). Furthermore, given the potential of deer to be infected with bovine pathogens, we tested for exposure to viruses of the bovine respiratory disease complex (bovine respiratory syncytial virus (BRSV), bovine viral diarrhea virus (BVDV), bovine parainfluenza-3 virus (PI3)) to characterize the pathogen transmission risk posed by uMD translocation to resident livestock (McMartin et al. 1977;
Morton et al. 1990; Tessaro et al. 1999; Van Campen et al. 1997). We also included testing for exposure to Toxoplasma gondii given its potential public health repercussions (Dubey et al. 2004;
Sacks et al. 1983). A better understanding of the driving factors of mule deer health will help qualify the health risk associated with future translocation projects as well as help define the role of infectious diseases in the decline of the nuMD deer population. We hypothesized that the health of the uMD is superior to the health of the non-urban population (lower pathogen exposure
30 rate, higher BCS, higher pregnancy rate) and that introduction of uMD to the nuMD population through translocation may negatively affect the health and survivorship of nuMD.
3.2 Materials & Methods
3.2.1 Study Area and Animals
This study was conducted in the Kootenay region of southeastern British Columbia,
Canada. Deer were captured by the British Columbia Ministry of Forests, Lands, Natural
Resource Operations and Rural Development (BC FLNRO) for two independent projects and sampled opportunistically for the present study (BC Wildlife Permit #CB16-224332 and #CB17-
260952). Deer were captured in urban areas (uMD) for translocation and in non-urban areas
(nuMD) for radio collar fitting for ecological studies (Figure 1). Urban MD were defined as deer captured inside of urban boundaries while nuMD were defined as deer captured outside of urban boundaries. Urban areas included Cranbrook (49°30' to 49°32'N, 115°44' to 115°46'W), Elkford
(49°59' to 50°1'N, 114°55' to 114°55'W), Invermere (50°29’ to 50°30’N, 116°01’ to 116°03’W),
Kimberley (49°40' to 49°41'N, 115°58' to 115°59'W) and Marysville (49°37' to 49°38'N, 115°56' to 115°57'W). These areas were typically characterized by residential areas, parks and green spaces. Non-urban areas included Grasmere (49°2' to 50°2'N, 114°55' to 116°26'W), Invermere
(50°3' to 50°33'N, 115°49' to 116°8'W), Newgate (49°3' to 49°27'N, 115°12' to 115°31'W) and
West Kootenay (49°0' to 50°19'N, 115°58' to 118°3'W). The non-urban areas were predominantly composed of montane forests and shrubland. Based on GPS collar data, the uMD and nuMD populations are known to intermingle in some areas (Adams 2018).
Biometric data, including age, sex, body condition score (BCS), was recorded post- capture. Age was estimated based on sequential development of dentition and wear, and categorized as follow: Fawn – 0-12 months old (mo); young adult – 13 mo to 3 years old (yo);
31 adult – 4yo-7yo; aged adult– 8yo and older (Severinghaus 1949). Age classification data was excluded for 26 nuMD. Those animals were captured by independent capture teams and consistency in age estimates could not be confirmed. Sex was determined by observation of external genitalia. Body condition assessment was performed by a single observer (PMS) and was scored on a five-point scale (1 – emaciated, 2 – poor, 3 – fair, 4 – good, 5 – excellent) (Shallow et al. 2015). Body condition scoring data was excluded for 15uMD and 38 nuMD. Those animals were captured by independent capture teams and consistency in BCS assessment could not be confirmed. A subset of 14 animals was weighed for the purpose of a coincident anesthesia study
(Mathieu et al. 2017).
3.2.2 Capture Techniques
Captures happened sporadically between December 2014 and April 2017. For the purpose of this study, capture seasons were defined as: winter 2014-15, winter 2015-16, winter
2016-17. Animals were either darted and anesthetized from the ground, net gunned from a helicopter, or captured in Clover traps and anesthetized (Barrett et al. 1982; Boesch et al. 2011;
Clover 1956). Deer were anesthetized with one of two drug protocols: medetomidine-azaperone- alfaxalone (Mathieu et al. 2017) or butorphanol-azaperone-medetomidine (Miller et al. 2009).
3.2.3 Sample Collection and Testing
5-10 mL of blood was collected via jugular venipuncture and transferred to Vacutainer® plain blood collection tubes. Blood tubes were later centrifuged, and serum was extracted, aliquoted into cryovials and stored under -20°C refrigeration for a period of 2 to 3 years.
For Pregnancy-Specific Protein B quantification using a sandwich ELISA that uses rabbit anti- moose PSPB polyclonal antibody coated plastic wells to bind cervid serum sample pregnancy
32 specific protein B (BioPRYN wild), serum samples were submitted to the BioTracking laboratory
(BioTracking Inc., 1150 Alturas Drive, Suite 105, Moscow, ID 83843, USA) (Duquette et al.
2012). Optical density (OD) cut-off value was set at OD > 0.21. For EHDV and BTV antibody titers determination using a competitive enzyme-linked immunosorbent assay (cELISA), serum samples were sent to the National Centre for Foreign Animal Disease (Canadian Food Inspection
Agency, 1015 Arlington Street, Winnipeg, MB R3E 3M4, Canada) (CFIA 2016a; CFIA 2016b).
Samples with a percent inhibition of ≥ 40% were reported as positive. Positive samples were then serotyped using a serum neutralization test (CFIA 2017). Sera were screened for evidence of cell toxicity at a final dilution of 1/10 after the addition of virus. For AHDV antibody titers determination using a deer adenovirus antigen specific ELISA, serum samples were submitted to the Oregon Veterinary Diagnostic Laboratory (College of Veterinary Medicine, Oregon State
University, 105 Magruder Hall, Corvallis, OR 97331, USA). This assay detects serum IgG antibodies against a synthetic DAV spike antigen expressed in Escherichia coli. Methods and other assay details have not yet been published and are proprietary information of Drs Rob
Bildfell and Ling Jin (College of Veterinary Medicine, Oregon State University). For BVDV,
BRSV and PI3 antibody titers determination using virus neutralization tests, serum samples were submitted to the Abbotsford Animal Health Center Laboratory (Ministry of Agriculture
Abbotsford Agricultural Centre, 1767 Angus Campbell Road, Abbotsford, BC V3G 2M3,
Canada) (OIE 2017). Sera were screened for evidence of cell toxicity at a final dilution of 1/10 after the addition of virus. For N. caninum antibody titers determination using a cELISA, serum samples were sent to Prairie Diagnostic Services (Western College of Veterinary Medicine,
University of Saskatchewan, 52 Campus Drive, Saskatoon, SK S7N 5B4, Canada) (VMRD).
Samples with a percent inhibition ≥ 30% were reported as positive. For T. gondii antibody titers determination using a commercial indirect ELISA optimized for use in cervids, serum samples
33 were submitted to the Department of Veterinary Microbiology (Dr. Emily Jenkins) of the Western
College of Veterinary Medicine (IDvet 2013). Samples with a sample-to-positive percentage
(S/P%) of ≥ 50% were reported as positive.
3.2.4 Statistical Analysis
Statistical analysis was performed using JMP 11 (SAS Institute Inc., Cary, North
Carolina, USA). Descriptive statistics were generated for the capture and biometric data.
Normality of continuous outcome data was assessed using standard graphical methods (Dohoo et al. 2003). Weight, BCS, PSBP and pathogen prevalence means and their corresponding 95%
Confidence Intervals were calculated for both uMD and nuMD populations.
A two-sample t-test was used to assess differences in mean weight between population types. The unadjusted associations between BCS and sex, population types and year were assessed using chi-square analyses. BCS categories 1 and 2 were merged so that the expected frequency count for each cell of the table was at least 5 to meet distributional assumptions for a chi-square analysis. BCS data for deer captured in winter 2014-15 were excluded from the temporal analysis of BCS since no uMD were sampled during that time.
PBSB results from samples collected prior to 40 days of gestation (i.e. prior to December
28th based on a mean conception date of November 18th (Dyer et al. 1984)) were excluded from the statistical analysis. All animals for which an age class was not specified were non-urban and sexually mature (young adult, adult or aged). Animals for which an age class was not defined were excluded for the age class analysis. All animals immobilized in winter 2014-15 were sampled prior to December 28th and were thus excluded from the PSPB analysis.
To assess the effect of the independent variables including age, population type, BCS, sampling year and pregnancy status and exposure to other pathogens, on each of the pathogen
34 exposure outcomes were evaluated using multivariable logistic regression models to control for potential confounding. Logistic regression models were fitted for each pathogen exposure outcome independently using a forward selection procedure. Significant association were presented as adjusted odds ratios from the logistic regression models. Significance was set at p<0.05.
3.3 Results
3.3.1 Animals Sampled
A total of 200 mule deer were captured over three years. Data pertaining to age, sampling year, BCS, sex, capture technique and capture area distribution of the sampled animals is presented in Table 1.
3.3.2 Biometrics
No physical abnormalities other than variations in body condition were noted on physical examination. Weights were documented for 28 female deer (14 uMD, 14 nuMD). Urban MD weighed 71.1±6.6 kg (mean ± SD) and nuMD weighed 70.1±9.6 kg. Young adults, adults and aged adults weighed 69.7 ± 9.3 kg, 72.8 ± 6.5 kg and 67.3 ± 3.9 kg, respectively. Mean weights did not differ significantly between populations (uMD: 71.1 ± 6.6 kg (95% CI: 67.6-74.6 kg); nuMD: 70.2 ± 9.6 kg (95% CI: 65.2-75.2 kg)) (p=0.779). BCS did not significantly differ between females (3.4 ± 0.8 (95% CI: 3.3-3.6) and males (3.2 ± 0.8 (95% CI: 2.8-3.6)) (p=0.064), and data was thus pooled across sexes for all subsequent BCS analyses. BCS for both populations and all sampling years averaged 3.4 ± 0.8 (95% CI: 3.3-3.6), and significantly differed between populations (nuMD 3.6 ± 0.8 (95% CI: 3.4-3.8); uMD 3.2 ± 0.8 (95% CI: 3.0-3.4), p<0.001) and sampling years (winter 2014-15: 3.8 ± 0.6 (95% CI: 3.6-4.0); winter 2015-16: 3.4 ± 0.8 (95% CI:
35
3.2-3.6); winter 2016-17: 3.0 ± 0.9 (95% CI: 2.7-3.3), p<0.001). NuMD were found to be in overall better body condition than uMD. The difference was particularly evident in winter 2016-
17 (uMD: 2.7 ± 0.6 (90% CI: 2.4-2.9); nuMD: 3.8 ± 1 (90% CI: 3.1-4.2)). Moreover, uMD were in significantly poorer body condition in winter 2016-17 than in winter 2015-16 (3.5 ± 0.8 (90%
CI: 3.3-3.8) (p<0.001). Mean BCS did not differ between uMD and nuMD sampled in winter
2015-16 (p>0.304).
3.3.3 Pregnancy-Specific Protein B Quantification
Of the 108 animals of reproductive age tested, 98 were deemed pregnant for an overall prevalence of 90.7% (95% CI: 83.3-95.0) (Table 2). Animals exposed to N. caninum were more likely to not be pregnant than animals not exposed to N. caninum (OR: 14.0, 95% CI: 2.7-105.6; p=0.003). Pregnancy status was not significantly associated with the age class, sampling year, population type, BCS exposure to all other pathogens tested.
3.3.4 Pathogen Exposure
3.3.4.1 Epizootic Hemorrhagic Disease Virus
176 deer were tested for exposure to EHDV. All results were negative (Table 3).
3.3.4.2 Bluetongue Disease Virus
Of the 176 deer tested for exposure to BTV, one was positive, for an overall prevalence of 0.6% (95% CI: 0.1-3.1) (Table 3). This sample had neutralizing antibodies to BTV-10 and
BTV-17, and was negative for BTV-2, -8, -11 and -13.
36
3.3.4.3 Adenovirus Hemorrhagic Disease Virus
Of the 177 deer tested for exposure to AHDV, 68 were positive for an overall prevalence of 38.4% (95% CI: 31.6-45.8) (Table 3). Exposure to AHDV was significantly associated with age, with older animals more likely to be exposed than younger animals (p=0.014). Compared to fawns, young adults, adults and aged adults were 2.5 (95% CI: 0.8-9.5), 5.1 (95% CI: 1.6-19.7) and 8.9 (95% CI: 1.6-63.0) times more likely to be exposed to AHDV, respectively. Exposure to
AHDv was significantly associated with the sampling year (p=0.032), with animals sampled in winter 2016-17 more likely to be exposed than animals sampled in winter 2014-15 (OR: 3.9, 95%
CI: 1.3-13.4; p=0.013). Animals exposed to PI3 were more likely to be exposed to ADHV (OR:
2.4, 95% CI: 1.2-4.8; p=0.009). Exposure to AHDv was not significantly associated with the population type, BCS and exposure to all other pathogens tested.
3.3.4.4 Bovine Respiratory Syncytial Virus
Of the 178 deer tested for exposure to BRSV, 15 were positive for an overall prevalence of 8.4% (95% CI: 5.2-13.4) (Table 3). Exposure to BRSV was not significantly associated with the age, sampling year, population type, BCS and exposure to all other pathogens tested.
3.3.4.5 Bovine Viral Diarrhea Virus
Of the 178 deer tested for exposure to BVDV, 2 were positive for an overall prevalence of 1.1% (95% CI: 0.3-4.0) (Table 3).
37
3.3.4.6 Bovine Parainfluenza-3 Virus
Of the 178 deer tested for exposure to PI3, 48 were positive for an overall prevalence of
27.0% (95% CI: 21.0-33.9) (Table 3). Exposure to PI3 was not significantly associated with the age class, sampling year, population type, BCS and exposure to all other pathogens tested.
3.3.4.7 Neospora caninum
Of the 158 deer tested for exposure to N. caninum, 35 were positive for an overall prevalence of 22.1% (95% CI: 16.4-29.2) (Table 3). Exposure to N. caninum was significantly associated with the sampling year, with animals sampled in winter 2014-15 more likely to be exposed than animals sampled in winter 2015-16 (OR: 4.6, 95% CI: 1.5-14.1; p=0.12). Exposure to N. caninum was not significantly associated with the age class, population type and BCS.
3.3.4.8 Toxoplasma gondii
Of the 158 deer tested for exposure to T. gondii, 13 were positive for an overall prevalence of 8.2% (95% CI: 4.9-13.6) (Table 3). uMD were more likely to be exposed to T. gondii than nuMD (OR: 3.5, 95% CI: 1.0-16.0; p=0.047).
3.4 Discussion
Knowledge of the health status of a wild animal population and of its determinants and comparison of health data over time and with other wildlife populations can assist wildlife managers in planning conservation management actions such as translocations. This study compared nutritional status, pregnancy rates and pathogen exposure of uMD and nuMD in the
Kootenay region over three years. Based on the present findings, it appears the health of the two mule deer populations does not significantly differ, suggesting that uMD translocations do not
38 pose a severe risk of transmission of the pathogens specifically studied here between mule deer populations. Comingling of uMD and nuMD in some areas likely contributes to exchange of pathogens and may explain why both deer populations have similar exposure levels to most pathogens. Creation of a robust predictive disease model is warranted to further characterize the health risk associated with uMD translocations and to inform wildlife management decisions.
Furthermore, the lack of significant differences the health of the two mule deer populations suggest that the pathogens screened for in this study do not play a major role in the decline of the nuMD population. The reasons for the recent decline of the nuMD population thus remain unclear. Further investigation is warranted to establish whether forage of limited quality and quantity, variations in predator-prey dynamics, and the harsh winter conditions are involved in the nuMD population decline (Mule Deer Working Group 2016).
3.4.1 Biometrics
Overall, body condition varied from emaciated to excellent, with an average score of 3.4 on a five-point scale. Scores significantly differed between population types and sampling year.
Non-urban MD were in overall better body condition than uMD. This difference in body condition means is attributable to both timing of captures and variations in winter conditions.
Most nuMD were captured in early winter (December and January) whereas all uMD were captured in late winter (February and March) when fat stores are typically depleted. Winter severity may have also contributed to lower BCS score of uMD in 2016-17 as southeastern
British Columbia experienced the harshest winter in 20 years, with below average temperatures and above average snowfall from December to February (BC MFLNRO 2017).
Mule deer are generally a highly prolific species with reports of pregnancy rates ranging from
78.8% to 100% (Andelt et al. 2004; Chattin 1948; Dyer et al. 1984; Robinette & Gashwiler 1950).
39
The present study revealed an overall pregnancy rate of 90.7% and statistical analysis did not detect differences in pregnancy rates between populations or sampling years. Given this lack of difference in pregnancy rates between populations, it is unlikely that fertility is a viable explanation for differences in demographics between uMD and nuMD. Investigation of recruitment failure (e.g. poor fawn survival) as a possible explanation for the nuMD population decline is recommended.
3.4.2 Pathogen Exposure
Epizootic hemorrhagic disease (EHD) and bluetongue are seasonal arthropod-borne diseases caused by two orbiviruses. In Canada, sporadic EHD outbreaks have been documented in southern portions of British Columbia, Alberta, Saskatchewan and Ontario (Ditchfield et al.
1964; Dulac et al. 1989; Dulac et al. 1992; Pybus et al. 2014). EHD and BT outbreaks occur during late summer and early fall during hot and dry weather when animals congregate near water sources and come in contact with their vectors, the hematophagous midges of the genus
Culicoides (Couvillion et al. 1981; Foster et al. 1977). Myers et al. (2015) and Dubay et al. (2004) observed an elevational difference in EHDV and BTV exposure in mule deer in Washington and
Arizona with a greater antibody prevalence at lower elevations than at higher elevations. They attributed this difference to a greater presence of Culicoides sp. in dry shrub-steppe habitats with standing bodies of water compared to alpine, subalpine and forested areas. Also, Culicoides sp. are not known to overwinter in Canada, and the temporal pattern of outbreaks reflects the vector’s presence in the western provinces only during the months of August to October. It has been proposed that seasonal wind patterns move the infected midges northwards from endemic northern American areas into Southwestern Canada, or that the midges’ range is expanding northwards as a result of climate change (Purse et al. 2005; Sellers & Maarouf 1991; Wittmann &
40
Baylis 2000). These observations are in agreement with the outbreaks in British Columbia that have been limited to the low elevation dry grasslands of the Okanagan Valley within 50 kilometers of the American border. Given the viruses’ predilection for low elevation dry shrub- steppe habitats with standing bodies of water, it is unlikely the high elevation forested Kootenay region is a suitable habitat for EHDV and BTV, which is supported by the minimal exposure prevalence to either viruses in our population sample. We propose this remains true regardless of the season.
Adenovirus hemorrhagic disease (AHD), a highly fatal disease of mule deer and white- tailed deer, has been responsible for several deer mortality outbreaks in Iowa, California, Oregon and Washington (Lapointe et al. 1999; ODFW 2003; Sorden et al. 2000; WDFW 2017; Woods et al. 1996). In Canada, AHDV was associated with the death of at least eight mule deer fawns in
Waterton Lakes National Park in Alberta (T. Shury, pers. comm.). In the subsequent years, evidence of AHDV exposure was reported in elk (Cervus canadensis) in the East Kootenay region of British Columbia (H. Schwantje, pers. comm.) Older animals were more likely to be exposed than younger animals. This is in accordance with observations that AHD has a high mortality rate in fawns and that adults exposed to AHDV can recover (Boyce et al. 2000). The significance of the association between positive PI3 and AHDV seroprevalence is unknown and may have been the result of a statistical fluke. The high seroprevalence may be real or be a consequence of the unknown specificity of the assay. Further investigation is required to assess the specificity and sensitivity of the ELISA in mule deer, to establish the true prevalence of
AHDV and to better understand its potential impact on the health of the translocated and resident animals.
Although evidence of exposure is commonly reported in wild cervids, the viruses of the bovine respiratory disease complex (BRDC) are not considered significant pathogens of cervids
41
(Dubay et al. 2015; Ingebrigtsen et al. 1986; Karstad 1981; Myers et al. 2015; Roug et al. 2012;
Thorsen & Henderson 1971; Van Campen & Early 2001). Experimental infection with BVDV only resulted in mild diarrhea, coronitis and laminitis in reindeer, and did not clinically affect red deer, elk, mule deer or white-tailed deer (McMartin et al. 1977; Morton et al. 1990; Tessaro et al.
1999; Van Campen et al. 1997). Further, ovine respiratory syncytial virus was experimentally shown to only induce asymptomatic pneumonic lesions in fawns (Bryson 1988). However, we screened for these pathogens given the high likelihood of wildlife and livestock interactions in our study regions and the economic losses that could result from livestock contracting diseases from deer. No statistically significant differences in exposure to respiratory viruses between uMD and nuMD were detected. While this data is preliminary, it suggests that uMD translocation do not pose a significant risk of transmission of BRDC viruses to the resident livestock. However, screening of resident livestock for pathogen exposure and assessment of the true interactions between livestock and wildlife is warranted to better characterize the risk of pathogen transmission between mule deer and livestock.
The role of wildlife in the epidemiology of N. caninum and the impact of the pathogen on wildlife health are poorly understood, with most reports on N. caninum infection in wildlife consisting of seroprevalence studies in asymptomatic animals (Anderson et al. 2007; Dubey et al.
1999; Dubey et al. 2009; Dubey et al. 2008; Dubey & Thulliez 2005; Ferroglio & Rossi 2001;
Lindsay et al. 2002; Malmsten et al. 2011; Myers et al. 2015; Tavernier et al. 2015; Tiemann et al. 2005; Woods et al. 1994). In Canada, evidence of N. caninum exposure was reported in mountain and boreal caribou (Rangifer tarandus caribou) in British Columbia, in woodland caribou in the Northwest Territories and in elk in Alberta (Mathieu et al. 2015; Pruvot et al. 2014;
Schwantje et al. 2014; Schwantje et al. 2016; Sifton 2005). The present study identified a relation between seropositivity to N. caninum and reduction in pregnancy rates, but no definitive link is
42 suggested. Neosporosis in cervids has been associated with fawns born dead, weak or with severe neurologic deficits (Basso et al. 2014; Dubey et al. 1996; Soldati et al. 2004; Woods et al. 1994).
With a prevalence of 22.1% and suspicion of detrimental reproductive effects, N. caninum likely is a determinant of mule deer health in the Kootenay region. However, given that exposure to N. caninum does not differ between uMD and nuMD, transmission of N. caninum through uMD translocations represents a low health risk for either deer population.
The protozoan parasite Toxoplasma gondii can infect most warm-blooded animals and relies on a felid intermediate host to complete its life cycle. While serological evidence of T. gondii exposure is abundant, the clinical significance of T. gondii in free-ranging mule deer remains poorly characterized (Lindsay et al. 2005). The overall seroprevalence for T. gondii in our population sample was low, but uMD were shown to be more likely to be exposed to T. gondii than nuMD. This observation is consistent with other studies that investigated urbanization as a risk factor for T. gondii exposure in wildlife (Ballash et al. 2015; Conrad et al. 2005; Lehrer et al. 2010). The increased seroprevalence in deer across the urban gradient was proposed to result from the greater risk of exposure to oocysts shed by domestic cats given their higher population density within city limits (Ballash et al. 2015). Bobcats and cougars are likely contributors to the environmental contamination in the Kootenay region in both urban and non- urban environments (Aramini et al. 1998; Bevins et al. 2012). While of apparent limited concern for the health of mule deer, T. gondii poses a significant public health risk to individuals that may consume viable tissue cysts in raw or undercooked mule deer venison (Dubey et al. 2004; Sacks et al. 1983), and uMD translocation may result in an increased prevalence of T. gondii in non- urban areas.
43
3.4.3 Study Limitations
Interpretation of serology results must be done with caution. Circulating antibodies may result from an active or latent infection or from a past exposure to a pathogen. As such, detection of antibodies does not characterize the nature of the infection, but merely indicates exposure to a pathogen. Since only survivors are available for testing, diseases with a high fatality rate are likely under-represented in this study. Serological assays vary in sensitivity and specificity and most are only validated in domestic species. Of all the assays used in this study, only the AHDV,
T. gondii and PSPB assays were validated in wildlife. PSPB may remain in circulation for several days to weeks after fetal death, resulting in false positives if sampling is done shortly following the loss of the fetus. In cattle, PSPB has a half-life of over 7 days (Kiracofe et al. 1993). Given these limitations, interpretation of serologic results should be nuanced, with more importance given to trends than to exact values.
While assessment of body condition was performed by a single observer to maximize consistency of observations, this method is subject to some variability. As such, the data reported here only partially reflects the nutritional status of the Kootenay mule deer population.
Given limited funds and access to carcasses, this study did not investigate the effect of parasitism on mule deer health despite early evidence of Echinococcus sp. and lungworm infections in nuMD (Mathieu et al. 2017). The role of parasitism on mule deer health warrants further investigations.
The limited sample size obtained for some of the variables makes differences more difficult to detect statistically. However, strong associations, such as the ones reported for N. caninum exposure and reduction in pregnancy rates and for T. gondii exposure and urban area, should not be masked by a small sample size. For example, even though BCS did not vary
44 significantly between males and females, this warrants further investigation as this difference may be due to the small number of males sampled and to the restricted age distribution of males.
3.5 Conclusion
BCS varied by sampling year and population type, pregnancy rates were overall high and antibodies were detected against AHDV, BTV, BRSV, BVDV, PI3, N. caninum and T. gondii.
No antibodies against EHDV were detected. Exposure to AHDV, PI3 and N. caninum was common whereas exposure to BTV, BRSV, BVDV and T. gondii was rare. Exposure to T. gondii was more prevalent in the uMD. Given the many limitations of serology, inclusion of additional health indicators, such as blood biochemistry and hematology, trace mineral levels, parasite burden, post-mortem examination findings of apparently healthy and diseased animals is advisable to better characterize mule deer health. The results of this study should be considered as part of a formal risk assessment for future uMD translocations in southeastern British Columbia.
45
Chapter 4. Predictive Models for Pathogen Transmission Risk Assessment Associated
with Urban Deer Translocations in Southeastern British Columbia, Canada
(Manuscript submitted in February 2018 for consideration to publish with PeerJ)
4.1 Introduction
As in many populated areas in North America, urban deer overabundance is a major wildlife management concern in the Kootenay region of southeastern British Columbia, Canada.
The provincial wildlife management agency performed a translocation to control the overabundant urban mule deer (Odocoileus hemionus; uMD) population. Several local communities are now considering using uMD translocations as a long-term wildlife management method. Given the potential health risks associated with intentional movement of animals, it is crucial to understand the risk of pathogen introduction and outbreak under a range of potential conditions, such as altered prevalence of specific diseases, prior to moving any animals. An assessment of the health risks to the translocated individuals and the recipient populations should be performed prior to the translocation in order to limit the potential for undesirable disease outcomes (IUCN/SSC 2013).
A health assessment of the urban and non-urban free ranging mule deer population of southeastern British Columbia conducted between 2014-2017 showed uMD were more likely to be exposed to Toxoplasma gondii than nuMD, but estimated the risk of disease transmission associated with urban deer translocation to be overall low (Mathieu et al. 2018). However, a limitation of the study preventing conclusive management recommendations was its moderate 46 sample size (n=200). Despite the efforts of dedicated agencies and appropriate funding, this limitation is a common denominator to many wildlife field studies due to the difficulties associated with free ranging animal captures resulting in low capture per unit effort ratios. Since mule deer management strategies are informed by the findings of the most applicable study, it is essential the data supporting those recommendations be as representative as possible of the actual population health status under a range of conditions. For most situations, extrapolation is used to bridge the gaps in the available data (Gerry & Michael 2002).
One strategy to extrapolate available data in a rigorous way is to create a predictive model.
Predictive disease modelling can inform practical management strategies, even when working with an incomplete or limited dataset. Data can be manipulated to produce variable estimates to fill data gaps and variables can be tested against a range of pathogen prevalence conditions to predict the influence they may have on a range of outputs (Heesterbeek et al. 2015). By using a combination of stochastic and predictive disease models, we established recommendations for translocation guidelines based on the known pathogen prevalence of mule deer in the Kootenay
Region and the predicted risk of pathogen transmission between its urban and non-urban mule deer populations. Pathogens of interest included epizootic hemorrhagic disease virus (EHDV), adenovirus hemorrhagic disease virus (AHDV), bluetongue virus (BTV), bovine respiratory syncytial virus (BRSV), bovine viral diarrhea virus (BVDV), bovine parainfluenza-3 virus (PI3),
Neospora caninum and Toxoplasma gondii. We used these outputs to advise on required targeted continuous disease monitoring efforts and priorities.
4.2 Materials and Methods
The disease models were built using the “Sample size: X-Sectional, Cohort, &
Randomized Clinical Trials” module of OpenEpi, an open source software for epidemiologic
47 calculations and data analysis (Dean et al. 2013). The baseline data used was sourced from
Mathieu et al. (2018). These models determined the predicted minimum sample size (PMSS) required to detect statistically significant pathogen exposure prevalence differences between mule deer populations to create recommendations for guidelines for assessment of the risk of pathogen transmission associated with uMD translocations. As such, PMSS served as an index for the magnitude of the risk of altering pathogen prevalence in either the uMD or the nuMD population.
A low PMSS indicated the pathogen exposure prevalence significantly differs between MD populations and an increased pathogen transmission risk. A high PMSS indicated the pathogen exposure prevalence does not significantly differ between MD populations and a low pathogen transmission risk.
All models used a 95% two-sided confidence level and a power of 0.8 to detect statistically significant pathogen exposure prevalence differences. The ratio of unexposed to exposed was calculated by dividing the unexposed group sample size by the exposed group sample size. The pathogen exposure prevalence for populations with a null baseline prevalence was set at 1% to create a valid model.
4.2.1 Model 1: Stochastic Risk of Pathogen Transmission Associated with uMD
Translocations (Table 4)
Under the assumption of an inverse correlation between the total PMSS and the magnitude of the risk of pathogen transmission, the estimates that were generated were used to categorize the pathogen transmission risk. Based on an exploratory analysis and a conservative approximation of the maximum number of deer that could be translocated, the threshold was set at a total PMSS of 1000 individuals. The following rules were established to interpret the random risk of pathogen transmission associated with uMD translocation based on the pathogen
48 transmission risk threshold: (i) a total PMSS of <1000 individuals was associated with an increased risk of pathogen transmission, and (ii) a total PMSS size of >1000 individuals was associated with an lower risk of pathogen transmission.
Urban MD were defined as the exposed group and nuMD were defined as the unexposed group. Percent of unexposed with outcome and percent of exposed with outcome were set according to the baseline prevalence data of the pathogen of interest. The pathogen exposure prevalence for populations with a null baseline prevalence was set at 1% to create a valid model.
The remaining parameters were filled automatically. The model could be fitted for AHDV, BTV,
BRSV, BVDV, PI3, Neospora caninum and Toxoplasma gondii, but not for EHDV. EHDV was excluded as the model generated was invalid given the equal pathogen exposure prevalence in both populations.
4.2.2 Model 2: Dynamic Risk of Pathogen Transmission Associated with uMD
Translocations (Table 5)
The dynamic risk of pathogen transmission associated with uMD translocation was predicted based on the PMSS required to detect statistically significant pathogen exposure prevalence differences between mule deer populations under a range of pathogen prevalence conditions (PMSSx, where x is the predicted pathogen exposure prevalence in the exposed population). The predicted pathogen exposure prevalence in the exposed population was set at
5%, 10%, 15%, 25%, 50% and 75% while the pathogen exposure prevalence was maintained at its baseline level in the unexposed population. Models to calculate the number of exposed animals needed to cause a significant risk of exposure from the altered exposure prevalence were created for EHDV, AHDV, BTV, BRSV, BVDV, PI3, N. caninum and T. gondii. The definition of exposed and unexposed group switched depending on the population of interest. When
49 simulating a range of pathogen prevalence conditions in the uMD population, the uMD and nuMD were defined as the exposed and unexposed groups, respectively. When simulating a range of pathogen prevalence conditions in the nuMD population, the nuMD and uMD were defined as the exposed and unexposed groups, respectively. The remaining parameters were filled automatically. The produced predicted risk of pathogen transmission was deemed substantial at the prevalence level for which the PMSS was reduced by ≥50% compared to the PMSS5%. This pathogen prevalence level is referred to as the PMSS50.
The risk of pathogen transmission was established to be of concern under current conditions if the predicted prevalence level associated to the PMSS50 was less than or equal to the observed prevalence in the exposed group. Conversely, the risk of pathogen transmission was established to not be of concern under the current conditions if the predicted prevalence level associated to the PMSS50 was greater to the observed prevalence in the exposed group. The risk of pathogen transmission was established to be of potential future concern if the predicted prevalence level associated to the PMSS50 was less than or equal to the current highest pathogen prevalence data published for cervids in a close geographic area or similar habitat (Dubey et al.
1999; Dubey et al. 2008; Myers et al. 2015). Conversely, the risk of pathogen transmission was established to not be of potential future concern if the predicted prevalence level associated to the
PMSS50 was greater than the current highest pathogen prevalence data published for cervids in a close geographic area or similar habitat.
4.2.3 Model 3: Continuous Pathogen Monitoring Guidelines (Table 6)
Continuous pathogen monitoring guidelines were generated from the predictive risk of pathogen transmission risk analysis results to make quantitative recommendations. Pathogens shown to be of current concern for mule deer health and pathogens associated with high
50 morbidity or mortality rates in cervids were prioritized over pathogens shown to be of potential future concern (Ditchfield et al. 1964; Dulac et al. 1989; Dulac et al. 1992; Lapointe et al. 1999;
ODFW 2003; Pybus et al. 2014; Sorden et al. 2000; WDFW 2017; Woods et al. 1996). Given the difficulties associated with wildlife disease monitoring, liberal targeted sampling recommendations were generated based on the highest pathogen prevalence data published for cervids in a close geographic area or similar habitat. The targeted sampling recommendations represent the PMSS required to detect realistic pathogen outbreaks in either mule deer population at any point in time.
The predicted pathogen exposure prevalence in the exposed population were set at 26% for
EHDV, 27% for BTV, 71% for BRSV, 57% for BVDV, 61% for PI3, 40.5% for N. caninum and
32.6% for T. gondii (Dubey et al. 1999; Dubey et al. 2008; Myers et al. 2015). No serosurvey data could be found in the current literature for AHDV in cervids so no PMSS could be calculated for this pathogen. The definition of exposed and unexposed group varied depending on the population of interest. When simulating a pathogen outbreak in the uMD population, the uMD and nuMD were defined as the exposed and unexposed groups, respectively. When simulating a pathogen outbreak in the nuMD population, the nuMD and uMD were defined as the exposed and unexposed groups, respectively. The remaining parameters were filled automatically.
4.3 Results
4.3.1 Model 1: Stochastic Risk of Pathogen Transmission Associated with uMD
Translocations (Table 4)
The total PMSS required to detect a statistically significant difference in exposure to
AHDV, BRSV and T. gondii between deer populations was <1000 individuals, thus violating the defined pathogen transmission risk threshold. Exposure to AHDV and T. gondii was higher in the
51 uMD population and that uMD translocations carried an increased risk of AHDV and T. gondii transmission for the nuMD population. Exposure to BRSV was higher in the nuMD population and that uMD translocations carried an increased risk of BRSV transmission for the uMD population. The total PMSS required to detect a statistically significant difference in exposure to
BTV, BVDV, PI3 and N. caninum between deer populations was >1000 individuals, thus below the defined pathogen transmission risk threshold. The pathogen associated with the smallest total
PMSS and the greatest risk of transmission was T. gondii. The pathogen associated with the largest total PMSS and the lowest risk of transmission was BTV.
4.3.2 Model 2: Dynamic Risk of Pathogen Transmission Associated with uMD
Translocations (Table 5)
The risk of BRSV transmission was of current concern for the uMD population and the risk of T. gondii transmission was of current concern for the nuMD populations. The risk of
EHDV, AHDV, BTV, BVDV, PI3 and N. caninum transmission was not of current concern for either populations. The risk of EHDV BTV, BRSV, BVDV and T. gondii transmission was of potential future concern for either populations, while the risk of PI3 and N. caninum transmission was not of potential future concern for either populations.
4.3.3 Model 3: Continuous Pathogen Monitoring Guidelines (Table 6)
The PMSS required to detect realistic pathogen outbreaks in either mule deer population at any point in time are presented in Table 6. Continuous pathogen monitoring guidelines could not be established for AHDV given the lack of comparative data. Pathogens associated with high morbidity or mortality rates in cervids included EHDV and BTV, and pathogen of current transmission concern included BRSV and T. gondii. Detection of EHDV and BTV outbreaks in
52 either MD population would require sampling of at least 61 individuals. Sampling of that many animals would also allow BRSV and BVDV outbreak detection. Detection of BRSV and T. gondii outbreaks in either MD population would require sampling of at least 133 individuals.
Sampling of that many animals would also allow for EHDV, BTV, BVDV, PI3 outbreak detection. Detection of outbreaks of all selected pathogens in either populations would require sampling of a total minimum of 265 individuals.
4.4 Discussion
This study used disease modeling tools to predict the stochastic and dynamic risks of pathogen transmission between mule deer populations associated with uMD translocations in southeastern British Columbia. By formulating guidelines for risk acceptability determination to interpret the model outputs, we proposed guidelines for uMD translocations and for continuous disease monitoring. Factors to consider when assessing the acceptability of risk of pathogen transmission in the context of wildlife translocation include the level of pathogenicity for each pathogen of concern and their potential impact on individual and population health, the degree of host susceptibility to each pathogen of concern, and the susceptible translocated or resident species involved in the translocation. For this study, we defined the acceptability of individual pathogen transmission risk based on the following assumptions: (i) For pathogens associated with minimal long-term repercussions on deer health at a population level (i.e. BRSV, BVDV, PI3, T. gondii), the acceptable risk of pathogen transmission is high and a liberal number of animals may be translocated; (ii) for pathogens associated with moderate long-term repercussions on deer health at a population level (i.e. EHDV, AHDV, BTV, N. caninum), the acceptable risk of pathogen transmission is moderate and conservative number of animals may be translocated; and
(iii) for pathogens associated with severe long-term repercussions on deer health at a population
53 level (i.e. bovine tuberculosis, chronic wasting disease, Brucella sp, Johne’s disease), the acceptable risk of pathogen transmission is null to minimal, and translocation of any animals is unadvisable.
4.4.1 Urban Mule Deer Translocation Guidelines
Model 1 established the risk of pathogen transmission associated with uMD translocations under the current conditions. The total PMSS required to detect a statistically significant difference in exposure AHDV, BRSV and T. gondii was above the defined pathogen transmission risk threshold, suggesting exposure to these pathogens differs between deer populations and that uMD translocations would result in an increased risk of AHDV, BRSV and
T. gondii transmission.
The risk of AHDV transmission was classified as moderately acceptable given AHDV is associated with high mortality rates in mule deer and white-tailed deer, but has not yet been associated with long-term negative repercussions at the population level (Lapointe et al. 1999;
ODFW 2003; Sorden et al. 2000; WDFW 2017; Woods et al. 1996). We propose a conservative approach to uMD translocations to mitigate the risk of AHDV transmission between MD populations and suggest no more than 344 uMD should be translocated under the current conditions.
The risk of BRSV transmission was classified as highly acceptable given BRSV, along with other viruses of the bovine respiratory disease complex such as PI3 and BVDV, are not associated with significant morbidity or mortality in cervids (McMartin et al. 1977; Morton et al.
1990; Tessaro et al. 1999; Van Campen et al. 1997). We propose the number of uMD to be translocated under the current conditions should not be limited by the risk of BRSV transmission between MD populations. Interestingly, while exposure to BRSV was not deemed to differ
54 between mule deer population in the health assessment study, this model suggests nuMD are more exposed to BRSV than uMD. This supports the hypothesis cohabitation of nuMD and cattle increases exposure to some of the viruses of the bovine respiratory disease complex (BRDC)
(Cantu et al. 2008; Roug et al. 2012; Wolf et al. 2008).
The risk of T. gondii was classified as highly acceptable given T. gondii is not associated with significant morbidity or mortality in cervids (Lindsay et al. 2005). However, transmission of
T. gondii to the non-urban habitat may result in an increased exposure risk for wild felids and an increased public health risk for individuals that consume deer venison (Aramini et al. 1998;
Bevins et al. 2012; Dubey et al. 2004; Sacks et al. 1983). We propose the number of uMD to be translocated under the current conditions should not be limited by the risk of T. gondii transmission between MD populations. However, the public should be made aware that the zoonotic risk associated T. gondii and the risk of being infected with this zoonotic parasite after eating deer venison hunted in non-urban environments may increase following uMD transocations. Interestingly, the calculated minimum sample size for T. gondii (n=336) is larger than the sample size that allowed for detection of a statistically difference in exposure to T. gondii in the health assessment study (n=158). This suggest the sample selected in the health assessment study was not representative of the population.
The total PMSS required to detect a statistically significant difference in exposure to BTV,
BVDV, PI3 and N. caninum was below the defined pathogen transmission risk threshold, suggesting exposure to these pathogens does not differ between deer populations and that uMD translocations would not result in an increased risk of BTV, BVDV, PI3 and N. caninum transmission. We propose the number of uMD to be translocated under the current conditions should not be limited by the risk of BTV, BVDV, PI3 and N. caninum transmission between MD populations.
55
4.4.2 Continuous Pathogen Monitoring Guidelines
Model 2 established the risk of pathogen transmission associated with uMD translocations under a range of pathogen prevalence conditions used in determining recommendation for continuous pathogen monitoring guidelines (Model 3). From those outputs, safeguards for the urban mule deer translocation guidelines were generated to account for the risk of pathogen transmission and its level of acceptability which may vary as a function of climatic and habitat change, population explosion and disease outbreaks (Institute of Medicine 2003). We found that the risk of pathogen transmission under a range of pathogen prevalence conditions is of current concern for BRSV and T. gondii. BRSV represents a risk for the uMD population while
T. gondii represents a risk for the nuMD population. The risk of EHDV, BTV, BRSV, BVDV and
T. gondii transmission is of potential future concern for both populations. The risk of EHDV,
AHDV, BTV, BVDV transmission is not of current concern for either populations. The risk of
PI3 and N. caninum transmission is not of current or potential future concern for either populations.
Based on the predictive pathogen transmission risk analysis results, we established liberal guidelines for the sample size required for detection of realistic pathogen outbreaks in mule deer in southeastern British Columbia. Calculations were based on deer populations that have a higher pathogen exposure prevalence than what was observed in the Kootenay region from 2014 to 2017 and used a power of 0.8. This allows wildlife managers the freedom to sample more individuals if additional resources are available. Model 3 revealed sampling of a minimum of a total of 265 individuals would allow for detection of realistic outbreaks of all selected pathogens. We proposed a more targeted and practical approach to continuous pathogen monitoring given the difficulties associated with wildlife sample collection by prioritizing pathogens shown to be
56 current and potential future concern only. We recommend sampling of a minimum of a total of
133 individuals. This would allow for detection of realistic outbreaks of all pathogens of concern including EHDV, BTV, BRSV, BVDV and T. gondii. Given the dynamic character of infectious diseases, we also recommend continuous pathogen monitoring of MD be performed as long uMD translocations are planned. Continuous pathogen monitoring of MD may be performed through ongoing research projects and opportunist sampling of deceased animals; but should be consistent in protocols to allow efficient and economic comparison between studies.
4.4.3 Model Limitations
These guidelines are subject to change as the baseline database expands with further development and implementation of continuous pathogen monitoring programs in the Kootenay region. The developed models could be further refined by the addition of environmental factors such as rainfall, temperatures, resource availability and known spatial considerations such as livestock densities and barriers to movement such as deer fences and topographical impediments.
4.5 Conclusion
While our original the health assessment only identified differences in population exposure to T. gondii, these models results suggest that BRSV and AHDV exposure also differs between
MD populations and that the overall pathogen transmission risk associated with translocation of uMD is likely greater than initially shown (Mathieu et al. 2018). The number of uMD to translocate under the current conditions in the East Kootenay region should be limited to 344 individuals to minimize the risk of ADHV transmission and its potential negative effects on the nuMD population. The risk of transmission of BRSV, T. gondii, BTV, BVDV, PI3 and N. caninum was established to be acceptable under the current conditions and to not further restrict
57 the number of uMD to be translocated. Continuous pathogen monitoring is required to ensure the translocation recommendations remain current and was shown to necessitate the sampling of a minimum of a total of 133 individuals to detect realistic outbreaks in all pathogens of concern, including EHDV, BTV, BRSV, BVDV and T. gondii. These models can be used to further inform practical mule deer management strategies in southeastern British Columbia as additional data becomes available.
58
Chapter 5. General Discussion
5.1 Introduction
Over the past decade, urban mule deer population growth and non-urban mule deer population decline in the Kootenay region of southeastern British Columbia, Canada, have become a growing concern for local wildlife managers (Mule Deer Working Group 2016). Urban deer translocation was proposed as a way to simultaneously reduce the uMD population and sustain the nuMD population by providing reproductively capable animals to depleted regions.
Given the inherent health and disease transmission risks to anthropogenic movement of animals, it is essential a health risk assessment be performed prior to translocating wildlife (Cunningham
1996). With little known about the drivers of mule deer health in the Kootenay region and the diseases that afflict them, a health assessment was needed to ensure the translocation project would not negatively affect the health of the translocated and resident animals. As such, free- ranging mule deer were captured and sampled in urban and non-urban environments in southeastern British Columbia from 2014 to 2017 to establish the determinants of their health, characterize the disease transmission risk associated with urban deer translocations and investigate the role of infectious sin the decline of the non-urban mule deer population (Mathieu et al. 2018).
5.2 General Hypothesis
This study addressed the hypothesis that the health of the urban deer population is superior to the health of the non-urban deer population and that introducing urban deer to a non- 59 urban environment through translocation may negatively affect the health and survivorship of the urban deer. The general hypothesis was largely refuted. However, constraints related to the study design and laboratory assays limit the interpretation of our findings.
5.3 Aims
5.3.1 Aim 1: To Characterize the Pathogen Exposure, Body Condition and Pregnancy
Rates of Urban and Non-Urban Mule Deer
Serum samples collected from urban and non-urban mule deer were tested for exposure to epizootic hemorrhagic disease virus (EHDV), bluetongue virus (BTV), adenovirus hemorrhagic disease virus (AHDV), bovine respiratory syncytial virus (BRSV), parainfluenza-3 virus (PI3), bovine viral diarrhea virus (BVDV), Neospora caninum and Toxoplasma gondii, and for quantification of the Pregnancy-Specific Protein B (PSPB).
Antibodies were detected for all selected pathogens in both populations, with the exception of EHDV. Pathogen exposure prevalence was overall low. Exposure to AHDV, PI3 and
N. caninum was common whereas exposure to BTV, BRSV, BVDV and T. gondii was rare.
Pathogen exposure prevalence did not significantly differ between populations, with the exception of T. gondii. Comingling of uMD and nuMD in some areas likely contributed to exchange of pathogens and may explain why both deer populations have similar exposure levels to most pathogens.
The sampled deer were found to be highly prolific, with similar pregnancy rates that are comparable to rates reported for other mule deer populations (Andelt et al. 2004; Chattin 1948;
Dyer et al. 1984; Robinette & Gashwiler 1950). Given this lack of difference in pregnancy rates between populations, it is unlikely that fertility is a viable explanation for differences in demographics between uMD and nuMD.
60
Investigation of the potential effects of pathogen exposure on body condition and fecundity revealed an association between exposure to N. caninum and reduction of fecundity, but no definitive link is suggested. No associations were established between pathogen exposure and body condition.
5.3.2 Aim 2: To Determine if the Health of the Urban Deer Population Differs from the
Health of the Non-Urban Deer Population and Characterize the Health Risk Associated with the Translocation Project
While differences were noted between deer populations for T. gondii exposure and body condition, comparison of pathogen exposure, body condition and pregnancy rates of urban and non-urban mule deer revealed their health is largely equivalent.
Urban deer were 3.5 times more likely to be exposed to T. gondii than non-urban deer.
This observation is consistent with other studies that investigated urbanization as a risk factor for
T. gondii exposure in wildlife (Ballash et al. 2015; Conrad et al. 2005; Lehrer et al. 2010). The increased seroprevalence in deer across the urban gradient was proposed to result from the greater risk of exposure to oocysts shed by domestic cats given their higher population density within city limits (Ballash et al. 2015). While of apparent limited concern for the health of mule deer in the Kootenay, T. gondii poses a significant public health risk to individuals that may consume viable tissue cysts in raw or undercooked mule deer venison (Dubey et al. 2004; Sacks et al.
1983). Translocation of urban deer may result in an increased prevalence of T. gondii in non- urban areas, but it remains unclear to what extent this would affect the health of non-urban mule deer and other resident animals. Further investigations into the role of T. gondii on the health of mule deer and sympatric wildlife is warranted to better characterize the health risk associated with translocation of seropositive urban mule deer.
61
A relation between seropositivity to N. caninum and reduction in pregnancy rates was suggested. However, the similarity in exposure to N. caninum and pregnancy rates between uMD and nuMD suggests the reproductive health of both population is affected equally by N. caninum and that transmission of N. caninum through uMD translocation represents a low health risk for either deer population. We recommend investigating recruitment failure (e.g. poor fawn survival) as a possible explanation for the nuMD population decline.
The statistically significant difference recorded in body condition scores between sampling areas and years were likely the consequence of timing of captures and variations in winter conditions. Most nuMD were captured in early winter (December and January) whereas all uMD were captured in late winter (February and March) when fat stores are typically depleted.
Winter severity may have also contributed to lower BCS score of uMD in 2016-17 as southeastern British Columbia experienced the harshest winter in 20 years, with below average temperatures and above average snowfall from December to February (Figure 2). No year variability was noted for any of the pathogen tested.
In light of these findings, it appears the intraspecific disease transmission risk associated with the introduction of urban deer to a non-urban environment through translocation is low.
However, further investigations are needed to fully characterize the health risk associated with urban deer translocations.
5.3.3 Aim 3: To Investigate the Role of Infectious Diseases in the Decline of the Non-
Urban Mule Deer Population
In parallel to characterizing the health risk associated with urban deer translocation, this study investigated the role of infectious diseases in the decline of the non-urban mule deer population. As mentioned above, the non-urban mule deer population experienced more than a
62
20% population reduction from 2011 to 2014 (Mule Deer Working Group 2016). The reasons for the decline remained speculative. It was proposed limited forage availability, variations in predator-prey dynamics, and the harsh winter conditions are involved, but infectious diseases were not considered (Mule Deer Working Group 2016). This study was unable to identify infectious causes for the decline of the non-urban population. While infectious diseases may in fact not have played an important role in the non-urban deer population decline, there are several reasons why this study could have failed to identify them. It is possible the diseases that afflicted nuMD from 2011-2014 were highly fatal and that most exposed deer died. Those animals were therefore not sampled, and evidence of previous exposure could not be detected. It also may be that the pathogens responsible for morbidity and mortality were not screened. As such, further investigations that include testing for exposure to additional pathogens and sampling of both live and dead diseased deer are warranted to better establish the potential role of infectious diseases in the nuMD population decline.
5.3.4 Aim 4: To Develop Predictive Models to Inform Management Decisions Related to
Urban Deer Translocations
Predictive disease models were developed to establish recommendations for the number of mule deer to be translocated under the current conditions based on the risk of pathogen transmission and for the number of mule deer to sample to allow for detection of realistic pathogen outbreaks in mule deer.
An increased risk of pathogen transmission was identified for AHDV, BRSV and T. gondii under the current conditions. However, only the risk of AHDV transmission was establishing to be of concern given its potential negative effects on mule deer health. Based on this finding and subsequent calculations, we proposed a conservative approach to uMD
63 translocations, focusing on the mitigation of the risk of AHDV transmission between MD populations, and suggested no more than 344 uMD should be translocated under the current conditions. Since pathogen transmission is a dynamic process, pathogen exposure prevalence change over time and continuous pathogen monitoring is necessary to ensure the translocation recommendations remain current and adequate. Based on modeling for realistic pathogen outbreaks, we established that the risk of EHDV, BTV, BRSV, BVDV and T. gondii transmission is of potential future concern for both mule deer populations, and that continuous monitoring for those pathogens should be prioritized. We calculated that sampling of a minimum of a total of
133 individuals would allow for detection of realistic outbreaks of all of those pathogens at any point in time.
5.4 Future Directions
From aims 1 and 2, it was found that inclusion of additional health indicators, such as blood biochemistry and hematology, trace mineral levels, parasite burden, post-mortem examination findings of apparently healthy and diseased animals, would be beneficial in gaining a better understanding of the determinants of mule deer health. Post-mortem examinations performed on apparently healthy individuals, on clinically disease individuals and on deceased animals would allow for more accurate disease diagnosis. Such information could be obtained from post-mortem investigations performed on well-preserved road-killed individuals, deceased radio-collared individuals and hunted individuals processed at the butcher shop. Urban deer culled for population management purposes also represent a valuable source of data. Furthermore, increasing the sample size would allow for a more accurate characterization of population health and for easier detection of statistically significant differences in health status between the two
64 deer populations. A follow-up study that incorporates post-translocation telemetry and mortality data would further our understanding of this complex disease dynamic.
From aim 3, it became apparent that targeted serological screening of live animals was not sufficient to identify infectious causes for the decline of the nuMD population. Mortality investigations, as mentioned above, may be more successful at identifying infectious causes. We also recommend investigation of recruitment rates and evaluation of habitat quality and suitability to try and elucidate the reasons for the decline of the nuMD population.
From aim 4, recommendations for guidelines for assessment of the risk of pathogen transmission associated with uMD translocations were created. Continuous pathogen monitoring of mule deer is required to ensure these translocation recommendations remain current and to build on the database we have used in the models.
The results of this mule deer health assessment provide a basis for a formal risk assessment for future uMD translocations in southeastern British Columbia. More detailed sampling and testing, as described above, would further our understanding of the determinants of mule deer health and allow for elaboration of more accurate pathogen transmission risk mitigation approaches.
5.5 Recommendations
Based on the preliminary findings of this study, it appears the intraspecific pathogen transmission risk associated with the introduction of urban deer to a non-urban environment through translocation is low. Predictive modelling suggests transmission of AHDV to the nuMD population is the highest risk factor under the current conditions, therefore a conservative approach to uMD translocations should be taken to mitigate the risk of AHDV transmission. We suggest no more than 344 uMD should be translocated under the current conditions until a
65 minimum of 133 individuals the mule deer populations can be resampled and retested so that translocation guidelines can be updated.
66
Appendix: Tables and Figures
67
Table 1. Characteristics of urban and non-urban free-ranging mule (Odocoileus hemionus) deer sampled in southeastern British Columbia, Canada.
uMD nuMD Total n (%) n (%) n (%) Overall - 88 (44.0) 112 (56.0) 200 (100) Fawn 20 (100) 0 (0) 20 (11.5) Young adult 31 (47.0) 35 (53.0) 66 (37.9) Age Adult 32 (42.1) 44 (57.9) 76 (43.7) Aged adult 5 (41.7) 7 (58.2) 12 (6.9) 2014-15 0 (0) 42 (100) 42 (21.0) Year 2015-16 63 (57.3) 47 (42.7) 110 (55.0) 2016-17 25 (52.1) 23 (47.9) 48 (24.0) Emaciated 1 (100) 0 (0) 1 (0.7) Poor 11 (64.7) 6 (35.3) 17 (11.6) BCS Fair 36 (61.0) 23 (39.0) 59 (40.1) Good 20 (34.5) 38 (65.5) 58 (39.5) Excellent 5 (41.7) 7 (58.3) 12 (8.2) Female 74 (39.8) 112 (60.2) 186 (93.0) Sex Male 14 (100) 0 (0) 14 (7.0) Darting 82 (77.4) 24 (22.6) 106 (53.0) Capture Net gunning 0 (0) 87 (100) 87 (43.5) technique Clover trapping 6 (85.7) 1(14.3) 7 (3.5) Cranbrook 25 (28.4) NA 25 (28.4) Elkford 15 (17.0) NA 15 (17.0) Urban areas Invermere 14 (15.9) NA 14 (15.9) Kimberley 19 (21.6) NA 19 (21.6) Marysville 15 (17.0) NA 15 (17.0) Invermere NA 29 (25.9) 29 (25.9) Non-urban West Kootenay NA 11 (9.8) 11 (9.8) areas Grasmere NA 39 (34.8) 39 (34.8) Newgate NA 33 (29.5) 33 (29.5)
68
Table 2. Prevalence of pregnancy based on Pregnancy-Specific Protein B levels in free- ranging mule deer (Odocoileus hemionus) in southeastern British Columbia, Canada.
PSPB
n % (95% CI)
Overall - 108 90.7 (83.3-95.0) Young adult 32 84.4 (68.2-93.1) Age Adult/Aged 53 96.2 (87.2-99.0) 2015-16 67 88.1 (78.1-93.8) Year 2016-17 30 96.7 (83.3-99.4) Population Non-urban 33 90.9 (76.4-96.8) Type Urban 64 90.6 (81.0-95.6) 1-2 14 100 (78.5-100) 3 32 90.6 (75.8-96.8) BCS 4 22 81.8 (61.5-92.7) 5 7 100 (64.6-100) Positive 0 0 (NA) EHDV Negative 95 90.5 (83.0-94.9) Positive 48 93.8 (83.2-97.8) AHDV Negative 49 87.8 (75.8-94.3) Positive 1 100 (20.6-100) BTV Negative 94 90.4 (82.8-94.9) Positive 11 90.9 (62.3-98.4) BRSV Negative 97 90.7 (83.3-95.0) Positive 1 100 (20.6-100) BVDV Negative 107 90.6 (83.6-94.8) Positive 35 85.7 (70.6-93.7) PI3 Negative 73 93.1 (84.9-97.0) N. Positive 18 72.2 (49.1-87.5) caninum Negative 75 97.3 (90.2-98.8) Positive 11 72.7 (43.4-90.2) T. gondii Negative 92 94.6 (87.9-97.6)
69
Table 3. Pathogen exposure prevalence in free-ranging mule deer (Odocoileus hemionus) in southeastern British Columbia, Canada.
EHDV AHDV BTV
n % (95% CI) n % (95% CI) n % (95% CI)
Overall - 176 0 (NA) 177 38.4 (31.6-45.8) 176 0.6 (0.1-3.1) Fawn 20 0 (NA) 20 20 (8.0-41.6) 20 0 (NA) Young 55 0 (NA) 56 33.9 (22.9-47.0) 55 0 (NA) Age adult Adult 68 0 (NA) 68 51.5 (39.8-62.9) 68 1.5 (0.3-7.9) Aged 9 0 (NA) 9 66.7 (35.4-87.9) 9 0 (NA) 2014-15 23 0 (NA) 23 21.7 (9.7-41.9) 23 0 (NA) Year 2015-16 105 0 (NA) 106 35.8 (27.4-45.3) 105 0.9 (0.2-5.2) 2016-17 48 0 (NA) 48 52.1 (38.3-65.5) 48 0 (NA) Population Non-urban 90 0 (NA) 90 33.3 (24.4-43.6) 90 0 (NA) Type Urban 86 0 (NA) 87 43.7 (33.7-54.1) 86 1.2 (0.2-6.3) 1-2 16 0 (NA) 17 52.9 (31.0-73.8) 16 0 (NA) 3 53 0 (NA) 54 40.7 (28.7-54.0) 53 1.9 (0.3-9.9) BCS 4 46 0 (NA) 45 40 (27.0-54.5) 46 0 (NA) 5 12 0 (NA) 12 33.3 (13.8-60.9) 12 0 (NA) Positive - - 0 - 0 - EHDV Negative - - 175 38.3 (31.4-45.7) 176 0.6 (0.1-3.1) Positive 67 0 (NA) - - 67 1.5 (0.3-8.0) AHDV Negative 108 0 (NA) - - 108 0 (NA) Positive 1 0 (NA) 1 100 (20.6-100) - - BTV Negative 175 0 (NA) 174 37.9 (31.0-45.3) - - Positive 13 0 (NA) 15 40 (19.8-64.2) 13 0 (NA) BRSV Negative 162 0 (NA) 162 38.3 (31.1-45.9) 162 0.6 (0.1-3.4) Positive 2 0 (NA) 2 100 (34.2-100) 2 0 (NA) BVDV Negative 173 0 (NA) 175 37.8 (30.9-45.1) 173 5.8 (0.1-3.2) Positive 47 0 (NA) 48 54.2 (40.3-67.4) 47 2.1 (0.3-11.1) PI3 Negative 128 0 (NA) 129 32.6 (25.1-41.0) 128 0 (NA) Positive 35 0 (NA) 35 28.6 (16.3-45.0) 35 0 (NA) N. caninum Negative 121 0 (NA) 123 43.9 (35.4-52.7) 121 0.8 (0.1-4.5) Positive 12 0 (NA) 13 53.8 (29.1-76.8) 12 0 (NA) T. gondii Negative 145 0 (NA) 145 38.6 (31.1-68.9) 145 0.7 (0.1-3.8)
70
Table 3 (cont.). Pathogen exposure prevalence in free-ranging mule deer (Odocoileus hemionus) in southeastern British Columbia, Canada.
BRSV BVDV PI3
n % (95% CI) n % (95% CI) n % (95% CI)
Overall - 178 8.4 (5.2-13.4) 178 1.1 (0.3-4.0) 178 27.0 (21.0-33.9) Fawn 20 5 (0.9-23.6) 20 0 (NA) 20 0 (NA) Young 56 7.1 (2.8-17.0) 56 0 (NA) 56 12.5 (6.2-23.6) Age adult Adult 68 8.8 (4.1-17.9) 68 1.5 (0.3-7.9) 68 39.7 (28.9-51.6) Aged 9 0 (NA) 9 0 (NA) 9 77.8 (45.3-93.7) 2014-15 23 8.7 (2.4-26.8) 23 0 (NA) 23 34.8 (18.8-55.1) Year 2015-16 107 7.5 (3.8-14.1) 107 1.9 (0.5-6.6) 107 25.2 (18.0-34.2) 2016-17 48 10.4 (4.5-22.2) 48 0 (NA) 48 27.1 (16.6-41.0) Population Non-urban 91 12.1 (6.9-20.4) 91 2.2 (0.6-7.7) 91 26.4 (18.4-36.2) Type Urban 87 4.6 (1.8-11.2) 87 0 (NA) 87 27.6 (19.3-37.8) 1-2 17 11.8 (3.3-34.3) 17 5.9 (1.0-27.0) 17 47.1 (26.2-69.0) 3 54 3.7 (1.0-12.5) 54 0 (NA) 54 27.8 (17.6-40.9) BCS 4 45 8.9 (3.5-20.7) 45 0 (NA) 45 28.9 (17.7-43.4) 5 12 8.3 (1.5-35.4) 12 0 (NA) 12 25 (8.9-53.2) Positive 0 - 0 - 0 - EHDV Negative 175 7.4 (4.4-12.3) 175 1.1 (0.3-4.1) 175 26.8 (20.8-3.9) Positive 68 8.8 (41-17.9) 68 2.9 (0.3-4.1) 68 38.2 (27.6-50.1) AHDV Negative 109 8.2 (4.4-15.0) 109 0 (NA) 109 20.2 (13.7-28.7) Positive 1 0 (NA) 1 0 (NA) 1 100 (20.6-100) BTV Negative 174 7.4 (4.4-12.4) 174 1.1 (0.3-4.1) 174 26.4 (20.4-33.4) Positive - - 15 0 (NA) 15 6.7 (1.2-29.8) BRSV Negative - - 163 1.2 (0.3-4.4) 163 28.8 (22.4-36.2) Positive 2 0 (NA) - - 2 100 (34.2-100) BVDV Negative 176 8.5 (5.2-13.6) - - 176 26.1 (20.2-33.1) Positive 48 2.1 (0.4-10.9) 48 4.1 (1.2-14.0) - - PI3 Negative 130 10.8 (6.5-17.3) 130 0 (NA) - - N. Positive 35 11.4 (4.5-26.0) 35 0 (NA) 35 25.7 (14.2-42.1) caninum Negative 123 6.5 (3.3-12.3) 123 1.6 (0.4-5.7) 123 28.4 (21.2-37.0) Positive 13 7.7 (1.4-33.3) 13 0 (NA) 12 46.2 (23.2-70.8) T. gondii Negative 145 6.9 (3.8-12.2) 145 1.4 (0.4-4.9) 145 26.2 (19.7-33.9)
71
Table 3 (cont.). Pathogen exposure prevalence in free-ranging mule deer (Odocoileus hemionus) in southeastern British Columbia, Canada.
N. caninum T. gondii
n % (95% CI) n % (95% CI)
Overall - 158 22.1 (16.4-29.2) 158 8.2 (4.9-13.6) Fawn 19 31.6 (15.4-54.0) 19 0 (NA) Young adult 50 24 (14.3-37.4) 50 10 (4.3-21.4) Age Adult 63 17.5 (10.0-28.6) 61 8.2 (4.3-21.4) Aged 8 37.5 (13.7-69.4) 8 37.5 (13.7-69.4) 2014-15 17 47 (26.2-69.0) 19 10.5 (2.9-31.4) Year 2015-16 93 16.1 (10.0-24.9) 91 7.7 (3.8-15.0) 2016-17 48 25 (14.9-38.8) 48 8.3 (3.2-19.6) Population Non-urban 76 19.7 (12.3-30.0) 77 3.9 (1.3-10.8) Type Urban 82 24.4 (16.4-34.7) 81 12.3 (6.8-21.3) 1-2 17 35.3 (17.3-58.7) 16 0 (NA) 3 51 13.7 (6.8-25.7) 50 12 (5.6-23.8) BCS 4 42 31 (19.1-46.0) 42 11.9 (5.2-25.0) 5 8 12.5 (2.2-47.1) 8 0 (NA) Positive 0 - 0 - EHDV Negative 156 22.4 (16.6-29.6) 156 22.4 (16.6-29.6) Positive 64 15.6 (8.7-26.4) 63 11.1 (5.5-21.2) AHDV Negative 94 25.6 (18.7-36.3) 95 6.3 (2.9-13.1) Positive 1 0 (NA) 1 0 (NA) BTV Negative 155 22.6 (16.7-29.8) 156 7.7 (4.4-13.0) Positive 12 33.3 (13.8-60.9) 11 9 (1.6-37.7) BRSV Negative 146 21.2 (15.4-28.6) 147 8.1 (4.7-13.7) Positive 2 0 (NA) 2 0 (NA) BVDV Negative 156 22.4 (16.6-29.6) 156 8.3 (4.9-13.7) Positive 44 20.4 (11.2-34.5) 44 13.6 (6.4-26.7) PI3 Negative 114 22.8 (16.1-31.3) 114 6.1 (3.0-12.1) N. Positive - - 35 14.3 (6.3-29.4) caninum Negative - - 121 6.6 (3.4-12.5) Positive 13 38.5 (17.7-64.5) - - T. gondii Negative 143 21 (15.1-28.4) - -
72
Table 4. Predicted minimal sample size (PMSS) required to detect statistically significant differences in pathogen exposure between urban and non-urban mule deer (Odocoileus hemionus) populations.
Baseline Actual sample PMSS prevalence size % n n uMD 1 86 NA EHDV nuMD 1 90 NA Total 176 NA uMD 43.7 87 344 AHDV nuMD 33.3 90 354 Total 177 698 uMD 1.2 86 42556 BTV nuMD 1 90 44683 Total 176 87239 uMD 4.6 87 211 BRSV nuMD 12 91 221 Total 178 432 uMD 1 87 1627 BVDV nuMD 2.2 91 1708 Total 178 3335 uMD 27.6 87 22911 PI3 nuMD 26.4 91 24056 Total 178 46967 uMD 24.4 82 1240 N. caninum nuMD 19.7 76 1153 Total 158 2393 uMD 12.3 81 172 T. gondii nuMD 3.9 77 164 Total 158 336
73
Table 5. Predicted minimal sample size for a range of pathogen prevalence conditions in the urban (A) and non-urban (B) mule deer (Odocoileus hemionus) populations.
A. PMSS (n) for a range of pathogen prevalence conditions in the uMD population EHDV PONE: 1% r: 1.05 UPE 5% UPE 10% UPE 15% UPE 25% UPE 50% UPE 75% Non-urban (51%) (UE) 288 102 60 31 13 7 Urban (49%) (E) 279 99 58 31 13 7 Total 567 201 118 62 26 14 AHDV PONE: 33.3% r: 1.03 UPE 5% UPE 10% UPE 15% UPE 25% UPE 50% UPE 75% Non-urban (51%) (UE) 32 50 87 487 139 23 Urban (49%) (E) 31 49 84 473 135 23 Total 63 99 171 960 274 46 BTV PONE: 1% r: 1.03 UPE 5% UPE 10% UPE 15% UPE 25% UPE 50% UPE 75% Non-urban (51%) (UE) 288 102 60 31 13 7 Urban (49%) (E) 279 99 58 31 13 7 Total 567 201 118 62 26 14 BRSV PONE: 12.1% r: 1.05 UPE 5% UPE 10% UPE 15% UPE 25% UPE 50% UPE 75% Non-urban (51%) (UE) 250 3709 2266 146 24 10 Urban (49%) (E) 238 3533 2158 139 23 10 Total 488 7242 4424 285 47 20 BVDV PONE: 2.2% r: 1.05 UPE 5% UPE 10% UPE 15% UPE 25% UPE 50% UPE 75% Non-urban (51%) (UE) 706 150 76 36 14 8 Urban (49%) (E) 672 143 73 35 13 7 Total 1378 293 149 71 27 15 PI3 PONE: 26.4% r: 1.05 UPE 5% UPE 10% UPE 15% UPE 25% UPE 50% UPE 75% Non-urban (51%) (UE) 48 90 203 16010 68 18 Urban (49%) (E) 46 86 193 15248 65 17 Total 94 176 396 31258 133 35 N. caninum PONE: 19.7% r: 0.93 UPE 5% UPE 10% UPE 15% UPE 25% UPE 50% UPE 75% Non-urban (48%) (UE) 73 208 980 929 38 13 Urban (52%) (E) 79 226 1065 1010 41 14 Total 152 434 2045 1939 79 27 T. gondii PONE: 3.9% r: 0.95 UPE 5% UPE 10% UPE 15% UPE 25% UPE 50% UPE 75% Non-urban (49%) (UE) 5636 271 108 44 15 8 Urban (51%) (E) 5933 285 114 46 16 8 Total 11569 556 222 90 31 16
PONE: Prevalence of outcome in the nuMD population UPE: Predicted exposure in the uMD population E: Exposed group UE: Unexposed group
74
Table 5 (cont.). Predicted minimal sample size for a range of pathogen prevalence conditions in the urban (A) and non-urban (B) mule deer (Odocoileus hemionus) populations.
B. PMSS (n) for a range of pathogen prevalence conditions in the nuMD population EHDV PONE: 1% r: 0.96 NUPE 5% NUPE 10% NUPE 15% NUPE 25% NUPE 50% NUPE 75% Non-urban (51%) (E) 293 104 61 32 13 7 Urban (49%) (UE) 285 101 59 31 13 7 Total 578 205 120 63 26 14 AHDV PONE: 43.7% r: 0.97 NUPE 5% NUPE 10% NUPE 15% NUPE 25% NUPE 50% NUPE 75% Non-urban (51%) (E) 20 28 41 103 991 40 Urban (49%) (UE) 20 27 40 100 961 39 Total 40 55 81 203 1952 79 BTV PONE: 1.2% r: 0.96 NUPE 5% NUPE 10% NUPE 15% NUPE 25% NUPE 50% NUPE 75% Non-urban (51%) (E) 334 111 64 33 13 7 Urban (49%) (UE) 324 107 62 32 13 7 Total 658 218 126 65 26 14 BRSV PONE: 4.6% r: 0.96 NUPE 5% NUPE 10% NUPE 15% NUPE 25% NUPE 50% NUPE 75% Non-urban (51%) (E) 47339 377 133 50 15 8 Urban (49%) (UE) 45446 362 127 48 16 8 Total 92785 739 260 98 31 16 BVDV PONE: 1% r: 0.96 NUPE 5% NUPE 10% NUPE 15% NUPE 25% NUPE 50% NUPE 75% Non-urban (51%) (E) 296 105 62 32 13 7 Urban (49%) (UE) 284 101 59 31 13 7 Total 580 206 121 63 26 14 PI3 PONE: 27.6% r: 0.96 NUPE 5% NUPE 10% NUPE 15% NUPE 25% NUPE 50% NUPE 75% Non-urban (51%) (E) 43 78 169 5055 77 18 Urban (49%) (UNE) 41 75 162 4853 74 18 Total 84 153 331 9908 151 36 N. caninum PONE: 24.4% r: 1.08 NUPE 5% NUPE 10% NUPE 15% NUPE 25% NUPE 50% NUPE 75% Non-urban (48%) (E) 51 104 279 93485 54 15 Urban (52%) (UE) 55 111 298 100029 58 16 Total 106 215 577 193514 112 31 T. gondii PONE: 12.3% r: 1.05 NUPE 5% NUPE 10% NUPE 15% NUPE 25% NUPE 50% NUPE 75% Non-urban (49%) (E) 5610 266 2578 154 25 11 Urban (51%) (UE) 5891 279 2449 146 24 10 Total 11501 545 5027 300 49 21
PONE: Prevalence of outcome in the nuMD population UPE: Predicted Exposure in the uMD population E: Exposed group UE: Unexposed group 75
Table 6. Predicted minimum sample size required to detect realistic pathogen outbreaks the urban (A) and non-urban (B) mule deer (Odocoileus hemionus) populations.
A. PMSS (n) required to detect realistic pathogen B. PMSS (n) required to detect realistic pathogen outbreaks in the uMD population outbreaks in the nuMD population EHDV PONE: 1% EHDV PONE: 1% r: 1.05 UPE 26% r: 0.97 NUPE 26% Non-urban (51%) (UE) 30 Non-urban (51%) (E) 31 Urban (49%) (E) 29 Urban (49%) (UE) 30 Total 59 Total 61 AHDV PONE: 33.3% AHDV PONE: 43.7% r: 1.03 NA r: 0.97 NA Non-urban (51%) (UE) NA Non-urban (51%) (E) NA Urban (49%) (E) NA Urban (49%) (UE) NA Total NA Total NA BTV PONE: 1% BTV PONE: 1.2% r: 1.03 UPE 27% r: 0.96 NUPE 27% Non-urban (51%) (UE) 30 Non-urban (51%) (E) 31 Urban (49%) (E) 29 Urban (49%) (UE) 30 Total 59 Total 61 BRSV PONE: 12.1% BRSV PONE: 4.6% r: 1.05 UPE 71% r: 0.96 NUPE 71% Non-urban (51%) (UE) 12 Non-urban (51%) (E) 9 Urban (49%) (E) 11 Urban (49%) (UE) 9 Total 23 Total 18 BVDV PONE: 2.2% BVDV PONE: 1% r: 1.05 UPE 57% r: 0.96 NUPE 57% Non-urban (51%) (UE) 12 Non-urban (51%) (E) 11 Urban (49%) (E) 11 Urban (49%) (UE) 11 Total 23 Total 22 PI3 PONE: 26.4% PI3 PONE: 27.6% r: 1.05 UPE 61% r: 0.96 NUPE 61% Non-urban (51%) (UE) 33 Non-urban (51%) (E) 36 Urban (49%) (E) 32 Urban (49%) (UNE) 35 Total 65 Total 71 N. caninum PONE: 19.7% N. caninum PONE: 24.4% r: 0.93 UPE 40.5% r: 1.08 NUPE 40.5% Non-urban (48%) (UE) 75 Non-urban (48%) (E) 128 Urban (52%) (E) 81 Urban (52%) (UE) 137 Total 156 Total 265 T. gondii PONE: 3.9% T. gondii PONE: 12.3% r: 0.95 UPE 32.6% r: 1.05 NUPE 32.6% Non-urban (49%) (UE) 29 Non-urban (49%) (E) 65 Urban (51%) (E) 30 Urban (51%) (UE) 68 Total 59 Total 133
PONE: Prevalence of outcome in the nuMD group PONE: Prevalence of outcome in the uMD group UPE: Predicted Exposure in uMD NUPE: Predicted Exposure in nuMD E: Exposed group E: Exposed group UE: Unexposed group UE: Unexposed group 76
Figure 1. Location of capture sites of urban and non-urban of free-ranging mule deer (Odocoileus hemionus) in southeastern British Columbia, Canada.
77
Figure 2. An urban mule deer (Odocoileus hemionus) forages on tree branches in April 2017 in Marysville, British Columbia, Canada.
78
References
Adams I. 2018. East Kootenay Urban Deer Translocation Trial. Cranbrook, British Columbia: VAST Resource Solutions Inc.
Andelt W, Pojar T, and Johnson L. 2004. Long-term trends in mule deer pregnancy and fetal rates in colorado. Journal of Wildlife Management 68:542–549.
Anderson T, Dejardin A, Howe D, Dubey J, and Michalski M. 2007. Neospora caninum antibodies detected in Midwestern white-tailed deer (Odocoileus virginianus) by Western blot and ELISA. Vet Parasitol 145:152-155.
Aramini J, Stephen C, and Dubey J. 1998. Toxoplasma gondii in Vancouver Island cougars (Felis concolor vancouverensis): serology and oocyst shedding. The Journal of Parasitology 84:pp. 438-440.
Armstrong D, R. J-H, and US. S. 2003. Animal movements and disease risk: A Workbook, 5th Edition. Apple Valley (MN): IUCN/SSC Conservation Breeding Specialist Group.
Ballash G, Dubey J, Kwok O, Shoben A, Robison T, TJ. K, and Dennis P. 2015. Seroprevalence of Toxoplasma gondii in White-Tailed Deer (Odocoileus virginianus) and Free-Roaming Cats (Felis catus) Across a Suburban to Urban Gradient in Northeastern Ohio. EcoHealth 12:359–367.
Barker I, Van Dreumel A, and Palmer N. 1992. Bluetongue and related diseases. In: KVF. J, Kennedy P, and Palmer N, eds. Pathology of domestic animals. 4 ed. San Diego (CA): Academic Press, Inc., 173-176.
Barrett M, and Chalmers G. 1975. A serologic survey of pronghorns in Alberta and Saskatchewan, 1970-1972. J Wildl Dis 11:157-163.
79
Barrett M, Nolan J, and Roy L. 1982. Evaluation of a hand-held net-gun to capture large mammals Wildlife Society Bulletin 10:108-114.
Basso W, Moré G, Quiroga M, Balducchi D, Schares G, and Venturini M. 2014. Neospora caninum is a cause of perinatal mortality in axis deer (Axis axis). Vet Parasitol 199:255-258.
Bender L, Carlson S, Schmitt M, and Haufler J. 2002. Production and survival of elk (Cervus elaphus) calves in Michigan. Am Midl Nat 48:163-171.
Beringer J, Hansen L, Wildling W, Fischer J, and Sheriff S. 1996. Factors affecting capture myopathy in white-tailed deer. J Wildl Manage 60:373-380.
Bevins S, Carver S, Boydston E, Lyren L, Alldredge M, Logan K, Riley S, Fisher R, Vickers T, Boyce W, Salman M, Lappin M, Crooks K, and VandeWoude S. 2012. Three pathogens in sympatric populations of pumas bobcats and domestic cats: implications for infectious disease transmission. PLoS One 7:1-10.
BioTracking. BioPRYN® Wild. Available at http://www.biotracking.com/wildlife/ (accessed June 2016.
Blecha F, Boyles S, and Riley J. 1984. Shipping suppresses lymphocyte blastogenic responses in Angus and Brahman X Angus feeder calves. J Anim Sci 59:576-583.
Boesch JM, Boulanger JR, Curtis PD, Erb HN, Ludders JW, Kraus MS, and Gleed RD. 2011. Biochemical variables in free-ranging white-tailed deer (Odocoileus virginianus) after chemical immobilization in clover traps or via ground-darting. Journal of Zoo and Wildlife Medicine 42:18-28. 10.1638/2009-0146.1
Boyce W, Woods L, Keel M, MacLachlan N, Porter C, and Lehmkuhl H. 2000. An epizootic of adenovirus-induced hemorrhagic disease in captive black-tailed deer (Odocoileus hemionus). J Zoo Wildl Med 31:370-373.
Bryson D, Evermann J, Liggitt H, Foreyt W, and Breeze R. 1988. Studies on the pathogenesis and interspecies transmission of respiratory syncytial virus isolated from sheep. Am J Vet Res 49:1424-1430.
80
Bryson D, McNulty M, Logan E, and Crush P. 1983. Respiratory syncytial virus pneumonia in young calves: clinical and pathologic findings. Am J Vet Res 44:1648-1655.
Butler J, Hamilton W, Sasser R, Ruder C, Hass G, and Williams R. 1982. Detection and partial characterization of two bovine pregnancy-specific proteins. Biol Reprod 26:925-933.
Cantu A, Ortega S, Mosqueda J, Garcia-Vazquez Z, Henke S, and George J. 2008. Prevalence of infectious agents in free-ranging white-tailed deer in northeastern Mexico. J Wildl Dis 44.
Carbyn L, and Watson D. 2001. Translocation of plains bison to Wood Buffalo National Park: economic and conservation implications. In: Maehr DS, Noss RF, Larking JL (eds.). Large mammal restoration: Ecological and sociological challenges in the 21st century, Washington (DC): Island Press. p. 189-204.
Caswell J. 2014. Failure of respiratory defenses in the pathogenesis of bacterial pneumonia of cattle. Vet Pathol 51:393-409.
Caughley G. 1981. Overpopulation. In: Jewell P, Holt S, and Hart D, eds. Problems in management of locally abundant wild mammals. New York (NY): Academic Press Inc., 7-19.
CFIA. 2016a. Bluetongue: competitive enzyme-linked immunosorbent assay for the detection of antibody to bluetongue virus. Version 4.3. Winnipeg (MB). 20 p.
CFIA. 2016b. Epizootic hemorrhagic disease virus: competitive enzyme-linked immunosorbent assay for the detection of antibody to the epizootic hemorrhagic disease virus. Version 4.3. Winnipeg (MB). 20 p.
CFIA. 2016c. Federally Reportable diseases for terrestrial animals in Canada. Available at http://www.inspection.gc.ca/ (accessed May 2016.
CFIA. 2017. Bluetongue virus (BTV) serotypes US-2, 8, 10, 11, 13, 17 and epizootic hemorrhagic disease of deer virus (EDHV) serotypes 1 (WISC) and 2 (ALTA) serum neutralization test in microplates. Version 3.4. Winnipeg (MB). 16 p. 81
Chattin J. 1948. Breeding season and productivity in the Interstate deer herd. California Fish and Game 34:25-31.
Chomel B, Carniciu M, Kasten R, Castelli P, Work T, and Jessup D. 1994. Antibody prevalence of eight ruminant infectious diseases in California mule and black- tailed deer (Odocoileus hemionus). J Wildl Dis 30:51-59.
Clark W. 1995. Capture and handling techniques for urban deer control. In: McAninch JB. Urban deer: A manageable resource? Proceedings of the Symposium of the North Central Section, The Wildlife Society. p. 81.
Clavijo A, Munroe F, Zhou E, Booth T, and Roblesky K. 2000. Incursion of bluetongue virus into the Okanagan Valley, British Columbia. Can Vet J 41:312-314.
Clover M. 1956. Single-gate deer trap. California Fish and Game 42:199-210.
Conover M. 1997. Monetary and intangible valuation of deer in the United States. Wildlife Society Bulletin 25:98-305.
Conover M. 2002. Resolving human-wildlife conflicts: The science of wildlife damage management. Boca Raton (FL): CRC Press LLC, Lewis Publishers.
Conover M, Pitt W, Kessler K, DuBow T, and Sanborn W. 1995. Review of human injuries, illnesses, and economic losses caused by wildlife in the United States. Wildlife Society Bulletin 23:407-414.
Conrad P, Miller M, Kreuder C, James E, Mazet J, Dabritz H, Jessup D, Gulland F, and Grigg M. 2005. Transmission of Toxoplasma: clues from the study of sea otters as sentinels of Toxoplasma gondii flow into the marine environment. Parasitic Zoonosis 35:1155–1168.
Cook J, Johnson B, Cook R, Riggs R, Delcurto T, Bryant L, and Irwin L. 2004. Effects of summer-autumn nutrition and parturition date on reproduction and survival of elk. Wildlife Monographs 155:1-61.
82
Cook R, Stephenson T, Myers W, Cook J, and Shipley L. 2007. Validating predictive models of nutritional condition for mule deer. J Wildl Manage 71:1934-1943.
Couvillion C, Nettles V, Davidson W, Pearson J, and Gustafson G. 1981. Hemorrhagic disease among white-tailed deer in the Southeast from 1971 through 1980. Proceedings of the 85th United States Animal Health Association:522-537.
Creacy G. 2006. Deer management within suburban areas. Available at https://tpwd.texas.gov/publications/pwdpubs/media/pwd_bk_w7000_1197.pdf/ (accessed June 2016.
Cunningham A. 1996. Disease Risks of Wildlife Translocations. Conservation Biology 10:349-353.
Dalziel A, Sainsbury A, McInnes K, Jakob-Hoff R, and Ewen J. 2017. A comparison of disease risk analysis tools for conservation translocations. EcoHealth 14:30-41.
Daszak P, Cunningham A, and Hyatt A. 2001. Anthropogenic environmental change and the emergence of infectious diseases in wildlife. Acta Tropica 78:103-116.
Davidson W, and Nettles V. 1992. Relocation of wildlife: identifying and evaluating disease risks. North American Wildlife and Natural Resources Conference. p 466- 473.
Dean A, Sullivan K, and Soe M. 2013. OpenEpi: Open Source Epidemiologic Statistics for Public Health, Version. Available at www.OpenEpi.com (accessed February 2, 2018).
Decker D, Brown T, Connelly N, Enck J, Pomerantz G, Purdy K, and Siemer W. 1992. Toward a comprehensive paradigm of wildlife management: Integrating the human and biological dimensions. In: Mangun W, ed. American fish and wildlife policy: the human dimension. Carbondale (IL): Southern Illinois University Press, 33-54.
Diehl S. 1988. Translocation of urban white-tailed deer captured in a Pisgah-Clover hybrid trap. In: Nielsen L, and Brown R, eds. Translocation of wild animals.
83
Milwaukee (WI): Wisconsin Humane Society, Inc. and The Caesar Kleberg Wildlife Research Institute, 249-257.
Ditchfield J, Debbie J, and Karstad L. 1964. The virus of epizootic hemorrhagic disease of deer. Transactions of the North American Wildlife Natural Resources Conference 29:196-201.
Dobson A, and Hudson P. 1995. Microparasites: Observed patterns in wild animal populations. In: Grenfell B, and Dobson A, eds. Ecology of infectious diseases in natural populations. Cambridge (UK): Cambridge University Press, 52-89.
Dohoo I, Martin W, and Stryhn H. 2003. Veterinary Epidemiologic Research. AVC Inc., Charlottetown, Prince Edward Island.
Drolet B, Reister L, Rigg T, Nol P, Podell B, Mecham J, VerCauteren K, van Rijn P, Wilson W, and Bowen R. 2013. Experimental infection of white-tailed deer (Odocoileus virginianus) with Northern European bluetongue virus serotype 8. Vet Microbiol 166:347-355.
Dubay S, Jacques C, Golden N, Kern B, Mahoney K, Norton A, Patnayak D, and Van Deelen T. 2015. Environmental factors influencing white-tailed Deer (Odocoileus virginianus) exposure to livestock pathogens in Wisconsin. PLoS One 10:e0128827. 10.1371/journal.pone.0128827
Dubay SA, deVos JC, Jr., Noon TH, and Boe S. 2004. Epizootiology of hemorrhagic disease in mule deer in central Arizona. Journal of Wildlife Diseases 40:119-124. 10.7589/0090-3558-40.1.119
Dubey J, Dennis P, Verma S, Choudhary S, Ferreira L, Oliveira S, Kwok O, Butler E, Carstensen M, and Su C. 2014. Epidemiology of toxoplasmosis in white tailed deer (Odocoileus virginianus): Occurrence, congenital transmission, correlates of infection, isolation, and genetic characterization of Toxoplasma gondii. Vet Parasitol 202:270-275.
Dubey J, Graham D, De Young R, Dah E, Eberhard M, Nace E, Won K, Bishop H, Punkosdy G, Sreekumar C, Vianna M, Shen S, Kwok O, Sumners J, Demarais S, Humphreys J, and Lehmann T. 2004. Molecular and biologic characteristics of
84
Toxoplasma gondii isolates from wildlife in the United States. Journal of Parasitology 9:67–71.
Dubey J, Hollis K, Romand S, Thulliez, P., OC K, Hungerford L, Anchor C, and Etter D. 1999. High prevalence of antibodies to Neospora caninum in white-tailed deer (Odocoileus virginianus). Int J Parasitol 29:1709-1111.
Dubey J, Jenkins M, Kwok O, Ferreira L, Choudhary S, Verma S, Villena I, Butler E, and Carstensen M. 2013. Congenital transmission of Neospora caninum in white- tailed deer (Odocoileus virginianus). Vet Parasitol 196:519-522.
Dubey J, Jenkins M, Kwok O, Zink R, Michalski M, Ulrich V, Carstensen M, and Thulliez P. 2009. Seroprevalence of Neospora caninum and Toxoplasma gondii antibodies in white-tailed deer (Odocoileus virginianus) from Iowa and Minnesota using four serologic tests. Vet Parasitol 161:330-334.
Dubey J, Mansfield K, Hall B, Kwok O, and Thulliez P. 2008. Seroprevalence of Neospora caninum and Toxoplasma gondii in black-tailed deer (Odocoileus hemionus columbianus) and mule deer (Odocoileus hemionus hemionus). Vet Parasitol 156:310-313.
Dubey J, and Odening K. 2001. Tissue inhibiting protozoans: Toxoplasmosis and related infections. In: Samuel W, Pybus M, and Kocan A, eds. Parasitic diseases of wild mammals, second edition. Ames (IA): Iowa State University Press, 478-520.
Dubey J, Rigoulet J, Lagourette P, George C, Longeart L, and LeNet J. 1996. Fatal transplacental neosporosis in a deer (Cervus eld isiamensis). J Parasitol 82:338- 339.
Dubey J, Schares G, and Ortega-Mora L. 2007. Epidemiology and control of neosporosis and Neospora caninum. Clin Microbiol Rev 20:323-367.
Dubey J, Thorne E, and Sharma S. 1980. Experimental toxoplasmosis in elk (Cervus canadensis). Am J Vet Res 41:792-793.
Dubey J, Thorne E, and Williams E. 1982. Induced toxoplasmosis in pronghorns and mule deer. J Am Vet Med Assoc 181:1263-1267. 85
Dubey JP, and Thulliez P. 2005. Prevalence of antibodies to Neospora caninum in wild animals. J Parasitol 91:1217-1218. 10.1645/GE-576R.1
Dulac G, Dubuc C, Myers D, Afshar A, Taylor E, Ward D, and Sterritt W. 1989. Incursion of bluetongue virus type 11 and epizootic hemorrhagic disease of deer type 2 for two consecutive years in the Okanagan Valley. Can Vet J 30:351.
Dulac G, Sterritt W, Dubuc C, Afshar A, Myers D, Taylor E, Jamieson B, and Martin M. 1992. Incursions of orbiviruses in Canada and their serologic monitoring in the native animal population between 1962 and 1991. In: Walton T, and Osburn B, eds. Bluetongue, African horse sickness, and related orbiviruses: Proceedings of the Second International Symposium. Boca Raton (FL): CRC Press LLC, 120- 127.
Duquette J, Belant J, Beyer D, and Svoboda N. 2012. Comparison of pregnancy detection methods in live white-tailed deer. Wildlife Society Bulletin 36:115-118.
Dyer O, Lincoln R, and Peatt A. 1984. Mule deer and white-tailed deer conception rates in the Okanagan region of British Columbia: 1981-1983. Internal Working Report, Ministry of Environment. Penticton (BC). 31 p.
Ewen J, Armstrong D, and Parker K. 2012. Reintroduction biology: Integrating science and management. New York (NY): Wiley-Blackwell.
Ferroglio E, and Rossi L. 2001. Prevalence of Neospora caninum antibodies in wild ruminants from the Italian Alps. Vet Rec 148:754-755.
Fletch A, and Karstad L. 1971. Studies on the pathogenesis of experimental epizootic hemorrhagic disease of white-tailed deer. Canadian Journal of Comparative Medicine 35:224-229.
Foley A, Pierce B, Hewitt D, DeYoung R, Campbell T, Hellickson M, Field J, Mitchell S, Lockwood M, and Miller K. 2008. Survival and movements of translocated white-tailed deer in South Texas. Proceedings of the Annu Conf SEAFWA. p 25- 30.
86
Foster N, Breckon R, Luedke A, and Jones R. 1977. Transmission of two strains of epizootic hemorrhagic disease virus in deer by Culicoides variipennis. J Wildl Dis 13:9-16.
Gaffuri A, Giacometti M, Tranquillo V, Magnino S, Cordioli P, and Lanfranchi P. 2006. Serosurvey of roe deer, chamois and domestic sheep in the central Italian Alps. J Wildl Dis 42:685-690.
Gerry P, and Michael J. 2002. Experimental design and data analysis for biologists: Cambridge University Press.
Gibbs E, and Greiner E. 1989. Bluetongue and epizootic hemorrhagic disease. In: Monath T, ed. The arboviruses: Epidemiology and ecology, Volume 2. Boca Raton (FL): CRC Press.
Gondim L. 2006. Neospora caninum in wildlife. Trends Parasitol 22:247-252.
Gortázar C, Ferroglio E, Höfle U, Frölich K, and Vicente J. 2007. Diseases shared between wildlife and livestock: a European perspective. Eur J Wildl Res 53:241– 256.
Green D, Askins G, and West P. 1997. Public Opinion: Obstacle or Aid to Deer Management? Wildl Soc Bull 25:367-370.
Griebel P, Hill K, and Stookey J. 2014. How stress alters immune responses during respiratory infection. Anim Health Res Rev 15:161-165.
Griffith B, Scott JM, Carpenter JW, Reed C. 1993. Animal Translocations and Potential Disease Transmission. Journal of Zoo and Wildlife Medicine 24:231-236.
Gutiérrez-Expósito D, Ortega-Mora L, Gajadhar A, García-Lunar P, Dubey J, and Alvarez-García G. 2012. Serological evidence of Besnoitia spp. infection in Canadian wild ruminants and strong cross-reaction between Besnoitia besnoiti and Besnoitia tarandi. Vet Parasitol 190:19-28.
87
Guzman E, and Taylor G. 2015. Immunology of bovine respiratory syncytial virus in calves. Mol Immunol 66:48-56.
Haddad J, Dohoo I, and VanLeewen J. 2005. A review of Neospora caninum in dairy and beef cattle--a Canadian perspective. Can Vet J 46:230-243.
Harder J, and Kirkpatrick R. 1994. Physiological methods in wildlife research. In: Bookhout T, ed. Research and management techniques for wildlife and habitats. Bethesda (MD): The Wildlife Society, 275-306.
Hartley M, and Sainsbury A. 2017. Methods of disease risk analysis in wildlife translocations for conservation purposes. EcoHealth 14:16–29.
Heesterbeek H, Anderson R, Andreasen V, Shweta Bansal S, De Angelis D, Dye C, Eames K, Edmunds J, Frost S, Funk S, Hollingsworth D, House T, Isham V, Klepac P, Lessler J, Lloyd-Smith J, Metcalf J, Mollison D, Pellis L, Pulliam J, Roberts M, Viboud C, and Collaboration INII. 2015. Modeling infectious disease dynamics in the complex landscape of global health. Science 347.
Hoff G, and Trainer D. 1972. Bluetongue virus in pronghorn antelope. Am J Vet Res 33:1013-1016.
Hoff G, and Trainer D. 1978. Bluetongue and epizootic hemorrhagic disease viruses: their relationship to wildlife species. Adv Vet Sci Comp Med 22:111-132.
Hoffman G. 1970. Intercontinental and trans-continental dissemination and transfaunation of fish parasites with emphasis on whirling disease. In: Snieszko S, ed. A symposium on diseases of fishes and shellfishes. Washington (DC): American Fisheries Society, 69-81.
Howard C. 2015. Urban deer translocation, September 2015 update. Utah Division of Wildlife Resources
Huang F, Cockrell D, Stephenson T, Noyes J, and Sasser R. 1999. Isolation, purification, and characterization of pregnancy-specific protein B from elk and moose placenta. Biol Reprod 61:1056-1061.
88
Huang F, Cockrell D, Stephenson T, Noyes J, and Sasser R. 2000. A serum pregnancy test with a specific radioimmunoassay for moose and elk pregnancy-specific protein B. J Wildl Manag 64:492-499.
IDvet. 2013. ID Screen® Toxoplasma indirect multi-species. Grabels, France. 4 p. Available at http://www.id-vet.com (accessed 10 October 2017).
Ingebrigtsen D, Ludwig J, and McClurkin A. 1986. Occurrence of antibodies to the etiologic agents of infectious bovine rhinotracheitis, parainfluenza 3, leptospirosis, and brucellosis in white-tailed deer in Minnesota. J Wildl Dis 22:83- 86.
Institute of Medicine, Committee on Emerging Microbial Threats to Health in the 21st Century. 2003. Factors in emergence. In: Smolinski M, Hamburg M, and Lederberg J, eds. Microbial Threats to Health: Emergence, Detection, and Response. Washington (DC): National Academies Press.
IUCN. 1987. The IUCN position statement on the translocation of living organisms: Introductions, reintroductions and re-stocking. Gland, Switzerland: IUCN Library.
IUCN/SSC. 2013. Guidelines for reintroductions and other conservation translocations. Gland, Switzerlan: IUCN Species Survival Commission.
Jakob E, Marshall S, and Uetz G. 1996. Estimating fitness: A comparison of body condition indices. Oikos 77:61-67.
Jakob-Hoff R, MacDiarmid S, Lees C, Miller P, Travis D, and Kock R. 2014. Manual of procedures for wildlife disease risk analysis. Paris, France: IUCN Species Survival Commission, the World Organisation for Animal Health (OIE).
Jones J, and Witham J. 1990. Post-translocation survival and movements of metropolitan white-tailed deer. Wildl Soc Bull 18:434-441.
Karstad L. 1981. Bovine virus diarrhea. In: Davis JW, Karstad LH, Trainer DO, ed Infectious disease of wild mammals Ames: Iowa State University Press, 209-211.
89
Keesing F, Belden L, Daszak P, Dobson A, Harvell C, Holt R, Hudson P, Jolles A, Jones K, Mitchell C, Myers S, Bogich T, and Ostfeld R. 2010. Impacts of biodiversity on the emergence and transmission of infectious diseases. Nature 468:647-652.
Kilpatrick H, LaBonte A, and Stafford K. 2014. The relationship between deer density, tick abundance, and human cases of Lyme disease in a residential community. J Med Entomol 51:777-784.
Kiracofe G, Wright J, Schalles R, Ruder C, Parish S, and Sasser R. 1993. Pregnancy- specific protein B in serum of postpartum beef cows. J Anim Sci 71:2199-2205.
Kocan A, Castro A, Shaw M, and Rogers S. 1987. Bluetongue and epizootic hemorrhagic disease in white-tailed deer from Oklahoma: Serologic evaluation and virus isolation. Am J Vet Res 48:1048-1049.
Kock R, Woodford M, and Rossiter P. 2010. Disease risks associated with the translocation of wildlife. Rev Sci Tech 29:329-350.
Kruse H, Kirkemo A, and K. H. 2004. Wildlife as source of zoonotic infections. Emerging Infectious Diseases 10:2067-2072.
Lankester M. 2001. Extrapulmonary lungworms of cervids. In: Samuel WM, Pybus MJ, Kocan AA (eds.). Parasitic diseases of wild mammals. 2nd ed. Ames (IA): Iowa State University Press. p. 228-278.
Lapointe J, Hedges J, Woods L, Reubel G, and MacLachlan N. 1999. The adenovirus that causes hemorrhagic disease of black-tailed deer is closely related to bovine adenovirus-3. Arch Virol 144:393-396.
Lehrer E, Fredebaugh S, Schooley R, and Mateus-Pinilla N. 2010. Prevalence of antibodies to Toxoplasma gondii in Woodchucks across an urban-rural gradient. Journal of Wildlife Diseases 46:977–980.
Leighton F. 2002. Health risk assessment of the translocation of wild animals. Rev Sci Tech 21:187-195.
90
Lillie L. 1974. The bovine respiratory disease complex. Can Vet J 15:233-242.
Lindsay D, Little S, and Davidson W. 2002. Prevalence of antibodies to Neospora caninum in white-tailed deer, Odocoileus virginianus, from the southeastern United States. J Parasitol 88:415-417.
Lindsay D, McKown R, DiCristina J, Jordan C, Mitchell S, Oates D, and Sterner M. 2005. Prevalence of agglutinating antibodies to Toxoplasma gondii in adult and fetal mule deer (Odocoileus hemionus) from Nebraska. Journal of Parasitology 91:1490-1491.
Loker C. 1996. Human dimensions of suburban wildlife management: Insights from three areas of New York State M.S. Cornell University.
Loudon A, McNeilly A, and Milne J. 1983. Nutrition and lactational control of fertility in red deer. Nature 302:145-147.
MacDiarmid S. 1997. Risk Analysis, international trade and animal health. In: Molak V, ed. Fundamentals of risk analysis and risk management. Boca Raton (FL): CRC Lewis Publishers, 377–387.
Malmsten J, Jakubek E, and Björkman C. 2011. Prevalence of antibodies against Toxoplasma gondii and Neospora caninum in moose (Alces alces) and roe deer (Capreolus capreolus) in Sweden. Vet Parasitol 177:275-280.
Mars M, Bruschke C, and van Oirschot J. 1999. Airborne transmission of BHV1, BRSV, and BVDV among cattle is possible under experimental conditions. Vet Microbiol 66:197-207.
Mathieu A, Carlsson A, and Kutz S. 2015. The Sahtu Wildlife Health Monitoring Program: Summary report of 2013-2014 pathogen screening. Government of the Northwest Territories 30 p.
Mathieu A, Caulkett N, Stent PM, and Schwantje HM. 2017. Capture of free-ranging mule reer (Odocoileus hemionus) with a combination of medetomidine, azaperone, and alfaxalone. Journal of Wildlife Diseases 53:296-303. 10.7589/2016-09-210 91
Mathieu A, Flint M, Stent P, Schwantje H, and Wittum T. 2018. Health assessment of urban and non-urban free-ranging mule deer (Odocoileus hemionus) in the Kootenay region, British Columbia, Canada. Under revision for publication in PeerJ.
Mauroy A, Guyot H, De Clercq K, Cassart D, Thiry E, and Saegerman C. 2008. Bluetongue in captive yaks. Emerg Infect Dis 14:675-667.
McCabe R, and McCabe T. 1984. Of slings and arrows: an historical retrospection. In: Halls L, ed. White-tailed deer: Ecology and management. Harrisburg (PA): Stackpole Books, 19-72.
McCullough D, Jennings K, Gates N, Elliot B, and Didonato J. 1997. Overabundant deer populations in California. Wildl Soc Bull 25:478-483.
McMartin DA, Snodgrass DR, and Corrigall W. 1977. Bovine virus diarrhoea antibody in a Scottish red deer. Vet Rec 100:187.
Medrano C, Boadella M, Barrios H, Cantú A, García Z, de la Fuente J, and Gortazar C. 2012. Zoonotic pathogens among white-tailed deer, northern Mexico, 2004-2009. Emerg Infect Dis 18:1372-1374.
Miller BF, Osborn DA, Lance WR, Howze MB, Warren RJ, and Miller KV. 2009. Butorphanol-azaperone-medetomidine for immobilization of captive white-tailed deer. Journal Wildlife Diseases 45:457-467. 10.7589/0090-3558-45.2.457
Miller M, and Thorne E. 1993. Captive cervids as potential sources of disease for North America’s wild cervid populations: avenues, implications and preventive management. Transactions of the North American Wildlife and Natural Resources Conference 58:460-467.
Miller P. 2007. Tools and techniques for disease risk assessment in threatened wildlife conservation programmes. Int Zoo Yearb 41:38-51.
Moore D, and Lee V. 1972. Antigenic relationship between the virus of epizootic haemorrhagic disease of deer and bluetongue virus. Brief report. Arch Gesamte Virusforsch 37:282-284. 92
Morton J, Evermann J, and Dietrich R. 1990. Experimental infection of reindeer with bovine viral diarrhea virus. Rangifer 10:75-77.
Mule Deer Working Group. 2015. New Mexico mule deer translocation update. Summary of the mule deer working group 2015 meeting. New Mexico Department of Game and Fish.
Mule Deer Working Group. 2016. 2016 Range-wide status of mule deer and black-tailed deer. Available at https://www.wafwa.org/Documents%20and%20Settings/37/Site%20Documents/W orking%20Groups/Mule%20Deer/Publications2/2016_Mule_Deer_and_BTD_Sta tus_Update_Final.pdf. (accessed 10 October 2016.
Mumford R, and Whitaker J. 1982. Mammals of Indiana. Bloomington (IN): Indiana University Press.
Murata H, and Hirose H. 1991. Suppression of bovine lymphocyte and macrophage functions by sera from road-transported calves. Br Vet J 147:455-462.
Murray J, and Trainer D. 1970. Bluetongue virus in North American elk. J Wildl Dis 6:144-148.
Murray N, MacDiarmid S, Wooldridge M, Gummow B, Morley R, Weber S, Giovannini A, and Wilson D. 2004. Handbook on import risk analysis for animals and animal products-introduction and qualitative risk analysis. Paris, France: World Organisation for Animal Health (OIE).
Myers W, Foreyt W, Talcott P, Evermann J, and Chang W. 2015. Serologic, trace element, and fecal parasite survey of free-ranging, female mule deer (Odocoileus hemionus) in eastern Washington, USA. J Wildl Dis 51:125-136.
Nettles S, Hylton S, Stallknecht D, and Davidson W. 1992. Epidemiology of epizootic hemorrhagic disease viruses in wildlife in the USA. In: Walton T, and Osbourn B, eds. Bluetongue, African horse sickness, and related orbiviruses: Proceedings of the Second International Symposium. Boca Raton (FL): CRC Press LLC, 238- 248.
93
Nettles V. 1988. Wildlife relocation: Disease implications and regulations. Proceedings of the 37th Ann Conf Wildlife Disease Association. Athens (GA). p 52.
Nettles V, and Stallknecht D. 1992. History and progress in the study of hemorrhagic disease of deer. Proceedings of the 57th North American Wildlife and Natural Resources Conference. p 499-516.
Noon T, Wesche S, Cagle D, Mead D, Bicknell E, Bradley G, Riplog-Peterson S, Edsall D, and Reggiardo C. 2002. Hemorrhagic disease in bighorn sheep in Arizona. J Wildl Dis 38:172-176.
O’Bryan M, and McCullough D. 1985. Survival of black-tailed deer following relocation in California. J Wildl Manage 49:115-119.
ODFW. 2003. Fast facts about adenovirus hemorrhagic disease. Available at http://www.dfw.state.or.us/wildlife/health_program/andevirus/ (accessed 10 June 2016).
OIE. 2015. Report of bluetongue in Canada. Available at http://www.oie.int/wahis_2/public/wahid.php/Reviewreport/Review?reportid=185 9/ (accessed May 2016.
OIE. 2017. Manual of diagnostic tests and vaccines for terrestrial animals 2017. Chapter 2.4.7. Bovine Viral Diarrhoea. Available at http://www.oie.int/en/international- standard-setting/terrestrial-manual/access-online/ (accessed 10 October 2017).
Oldroyd B. 1999. Coevolution while you wait: Varroa jacobsoni, a new parasite of western honeybees. Trends Ecol Evol 14:312-315.
Onderka D, Rawluk S, and Wishart W. 1988. Susceptibility of Rocky Mountain bighorn sheep and domestic sheep to pneumonia induced by bighorn and domestic livestock strains of Pasteurella haemolytica. Can J Vet Res 52:439-444.
Parsonson I. 1990. Pathology and pathogenesis of bluetongue infections. Curr Top Microbiol Immunol 182:119-141.
94
Pasick J, Handel K, Zhou E, Clavijo A, Coates J, Robinson Y, and Lincoln B. 2001. Incursion of epizootic hemorrhagic disease into the Okanagan Valley, British Columbia in 1999. Can Vet J 42:207-209.
Peig J, and Green A. 2009. New perspectives for estimating body condition from mass/length data: the scaled mass index as an alternative method. Oikos 118:1883-1891.
Phoofolo P. 1993. Epidemics and revolutions: the rinderpest epidemic in late nineteenth- century southern Africa. Past & Present 138:112-143.
Piesman J, Spielman A, Etkind P, Ruebush Tn, and Juranek D. 1979. Role of deer in the epizootiology of Babesia microti in Massachusetts, USA. J Med Entomol 15:537- 540.
Pruvot M, Kutz S, van der Meer F, Musiani M, Barkema H, and Orsel K. 2014. Pathogens at the livestock-wildlife interface in Western Alberta: does transmission route matter? Veterinary Research 45.
Purse B, Mellor P, Rogers D, Samuel A, Mertens P, and Baylis M. 2005. Climate change and the recent emergence of bluetongue in Europe. Nature Reviews Microbiology 3:171-181.
Pybus M, Ravi M, and Pollock C. 2014. Epizootic hemorrhagic disease in Alberta, Canada. J Wildl Dis 50:720-722.
Rickard L, Siefker C, Boyle C, and Gentz E. 1999. The prevalence of Cryptosporidium and Giardia spp. in fecal samples from free-ranging white-tailed deer (Odocoileus virginianus) in the southeastern United States. Journal of Veterinary Diagnostic Investigation 11:65-72.
Rijks J, Roest H, van Tulden P, Kik M, IJzer J, and Gröne A. 2011. Coxiella burnetii infection in roe deer during Q fever epidemic, the Netherlands. Emerg Infect Dis 17:2369-2271.
Robinette W, and Gashwiler J. 1950. Breeding season, productivity, and fawning period of the mule deer in Utah. J Wildl Manage 14:457-446. 95
Rodriguez M, Hooghuis H, and Castaño M. 1992. African horse sickness in Spain. Vet Microbiol 33:129-142.
Ross R, Stec L, Werner J, Blumenkranz M, Glazer L, and Williams G. 2001. Presumed acquired ocular toxoplasmosis in deer hunters. Retina 21:226-229.
Roth E. 1970. Leptospirosis. In: Davis J, Karstad L, and Trainer D, eds. Infectious diseases of wild mammals. Ames (IA): Iowa State University Press, 293-303.
Roug A, Swift P, Torres S, Jones K, and Johnson C. 2012. Serosurveillance for livestock pathogens in free-ranging mule deer (Odocoileus hemionus). PLoS One 7:e50600.
Sacks J, Delgado D, Lobel H, and Parker R. 1983. Toxoplasmosis infection associated with eating undercooked venison. American Journal of Epidemiology 118:832– 838
Sainsbury A, and Vaughan-Higgins R. 2012. Analyzing disease risks associated with translocations. Conserv Biol 26:442-452.
Sasser R, and Ruder C. 1987. Detection of early pregnancy in domestic ruminants. J Reprod Fertil Suppl 34:261-271.
Sasser R, Ruder C, Ivani K, Butler J, and Hamilton W. 1986. Detection of pregnancy by radioimmunoassay of a novel pregnancy-specific protein in serum of cows and profile of serum concentrations during gestation. Biology of Reproduction 35:936- 942.
Schwantje H, Macbeth B, Kutz S, and Elkin B. 2014. British Columbia Boreal Caribou Health Program, Progress Report: Year 1 (November 1, 2013 - December 31, 2014). The British Columbia Boreal Caribou Health Research Program Working Group 78 p.
Schwantje H, Macbeth B, Kutz S, and Elkin B. 2016. British Columbia Boreal Caribou Health Program, Progress Report: Year 2 (February 1, 2015 - March 31, 2016). The British Columbia Boreal Caribou Health Research Program Working Group 49 p.
96
Sellers R, and Maarouf A. 1991. Possible introduction of epizootic hemorrhagic disease of deer virus (serotype 2) and bluetongue virus (serotype 11) into British Columbia in 1987 and 1988 by infected Culicoides carried on the wind. Can J Vet Res 55:367-370.
Severinghaus C. 1949. Tooth development and wear as criteria of age in white-tailed deer. Journal of Wildlife Management 13:195–216.
Shallow J, Hurley M, Monteith K, and Bowyer R. 2015. Cascading effects of habitat on maternal condition and life-history characteristics of neonatal mule deer. Journal of Mammalogy 96:194-205.
Sifton E. 2005. Disease risk assessment for an experimental captive breeding program of mountain caribou in British Columbia. British Columbia Ministry of Environment, Lands and Parks 83 p.
Soldati S, Kiupel M, Wise A, Maes R, Botteron C, and Robert N. 2004. Meningoencephalomyelitis caused by Neospora caninum in a juvenile fallow deer (Dama dama). J Vet Med A 51:280-283.
Sorden S, Woods L, and Lehmkuhl H. 2000. Fatal pulmonary edema in white-tailed deer (Odocoileus virginianus) associated with adenovirus infection. J Vet Diagn Invest 12:378-380.
Spraker T, Collins J, Adrian W, and Olterman J. 1986. Isolation and serologic evidence of a respiratory syncytial virus in bighorn sheep from Colorado. J Wildl Dis 22:416-418.
Stallknecht D, and Howerth E. 2004. Epidemiology of bluetongue and epizootic haemorrhagic disease in wildlife: Surveillance methods. Vet Ital 40:203-207.
Stephen C. 2014. Toward a modernized definition of wildlife health. J Wildl Dis 50:427- 430.
Swihart R, Picone P, DeNicola A, and Cornicelli L. 1995. Capture and handling techniques for urban deer control. In: McAninch J, ed. Urban deer: A manageable
97
resource? : Proceedings of the Symposium of the North Central Section, The Wildlife Society, 35-44.
Tabachnick W. 1996. Culicoides variipennis and bluetongue-virus epidemiology in the United States. Annu Rev Entomol 41:23-42.
Tavernier P, Sys S, De Clercq K, De Leeuw I, Caij A, De Baere M, De Regge N, Fretin D, Roupie V, Govaerts M, Heyman P, Vanrompay D, Yin L, Kalmar I, Suin V, Brochier B, Dobly A, De Craeye S, Roelandt S, Goossens E, and Roels S. 2015. Serologic screening for 13 infectious agents in roe deer (Capreolus capreolus) in Flanders. Infect Ecol Epidemiol 5.
Telford Sr, Mather T, Moore S, Wilson M, and Spielman A. 1988. Incompetence of deer as reservoirs of the Lyme disease spirochete. Am J Trop Med Hyg 39:105-109.
Tessaro SV, Carman PS, and Deregt D. 1999. Viremia and virus shedding in elk infected with type 1 and virulent type 2 bovine viral diarrhea virus. J Wildl Dis 35:671- 677. 10.7589/0090-3558-35.4.671
The Spokesman Review. 2015. Bluetongue disease killing deer in the Northwest. Available at http://www.bendbulletin.com/outdoors/3574859-151/bluetongue- disease-killing-deer-in-the-northwest. (accessed May 2016.
Thomas F, Skinner D, and Samagh B. 1982. Evidence for bluetongue virus in Canada: 1976-1979. Can J Comp Med 46:350-353.
Thorne E. 1982. Bluetongue and epizootic hemorrhagic disease. In: Thorne E, Kingston N, Jolley W, and Bergstrom R, eds. Diseases of wildlife in Wyoming. Cheyenne (WY): Wyoming Game and Fish Department, 5-10.
Thorne E, Williams E, Spraker T, Helms W, and Segerstrom T. 1988. Bluetongue in free- ranging pronghorn antelope (Antilocapra americana) in Wyoming: 1976 and 1984. J Wildl Dis 24:113-119.
Thorsen J, and Henderson J. 1971. Survey for antibody to infectious bovine rhinotracheitis (IBR), bovine virus diarrhea (BVD) and parainfluenza 3 (PI3) in moose sera. J Wildl Dis 7:93-95. 98
Tiemann J, Souza S, Rodrigues A, Duarte J, and Gennari S. 2005. Environmental effect on the occurrence of anti-Neospora caninum antibodies in pampas deer (Ozotoceros bezoarticus). Vet Parasitol 134:73-76.
Trees A, and Williams D. 2005. Endogenous and exogenous transplacental infection in Neospora caninum and Toxoplasma gondii. Trends Parasitol 21:558-561.
Triezenberg H, Lauber T, and Knuth B. 2012. Policy learning and deer co-management in New York State: Insights from newspaper coverage. J Wildl Manage 76:1317- 1325.
Urban-Chmiel R, Wernickia A, Stęgierskaa D, Marczukb J, Rolac J, Sochac W, and Valverde Piedrad J. 2017. Detection of BHV-1 and BRSV viruses in European bison in the Białowieża forest: A preliminary study. J Appl Anim Res 45:170-172.
Van Campen H, and Early G. 2001. Orthomyxovirus and paramyxovirus infections. In: Williams E, and Barker I, eds. Infectious diseases of wild mammals, third edition. Ames (IA): Iowa State University Press, 271-227.
Van Campen H, Williams ES, Edwards J, Cook W, and Stout G. 1997. Experimental infection of deer with bovine viral diarrhea virus. Journal of Wildlife Diseases 33:567-573. 10.7589/0090-3558-33.3.567
Van der Poel W, Kramps J, Middel W, Van Oirschot J, and Brand A. 1993. Dynamics of bovine respiratory syncytial virus infections: a longitudinal epidemiological study in dairy herds. Arch Virol 133:309-321.
Verme L. 1969. Reproductive patterns of white-tailed deer related to nutritional plane. J Wildl Manag 33:881-887.
Viggers K, Lindenmayer D, and Spratt D. 1993. The importance of disease in reintroduction programs. Wildl Res 20:687-698.
VMRD. Neospora caninum antibody test kit, cELISA. Pullman, Washington. 4 p. Available at http://www.vmrd.com/ (accessed 10 October 2017).
99
Walker S, Bosch J, James T, Litvintseva A, Valls J, Piña S, García G, Rosa G, Cunningham A, Hole S, Griffiths R, and Fisher M. 2008. Invasive pathogens threaten species recovery programs. Curr Biol 18:853-854.
Wapenaar W, Barkema H, Schares G, Rouvinen-Watt K, Zeijlemaker L, Poorter B, O’Handley R, Kwok O, and Dubey J. 2007. Evaluation of four serological techniques to determine the seroprevalence of Neospora caninum in foxes (Vulpes vulpes) and coyotes (Canis latrans) on Prince Edward Island, Canada. Vet Parasitol 145:51-58.
Wapenaar W, Jenkins M, O’Handley R, and Barkema H. 2006. Neospora caninum-like oocysts observed in feces of free-ranging red foxes (Vulpes vulpes) and coyotes (Canis latrans). J Parasitol 92:1270-1274.
WDFW. 2017. Adenovirus hemorrhagic disease (AHD). Available at http://wdfw.wa.gov/conservation/health/ahd/ (accessed 20 September 2017).
Whisson D, Holland G, and Carlyon K. 2012. Translocation of overabundant species: Implications for translocated individuals. J Wildl Manag 76:1661-1669.
Whitney G. 1994. From coastal wilderness to fruited plain: A history of environmental change in temperate North America from 1500 to the present. Cambridge (MA): Cambridge University Press.
Williams E, Mills K, Kwiatkowski D, Thorne E, and Boerger-Fields A. 1994. Plague in a black-footed ferret (Mustela nigripes). J Wildl Dis 30:581-585.
Winter M. 1999. Deer Ahead! ALS NEWS, Cornell University, College of Agriculture and Life Sciences, p. 1.
Wittmann EJ, and Baylis M. 2000. Climate change: effects on culicoides-transmitted viruses and implications for the UK. The Veterinary Journal 160:107-117. 10.1053/tvjl.2000.0470
Wobeser G. 2006. Essentials of disease in wild animals. Ames (IA): Blackwell Publishing.
100
Wolf K, DePerno C, Jenks J, Stoskopf M, Kennedy-Stoskopf S, Swanson C, Brinkman T, Osborn R, and Tardiff J. 2008. Selenium status and antibodies to selected pathogens in white-tailed deer (Odocoileus virginianus) in Southern Minnesota. J Wildl Dis 44.
Wolfe B. 2003. Toxoplasmosis. In: Fowler M, and Miller R, eds. Zoo and wild animal medicine, Volume 5. St. Louis (MO): Elsevier, 745-749.
Woodford M. 2000. Quarantine and health screening protocols for wildlife prior to translocation and release into the wild. : Published jointly by the IUCN Species Survival Commission’s Veterinary Specialist Group (Gland, Switzerland), the Office International des Epizooties (OIE) (Paris, France), Care for the Wild (UK), and the European Association of Zoo and Wildlife Veterinarians (Switzerland).
Woodford M, and Rossiter P. 1993. Disease risks associated with wildlife translocation projects. Revue Scientifique et Technique 12:115-135.
Wooding F, Roberts R, and Green J. 2005. Light and electron microscope immunocytochemical studies of the distribution of pregnancy associated glycoproteins (PAGs) throughout pregnancy in the cow: possible functional implications. Placenta 26:807-827.
Woods L, Anderson M, Swift P, and Sverlow K. 1994. Systemic neosporosis in a California black-tailed deer (Odocoileus hemionus columbianus). J Vet Diagn Invest 6:508-510.
Woods L, Lehmkuhl H, Swift P, Chiu P, Hanley R, Nordhausen R, Stillian M, and Drew M. 2001. Experimental adenovirus hemorrhagic disease in white-tailed deer fawns. J Wildl Dis 37:153-158.
Woods L, Swift P, Barr B, Horzinek M, Nordhausen R, Stillian M, Patton J, Oliver M, Jones K, and MacLachlan N. 1996. Systemic adenovirus infection associated with high mortality in mule deer (Odocoileus hemionus) in California. Vet Pathol 33:125-132.
101