The IUCN/SSC Canid Specialist Group's
African Wild Dog Status Survey and Action Plan (1997)

Chapter 8
Research and Monitoring:
Information for Wild Dog Conservation

by Joshua R. Ginsberg & Rosie Woodroffe

While a great deal of information about wild dog ecology has become available recently, further research will allow more effective wild dog management.

Such information will help to determine the combination of husbandry practices, local legislation, compensation and education needed to allow wild dogs and people to coexist.

Background

In previous chapters, we have formulated plans for wild dog conservation using the best information available to us. However, in several cases we have found that more research would enhance the creation and implementation of effective management strategies. A great deal of research has been carried out on wild dogs recently (See Appendix3), so that wildlife managers are now much better equipped to conserve wild dogs than they were ten, or even five years ago. Nevertheless, there are still areas where more information would be extremely valuable. In this chapter, we summarize the research we feel would facilitate wild dog conservation. Techniques for carrying out some of these projects are described in Appendix2.

This chapter is divided into sections, dealing with broad research topics. We have arranged these in an order which reflects the structure of the Action Plan, rather than any priority. However, within each section we have highlighted the topics that we consider need most urgent attention.

Taxonomy

Despite extensive research, some questions remain about the taxonomic status of wild dogs and what, if anything, constitutes a wild dog sub-species. Resolution of this question is important for two reasons. First, the maintenance of genetic diversity is an important component of biodiversity conservation. Genetic analyses indicate that some populations - such as the one in Kruger National Park - may contain genotypes not found elsewhere (Chapter2). Analysis of DNA taken from a museum skin suggests that wild dogs in West Africa might also be genetically distinct from those in East and southern Africa (Chapter2, Roy et al. 1994). Such distinctiveness may place a high conservation value on certain populations, and yet no research has been carried out on the genetics of several wild dog populations, especially those in West and central Africa.

While we question the universal value of reintroduction as a conservation tool for wild dogs (Chapter7), wherever possible, wild dogs released into a given area should be as similar as possible, genetically and morphologically, to those dogs which originally occured at the release site. In practice, it may be difficult to determine the genotypes or phenotypes appropriate for specific release sites without better information. For example, in Chapter7 we suggested that Selous might represent a source of wild-caught animals for translocation to other parts of East Africa, but recent work suggests that this population is genetically closer to those in southern Africa than to others in East Africa (Chapter2). Similarly, re-establishment of populations in West Africa is a priority, but we know little about the genetic and phenotypic characteristics of West African wild dogs.

The importance of such genetic considerations to reintroduction programmes must be considered in this and other species - if no animals of the appropriate genotype are available to reintroduction programmes, is the release of animals with 'foreign' genotypes an acceptable alternative? The answer to this question depends, in part, upon the adaptive basis of genotypic variation. Since animals with foreign genotypes might not be adapted to local ecological conditions, reintroduction programmes which used them could, theoretically, end in failure. It would be very difficult, however, for field projects to determine whether populations which differed in their genetic makeup also differed in their behaviour and physiology. Further morphological work on museum specimens might go some way to solving this problem.

Distribution

Wild dogs' status in East and southern Africa is fairly well known, but basic surveys are still needed in several other areas, especially central Africa. There is an accepted protocol for the use of photographic surveys to census wild dog numbers (e.g. Maddock & Mills 1994) but we lack simple, inexpensive, but effective mechanisms with which to carry out preliminary censuses or long-term monitoring of wild dog populations. Postal surveys, such as those presented in Chapter3, are effective tools for assessing status, but they cannot substitute for sustainable local efforts administered either by government departments or a local non-governmental organizations.

In Chapter9, we list country-by-country priorities for action, many of which include census and survey activities. Some of the highest priority activites are:

  1. Is there really a relict population of wild dogs in the Teffedest Mountains, Algeria? If this isolated population really exists, it is likely to be genetically (and perhaps ecologically) distinct from other populations, and would have a very high conservation value.
  2. What is the status of the wild dog populations in Cameroun and the Central African Republic? Very little is known about these populations, which may represent a reservoir of wild dogs in central Africa.
  3. What is the status of the wild dog population in southern Sudan? Little is known about this population, but it may be the source of wild dogs sighted recently in northern Uganda. If so, it could link the populations in southern Ethiopia and northern Kenya with those in central Africa.
  4. What is the status of the wild dog population in southern Ethiopia?
  5. Where wild dogs are sighted fairly regularly, more intensive surveys would be useful. Photographic surveys based (in part) upon pictures taken by tourists have been set up in a number of countries, including South Africa (Maddock & Mills 1994), Tanzania (Burrows 1995; Creel & Creel 1993) and Zimbabwe (J.R.G., Unpublished data). Countries such as Kenya and Zimbabwe have a fairly large volume of tourists visiting networks of protected areas. In these countries, nationwide photographic surveys could help to give a better estimate of wild dog numbers, and to assess the degree to which animals move between protected areas.
  6. Even where such surveys are already in place, better coordination between projects in neighbouring countries, more involvement of local people, and better advertising of such projects in both the range states and the tourists' home countries would all contribute to the accumulation of more useful data.

Ecological Monitoring

Wild dogs' conservation requirements are now much better known than they were 10 years ago, principally as a result of ecological research on populations in several parts of Africa. Such work has identified the main threats to wild dog populations, and therefore forms the basis of this Action Plan. Continued study of these populations will contribute to wild dog conservation biology by monitoring exisiting threats and, perhaps, by identifying new ones. They will also help to determine the factors which cause wild dog populations to rise and fall in different areas.

Disease is one threat which requires continued monitoring. Not only will this will allow the identification of new disease risks that emerge, but repeated samples taken from the same population - or, better still, from the same individual - will provide extremely important data upon disease dynamics within wild dog populations. For example, the pattern of seroprevalence in different age classes can help to determine whether animals are facing a chronic disease threat, or whether seropositive animals simply represent a record of past epidemics (Thrusfield 1986). No wild dog should ever be immobilized without being screened for disease. Any wild dog found dead should be necropsied and screened for disease - even if disease is not suspected as a cause of death. Such monitoring will help to determine the threats posed by diseases such as parvovirus, adenovirus, coronavirus and herpesvirus, whose impact on wild dog populations is not yet clear (Chapter4). Wherever possible, domestic dogs and wild carnivores living in wild dog areas should also be screened for disease.

Long-term ecological monitoring will also help to determine the resilience of wild dog populations. Ecological studies have established that competition with larger predators is likely to limit wild dog numbers over the long term (Creel & Creel 1996; Fuller et al. 1992; Mills & Biggs 1993), but they have not yet determined how wild dog populations recover from episodes of high mortality. Our simulations of wild dog populations suggest that their large litter sizes should equip them to recover rapidly from perturbation (Chapter5), but empirical studies have not yet documented any such recoveries. Empirical evidence would help to test the reliability of our simulations and, therefore, the validity of our conclusions. The recent loss of several whole packs in Northern Botswana may provide an opportunity to monitor the recovery of a study population.

Conflicts between Wild Dogs and People

Despite the fact that persecution remains one of the most important threats faced by wild dog populations, little is known about the precise circumstances under which people come into conflict with wild dogs.

  1. When do wild dogs stop ignoring livestock (as they did in the area of the Masai Mara, Fanshawe 1989; Fuller & Kat 1990) and start to kill them? Are livestock taken only when wild ungulate prey have been depleted? Are more livestock taken during the denning period, when the dogs' movements are restricted?
  2. What tactics do wild dogs use to hunt livestock? This would help with the development of techniques to protect livestock from wild dogs.
  3. How serious are the economic losses caused by wild dog predation? Are there persistent losses in some areas (e.g. on the borders of reserves with substantial wild dog populations), or are losses sporadic?
  4. What is the public attitude to wild dogs in areas that they use regularly? Does the local attitude reflect the real losses that wild dogs cause?
  5. Can husbandry techniques be modified to mitigate losses to wild dogs in areas where predation on livestock is a serious problem? Would confining livestock to bomas at night, or better-designed bomas, help to reduce losses? Would the use of guard dogs help (if disease could be controlled adequately)?
  6. Would compensation schemes help to reduce local peoples' hostility to wild dogs? Would funds be available, and could such schemes be implemented realistically?
  7. Where communal and private lands have been converted to wildlife use, wild dogs' prey species become a valuable commodity both for consumption and for game viewing. Some of these uses, such as photo-tourism, may benefit wild dogs but others, such as game ranching and hunting, may place wild dogs in real or perceived competition with humans for wild ungulates. Can we develop land-use zoning plans which provide a clear definition of where predators will, and will not, be tolerated?

Further research to answer these questions is a high priority for wild dog conservation, especially for populations that use livestock areas on the borders of reserves, and for those that persist outside protected areas.

Strategies for Disease Control

Disease represents a serious threat to several wild dog populations, but in no case are wildlife managers fully equipped to deal with the problem. Research is needed in several areas to help devise better strategies for disease control in wild dogs.

Protocols for Rabies Vaccination in Wild Dogs

Rabies has spilled over into wild dog populations in the past, and it is likely that this will happen again. For example, rabies is endemic in jackals and domestic dogs in many parts of Zimbabwe, with no immediate prospect of a control programme (Bingham 1995; Bingham et al. 1995). It may be just a question of time before wild dogs in Zimbabwe become infected. In the past, some researchers faced with proven risks of rabies infection have vaccinated wild dogs (Appendix1). However, the death from rabies of some of the vaccinated animals has led several authors to question the value of rabies vaccination as a tool in wild dog management (Burrows 1992; Burrows et al. 1994).

The rabies vaccination programmes that have been carried out on free-ranging wild dogs are discussed in detail in Appendix1. In summary, however, the most likely cause of the vaccine failures lies in the vaccination protocols used. Each wild dog was given only a single dose of vaccine. However, administration of single doses of inactivated rabies vaccine to wild dogs held in captivity in Tanzania failed to bring about seroconversion (Visee 1996), and preliminary vaccine trials in South Africa suggest that two doses must be given in order to achieve and maintain protective antibody levels (G.Thomson, pers. comm.). Further vaccine trials are urgently needed to determine the best protocol. In particular, they need to ask:

  1. Are two or more doses of vaccine, given 2-8 weeks apart, needed to establish high circulating levels of rabies neutralizing antibodies? How often must boosters be given thereafter?
  2. Does vaccination by dart produce as strong an immune response as vaccination of immobilized animals by hand?
  3. It has been suggested that handling stress could have compromised wild dogs' cell-mediated immune response to rabies infection (Burrows et al. 1994) - does vaccination induce a cell-mediated immune response? Cell-mediated immunity can be assayed in the laboratory from blood samples (Gerber et al. 1985; Jayakumar & Ramadass 1990).

The ultimate test of vaccine efficacy is challenge with a dose and strain of rabies virus known to be lethal to unvaccinated animals. However, establishing the necessary challenge conditions, followed by carrying out the challenge experiments themselves, would necessitate killing at least 20-30 captive wild dogs. The consensus of vets and biologists involved in research on rabies in wild dogs and other carnivores is that challenges would be both unnecessary and unethical - for this reason, applications for government licences to carry out such experiments would probably be unobtainable (M.Artois pers. comm.; S.Cleaveland, pers. comm.; G.Thomson, pers. comm.). Nevertheless, the experiments suggested above would answer most of the questions that have been raised concerning the efficacy of inactivated rabies vaccines, without the need for carrying out challenge experiments.

Vaccination of Wild Dogs against Canine Distemper Virus

Canine distemper may represent a serious threat to wild dog populations. However, experimental administration of live CDV vaccines to captive wild dogs has, on occasion, found them to be ineffective or even dangerous. More research is needed to answer the following questions:

  1. How serious is the risk of vaccine-induced distemper? While live CDV vaccines have induced distemper in several cases (Durchfeld et al. 1990; McCormick 1983; van Heerden et al. 1989), some captive facilities vaccinate their wild dogs routinely without reporting any ill effects (van Heerden 1986). No informed decision about further use of live CDV vaccines can be taken without detailed knowledge of how often they cause distemper, and the circumstances under which this occurs. For example, are adults as vulnerable as pups (all recorded cases of vaccine-induced distemper have involved pups, Durchfeld et al. 1990; McCormick 1983; van Heerden et al. 1989)? A postal survey of zoos holding wild dogs might easily answer this question.
  2. Does the administration of live CDV vaccines bring about seroconversion? One study, of three litters of pups, found no evidence of seroconversion (van Heerden et al. 1980), while another found that adults given booster vaccinations did seroconvert (Spencer & Burroughs 1992). These results provide circumstantial evidence that, as suspected for rabies vaccination, more than one dose of vaccine might be needed to achieve and maintain protective antibody levels. In zoos that vaccinate wild dogs against CDV routinely, more studies could be carried out to assess the efficacy of different protocols. As for rabies, it would be useful to know whether multiple doses of vaccine are more effective than a single dose, whether dart-vaccination is as effective as vaccination by hand, and how often boosters must be given.
  3. Do inactivated vaccines represent a viable alternative to live CDV vaccines? Inactivated CDV vaccines do not trigger seroconversion in several other wild canid species (Montali et al. 1983), and caused seroconversion in only 3/12 (25%) captive wild dogs in Tanzania (Visee 1996). Nevertheless, further experiments, perhaps involving the administration of multiple doses, are needed to determine whether inactivated vaccines have any value for CDV control in wild dogs.

Possibilities for Disease Control in Reservoir Hosts

In some circumstances, controlling disease in its reservoir hosts could be a better long-term solution than vaccinating wild dogs themselves (Chapter6). For example, rabies control in domestic dogs would protect people and their livestock as well as wild dogs. In other cases, however, it is not always clear that attempts to control disease in other species will provide effective protection for wild dogs. This highlights the need for more research, to address the following questions:

  1. How does interaction with wildlife affect the epidemiology and control of rabies in domestic dogs? As far as we are aware, all mathematical models of rabies control in domestic dogs have considered the dog population in isolation (e.g.Coleman & Dye 1996). For the wild dog-domestic dog interaction, this may be a reasonable approximation: domestic dogs encounter one another far more often than they encounter wild dogs, and it is unlikely that transmission from wild dogs to domestic dogs would be an important component of rabies epidemiology. However, where rabies affects wild dogs, it also affects other wild carnivores such as bat-eared foxes (Cleaveland & Dye 1995) which live at much higher densities than do wild dogs. Interactions with such species might contribute to the persistence of the disease in domestic dogs, making it more difficult to eradicate. Empirical and theoretical research is needed to establish whether vaccination of domestic dogs can protect wildlife, and whether a higher proportion of dogs must be vaccinated than is necessary when domestic dogs are considered in isolation.
  2. In areas where rabies occurs in domestic dog populations, why does the infection appear not to affect wild canids in some areas (as, for example, in Kruger and Hluhluwe-Umfolozi, M.G.L. Mills pers. comm.), but spill over into bat-eared foxes and jackals elsewhere (as, for example, in Serengeti and Etosha, Cleaveland & Dye 1995; Scheepers & Venzke 1995)?
  3. If vaccination programmes aim to establish a cordon sanitaire around wild dog areas, how wide must the cordon be? A pilot scheme in the Masai Mara vaccinated domestic dogs in a belt 15km wide (R.Kock, pers. comm.), but this might not be wide enough if domestic dogs and wildlife range over longer distances.
  4. Can rabies be controlled in wildlife reservoirs? Domestic dogs are important rabies reservoirs in East Africa, but in southern Africa wild species such as bat-eared foxes and jackals may be more important. Achieving anything approaching adequate vaccination cover in these species would be impossible if vaccines had to be delivered by hand, but oral vaccination is a possible alternative. This method of vaccine delivery has successfully eradicated rabies from red foxes in some parts of Europe and North America (Wandeler 1993). However, although experimental administration of live oral vaccines to black-backed and side-stripe jackals has been shown to confer protection from rabies, the strain used proved highly pathogenic to baboons (Bingham et al. 1995). Thus, more (ongoing) research, using other strains, is needed to perfect a method for vaccinating wild canids safely and effectively.
  5. What is the reservoir host for CDV? Although domestic dogs seem to be the reservoir in the Serengeti ecosystem (Alexander & Appel 1994; Roelke-Parker et al. 1996), in Selous the disease appears to persist in wildlife in the absence of domestic dogs (Creel et al. in prep.). Research is needed to identify the wildlife reservoir(s) in systems of this kind.
  6. What is the critical vaccination cover needed to eradicate CDV from domestic dog populations? Very little is known about the epidemiology of CDV in domestic dogs, and there are no published mathematical models. This makes it very difficult to formulate targets for vaccination cover. The possibility that the disease might also persist in wildlife species adds another complication to the epidemiological picture that needs addressing. More work is needed to formulate epidemiological models of CDV in domestic dog populations.
  7. Can the population density of reservoir hosts be reduced? In principal, reducing the density of reservoir hosts could lead to lower transmission rates and prevent disease from persisting in the population. The practical possibilities of doing this depend upon a number of factors. If the reservoir host was a wildlife species, controlling population size would rarely be possible. For domestic dogs, the possibilities would depend upon local peoples' requirement for those dogs.
  8. Can contact between wild dogs and domestic dogs be minimized? Again, this would depend upon local peoples' need for domestic dogs. More research is needed to determine whether domestic dogs' movements could be restricted by, for example, requiring that owned dogs be collared, that dogs be tied up at night, and shooting unaccompanied dogs.
  9. Could eradicating disease affect mortality in domestic dog populations? The mortality caused by CDV is poorly known, but it is conceivable that the disease is important in limiting the numbers of domestic dogs. If this were the case, then eradicating CDV could bring about an increase in the domestic dog population. This could present two further problems for wild dogs. First, if the domestic dog population was larger, other diseases might be able to persist when this was previously impossible. Second, if vaccination had to be stopped - perhaps due to lack of funds - a high proportion of domestic dogs would soon become susceptible. This would set the stage for a severe epidemic with an increased probability of transmission to wildlife.

Conclusions

A great deal of information about wild dog ecology has become available in recent years. Many of the research questions raised at the IUCN/SSC Canid Specialist Group's 'Workshop on the Conservation & Recovery of the African Wild Dog' (Ginsberg 1992) have now been answered, and generated a new set of research priorities. Persecution remains a serious threat, and work is urgently needed to devise ways of resolving conflict between the interests of wild dogs and those of livestock farmers. A substantial volume of research is also needed into disease control - it was not until the wild dog study populations disappeared from the Serengeti ecosystem that it became clear just how severe a threat disease could pose to wild dogs. We still cannot determine the best strategy for controlling disease - and at present we are not fully equipped to carry out any of them.

Return to Wild Dog Action Plan Table of Contents, AWD Species Account, CSG Publications or CSG Home Page.

© 1997 International Union for the Conservation of Nature and Natural Resources.