The recent article by Chauvenet et al. “Unintended consequences of conservation actions: managing disease in complex ecosystems” advocates for holistic approaches to disease management through examination of the relationships between cheetah and lion populations in the Serengeti ecosystem and to infer potential impacts arising from domestic dog vaccination campaigns against canine distemper virus (CDV). Disease management in complex ecosystems is an important topic and warrants thoughtful discussion among stakeholders. Rinderpest provides a clear example whereby a disease intervention aimed at protecting livestock economies had unintended consequences for the Serengeti, with mass livestock vaccinations resulting in a 5-fold increase in wildebeest numbers (Sinclair, 1979) and subsequent trophic cascades (Holdo et al., 2009). Developing scientific approaches to understand, anticipate and, if necessary, mitigate these consequences is clearly a challenge. Unfortunately, the study of Chauvenet et al. seriously misrepresents the context in which they have chosen to set this issue.

The authors question how a vaccination campaign against CDV implemented in domestic dogs could affect the viability of cheetah populations through its impact on lions. However, the authors present incorrect correlative inferences between dog vaccinations and changes in lion numbers then fit a questionable relationship between lions and cheetahs to infer negative impacts on cheetah populations.

The authors assume that CDV vaccination of domestic dogs between 1996 and 2003 was responsible for the increase in the lion population from 1995 (Fig.1 in the article). However, this temporal association does not represent a causal relationship.
First, we informed the authors that the initial CDV efforts in 1996 were extremely limited in geographic scope, focusing on <7,500 dogs per year in a few villages in a 250 km2 area to the west of the Serengeti National Park (SNP) (as described in Cleaveland et al., 2003). CDV age-seroprevalence data indicated that infection patterns differed little between vaccinated villages and adjacent control (unvaccinated) villages during this phase of the program (Lembo, 2007). We emphatically informed the authors that the cordon sanitaire, involving 187 villages and ~40,000-50,000 dogs per year was not established until 2003, well after the increase in lion numbers. Second, we also informed the authors of published evidence showing that CDV swept through the Serengeti lions again in 1999 and in 2007 (Munson et al., 2008), indicating that neither the early vaccination efforts or the later ring-vaccination program prevented circulation of CDV in Serengeti’s wildlife. Further, there are obvious alternative explanations for the observed increase in lion numbers prior to the large-scale CDV vaccination. A comprehensive ecological analysis of lion population dynamics showed that the population increase on the plains, like a similar increase in the woodlands ten years earlier, resulted from increased vegetative cover, thereby increasing the feeding success of the plains lions (Packer, et al. 2005).

The authors concede that the relationship between lion density and cheetah cub survival is not known, but they proceed by assuming it is negative, and succeed only in explaining an additional 4% of the variation in cheetah numbers by freely fitting this presumed relationship within the confines of their formulation. The authors made no attempt to assess the goodness of fit under the assumption that cheetah numbers vary independently of lion numbers, so they cannot claim that accounting for lion density is ‘key to cheetah population modelling’.

While complex ecosystems clearly show potential for unintended consequences from large-scale disease management efforts, the dog vaccination program around the Serengeti provides no concrete evidence of (a) a positive impact on lion numbers or (b) a negative effect on cheetah.

Cleaveland S., Kaare M., Tiringa P., Mlengeya, J. Barrat (2003) A dog rabies vaccination campaign in rural Africa: impact on the incidence of dog rabies and human dog-bite injuries. Vaccine, 21: 1965-1973
Holdo RM, Sinclair ARE, Dobson AP, Metzger KL, Bolker BM, Ritchie ME, Holt RD (2009) A disease-mediated trophic cascade in the Serengeti and its implications for ecosystem C. PLoS Biol 7:e1000210.
Lembo T. (2007) An investigation of disease reservoirs in complex ecosystems: rabies and canine distemper in the Serengeti. PhD Thesis, University of Edinburgh.
Munson, L., K.A. Terio, R. Kock, T. Mlengeya, M.E. Roelke, E. Dubovi, B. Summers, A.R.E. Sinclair & C. Packer. 2008. Climate extremes and co-infections determine mortality during epidemics in African lions. PLoS-One 3, e2545.
Packer, C., R. Hilborn, A. Mosser, B. Kissui, J. Wilmshurst, M. Borner, G. Hopcraft & A.R.E. Sinclair. 2005. Ecological change, group territoriality and non-linear population dynamics in Serengeti lions. Science 307, 390-393.
Sinclair, A.R.E. (1979). The eruption of the ruminants. In Serengeti: Dynamics of an Ecosystem, eds. A. R. E. Sinclair, and M. Norton-Griffiths. Chicago: University of Chicago Press.


Craig Packer
Department of Ecology, Evolution & Behavior
University of Minnesota
St Paul, MN 55108

Sarah Cleaveland
Katie Hampson
Dan Haydon
Felix Lankester
Tiziana Lembo
Institute of Biodiversity, Animal Health and Comparative Medicine
College of Medicine, Veterinary Medicine and Life Sciences
University of Glasgow Glasgow G12 8QQ

Andrew Dobson
Department of Ecology and Evolutionary Biology
117 Eno Hall
Princeton, NJ 08544