The Conservation Relevance of Epidemiological Research into Carnivore Viral Diseases in the Serengeti

Abstract:  Recent outbreaks of rabies and canine distemper in wildlife populations of the Serengeti show that infectious disease constitutes a significant cause of mortality that can result in regional extirpation of endangered species even within large, well-protected areas. Nevertheless, effective management of an infectious disease depends critically on understanding the epidemiological dynamics of the causative pathogen. Pathogens with short infection cycles cannot persist in small populations in the absence of a more permanent reservoir of infection. Development of appropriate interventions requires detailed data on transmission pathways between reservoirs and wildlife populations of conservation concern. Relevant data can be derived from long-term population monitoring, epidemic and case-surveillance patterns, genetic analyses of rapidly evolving pathogens, serological surveys, and intervention studies. We examined studies of carnivore diseases in the Serengeti. Epidemiological research contributes to wildlife conservation policy in terms of management of endangered populations and the integration of wildlife conservation with public health interventions. Long-term, integrative, cross-species research is essential for formulation of effective policy for disease control and optimization of ecosystem health.     

Local‐Scale Dynamics and Local Drivers of Bushmeat Trade

Bushmeat management policies are often developed outside the communities in which they are to be implemented. These policies are also routinely designed to be applied uniformly across communities with little regard for variation in social or ecological conditions. We used fuzzy‐logic cognitive mapping, a form of participatory modeling, to compare the assumptions driving externally generated bushmeat management policies with perceptions of bushmeat trade dynamics collected from local community members who admitted to being recently engaged in bushmeat trading (e.g., hunters, sellers, consumers). Data were collected during 9 workshops in 4 Tanzanian villages bordering Serengeti National Park. Specifically, we evaluated 9 community‐generated models for the presence of the central factors that comprise and drive the bushmeat trade and whether or not models included the same core concepts, relationships, and logical chains of reasoning on which bushmeat conservation policies are commonly based. Across local communities, there was agreement about the most central factors important to understanding the bushmeat trade (e.g., animal recruitment, low income, and scarcity of food crops). These matched policy assumptions. However, the factors perceived to drive social‐ecological bushmeat trade dynamics were more diverse and varied considerably across communities (e.g., presence or absence of collaborative law enforcement, increasing human population, market demand, cultural preference). Sensitive conservation issues, such as the bushmeat trade, that require cooperation between communities and outside conservation organizations can benefit from participatory modeling approaches that make local‐scale dynamics and conservation policy assumptions explicit. Further, communities’ and conservation organizations’ perceptions need to be aligned. This can improve success by allowing context appropriate policies to be developed, monitored, and appropriately adapted as new evidence is generated. Dinámicas a Escala Local y Conductores Locales del Mercado de Carne de Caza     

A generalized approach for producing,quantifying, and validating citizen science data from wildlife images

Citizen science has the potential to expand the scope and scale of research in ecology and conservation, but many professional researchers remain skeptical of data produced by nonexperts. We devised an approach for producing accurate, reliable data from untrained, nonexpert volunteers. On the citizen science website www.snapshotserengeti.org , more than 28,000 volunteers classified 1.51 million images taken in a large‐scale camera‐trap survey in Serengeti National Park, Tanzania. Each image was circulated to, on average, 27 volunteers, and their classifications were aggregated using a simple plurality algorithm. We validated the aggregated answers against a data set of 3829 images verified by experts and calculated 3 certainty metrics—level of agreement among classifications (evenness), fraction of classifications supporting the aggregated answer (fraction support), and fraction of classifiers who reported “nothing here” for an image that was ultimately classified as containing an animal (fraction blank)—to measure confidence that an aggregated answer was correct. Overall, aggregated volunteer answers agreed with the expert‐verified data on 98% of images, but accuracy differed by species commonness such that rare species had higher rates of false positives and false negatives. Easily calculated analysis of variance and post‐hoc Tukey tests indicated that the certainty metrics were significant indicators of whether each image was correctly classified or classifiable. Thus, the certainty metrics can be used to identify images for expert review. Bootstrapping analyses further indicated that 90% of images were correctly classified with just 5 volunteers per image. Species classifications based on the plurality vote of multiple citizen scientists can provide a reliable foundation for large‐scale monitoring of African wildlife.     

Long-Term Ecosystem Dynamics in the Serengeti: Lessons for Conservation

Abstract:  Data from long-term ecological studies further understanding of ecosystem dynamics and can guide evidence-based management. In a quasi-natural experiment we examined long-term monitoring data on different components of the Serengeti-Mara Ecosystem to trace the effects of disturbances and thus to elucidate cause-and-effect connections between them. The long-term data illustrated the role of food limitation in population regulation in mammals, particularly in migratory wildebeest and nonmigratory buffalo. Predation limited populations of smaller resident ungulates and small carnivores. Abiotic events, such as droughts and floods, created disturbances that affected survivorship of ungulates and birds. Such disturbances showed feedbacks between biotic and abiotic realms. Interactions between elephants and their food allowed savanna and grassland communities to co-occur. With increased woodland vegetation, predators' capture of prey increased. Anthropogenic disturbances had direct (hunting) and indirect (transfer of disease to wildlife) effects. Slow and rapid changes and multiple ecosystem states became apparent only over several decades and involved events at different spatial scales. Conservation efforts should accommodate both infrequent and unpredictable events and long-term trends. Management should plan on the time scale of those events and should not aim to maintain the status quo. Systems can be self-regulating through food availability and predator-prey interactions; thus, culling may not be required. Ecosystems can occur in multiple states; thus, there may be no a priori need to maintain one natural state. Finally, conservation efforts outside protected areas must distinguish between natural change and direct human-induced change. Protected areas can act as ecological baselines in which human-induced change is kept to a minimum     

Siblicide in Serengeti spotted hyenas: a long-term study of maternal input and cub survival

In the Serengeti National Park, Tanzania, large fluctuations of prey abundance alters the frequency at which spotted hyena (Crocuta crocuta) cubs are nursed and thus the total level of maternal input available to them. Maternal input is high when mothers feed on high densities of locally available migratory herbivores and low when mothers travel up to 70 km to forage. Using data from 19 cub cohorts on the incidence of siblicide (from monitoring the survival of 609 cubs in twin litters) and cub growth rates (from 195 cubs in twin litters) as a measure of maternal input, we demonstrate that the incidence of siblicide increased as average cohort growth rate declined. In total, there were 37 siblicides in 384 litters (9% of litters). When both cubs were alive, total maternal input in siblicidal litters was significantly lower than in non-siblicidal litters and the mean share of the dominant sib of 64.6% was significantly higher than the mean of 52.1% for dominant sibs of non-siblicidal litters. After siblicide, growth rates of siblicide victors substantially increased, demonstrating that mothers did not reduce maternal input after litter reduction. As a result, siblicide victors achieved a long-term growth rate similar to that of singletons and thus significantly increased their expected survival. We conclude that in spotted hyenas, high maternal input in lactation has favoured the evolution of facultative siblicide in populations inhabiting areas with low or fluctuating food resources.     

Exploring potential socio-ecological impacts of changes to the Loliondo Gamed Controlled Area,Northern Tanzania: the case of the pastoral village Ololosokwan

Recent plans to alter the Loliondo Game Controlled Area (GCA), a nature conservation area located in Northern Tanzania, would result in substantial reduction of rangelands in the region. We quantify the current and hypothetical levels of the aboveground Human Appropriation of Net Primary Production (aHANPP) in one of the affected villages, and estimate the maximum exploitability rate of rangelands by livestock in the region. We find that the current aHANPP of the village amounts to 34–38% of the potential productivity, which could increase to 59–67% due to the altered GCA. On rangelands, livestock-induced aHANPP would increase from the current level of 30–34% to 54–61%, which is far above a maximum exploitability rate of 40–41%. Our results reveal that the intended changes to the Loliondo GCA will severely affect the current livelihood strategy of the Maasai, which is based on pastoralism.     

Individual heterogeneity in recapture probability and survival estimates in cheetah

Accurate estimates of demographic parameters are key for understanding and predicting population dynamics and for providing insights for effective wildlife management. Up until recently, no suitable methodology has been available to estimate survival probabilities of species with asynchronous reproduction and a high level of individual variation in capture probabilities. The present work develops a capture-mark-recapture model for cheetahs in the Serengeti National Park, Tanzania, which (a) deals with continuous reproduction, (b) takes into account the high level of individual heterogeneity in capture probabilities and (c) is spatially explicit. Results show that (1) our approach, which is an extensive modification of the Cormack-Jolly-Seber model, provides a lower female adult survival estimate and a higher male adolescent survival estimate than previous approaches to estimate cheetah survival in the area, (2) using sighting location alone is not sufficient to capture the individual variation in resighting probabilities for both sexes, and (3) precision in estimated survival probabilities is generally increased. Species which are individually recognizable, wide-ranging and/or where individuals differ substantially in sightability are particularly appropriate to our modelling approach, and our methodology would thus be appropriate for a wide number of species to provide more accurate estimates of survival.     

Modeling the Effects of Trophy Selection and Environmental Disturbance on a Simulated Population of African Lions

Abstract:  Tanzania is a premier destination for trophy hunting of African lions (Panthera leo) and is home to the most extensive long-term study of unhunted lions. Thus, it provides a unique opportunity to apply data from a long-term field study to a conservation dilemma: How can a trophy-hunted species whose reproductive success is closely tied to social stability be harvested sustainably? We used an individually based, spatially explicit, stochastic model, parameterized with nearly 40 years of behavioral and demographic data on lions in the Serengeti, to examine the separate effects of trophy selection and environmental disturbance on the viability of a simulated lion population in response to annual harvesting. Female population size was sensitive to the harvesting of young males (≥3 years), whereas hunting represented a relatively trivial threat to population viability when the harvest was restricted to mature males (≥6 years). Overall model performance was robust to environmental disturbance and to errors in age assessment based on nose coloration as an index used to age potential trophies. Introducing an environmental disturbance did not eliminate the capacity to maintain a viable breeding population when harvesting only older males, and initially depleted populations recovered within 15–25 years after the disturbance to levels comparable to hunted populations that did not experience a catastrophic event. These results are consistent with empirical observations of lion resilience to environmental stochasticity .