Quantifying Mammalian Interactions and Distributions to Inform Conservation Planning in Mozambique
Date of Final Oral Examination (Defense)
Type of Culminating Activity
Master of Science in Biology
Neil Carter, Ph.D.
Amy C. Ulappa, Ph.D.
Jesse R. Barber, Ph.D.
Protected areas are a staple in conservation, but human activities outside of protected areas drive species interactions, compositions, and distributions. Research is especially needed in these multi-use landscapes to maintain habitat connectivity for entire wildlife communities between protected areas. Yet, such research is lacking in areas it is needed most, such as in sub-Saharan Africa, where human populations are expected to double by 2050. My objectives were to quantify mammal distributions, interactions, community compositions, and their relationships with human and natural factors within a sustainable-use forestry concession outside of Gorongosa National Park. I used recently developed multispecies occupancy models to analyze presence/absence data from 75 motion-detecting camera traps.
First, I wanted to know if small, sympatric carnivores avoided each other in time or space in a human-modified landscape with few apex predators. I examined activity patterns, habitat preferences, tolerance to people and potential for intraguild competition among three common, but understudied African carnivores: African civets (Civettictis civetta), bushy-tailed mongoose (Bdeogale crassicauda), and large-spotted genets (Genetta maculata). I hypothesized that all three species would overlap in time, but in space, genets and mongoose would both avoid civets due to being smaller; and genets and mongoose, being roughly the same size, would not affect each other. I used the time stamps from each species’ detections and found that these three species exhibited strong temporal overlap. I then used N-mixture models in a Bayesian framework to measure these species’ spatial relationships. I found that civets and mongoose avoid each other, indicated by the strong negative relationship between their predicted abundances at each camera trap site. In contrast, genets and mongoose exhibited a positive relationship, and there was no significant relationship between genets and civets. Civets and mongoose may be further limited in space through the avoidance of human settlements if they are also competing with each other, while genets were unaffected by human presence. Such interspecific interactions are important to consider for multispecies conservation planning in multi-use landscapes, as these relationships may change as human populations grow.
Second, I investigated how natural and anthropogenic factors influence animal space use and richness in a multi-use area. We used hierarchical, multispecies models to quantify species and species groups, and community level spatial relationships with human and environmental variables for 30 detected mammals. We modelled species occupancies when separated into two different groups: 1) taxonomic/functional groups consisting of carnivores, ungulates, primates, other foragers, and insectivores, and 2) body size groups, consisting of small species (200kg). We also quantified occupancy probabilities and richness for the entire community to determine where species richness was greatest and inform biodiversity conservation efforts. We predicted that carnivores and large mammals would be the most sensitive to anthropogenic features. In partial support of this hypothesis, increasing distance from settlements positively affected the occupancies of carnivores, as well as primates and other foragers, and large mammals, as well as all three of the other body size groups. However, active roads and human activity rates did not have a statistically significant relationship with any species’ occupancies or detection rates, respectively. Overall, mammalian richness was highest far from human settlements in the concession and close to rivers. Our results have important implications for connectivity planning for multispecies conservation outside of Gorongosa National Park, and provide a starting point for prioritizing these efforts.
Easter, Tara, "Quantifying Mammalian Interactions and Distributions to Inform Conservation Planning in Mozambique" (2018). Boise State University Theses and Dissertations. 1427.