The application of spatial capture-recapture models to investigate leopard ecology and conservation in South Africa

Rogan, Matthew S

The application of spatial capture-recapture models to investigate leopard ecology and conservation in South Africa

Doctoral Thesis

2021

Abstract

Population monitoring is essential to wildlife conservation and management. Rare and elusive species are difficult to observe, and hence monitor, in wild populations. Leopards (Panthera pardus) are an iconic and threatened species whose conservation is hampered by a lack of robust population data, in part due to their sparse populations and cryptic nature. I used cameratrap surveys from 27 protected areas in northeastern South Africa to make inferences about the status and conservation needs of leopards. I first evaluated the relationship between leopard density and proportion of area used within protected areas to determine if the latter could serve as a more efficient yet robust proxy for the former. I found that the relationship was too imprecise to be informative, that many populations of varying density used all the space available, and that the scale of individual movement strongly influenced the proportion of area used regardless of density. I then fit multisession spatial capture-recapture models to time series data from seven of these leopard populations to assess their threat level based on the estimated probability of population declines. I found some evidence of decline in six of the seven populations and found that the population at one site has a 75% chance of declining by 80% over three leopard generations. Lastly, I investigated the relative influence of bottom-up ecological factors and top-down anthropogenic factors as possible determinants of leopard density to identify what conditions are most suitable for conserving leopard populations. I found that while habitat and management characteristics of protected areas matter, human impacts around and within protected areas are the primary drivers of variation in leopard density. Based on these analyses, I conclude that South African protected areas are not conferring sufficient protection to leopard populations and that more must be done to mitigate human impacts inside protected areas. I also show that the leopard monitoring program would benefit from longer surveys with more sampling locations to increase the statistical power for detecting changes. This thesis demonstrates the capacity for large-scale monitoring programs to greatly expand our understanding of the conservation needs of a cryptic species.

Keywords

Biological Sciences

Reference:

Collections

PhD / Doctoral

Full item page