Population densities of mesocarnivores across protected
and non-protected private landscapes in the Eastern Cape
of South Africa

Abstract

Protected areas can be an effective tool for carnivore conservation, but increasing human populations are intensifying pressures and increasing negative interactions with wildlife. The conservation potential of private landscapes is being increasingly recognised to provide vital
habitat and connectivity for wildlife. Due to their ecological and ethological plasticity, mesocarnivores thrive on a wide range of landscapes, but are frequently subject to intensive persecution outside of protected areas. Development of reliable and accurate monitoring methods is therefore vital to enable evidence-based decisions for achieving sustainable population management. Novel techniques have been developed to estimate population
densities of unmarked species using camera trapping, but many of these methods require field validation. Using one such method, distance sampling with camera traps (CTDS), population densities of mesocarnivores were estimated across two adjacent properties in the Eastern Cape of South Africa: a mixed used agricultural area, and a ‘Big Five’ private game reserve. Evidence of a high density brown hyaena (Parahyaena brunnea) population was recorded within the protected site, whereas insufficient captures were obtained on the mixed-use agricultural area. Conversely, lower densities of black-backed jackal (Canis mesomelas) were found in the reserve than the agricultural landscape, indicating variable effects of land use on mesocarnivore density. However, high imprecision in density estimates was evident, primarily attributed to spatial heterogeneity in encounter rates. This study reinforces the importance of evaluating camera trap performance metrics for more accurate measures of sampling effort and suggests how a site-specific approach can increase the efficacy of CTDS for population monitoring when camera performance is extremely variable. Improving precision in estimates is vital to validate CTDS as a reliable monitoring practice, and future studies should aim to test ways in which survey design and sampling effort can affect confidence intervals, as well as evaluating performance against known population sizes or established density estimation techniques.