Parameterizing the random encounter staying time model to generate mean and variance ungulate density and abundance estimates

Carswell, Brendan M (2024) Parameterizing the random encounter staying time model to generate mean and variance ungulate density and abundance estimates. Masters thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (2MB)

Abstract

Understanding wildlife density and abundance is perhaps the most important and universal concept across facets of wildlife management, conversation, and research. Despite the significance of understanding species densities in ecology, methods for estimating density for large terrestrial mammals in Canada continue to have high levels of inaccuracy in addition to being a costly, exclusive practice. From the mass use of remote camera traps in wildlife life science came a series of camera trap-based density estimation models, known as viewshed density estimators, which could allow practitioners to estimate wildlife density from camera trap data. Despite the cost-effective and accessible framework, viewshed density estimators remain analytically challenging to parameterize and implement. To accurately estimate density, viewshed density estimators require a precise metric of the physical area camera traps monitor, a highly variable number that can be challenging to quantify. Here, I tested a field and analytical framework that can be used to accurately estimate the spatial footprint of camera traps with a 100% capture probability, the Effective Capture Area. Next, I use the Effective Capture Area to parameterize the Random Encounter Staying Time model of density estimation for generating density and abundance estimates for moose (Alces alces) and elk (Cervus canadensis) across camera trap grids in Riding Mountain National Park, Canada. I show that, given adequate spatial and temporal sampling periods, the Random Encounter Staying Time model produces density and abundance estimates that correlate well with historic aerial flight surveys on both fine- and coarse-spatial scales. Finally, I comment on how viewshed density estimators can improve our understanding of wildlife density and abundance estimation, as well as provide novel insights in many areas of ecological study.

Actions (login required)

View Item