OPTIMIZING CAMERA TRAP SAMPLING FOR ABUNDANCE ESTIMATION USING ANIMAL MOVEMENT BEHAVIOR

Author

• Katherine Grace Garrett, University of MontanaFollow

Year of Award

2026

Document Type

Thesis

Degree Type

Master of Science (MS)

Degree Name

Wildlife Biology

Department or School/College

Wildlife Biology Program

Committee Chair

Dr. Mahdieh Tourani

Commitee Members

Dr. Paul Lukacs, Dr. Sarah Sells, Dr. David Messmer

Keywords

Montana, camera trap, ungulate

Subject Categories

Terrestrial and Aquatic Ecology

Abstract

Unmarked camera-based abundance estimators offer ways of monitoring multiple species simultaneously without requiring individual identification. Some assumptions of unmarked estimators rely on the adherence of animal behavior and implementation of appropriate sampling designs. One shared assumption among unmarked abundance estimators is that camera traps are placed randomly with respect to animal space-use. Simple random sampling, especially in areas of rugged terrain, tall vegetation, or inaccessible roads, are often costly for long-term monitoring. Stratified random sampling could be an alternative strategy for optimizing sampling, where the strata is informed by animal space-use. Optimal design often consists in gaining the highest amount of information at a lower cost. In multi-species monitoring, all target species space-use must be considered in the study design and quantifying multispecies space-use is the first necessary step. This thesis focuses on the creation of a basis that managers can use to examine various sampling designs that may fit the needs of a long-term population monitoring study.

In northwestern Montana, the development of a long-term monitoring project motivated a need for prioritizing low-cost deployment locations while upholding estimator assumptions for all target species. Currently, the project conducts broadscale simple random sampling for estimating population size of elk, Cervus canadensis, mule deer, Odocoileus hemionus, and white-tailed deer, O. virginianus. Here, we first quantify space-use of the three sympatric ungulate species using habitat analysis and process the models outputs to create a multispecies composite map based on three weighting scenarios. Areas of high information value for one species does not necessarily have the same information value for another species. Second, we analyze GPS track data of technicians hiking to previous camera locations during sampling using habitat analysis and produce a cost map of deployments that represents ease of movement in the study area. Third, we weighted the composite ungulate space-use maps with the cost map in various ways based on species priority. Finally, through a simulation study we investigated how common abundance estimators, i.e., space-to-event and instantaneous sampling, may be biased under a common scenario where camera deployment is heavily informed by animal space-use. The outputs of this thesis could guide future sampling designs to prioritize low-cost deployment locations while upholding estimator assumptions.

Recommended Citation

Garrett, Katherine Grace, "OPTIMIZING CAMERA TRAP SAMPLING FOR ABUNDANCE ESTIMATION USING ANIMAL MOVEMENT BEHAVIOR" (2026). Graduate Student Theses, Dissertations, & Professional Papers. 12706.
https://scholarworks.umt.edu/etd/12706