Date of Award

2026

Degree Name

Biological Sciences

College

College of Science

Type of Degree

M.S.

Document Type

Thesis

First Advisor

Dr. Shane Welch

Second Advisor

Dr. Jayme Waldron

Third Advisor

Dr. Kyle Palmquist

Fourth Advisor

Dr. Anne Axel

Abstract

Reliable population monitoring is essential for evaluating wildlife translocations and informing conservation management, yet traditional monitoring approaches are often labor intensive and require prolonged sampling. This study evaluated marking techniques, survey methods, and population estimation approaches for monitoring a translocated population of southern fox squirrels (Sciurus niger niger) at Marine Corps Recruit Depot Parris Island, South Carolina, USA (MCRDPI). From January-June 2023, I conducted six live trapping and camera trapping sessions across five survey sites. I compared the performance of color-coded collars and tail shaves for individual identification and the recapture probability between live traps and camera traps. Logistic mixed-effects models were used to evaluate recapture probabilities and observer identification error. Camera traps produced significantly greater recapture probabilities than live traps, with live-trapping reducing recapture probability by 37% (OR=0.63, 95% CI=0.42–0.85, p < 0.05). Marking method also influenced identification accuracy, as collars had 4.38 times greater odds of correct identification than tail shaves (95% CI= 1.13–17.06, p < 0.05). Tail shaves and passive integrated transponder (PIT) tags persisted throughout the study, whereas collard exhibited moderate losses due to chewing damage but remained the most reliable temporary mark for camera-based identification. In Chapter 2, I compared rapid assessment and seasonal sampling approaches for estimating population density. Four weeks of camera trapping data analyzed with a modified Huggins density estimator produced a mean density estimate of 1.22 squirrels/ha (SE=0.51, 95% CI=0.54–2.75), while six months of live trapping data analyzed with a POPAN open-population model yielded a comparable estimate of 1.05 squirrels/ha (SE=0.34, CI=0.56–1.96). Estimated superpopulation sizes for suitable habitat on MCRDPI were similarly consistent, with approximately 499 squirrels (CI=221.59-1122.94) from rapid assessment and 473 (CI=229.30-799.67) from the seasonal sampling. Together, these findings demonstrate that camera trapping paired with reliable individual marking can improve monitoring efficiency while providing population density estimates comparable to more laborintensive approaches. Rapid camera trap surveys therefore represent an effective and minimally invasive alternative for monitoring translocated wildlife populations inhabiting managed forest ecosystems.

Subject(s)

Ecology.

Wildlife conservation.

Wildlife management.

Squirrels.

Population density.

Longleaf pine.

Forests and forestry.

Habitat (Ecology)

Data transmission systems.

Environmental monitoring.

Mammal surveys.

Download

Share

COinS