Date of Award

2024

Degree Name

Biological Sciences

College

College of Science

Type of Degree

M.S.

Document Type

Thesis

First Advisor

Dr. Jayme Waldron, Committee Chairperson

Second Advisor

Dr. Shane Welch

Third Advisor

Dr. Anne Axel

Fourth Advisor

Dr. Kyle Palmquist

Abstract

Eastern diamondback rattlesnakes (Crotalus adamanteus; EDB) are an imperiled species native to the Southeastern United States longleaf pine (Pinus palustris) ecosystem. Several markrecapture methods have been used range-wide in hopes of increasing knowledge on their population numbers. I explored the efficacy of two mark-recapture methods (scale cauterization and dorsal head photography) using the Marine Corps Recruit Depot, Parris Island, SC EDB population in this multi-part study. The first method I evaluated was ventral scale branding, which has been widely implemented without knowledge of brand permanence. I classified misidentification errors at capture status (new versus recapture) and individual (cautery identification number) levels, confirming snake identity based on the presence of a passive integrated transponder (PIT) tag. I used logistic regressions to examine covariate effects on the probability of surveyors committing Type I (uncauterized snakes incorrectly identified as recaptures) and Type II (cauterized snakes incorrectly identified as new captures) capture status misidentification errors. I quantified individual-level misidentification errors when surveyors incorrectly interpreted the unique cautery identification numbers given to a snake. Surveyors were more likely to commit Type II errors than Type I errors, and the probability of misidentifying capture status averaged 17.5% when snakes were previously cauterized, as compared to 3.1% when snakes were uncauterized. Ecdysis cycle was the most important predictor of individual-level misidentification errors, and snakes that were in ecdysis at the time they were evaluated were more likely to be misidentified. High misidentification rates do not negate the use of scale cauterization in snake monitoring programs. Rather, knowledge of misidentification errors improves the utility of snake mark-recapture data by enabling researchers to select appropriate demographic models that include considerations for misidentification errors. In Chapter 2, I investigated the use of dorsal head photographs of EDBs to evaluate the suitability of unique head scale patterns for identifying individuals through photographic mark-recapture (PMR). This method has been used with other snake taxa, with unique scale patterns, but no published studies have evaluated this method in EDBs. Head scale patterns were assessed and matched by creating a database of 480 photos that were taken in either a controlled environment using smartphones or captured in adapted-Hunt drift fence technique (AHDriFT) camera trap arrays. I used two computer-assisted pattern recognition software programs (Wild-ID and HotSpotter) to match samples photos (recapture photos of individuals) to reference photos (preexisting photos in the database). Using a logistic regression analysis, I examined covariate effects of misidentification errors when the software programs failed to correctly rank a photo or photos of the same individual as the top match or matches. HotSpotter (12% error rate) outperformed Wild-ID (14.3% error rate). Ecdysis cycle was the only significant predictor of misidentification errors, indicating that photos taken of snakes in ecdysis were more likely to be incorrectly matched to a photo of a different individual. The use of natural head pattern markings is a reliable method of identifying individual EDBs but has limited utility for estimating population demographics due its inability to provide data such as sex or mass. Therefore, unique head patterns are likely best suited for supplementing traditional mark-recapture methods as a backup approach.

Subject(s)

Cautery.

Eastern diamondback rattlesnake.

Longleaf pine.

Ecosystem.

Eastern diamondback rattlesnake -- Population.

Wildlife conservation.

Wildlife management.

Download

Share

COinS