Title:
Identification of Notophthalmus viridescens Individuals from Photographs Using the Interactive Individual Identification System (I3S) Software
Author(s):
Aaron Goldfarb
Author Email:
agoldarb@unca.edu
Department:
ENVIRONMENTAL STUDIES
Faculty Mentor(s):
Andrew Laughlin
Abstract / Summary:
More efficient methods for estimating the population size and density of salamander species are becoming increasingly important. Amphibian species across the world are experiencing massive population declines and extinctions due to factors such as habitat loss, climate change, and diseases such as Batrachochytrium dendrobatidis (Bd), or the chytrid fungus. Batrachochytrium salamandrivorans (Bsal) is an emerging pathogenic chytrid fungus that has been documented in both wild and captive salamanders across Europe and is likely to be spread across the world through the pet trade, posing a significant threat to salamander populations. Estimates of population size (such as Mark-Recapture) are important to track changes in populations over time. However, conventional methods of population size estimation may suffer from drawbacks such as being labor intensive, being invasive, or causing mortality. I tested the effectiveness of the Interactive Individual Identification System (I3S) software at identifying Notophthalmus viridescens (Eastern Newt) individuals from images. I caught and photographed the ventral surfaces of 33 Eastern Newts across six survey days between the months of March and July 2023. I captured all newts at Sandy Bottom Preserve, located south of Asheville, North Carolina. I took at least three photographs of each individual. I used the software I3S Straighten to ensure all individuals were straight and facing the same direction in images. I then tested the effectiveness of the software I3S Pattern+ in correctly matching images of the same individual. I found that the software correctly matched images of the same individual in the first match result 91% of the time. I also found match accuracy to be directly related to image quality. Additionally, I provide future recommendations for conducting Mark-Recapture studies on this species at Sandy Bottom Preserve and effective software use.
Publication Date:
Dec-12-2023
Documents: