Date of Award
12-2024
Culminating Project Type
Thesis
Styleguide
apa
Degree Name
Biological Sciences - Ecology and Natural Resources: M.S.
Department
Biology
College
College of Science and Engineering
First Advisor
Jennifer Y. Lamb
Second Advisor
Matthew P. Davis
Third Advisor
Sarah Z. Gibson
Fourth Advisor
Jeremy B. Cable
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Keywords and Subject Headings
Four-toed salamander, eDNA surveys, Environmental DNA, Minnesota, Traditional surveys, Hemidactylium
Abstract
Successful habitat management for amphibians depends on accurate detection and monitoring of a target species, especially when it is rare or illusive. The Four-toed Salamander (Hemidactylium scutatum) is listed as a species of special concern in Minnesota. Traditional surveys for this species can be time-consuming, challenging, and may damage the wetlands these salamanders use for reproduction. Environmental DNA (eDNA) methods could be an effective alternative to traditional methods. For my thesis, I asked whether eDNA survey methods are as effective as traditional methods in detecting the Four-toed Salamander in Minnesota. I designed and evaluated eDNA assays for segments of two mitochondrial genes, cytochrome oxidase subunit 1 (Cox1) and cytochrome b (Cytb) (Chapter 1). Each assay consists of forward and reverse primers, and an internal fluorescing probe with a BHQ tag. I used quantitative Polymerase Chain Reaction (qPCR) to assess the efficiency, limit of quantification (LOQ), and limit of detection (LOD) for each assay. Both assays were tested for specificity against non-target amphibian species in silico and in vitro. Both were found to be species-specific for DNA from H. scutatum and no other non-target species. The Cox1 assay performed better than the Cytb assay, with an LOD of 10-5 and an efficiency of 0.973. In the spring and summer of 2023 and 2024, I completed nest search surveys, larval surveys, and eDNA surveys at sites across five counties in the known range of H. scutatum in Minnesota. Three water samples and a field negative sample were filtered using a GroverGo pump and Smith-Root filter pack at each site. All water samples were tested in triplicate after eDNA extraction via quantitative Polymerase Chain Reaction (qPCR). Many samples were inhibited so all samples were treated with a Zymo inhibitor removal kit. The Cox1 assay found eDNA from H. scutatum at a total of 7 of 24 sites sampled during the nesting period in 2023 (naïve occupancy = 0.29). Traditional nest searches encountered H. scutatum at 19 of 24 of these sites (naïve detection rate = 0.79). I used Dorazio and Erickson’s Bayesian multi-scale occupancy modeling approach to understand what covariates might be affecting occupancy and detection with eDNA assays. Models included the covariates site ID, known salamander presence, salamander abundance (RAI), volume of water filtered (mL), and wetland surface area (m2) at the site (ψ) or sample (θ) level. No models included covariates at the qPCR technical replicate level (p). Models were run using the eDNAoccupancy package in R (Version 4.3.2) using short (11,000) MCMC chains, and three top models were later run using long (110,000) MCMC chains. Model 1 (null model) results were used to calculate cumulative detection probabilities. At least five samples need to be collected per site, and at least four qPCR replicates are needed to detect H. scutatum eDNA at the 95% detection rate. Bayesian model selection criteria (WAIC and PPLC) identified four top models, including the null model (Model 1) and three models with the covariates salamander presence (Model 3), salamander abundance (Model 9), and the volume of water filtered (Model 8) included at the site level (ψ). The covariates applied to psi (ψ) for the top three models did not improve on the null model, based on low confidence and weak relationships between probability of eDNA capture and covariates. Based on this research, traditional surveys are more effective than are eDNA surveys for H. scutatum during the nesting season. Understanding the limitations of eDNA survey methods will assist land managers with determining its use and application for H. scutatum in Minnesota.
Recommended Citation
Roberts, Alyssa, "Using Environmental DNA to Detect the Four-toed Salamander (Hemidactylium scutatum) in Minnesota" (2024). Culminating Projects in Biology. 87.
https://repository.stcloudstate.edu/biol_etds/87
Comments/Acknowledgements
I would like to express immense gratitude to Dr. Jennifer Y. Lamb (SCSU) for tirelessly working with me on this project. She and I learned a lot together while working on this research, and our working relationship was always positive. I appreciated her professionalism, respect and willingness to learn from my experiences. From her I learned even more than I had anticipated, while also learning to embrace the growth I was experiencing as a person. I would also like to thank my committee members, Dr. Matthew P. Davis (SCSU), Dr. Sarah Z. Gibson (SCSU), and Jeremy B. Cable (US Forest Service). I greatly appreciate their feedback and support throughout the development of this project. A big thanks goes out to Dr. Davis for allowing me to use equipment like the Quantus™ Fluorometer. I benefitted from the perspective and ideas from Dr. Gibson from outside the realm of herpetology, in developing my thesis research. I would like to extend my thanks to Jeremy Cable, who also supervises my position with the US Forest Service. Without his flexibility and complete understanding, I would have experienced extreme difficulty in completing my thesis and obtaining my degree while also working a full-time job.
I would like to give immense personal thanks to my family and friends who not only helped support me through this process, but also lifted me up when life’s challenges tried to get in the way of me completing my thesis. I especially want to thank my partner, David Moody, who provided more than what was expected in the form of emotional support, especially when things were difficult. For him, I am eternally grateful, and I even forgive him for losing that newt. I would like to thank my friends and co-workers with the US Forest Service, Adam Myers, Melissa Gabrielson, Mary LaPlant, Devona Berndt, and David Hainlen, for helping me out in the field with surveys, and with organizing my field data. I have immense gratitude for Andrew Herberg (Minnesota DNR) for not only helping me collect field data, but for nurturing the spark that ignited the idea for this research in the first place. Not only did he introduce me to Dr. Lamb as a potential advisor while I was considering going back to school, but he also suggested that I be the supported graduate student in this research partnership. It is because of Andrew that everything so serendipitously worked out for me to return to school without moving away from home, while also working with a species that has been so much a part of my life. Andrew also created the funding opportunity that allowed for this research to happen.
This work was primarily funded by a Minnesota Interagency Agreement between the Minnesota DNR and St. Cloud State University. Funding was provided to me by the St. Cloud State University’s Student-Mentor Collaboration Grant in 2023 and 2024, and a 2023 Minnesota Herpetological Society Grant co-authored with Dr. Lamb. This work was also supported by a 2022 SCSU Early Career Grant to Dr. Lamb. All work was conducted according to research permits from the Minnesota Department of Natural Resources (MN DNR) (Special Permit No. 35353), the MN DNR Parks and Trails Division (No. 202349, 202443), and SCSU’s Institute for Animal Care and Use Committee (No. 17-149, Appendix C).