Return to search

Conservation at the speed of light: Applications of non-invasive technologies for assessing physiological phenomena in amphibians

The Anthropocene epoch in which we are currently living, also known as the Holocene, has brought about unprecedented losses in planet Earth’s biodiversity. Numerous extirpations of floral and faunal species have been influenced by human encroachment and more specifically, the exploitation of such species and the respective habitats in which they reside. It is this notion that has propelled many scientists to take up intellectual arms in an effort to protect these invaluable resources. The purpose of this research was to develop technologies to measure and evaluate various variables that influence animal physiology, specifically in amphibians who represent the most threatened class of all animal taxa. Species-specific knowledge including life history and an understanding of evolutionary traits are often needed to effectively guide the management decisions surrounding any given animal population. Specific objectives of this project were to develop non-invasive methods, such as hormone monitoring, machine learning-aided ultrasonography, and near-infrared spectroscopy (NIRS), to assess vital physiological traits, such as biological sex, reproductive status, and chytrid fungus pathogen detection in threatened amphibian species. The novel technologies developed and applied in amphibians here may provide insights for addressing conservation related questions in other animal as well as plant species. Additionally, automation of physiological monitoring techniques through the use of machine learning methods reduces barrier to entry and enables these technologies to be operated by a larger practitioner base. This research also serves to advance methods surrounding chemometric analyses as it pertains to the discipline of wildlife spectroscopy, where large multivariate datasets require data manipulation strategies to produce robust prediction models for the physiological trait of interest for qualitative or quantitative assessment. To that end, a multi-model framework is provided for optimizing predictive outcomes to address questions relating to wildlife management and conservation initiatives.

Identiferoai:union.ndltd.org:MSSTATE/oai:scholarsjunction.msstate.edu:td-7103
Date10 May 2024
CreatorsChen, Li-Dunn
PublisherScholars Junction
Source SetsMississippi State University
Detected LanguageEnglish
Typetext
Formatapplication/pdf
SourceTheses and Dissertations

Page generated in 0.0023 seconds