The full text of this item is not available at this time because the student has placed this item under an embargo for a period of time. The Libraries are not authorized to provide a copy of this work during the embargo period, even for Texas A&M users with NetID.
Using Species Distribution Models and Machine Learning to Determine Range Extent, Reduction, and Expansion of Unionid Freshwater Mussels
Abstract
Understanding how the distribution and abundance of a species is influenced by geographical and environmental processes is a fundamental goal of basic and applied ecological research. Freshwater riverine ecosystems exhibit extreme diversity harboring more than 100,000 species of insects, mollusks, and vertebrates, despite comprising less than one percent of the Earth’s non-glaciated surface. However, human alteration of riverine ecosystems has been linked with systemic declines in species richness and abundance at rates above those seen in tropical rainforests. Freshwater mussels (Bivalvia: Unionidae) are among the most imperiled aquatic fauna in the world. Declines stem from a combination of human mediated impacts to water quality and quantity, and the inability of mussels to cope with or avoid them. The results of this dissertation indicate that mussel occupancy and range distribution is influenced by environmental variables (i.e., climate, elevation, and flow). Machine learning models, species distribution models, and ensemble species distribution models can be used to explain the extent, expansion, retraction, and limitations of a species geographic range as a response to these environmental characteristics. In this study, it is shown how major dimensions of flow, climate, and topography shape the occurrence of mussels. Additionally, these models show that there is a difference in impact of specific variables to individual species. As such, modeling of predicted distributions is most effective at the species level. The creation of the Gridded River Identification System (GRIS) allows for the spatially explicit and reproducible framework to base future management and conservation actions on. Using the GRIS and projecting species distributions to different climate scenarios allows resource managers to make informed decisions on the placement and allocation of conservation resources to ensure their utility under changing conditions. Furthermore, modeling under varied climate conditions allows resource managers the ability to identify individual species that maybe in greater need of conservation under future climate change. The approaches documented here could be easily expanded to stream systems in other states, regions, and countries, providing a consistent workflow for conservation planning of imperiled species or identification of high priority biologically diverse areas.
Subject
Species Distribution ModelUnionid
Freshwater Mussels
Ensembles
Machine Learning
Random Forest
Habitat
Conservation
Systematic Prioritization
Flow
Climate Change
Suitability
Citation
Kiser, Alexander Hendrix (2023). Using Species Distribution Models and Machine Learning to Determine Range Extent, Reduction, and Expansion of Unionid Freshwater Mussels. Doctoral dissertation, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /199080.