This thesis is submitted in two chapters. The first chapter contains background research, literature review, relevant information, and justification for the primary study. The second chapter is written as a standalone paper for submission to Ecology and Evolution, that can be read and interpreted independently of the first chapter. Specialization is a widespread but highly ambiguous and context dependent ecological concept. Here, we construct a framework to assess specialization using an objective, metric-based approach, utilizing study system Quercus. To create Metric-Based Specialization rankings, metrics related to specialization were gathered for 141 Quercus species (Number of distinct inhabited ecoregions, extent of occurrence, plasticity of functional leaf traits, presence of domatia, number of notable documented interspecies interactions), and a model selection process was utilized to determine which were the best predictors of species threat level as determined by the International Union for Conservation of Nature (IUCN Red List) and the results of a specialization survey, where experts familiar with Quercus were asked to score species based on their level of ecological specialization. Alignments between Metric-Based Specialization Rankings, Quercus experts, and IUCN data show that specialization studies can be standardized, allowing for easier meta-analyses and comparisons across studies. Rankings, and the metrics they are comprised of, were analyzed for evolutionary trends using Phylogenetic Generalized Least Squares, Ancestral Character State Reconstruction, and PhylogeneticEM, for shift detection. Occurrence data were mapped to assess geographic distributions of species. Clustering analyses were performed on mean Metric-Based Specialization Rank and the number of distinct species by ecoregion for the continental United States, Mexico, and Central America, showing clear delineations in Rank across regions. Species tend to rank as more specialized in climates with extreme water availability or precipitation seasonality such as the tip of Florida, California, and Southern Mexico, while Eastern North American species are largely generalist. While the metrics that Metric-Based Rankings are comprised of show evolutionary relationships, sorting by region and environmental factors such as precipitation seasonality (Bioclimatic Variable 15) shows a stronger influence on specialization. Novel use of leaf functional traits extracted from herbarium specimens is shown to be of value to large-scale investigations of a clade. Metric-based systems are shown to be useful tools for bulk identification of at-risk species, and experts are shown to be reliable when evaluating the specialization level of a species, though it seems they largely rely upon native ranges for their determinations. We find that specialization acts as an emergent property of a species’ native region, and similar ranking systems should be utilized to better identify and preserve threatened species, especially under conditions of accelerated climate change.


Matthew Kaproth

Committee Member

Christopher Ruhland

Committee Member

Michael Minicozzi

Date of Degree




Document Type



Master of Science (MS)

Program of Study

Biological Sciences


Biological Sciences


Science, Engineering and Technology



Rights Statement

In Copyright