Abstract
Generating predictions about how species will respond in a changing world requires an understanding of the processes underlying how species tend to move through space, split, and become extinct. Historical biogeographers study these processes by applying mathematical models to reconstruct how the ancestors of living species have diversified and dispersed. These mathematical models allow researchers to use data about the present to investigate the past. Many different models have been created for the purposes of explaining how species both evolve over time and move through space. These models vary widely in function and application. However, capturing the complexities of the biological world is a perpetual challenge, necessitating the construction of new models to address biogeographic questions that could not previously be answered. In this dissertation, I sought to develop phylogenetic methods for historical biogeography incorporating multiple environmental features, paleogeological evidence, and important spatial properties of species ranges. Additionally, I investigated computational workshops as a mechanism for communicating novel biogeographic models to other researchers. In Chapter 2, I introduced the Multiple Feature-Informed GeoSSE (MultiFIG) model, a biogeographic model that uses information about the environmental features of a set of discrete regions to investigate the relationship between the regions where a lineage is evolving and rates of speciation, extinction, and dispersal. I then applied the model to the South American lizard genus Liolaemus, uncovering a negative relationship between distance and dispersal, and a negative relationship between region size and extinction. In Chapter 3, I created the Ellipse Model for Phylogenetic Inference of Range Evolution (EMPIRE), a model of range evolution in continuous space. Instead of assigning species to a set of discrete regions, EMPIRE models whole species ranges as they move and change shape, size, and orientation in continuous space. It also incorporates multiple range inheritance scenarios at cladogenesis. I applied this model to the Australian Sphenomorphine skinks, reconstructing ancestral ranges across the continent. I then combined these reconstructions with paleoclimate data. I found that aridification was an important predictor of speciation, although daughter lineages after speciation did not appear to partition environmental conditions. In Chapter 4, I presented a new framework for performing likelihood-based biogeographic model selection that includes both biogeographic processes and relevant paleogeological processes. These joint models consider multiple sources of evidence for reconstructing the past, constraining biogeographic inferences to consider the most realistic paleogeographic scenarios by appropriately weighting geographical data. I applied this new framework to the Hawaiian Psychotria, and found that models with realistic paleogeology were preferred, producing more reasonable paleogeographic and biogeographic inferences than a traditional geologically-uninformed model comparison experiment. These phylogenetic models are generalizable to other study systems, intended to be used by other researchers on their own datasets. In Chapter 5, I co-instructed a workshop on phylogenetic biogeography for other researchers, and administered a survey to assess whether the workshop encouraged participants to engage with biogeographic models and which activities they found helpful. Participants reported that they were more likely to use biogeographic models after attending the workshop, and gained confidence in several key areas. The workshop activities were generally perceived positively. However, despite the participants’ investment in learning new software tools, traditional lectures were still preferred over coding activities.
Committee Chair
Michael Landis
Committee Members
Christine Edwards; Jonathan Myers; José Figueroa-López; Sebastián Tello; Toby Pennington
Degree
Doctor of Philosophy (PhD)
Author's Department
Biology & Biomedical Sciences (Evolution, Ecology & Population Biology)
Document Type
Dissertation
Date of Award
8-12-2026
Language
English (en)
DOI
https://doi.org/10.7936/qm49-zb87
Recommended Citation
Swiston, Sarah Kathryn, "How Species Evolve Through Space and Time: Designing and Disseminating Phylogenetic Methods for Historical Biogeography" (2026). Arts & Sciences Graduate Student Theses and Dissertations. 3859.
The definitive version is available at https://doi.org/10.7936/qm49-zb87