Modern analyses of complex datasets in plant ecology and conservation




Northup, Alison Pechin

Journal Title

Journal ISSN

Volume Title



Ecological data are often difficult to analyze, with complexity inherent in their real-world provenance, and with technological advancements and the passage of time resulting in publicly-available ecological datasets that are constantly growing in size. Timely ecological questions and large, complex datasets demand modern, innovative analytical approaches. This research comprises three cases where creative analytical approaches are used to tease out answers to questions about conservation and climate change impacts. The first case attempts to resolve a standing question: which of the two dominant tree species on the Edwards Plateau, TX, is more resistant to drought? Starting with an erratic sap flow dataset, environmental data are used to select the most relevant portions of the time series, and the application of a Bayesian model creates enough statistical power to suggest an answer to the question, helping resolve inconsistencies in prior literature. The second case makes use of herbarium sheet images and a spatially-explicit Bayesian model to explore continental-scale flowering phenology among a subfamily of trees in the New World Tropics, with surprising results. The third case uses environmental raster datasets and spatial optimization software to determine which areas within a region of West Texas are most vulnerable to future oil and gas development from a biodiversity and ecosystem services perspective. Together, these cases show the power that modern and creative analytical techniques can bring when applied to complex data sets.


LCSH Subject Headings