Date of Award


Degree Name

Doctor of Philosophy


Plant Biology

First Advisor

Gibsno, David


During European settlement, the vast majority of grasslands in Illinois were converted for agricultural purposes. Some of the remaining natural areas in southern Illinois include natural xeric forest openings (i.e., barrens, glades, outcrops), that have transitional community compositions representative of previously extensive grasslands and adjacent hardwood forests. Previous research in these forest openings show that the communities are largely driven by edaphic conditions and vary spatially across southern Illinois. While most of these communities are currently protected and established as nature preserves, threats to these natural xeric forest openings continue to persist, such as climate change and exotic invasion. These threats are capable of altering taxonomic, phylogenetic, and functional diversity and community composition. The overall goals of this research were to 1) determine metapopulations and metacommunity structure in a local group of sandstone outcrop communities (a subset of natural xeric forest opening communities), 2) expand resolution to a regional scale to include natural xeric forest openings with several substrates to investigate changes in taxonomic, phylogenetic, and functional diversity and community composition since surveys conducted in 1988, and 3) include spatial analyses to characterize autocorrelation structure of diversity and environmental variables and biological turnover of natural xeric forest openings at a global scale. Sandstone outcrop communities at Jackson Hollow in Pope County, Illinois had several metapopulation species based on Hanski’s incidence functional model. These metapopulation species were more often members of the Asteraceae and Poaceae than members of other plant families, and were often exotic in origin with short-lived lifecycles. These metapopulation species were also neither dominant species or singletons within sandstone patches. Based on the Elements of Metacommunity (EMS) Framework, positive coherence, species turnover, and boundary clumping indicate that these sandstone outcrop communities are predictable communities where species replace one each other regularly as groups of species and respond similarly to environmental gradients. Furthermore, diversity metrics were all positively correlated with each other, but not with patch characteristics. In addition, total species and metapopulation species were positively associated with phylogenetic and functional diversity, but metapopulation species were positively associated with non-standardized phylogenetic and functional indices. When looking at several natural xeric forest openings across southern Illinois and comparing them based on substrate types and between surveys in 1988 and 2019, taxonomic, phylogenetic, and functional compositions were distinct based on substrates, but were stable and resilient across surveys. Sandstone and shale communities were the most similar based on composition, followed by limestone communities compared to sandstone and shale communities, and then loess communities being the most dissimilar from all other communities. Environmental variables that best explained differences in community composition were canopy cover, soil acidity, photosynthetically active radiation, and soil depth. Diversity variables that best explained differences in community composition were phylogenetic nearest taxon index, Faith’s phylogenetic diversity, and dominant species richness. Ninety-four different species were either significant indicators for specific substrates, or 2019 surveys. While no cases of phylogenetic signal were observed based on functional traits, substrate types, or surveys, three cases of functional signal based on dominant communities were observed in sandstone communities in 2019, shale communities in 2019, and overall shale communities across 1988 and 2019 surveys. When applying spatial analyses to these natural xeric forest openings to understand autocorrelation structure and biological turnover, soil depth was the only environmental variable that exhibited significant spatial autocorrelation, as previous glacial events caused loess hill prairies in the northwestern extent of this study to have deeper soil due to Quaternary loess deposition. However, several diversity metrics exhibited spatial structure based on 1988 and 2019 surveys (1988: dominant species richness, Pielou’s evenness, Shannon-Weiner diversity, Faith’s phylogenetic diversity, and distance-based functional diversity; 2019: dominant species richness, Pielou’s evenness, Shannon-Weiner diversity, Faith’s phylogenetic diversity, and phylogenetic nearest taxon index). In addition, climate variables daily mean temperature and total annual precipitation exhibited spatial structure. Most variables were spatially clustered at local site scales and spatially dispersed at larger scales and spatially over-dispersed at a regional scale. Generalized dissimilarity models constructed based on elevation, soil, and climate variables showed that models based on both functional composition and 2019 surveys best explained biological turnover compared to taxonomic and phylogenetic and 1988 models. Several variables included in models differed based on aspects of biological turnover or surveys, but temperature annual range and soil bulk density variables were common across all models. However, differences between total deviance explained and null deviances show that geographic distance between natural xeric forest openings was overwhelmingly the most influential variable contributing to biological turnover. Given these observations, natural xeric forest openings at local scales persist as distinct habitat patches amidst a landscape that largely constitutes an inhospitable matrix to colonizing species, yet certain species are able to migrate between patches. The resulting community assembly of individual patches is determined by both colonizing species and environmental gradients across the landscape. Despite simultaneous threats to diversity and composition in natural xeric forest openings at a regional scale, substrate continues to drive community assembly, in that certain species are characteristic indicators of these substrates. Furthermore, the most recent surveys in sandstone and shale communities exhibited functional signals. However, these natural xeric forest openings are spatially structured based on soil depth due to glacial history, certain metrics of diversity, and climate variables. Along with these occurrences of spatial autocorrelations, different aspects of biological turnover are best predicated by geographic distance as well as unique combinations of climate and soil variables between different aspects of diversity and surveys.




This dissertation is only available for download to the SIUC community. Current SIUC affiliates may also access this paper off campus by searching Dissertations & Theses @ Southern Illinois University Carbondale from ProQuest. Others should contact the interlibrary loan department of your local library or contact ProQuest's Dissertation Express service.