Abstract

Understanding the drivers of community assembly is a key goal in ecology and conservation. One way to better understand community assembly is to understand how the presence or absence of dominant species– highly abundant species with a proportionally large effect on the ecosystem–affects assembly. This approach allows for a more generalized understanding of community assembly based on the presence or abundance of one species rather than pairwise interactions among all species in a community. In forests, studies on dominant tree species have focused on how they affect patterns of species diversity. In contrast, surprisingly little is known about how dominant species influence key processes of community assembly, such as non-random (deterministic) selection resulting from niche differences among species, random (stochastic) changes in species relative abundance (ecological drift), and dispersal. Additionally, the herbaceous layer of forests often does not receive as much attention as dynamics among trees, despite herbs making up a majority of the species diversity in temperate forests. Thus, my dissertation focuses on how a geographically widespread and locally dominant tree species–pawpaw (Asimina triloba, Annonaceae)–affects the assembly, biodiversity, and abundance of forest understory herbs in a temperate oak-hickory forest. My dissertation aims to accomplish three goals: 1) determine how the presence of dominant species influences patterns of herb species composition, diversity, and abundance across spatial scales, 2) determine how dominant species alter the relative strength of the high-level assembly processes of selection, ecological drift, and dispersal to produce these patterns, and 3) elucidate the biological mechanisms by which dominant species are altering these patterns and processes. In Chapter 2, I determined the relative importance of deterministic and stochastic processes in assembling the herbaceous layer inside and outside of pawpaw patches by using a combination of observed local diversity (alpha-diversity), local community size (total abundance of all species), and observed and null-model simulated compositional variation (beta-diversity). I found that inside pawpaw patches, there was lower local diversity, lower community size and lower abundances (higher rarity) of a large number of herb species, and beta-diversity that was both higher and closer to the null (stochastic) expectation compared to outside pawpaw patches, all suggesting that community assembly is more random inside pawpaw patches, though deterministic assembly still plays an important role. In Chapter 3, I investigated how pawpaw influences deterministic or stochastic assembly through abiotic selection or effects on community size. I used variation partitioning to explain herb composition as a function of local environmental conditions (light levels, soil properties, fire disturbance) and spatial processes (geographic distance) inside and outside pawpaw patches. I found that local environmental conditions explain less variation in herb composition inside compared to outside pawpaw patches, suggesting weaker niche selection (species sorting) and stronger ecological drift inside pawpaw patches with smaller community size. Community size increased with light availability, suggesting that pawpaws increase ecological drift by decreasing light availability and increasing rarity of herb species inside pawpaw patches. Soil variables were significant predictors of herb composition, suggesting that spatially variable selection outside of pawpaw patches is associated with species sorting along local soil-resource gradients, but that pawpaws weaken these plant-soil associations. Finally in Chapter 4, I experimentally tested the degree to which the rarity of three herb species inside pawpaw patches is caused by three types of recruitment limitation: seed limitation, establishment limitation, and limited reproduction (flowering). In one herb species (Desmodium glutinosum), I found that local abundances are seed-limited across both patch types, but that pawpaw does not appear to alter the degree of seed limitation. There was some evidence for establishment limitation for Desmodium glutinosum seedlings being higher inside pawpaw patches, but establishment for transplanted seedlings was not significantly different in the pawpaw patches for any of the three species. Flowering was significantly reduced inside pawpaw patches for two of the three species investigated. This suggests that recruitment inside pawpaw patches is most limited by reproduction and that the presence of pawpaws may promote source-sink dynamics. Together, my dissertation illustrates how a locally dominant tree species increases stochastic community assembly, alters abiotic conditions and species sorting, and increases local rarity of forest herbs by limiting establishment and reproduction. While pawpaw trees are part of a healthy landscape, their impact on understory herbs should not be ignored in land-management plans and conservation. These findings could potentially be generalizable to other dominant species across forest types and other ecosystems.

Committee Chair

Jonathan Myers

Committee Members

Christine Edwards; Kim Medley; Matthew Albrecht; Rachel Penczykowski

Degree

Doctor of Philosophy (PhD)

Author's Department

Biology & Biomedical Sciences (Evolution, Ecology & Population Biology)

Author's School

Graduate School of Arts and Sciences

Document Type

Dissertation

Date of Award

8-18-2025

Language

English (en)

Author's ORCID

https://orcid.org/0000-0002-7778-4996

Download

Included in

Biology Commons

Share

COinS