Date of Award

Fall 2022

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Geology and Geophysics

First Advisor

Hull, Pincelli

Abstract

Since the proliferation of animal life during the Cambrian (ca. 541-485 million years ago), the biosphere has changed considerably, including long-term trends towards increasing taxonomic and functional diversity within and across ecosystems. Whether and how long-term trends in the evolution of life affected the Earth system is an active area of research, in part because the function and dynamics of ecosystems cannot be inferred by simply summing up the effect of all individuals present. Instead, the effect of diversity on ecosystem-level properties is modulated by the network of links between individuals, including trophic (e.g., predator-prey) and non-trophic (e.g., symbiosis and competition) interactions. For this reason, network-based approaches provide a means of reconstructing ecological systems through time to understand whether the ways in which communities are built and interact has changed and, if so, how these changes affect emergent properties such as ecosystem function and stability. However, there is currently no framework that directly connects species and paleocommunity dynamics, which is necessary for future studies of how community processes contribute to macroecological and macroevolutionary processes. The goal of this dissertation is to provide such a framework and to pave the way for research connecting those dynamics to biospheric processes. Evaluating paleocommunities requires an understanding of the nature of the underlying data. Fossil data are biased in several ways, including the preferential preservation of certain environments and organisms, which skews apparent faunal (and ecological) composition. My estimates of how many organisms in a marine assemblage would likely leave fossil evidence and how this varies with environment and substrate permits an assessment of information loss which can be used to moderate comparisons of the diversity of different fossil assemblages. I build upon this work by considering how such information loss scales from skewing diversity estimates to skewing apparent community structure. I show how the preferential preservation of biomineralizing organisms alters inferences of food web structure in predictable ways because the taxa preferentially lost (i.e., soft-bodied organisms) play specific ecological roles that are not represented by taxa that typically leave fossil evidence. This indicates that soft-bodied organisms play important roles in marine communities and suggests that correlations between preservation potential and ecology have pervasive effects on our understanding of community structure. With an understanding of the nature of paleocommunity data, I present and test a method of inferring ancient food webs using fossil trait data. I also consider how a network of feasible trophic interactions (i.e., a metaweb) compares to a typical food web based on observed (i.e., realized) interactions. To reconstruct this “typical” food web I apply a previously hypothesized distribution of interactions to metawebs in order to replicate the distribution of realized interactions in modern communities. Metawebs and “hypothetical realized webs” have distinct uses: the former for exploring eco-evolutionary processes, the latter for identifying structural differences with modern systems. The predictive accuracy of this inference method suggests that readily available fossil data can be used to generate and compare ancient food webs across deep time. I use these methods to demonstrate significant changes in trophic structure across the Phanerozoic. The results reveal fundamental differences in how communities assembled and operated in the past, which may indicate that data from ancient systems are critical to developing a comprehensive theory of macroecology. Interactions are distributed more evenly in modern communities than ancient systems, suggesting that communities developed greater resistance to perturbations through time. In sum, this research has improved our understanding of the nature of paleocommunity data and demonstrated that such data can be used for robust investigations of how species- and community-level dynamics have changed the course of animal evolution. The new approach to inferring ancient food webs reveals significant differences between the structures of ancient and modern trophic systems, but more work is necessary (e.g., analyses of systems between the Cambrian and present day) to determine the mechanisms underlying such differences. More broadly, my results lay the foundation for further research connecting individual-, community-, and perhaps even ecosystem-level processes across deep-time. Such work will help to identify how emergent properties, such as stability, relate to macroecological and macroevolutionary trends through animal history.

Download

Share

COinS