Modeling Stratigraphic Paleobiology in R
Introduction
Stratigraphic paleobiology is a methodological framework that applies physical stratigraphy - which includes event deposition, sequence-stratigraphic architecture, and sedimentary basin analysis - to the interpretation of the fossil record (Patzkowsky and Holland 2012; Holland, Patzkowsky, and Loughney 2024). The discipline is founded on the principle that the fossil record is a joint expression of ecological, taphonomic, evolutionary, and stratigraphic processes, meaning that fossil occurrences are controlled not only by where organisms lived and when they evolved, but also by when and where sediment capable of preserving them accumulated (Holland and Patzkowsky 2015; Holland, Patzkowsky, and Loughney 2024). Because species are distributed along environmental gradients such as water depth, and because sequence-stratigraphic architecture systematically controls sediment deposition, first and last occurrences of fossil taxa cluster predictably at sequence boundaries, flooding surfaces, and surfaces of forced regression, even when no change in extinction or origination rate has occurred (Holland 2020, 1995).
Stratigraphic paleobiology has changed our understanding of mass extinctions: simulation studies and field tests (e.g. Italy’s Po Plain) demonstrated that what appeared to be sudden, stepwise extinction events may actually reflect long intervals of elevated extinction rate compressed by stratigraphic architecture into apparent clusters of last occurrences (Holland 2020; Holland and Patzkowsky 2015).
What do you need to do stratigraphic paleobiology?
The first studies focused on first and last occurrence data (e.g. Holland and Patzkowsky (2015); Nawrot et al. (2018); Hohmann (2021)), but subsequent research extended it to occurrences in bin (Jarochowska et al. 2018), diversity (Loughney et al. 2021), abundance (Danise and Holland 2017), trait value measurements (Hannisdal 2006; Hohmann et al. 2024) and assemblage composition (Belanger and Bapst 2023) - all types of data representing fossils or other variables stored in the physical geological record are potentially subject to stratigraphic paleobiology. Secondly, a representation of the stratigraphy is needed. This can be: lithofacies, systems tracts, estimates of depositional rate, or another class. Here we advocate for using an age-depth model and show how they can be used even when no data on absolute age is available.
Depositional systems
All depositional systems have their defining physical, chemical and biological processes; the same biological event will be preserved differently on a clastic shelf than on a carbonate shelf or in a river delta (Holland 2023). Much previous work focused on siliciclastic marine settings (Hannisdal 2006; Holland and Patzkowsky 1999, 2015) and some on terrestrial (Holland 2021, 2022) and mixed, cool-water carbonates (Scarponi et al. 2017; Nawrot et al. 2018). Also a transition between systems will have a strong effect on fossil preservation (Danise and Holland 2017). Here we will use tropical carbonate platforms (Hohmann et al. 2024), because the effects in them have been largely predicted, but not well tested yet. If you are interested in modeling a different depositional setting, please discuss with us and we may be able to point you to a suitable model.
Learning Objectives
By the end of this workshop, you will be able to
build age-depth models to quantify stratigraphic incompleteness and hiatus distributions across depositional environments,
simulate and analyze different modes of trait evolution (stasis, random walk, and Ornstein-Uhlenbeck processes) and identify how stratigraphy affects their expression in the fossil record,
model ecological niche preferences and understand how environmental gradients control fossil abundance patterns, and
simulate last occurrence data to recognize how stratigraphic phenomena like hiatuses and condensation surfaces can create artifactual “mass extinction” signals.
You will develop proficiency in using R’s pipe operator |> to build reproducible, modular analysis workflows that combine the StratPal and admtools packages.
Structure
The webpage and workshop is structured as follows:
Getting Started covers software dependencies, package installation and loading, and a brief introduction to the pipe operator in base R (10 min).
Stratigraphic Architectures provides an overview of the example stratigraphic architecture used in the workshop, and introduces age-depth models (45 min).
Trait Evolution explores how stratigraphic effects changes patterns of phenotypic evolution (45 min).
Event Data discusses how data such as fossil abundance or last occurrences are modified by ecological, taphonomic, and stratigraphic effects (45 min).
Advanced Topics covers miscellaneous topics, e.g., combining niche models with trait evolution, and modification of phylogenetic trees.
Contributing
Did you find a typo, a mistake, or do you thing some topics are not sufficiently covered? Please post your issue in the GitHub repo.
Funding
Funded by the European Union (ERC, MindTheGap, StG project no 101041077). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Research Council. Neither the European Union nor the granting authority can be held responsible for them. 