Bringing Realistic Stratigraphy into Evolutionary Biology through Age-Depth Modeling

Presentation "Bringing Realistic Stratigraphy into Evolutionary Biology through Age-Depth Modeling" held at the "Stratigraphy in Modern Geoscience: methodologies, applications and future relevance" meeting London, 23rd of January 2024) organized by The Geological Society.

Abstract text:

The stratigraphic record allows observation of fossil evolution over longer time scales than are accessible to direct human observation. However, direct dating of fossils in deep time is impossible, making it necessary to estimate their age from indirect methods. The age inference is complicated by stratigraphic effects such as changing sedimentation rates and hiatuses. Age-depth models assign chronological ages to stratigraphic positions, thus establishing a temporal framework for the fossils they contain, but are themselves dependent on stratigraphic architectures. A robust understanding of these architectures is fundamental for establishing realistic, accurate age-depth models, and reconstructing evolutionary change from the fossil record.

We present a novel mathematical framework for age-depth modeling implemented in the R software. It includes two non-parametric methods to estimate age-depth models from complex stratigraphic data and to quantify the associated geochronological uncertainties. In contrast to other age-depth modeling frameworks, it allows to directly transform complex data structures and mathematical objects such as phylogenetic trees and time series from the time domain to the stratigraphic domain and vice versa. As examples of the developed framework, we

1.    construct age-depth models for Devonian strata in the La Thure section, Belgium, using sedimentation rates constrained by cyclostratigraphic methods.

2.    use measurements of extraterrestrial ³He from ODP site 960 (Maud Rise, Weddell Sea) to construct age-depth models for the Paleocene–Eocene thermal maximum.

3.    examine the preservation of phylogenetic relationships along the onshore-offshore gradient in a carbonate platform simulated in the CarboKitten model.

The presented framework simplifies the interface between stratigraphy and other domains of science, making it easier for communities outside of sedimentology to understand and use stratigraphic information in inter-disciplinary approaches. It demonstrates that a modern, model-driven understanding of stratigraphy is highly relevant to all disciplines drawing information from geohistorical records.