Stratigraphy provides the best record of Earth history, tracking the coevolution of surface geochemistry, biology, geodynamics, and climate. Stratigraphic information, however, is often utilized only qualitatively, lacking straightforward pathways for quantitative use. The unifying project of the chapters in this dissertation is the quantification of stratigraphy in space and time and the novel insights that can be obtained through this quantification on questions ranging from planetary geodynamics to modes of climate sensitivity during glacial states.
Chapter 1 addresses the spatial evolution of Laurentian macrostratigraphy over the Phanerozoic. In particular, I present a unifying model for the well-known Sloss Sequences, which capture several episodes of shallow marine flooding of Laurentia. Using Macrostrat, a geospatial database of geology, I show that the geometry of each marine flooding episode associated with a Sloss Sequence is significantly correlated with subduction under a nearby margin of Laurentia. This result is consistent with the model that subduction generates dynamic subsidence of continental interiors. A notable exception to this pattern is the first major flooding event, the Sauk Transgression, which seals the Great Unconformity, yet occurs when Laurentia lacks any convergent margins. Despite the spatiotemporally heterogeneous pattern of marine flooding, the total marine flooding of Laurentia reproduces globally-derived estimates for sea level evolution over the Phanerozoic. The resolution to this apparent paradox rests in the geodynamic mechanism for global sea level evolution: dynamic subsidence of continents in response to subduction must be balanced (by mass conservation) by uplift elsewhere on Earth, namely at oceanic spreading ridges. The subduction-driven flooding of Laurentia is therefore part-and-parcel of the "hypsometric see-saw" that dictates global sea level, explaining this correlation. The hypsometric see-saw also explains, for the first time, the Sauk Transgression: while Laurentia lacked convergent margins, newly-formed Gondwana was ringed by approximately 35,000 km of subducting margins. This subduction caused dynamic subsidence of the supercontinent, balancing uplift in oceanic spreading centers and displacing seawater globally, including onto Laurentia, producing the Sauk Transgression.
Chapter 2 focuses on the statistical distribution of time in glaciomarine stratigraphy, which can be tied back to the spatial geometry of sedimentary dynamics. I draw on recent theory developed around fundamental statistical quantities—completeness and the compensation timescale and length scale—that control the distribution of time in stratigraphy. This theory, while general to any depositional setting, has largely been developed in the context of fluvial systems. I apply these concepts for the first time to the glaciomarine realm, hypothesizing that the compensation length scale—which is set by the deepest channel depth in a fluvial setting—is set by the vertical amplitude of grounding line motion. Furthermore, when ice sheets are sensitive to orbital forcings, the amplitude of grounding line motion is large, dozens to hundreds of meters, making a prediction for the value of the compensation length scale. I use the ANDRILL AND-2A Antarctic sediment core to test this hypothesis. I develop a novel Bayesian age modeling methodology that probabilistically infers sedimentation rates within beds and hiatus durations between beds in stratigraphy, thereby also yielding a new probabilistic age model for the core. The posterior distribution of hiatuses constrains a compensation timescale of 129–402 kyr for the orbitally-forced Early Miocene interval than spans most of the core, corresponding to a compensation length scale of 21–67 m. This range of values is consistent with the prediction generated by my hypothesis. This work provides a quantitative benchmark and model for evaluating the role of orbital forcing in deeper time glaciomarine stratigraphy.
Chapter 3 presents novel geochronology and quantification of Cryogenian glaciomarine stratigraphy from Fransfontein Ridge in Namibia. I present three new U-Pb CA-ID-TIMS ages from the Ghaub Formation, which provides the sedimentary record of the Marinoan Snowball Earth glaciation. Based on detailed mapping and stratigraphy, I demonstrate that the oldest of these ages at ca. 639 Ma provides a direct constraint on the onset of the Marinoan Snowball, which presently has 11 Myr of uncertainty. This result yields a ca. 4 Myr duration for the Marinoan Snowball, which is an order of magnitude shorter than the preceding 56 Myr Sturtian Snowball. In addition to providing dateable ashes, the Ghaub Formation at Fransfontein Ridge presents beautiful, nearly 100% exposed outcrop of grounding line advance-retreat cycles. Previous proposals for orbitally-forced ice sheets have been proposed to explain this style of Cryogenian sedimentary cyclicity. The ages I report constrain the timescales of this grounding line motion, suggesting that the ca. 10 cycles, if accumulated over the entire 4 Myr Marinoan Snowball, are unlikely to be orbital in origin. Furthermore, because the excellent outcrop affords the opportunity to directly quantify the compensation length scale from stratal geometries, I use the results and insights from Chapter 2 to compare the Ghaub Formation with the AND-2A record, which is known to record the dynamics of orbitally-forced ice sheets. Using hundreds of bed traces on 5 mm/px drone imagery over the best exposures of the Ghaub Formation, I constrain a compensation length scale of 3–20 m. This range of values is significantly smaller than for AND-2A, demonstrating that the grounding line motion recorded in the Ghaub Fm was greatly subdued in comparison. This result is inconsistent with strong sensitivity to orbital forcing during the Marinoan Snowball glaciation.
