The sedimentary record is a crucial archive of past surface processes, including ancient river dynamics, on Earth. Rivers are sensitive to allogenic (external) forcings, such as changes to tectonic and climatic boundary conditions, and can respond to these forcings by propagating environmental signals, such as changes to sediment supply and grain-size, throughout fluvial networks. In theory, environmental signals associated with these allogenic forcings are preserved in depositional stratigraphy. However, rivers are also sensitive to autogenic (internal) forcings, such as channel migration and avulsion, which generates “noise” in depositional stratigraphy. Stratigraphy therefore represents the time-integrated product of the movement of water and sediment, in response to both allogenic and autogenic forcing, across Earth’s surface in the geological past. The ability to reconstruct mass fluxes in time and space from the continents, and to extract these signals from fluvial strata, provides unique insights into the dynamics and behaviour of the Earth system in the geological past.
In this thesis, I explore methods to investigate the dynamics and behaviour of ancient fluvial systems from fluvial strata. While qualitative methods (e.g., facies mapping, logging) provide useful insights, these insights are limited in the extent to which they can be used to investigate ancient river dynamics. Here I explore and develop quantitative methods to investigate ancient fluvial systems. Quantitative insights are crucial to constrain total water and sediment discharges, spatial and temporal trends in river dynamics, and the magnitudes and frequencies of river response to forcing. At large spatial and temporal scales, I present a new method to reconstruct water and sediment discharges in palaeo-catchments — this method exploits access to palaeo-digital elevation models and general circulation climate model results, both of which are now becoming increasingly sophisticated. At smaller spatial and temporal scales, I present a framework to reconstruct morphologic (e.g., flow depths, slopes) and hydrodynamic (e.g., flow velocities, water discharges) parameters from fluvial strata, and I present a new method to reconstruct river planform from field-derived observations. These methods and frameworks can provide a range of insights into ancient fluvial systems in general and, in this thesis, I successfully apply them to ancient fluvial systems in the Late Cretaceous North American continent.
Beyond the development of new methods and frameworks to investigate the behaviour of ancient fluvial systems, this thesis also presents unambiguous stratigraphic evidence for bedform preservation in non-steady, or disequilibrium, conditions for Upper Cretaceous fluvial deposits in North America. These observations challenge the use of steady-state bedform preservation models in palaeohydraulic reconstructions. My results highlight the importance of making systematic field measurements of cross-set geometries in fluvial strata to determine the nature of bedform preservation, and the importance of considering disequilibrium dynamics for palaeohydraulic reconstructions. Further, these results provide a potentially powerful avenue to quantify flood variability from fluvial strata, and I explore the necessary future work to achieve this research goal.
Together, the methods and frameworks that I present in this thesis advance our ability to extract quantitative information from fluvial strata. Further, in exemplifying these methods and frameworks for ancient fluvial systems of the Late Cretaceous North American continent, I highlight the advantages, limitations, and best practices associated with these approaches. These methods and frameworks can be implemented at a variety of spatial and temporal scales and can provide sophisticated insights into the dynamics and behaviour of ancient fluvial systems, both on Earth and other planets.