Paleohydraulics of Extreme Coastal Events: Using Laboratory Experiments To Evaluate Suspended Sediment Inverse Models

Past occurrences of extreme events such as climate-influenced storm surges are often recorded, imperfectly, in coastal sediment deposits. Event magnitudes are particularly challenging to quantitatively decipher from the stratigraphic record. Although flow hydrodynamics appear to influence deposit grain size distributions (GSDs) and thicknesses, most sediment transport-based inverse models for constraining paleohydraulics have not been rigorously evaluated or validated. We conducted idealized flume experiments to evaluate how sorting is controlled by suspended sediment transport and settling, how source GSDs influence deposit GSDs, and how well a proposed advection-settling model can predict flow depths and velocities. Scaled bores (representing storm surges or tsunamis) were created in a 32 m flume by raising a computer-controlled lift gate and releasing ~6.5 m³ of impounded water. The resulting flow entrained sand from an upstream “source dune” and deposited it along the flume (“inland”) due to both bedload and suspended load transport. Source GSDs and ponded water depths were systematically varied. Advection-settling models have previously assumed that coarse size fractions such as D₉₅ will best predict flow depths and velocities. However, our experiments show that intermediate GSD percentiles such as D₅₀ (the median diameter) are often more appropriate, because the spatial distribution of coarser size fractions in deposits are strongly influenced by longitudinal dispersion caused by turbulence-controlled mixing. The availability of sizes in the source GSD also can influence model predictions. We put rational constraints on uncertainties for depth and velocity predictions. Finally, we present geometric and dynamic (undistorted Froude) scaling to show how the experiments compare to natural events, and generalize our results to other coastal and depositional systems in which stratigraphy may be useful for inferring flow characteristics.