EP53B-1024
Modeling the cliff retreat response to base-level change in layered rocks, Colorado Plateau, USA
Friday, 18 December 2015
Poster Hall (Moscone South)
Dylan Ward and Christopher Sheehan, University of Cincinnati, Cincinnati, OH, United States
Abstract:
The retreat of cliffs is an important mode of erosion in layered rocks of variable strength. For example, the iconic Colorado Plateau landscapes of Grand Canyon, Canyonlands, and Monument Valley owe their unique forms to this process. These landscapes are the end result of incision by trunk streams followed by cliff retreat. Local interactions between stochastic rockfall and first-order channels draining a cliff regulate the cliff retreat response to a base-level fall. However, nonlinear transport dynamics, steep slopes, and variable rock strength challenge the modeling of landscape evolution in these settings. Here, we employ structure-from-motion photogrammetry to generate high-resolution DTMs of a natural experiment site on the Colorado Plateau. The site features a simple, sandstone-over-shale stratigraphy with a continuous gradient in cliffband height and evidence for an ongoing transient response to base level fall. The terrain data inform a high-resolution (dx=5 m), 2D numerical model of cliffband erosion. The model simulates the interaction of three primary processes: fluvial erosion and sediment transport; hillslope transport of regolith, including shallow landsliding; and rockfall from resistant units. Crucially, the model allows us to modify stratigraphy arbitrarily to examine the landscape response to parameters such as thickness, spacing, and dip of resistant units. Results indicate that the contrast in fluvial erodibility sets the pattern of emergence of cliffs as a resistant layer is exhumed, while the difference in weathering rates across rock types sets the rate at which cliffs emerge. Once rockfall begins, erosion rates are modified by the thickness of the resistant layer, which sets the volume of rockfall debris reaching the channels below the cliff. The modeling highlights the need for process-based understanding of the conditions for cliff failure by rockfall and redistribution of the debris in very steep, rapidly-eroding landscapes.