How well can hillslope evolution models “explain” topography? Simulating soil transport and production with high-resolution topographic data

The morphology of hillslopes is a direct refl ection of tectonic forcing and climatic and biologic processes that drive soil production, mobilization, and transport. Soil transport on hillslopes affects river incision by providing tools for channel abrasion and controls the distribution of sediment that infl uences aquatic habitat. Although numerous hillslope transport relationships have been proposed over the past 60+ years, a comprehensive analysis of model predictions for a real landscape has not been performed. Here, we use high-resolution topographic data obtained via airborne laser swath mapping (ALSM) to simulate the long-term evolution of Oregon Coast Range hillslopes and test three published transport models and a new model that accounts for nonlinear depth- and slopedependent transport. Analysis of one-dimensional, steady-state solutions for these four models suggests that plots of gradient-curvature may be diagnostic for distinguishing model predictions. To evaluate two-dimensional model predictions for our fi eld site, we assumed local steady-state erosion for a 72,000 m 2 sequence of hillslopes and valleys. After calibrating each of the four models, we imposed constant base-level lowering for cells within the valley network, simulated 500,000 yr of soil production and transport, and determined which transport model best preserved morphologic patterns that describe the current landscape form. Models for which fl ux varies proportionally with hillslope gradient generated broadly convex hilltops inconsistent with the sharp-crested, steep-sided slopes of our study site, whereas the two nonlinear slope-dependent models produced convex-planar slopes consistent with current hillslope form. Our proposed nonlinear slope- and depth-dependent model accounts for how soil thickness controls the magnitude of biogenic disturbances that drive transport; this model best preserved the current landscape form, particularly the narrow, sharply convex hilltops characteristic of the Oregon Coast Range. According to our formulation, which provides an explicit linkage for relating the distribution of biota to hillslope processes, the degree of hilltop convexity varies nonlinearly with the ratio of erosion rate to maximum soil production rate, highlighting the profound infl uence of soil depth on hillslope evolution.