Soil Production Limits and the Transition to Bedrock Dominated Landscapes

The prevalence and persistence of the Earth's soil mantle depends on a long-term balance between soil production and erosion. Higher soil production rates under thinner soils provide a critical stabilizing feedback mechanism, and previous work established a paradigm that climate- and lithology-controlled soil production sets the upper limit for steady-state hillslope erosion. In such a model, erosion rates exceeding the maximum soil production rate could only be due to bedrock mass movements. Yet, observation of pervasive, if patchy, soil cover in areas of rugged topography and rapid erosion suggests that this framework is incomplete. Here we use new and published Be-10 soil production and detrital erosion rates to show that soil production rates increase with increasing catchment-averaged erosion rates, enhancing the persistence of soil cover. Specifically, we focus on upland, colluvial soils and present a new dataset of 58 soil production rates quantified from Be-10 concentrations in soil pits across the San Gabriel Mountains (SGM) of California: from catchments spanning two orders of magnitude in erosion rate as the landscape varies from gentle, soil-mantled, and creep-dominated in the west to steep, rocky and landslide-dominated in the east. These rates are the first to cross the transition from soil-mantled to rocky hillslopes. They clearly document local soil production rates in steep, rapidly eroding terrain that greatly exceed the maximum soil production rate (SPmax) determined for low-relief, soil-mantled areas with similar climate and lithology. We show that a process transition to landslide-dominated erosion in steeper, more rapidly eroding catchments results in thinner, patchier soils and rockier topography, but find that there is no sudden transition to bedrock landscapes. Instead, using a global data compilation, we suggest that soil production processes may increase in frequency and magnitude to keep up with increasing erosion rates. These data imply that existing models greatly exaggerate changes in critical zone processes in response to tectonic uplift. We complement this interrogation of process rates with analysis of high-resolution LiDAR topographic data and adapt a numerical model to explore the role of episodic landslides in the gradual transition from soil-mantled to bedrock dominated landscapes.