Physically-based Hydrologic-response Simulation of a Steep, Zero-Order Catchment

Near-surface hydrologic response plays a critical role in landscape evolution, particularly in steep terrain where hydrologically-driven slope instability is the predominant mechanism of sediment removal. Despite the importance of hydrology in governing the timing and extent of slope failure, the majority of slope stability modeling efforts utilize steady-state approximations to the 1D or 2D saturated subsurface flow equations to generate pore pressures. The effort reported here employs the comprehensive, physics-based Integrated Hydrology Model (InHM) to rigorously simulate near-surface hydrologic response. InHM simulates, in a fully-coupled approach, 3D transient variably-saturated flow and solute transport in porous media and macropores and 2D transient flow and solute transport over the land surface and in open channels. Our modeling approach is tested with the extensive data sets from the 860 m2 Coos Bay experimental catchment (CB1) in the Oregon Coast Range for both sprinkling experiments (event-based simulation) and natural storms (continuous simulation). The instrumentation at CB1 for characterizing the spatial and temporal variability in hydrologic response includes an exhaustive array of rain gages, piezometers, tensiometers, TDR wave guide pairs, lysimeters, meteorological sensors, and two weirs all monitored during three, week-long sprinkling experiments. Continuous measurements of rainfall, discharge, and total head (from selected piezometers) are available from 1990 through 1996. Extensive site characterization at CB1 provides high resolution topography and soil characteristics (e.g., geometry/thickness, saturated hydraulic conductivity, soil-water content, porosity, and hysteretic capillary pressure relationships). CB1 provides one of the most comprehensive hydrologic response data sets in existence for a steep catchment that has experienced slope failure. Results from 3D InHM simulations include comparisons of observed versus simulated runoff, pressure head, soil-water content, solute concentration, and pore pressure that facilitates a quantitative assessment of the importance of both convergent flow in the saturated subsurface and flow through the unsaturated zone in controlling the transient dynamics of near-surface hydrologic response, with implications for better understanding hydrologically-driven landslide initiation. Analyses of the simulation results indicate that (i) the unsaturated zone is important in controlling the timing of pore pressure generation at the soil-bedrock interface, (ii) temporal and spatial variability in input data (sprinkling/rainfall) cannot be neglected, and (iii) steady-state approximations to the subsurface flow equations do not adequately represent hydrologic response for steep, highly conductive hillslopes.