APPLICATION OF A GRID-BASED TRANSIENT RAINFALL INFILTRATION-SLOPE STABILITY MODEL IN SOUTHWESTERN SEATTLE, WASHINGTON

We have developed a model to simulate the effect of transient pore pressure response due to vertical infiltration of rainfall on the initiation of shallow landslides that can be applied over a broad region. TRIGRS (Transient Rainfall Infiltration and Grid-based Regional Slope-Stability analysis) implements a modified version of Iverson’s (2000) approximate analytic solution to Richards equation in one dimension coupled with a simple infinite-slope model. This solution for pore-pressure response may be superposed on any steady-state groundwater flow field that is consistent with the model assumptions. Excess rainfall is routed over the surface to adjacent downslope cells where it can infiltrate with a simple flow-routing scheme that maintains mass balance. Applied over digital topography, the model computes a factor of safety for each grid cell at any time during a rainfall event. The problem domain is a layer of finite thickness. The upper boundary condition is a specified flux (rainfall infiltration) at the ground surface. The lower boundary is impermeable. By assuming the hillslope is initially tension saturated the model represents infiltration as a linear process. Input parameters may vary from cell to cell and the precipitation rate can vary in both space and time. The results are a temporally and spatially varying factor of safety that can be easily viewed in a grid-based GIS.We applied the model on a cell-by-cell basis over a 10 m DEM in a 25 km² area of southwestern Seattle, Washington. Material parameters were determined from field measurements of saturated hydraulic conductivity and laboratory tests of material strength properties and soil-moisture characteristic curves. The boundary flux was described by measured rainfall histories that triggered shallow landslides in the study area. The model simulates the progressive development of instability on steep slopes during rainfall events in Seattle. A comparison of simulated instability with the timing and location of documented landslides was used to assess the model results. This approach to landslide susceptibility mapping has advantages over steady-state approaches in that it does not restrict failure locations to areas with specific topographic attributes and captures the transient nature of the shallow landslide process with modest data requirements.