GSA Annual Meeting in Denver, Colorado, USA - 2016

Paper No. 302-15
Presentation Time: 5:00 PM

USING PAST EVENTS TO VALIDATE AND PREDICT POST-FIRE EROSION IN RANGELAND ECOSYSTEMS


GIBBLE, Katie, Department of Geosciences, Boise State University, 1910 University Dr, Boise, ID 83725 and PIERCE, Jennifer L., Department of Geosciences, Boise State University, Boise, ID 83725, katiegibble@u.boisestate.edu

Planning for larger, more severe fires and prolonged fire seasons requires consideration of the increased risk of fire-induced debris flows. Like many cities in the western US (e.g. Denver, Reno, Salt Lake City), the Boise Wildland Urban Interface (WUI) is located at the base of a mountain front. Range front vegetative communities in these western US WUI’s are typified by shrubs (sagebrush steppe) and grasslands at lower elevations, with open forests at higher elevations and on north-facing slopes. Despite WUI development increasing concurrently with fire and erosion hazards in these ecosystems, most erosion and fuel models tend to be developed for forested ecosystems. Here we present post-fire debris flow hazard scenarios for the Boise, Idaho foothills. We vary fire severity and precipitation intensity to create modeled scenarios from which land managers at local and state levels can plan for a variety of potential post-fire erosion scenarios. We validate the model’s estimate of sediment volume by comparing model predictions to a historic mudflow event that occurred shortly after a 7000 acre fire in the Boise foothills in 1959. We found that volume estimates fell well within an order of magnitude of the recorded volumes eroded from the foothills during the 1959 event. These results validate the use of post-fire debris flow models in rangeland (vs.forest) settings, and are encouraging for future decision-making use within a rapidly growing city, where development often occurs within steep erosive foothills known to have produced debris flows in the past. This work provides an example of how post-fire erosion models can be applied in pre-fire settings for planning purposes, allowing time to validate model results with locally relevant historic records.