HR: 17:00h
AN: H34C-05    [Abstracts]
TI: Large Scale Predictions of Potential Post-fire Erosion
AU: * Miller, M E
EM: memiller@buffalo.edu
AF: Colorado State University, Department of Forest, Rangeland, and Watershed Stewardship, Fort Collins, CO 80523 United States
AU: MacDonald, L H
EM: leemac@cnr.colostate.edu
AF: Colorado State University, Department of Forest, Rangeland, and Watershed Stewardship, Fort Collins, CO 80523 United States
AB: High-severity wildfires are of increasing concern because of their potential for initiating flash floods and surface erosion, degrading water quality, and reducing reservoir capacity. In many areas fire suppression has increased fuel accumulations and hence the potential for high-severity wildfires. Land management agencies are undertaking programs to reduce fuel loadings and the associated risk of high-severity wildfires, but the areas needing treatment greatly exceed the available funding. It is therefore necessary to determine which areas should have a higher priority for such treatments. Similarly, when wildfires do occur there is an immediate need to determine which areas should have the highest priority for post-fire rehabilitation treatments. One criterion for allocating treatments is the potential risk of post-fire erosion, but to be effective this assessment needs to be carried out at a broad scale. This paper presents a procedure and initial results for predicting spatially-explicit, post-fire erosion risks at the hillslope scale for forest and shrub lands across the western U.S. Our approach utilizes existing physical models and datasets in a GIS framework. The model for predicting erosion is GeoWEPP, the Geographical interface for the Water Erosion Prediction Project (WEPP). The primary inputs for GeoWEPP include climate, topography, soils, and land cover/land use. Daily climate inputs were generated with Cligen, which is a stochastic weather generator distributed with WEPP. A 30-m digital elevation model, STATSGO-derived soils data, and vegetation cover were obtained from the U.S. Forest Service's LANDFIRE project. Since recent research has shown that percent ground cover is a dominant control on post-fire erosion rates, we generated a spatially-explicit map of post-fire ground cover by first using historic weather data to determine the 1000-hr fuel moisture values when fuel conditions were at 98-100% ERC (Energy Released Component). These fuel moisture values were fed into FOFEM (First Order Fire Effects Model) to obtain spatially-explicit predictions of percent ground cover, and this provided the additional land cover/land use information needed by GeoWEPP. The predicted erosion rates are comparable to measured values in the Colorado Front Range, but are much too high for the higher rainfall areas along the Pacific Coast. This pattern indicates that precipitation is having a pre-dominant effect on predicted post-fire erosion rates, especially in areas that are projected to burn at low severity. Hence the predicted erosion rates will be most useful in relative terms at the local and possibly regional scale, while comparisons between regions may be of more limited validity.
DE: 1815 Erosion
DE: 1819 Geographic Information Systems (GIS)
DE: 1826 Geomorphology: hillslope (1625)
DE: 1847 Modeling
DE: 1879 Watershed
SC: Hydrology [H]
MN: Fall Meeting 2005