 |
| Titel |
Predicting Post-fire Flooding and Sediment Delivery at the Watershed Scale: An Urgent Need for Upscaling |
| VerfasserIn |
Lee MacDonald, Joe Wagenbrenner, Peter Nelson, Dan Brogan |
| Konferenz |
EGU General Assembly 2013
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 15 (2013) |
| Datensatznummer |
250077847
|
|
|
|
|
|
| Zusammenfassung |
| Over the past 10-15 years tremendous advances have been made in understanding the effects
of fires on runoff and erosion processes, the effectiveness of various post-fire rehabilitation
treatments, and the prediction of these changes. The problem is that nearly all of this work
has been done at the plot and hillslope scale, while it is the larger-scale issues of
flooding, water quality, and sedimentation that are of primary concern to resource
managers and the public. In most cases these larger-scale changes are predicted either
by simple lumped models, such as the curve number technique, or by summing
up the hillslope-scale responses. These approaches can greatly overestimate the
downstream effects because they do not account for the spatial variability of rainfall or the
complexities of routing and storage. While post-fire stormflows may be efficiently
routed downstream, our observations from fires across the western US indicate very
large differences in the amount of sediment delivered to downstream areas. As one
example, a large storm after the 2012 High Park fire near Fort Collins generated huge
amounts of sediment, but relatively little of this was delivered to the Cache la Poudre
River; in the case of the 2002 Hayman fire near Denver even moderate-sized storms
delivered enough sand and fine gravel to temporarily dam up the South Platte River
and over time around 750,000 m3 of sediment was deposited into Strontia Springs
Reservoir.
In this paper we hypothesize that relatively simple set of tools can greatly improve our ability
to predict post-fire runoff and sediment delivery at the watershed scale (5-100 km2). In areas
dominated by convective storms post-fire flood risks should be modified according
to the size of those storms relative to the upslope contributing area and extent of
high or moderate burn severity. The potential delivery of post-fire sediment can
be improved by combining the predicted flood risk with empirical adjustments
based on valley confinement, valley slope, and the density of the residual riparian
vegetation. Particle size is hypothesized to be a lesser control because most of the
deposited post-fire sediment is easily transported. Valley slope can be estimated
from existing DEMs, while the accurate characterization of valley confinement and
riparian vegetation requires higher-resolution imagery such as lidar and Quickbird or
SPOT. The development of such tools should greatly improve our ability to predict
the larger-scale risks of flooding and sedimentation, and more efficiently allocate
post-fire treatments to watersheds that have the highest sediment delivery potential. |
|
|
|
|
|
|