Title

Rapid response tools and datasets for post-fire hydrological modeling applied to the High Park Fire

Document Type

Conference Paper/Presentation

Publication Date

2016

Abstract

Post-fire flooding and erosion can pose a serious threat to life, property and natural resources. Time is critical in post-fire remediation as plans and treatments must be developed and deployed before the first major storms to be effective. We have developed an interactive database for the continental United States to facilitate rapid assessment of runoff and erosion risks from burned watersheds. This interactive database allows modelers to upload earth observations of soil burn severity and quickly download spatially-explicit model inputs at either 10- or 30-m resolution for process-based models, particularly WEPP based models (http://geodjango.mtri.org/geowepp/). Other modeling applications include agriculture, construction, or mining. The online database has allowed post-fire remediation teams in the western U.S. to rapidly predict post-fire erosion and effects of mulching treatments for over a dozen fires ranging in size from 4-540 km2. There is an urgent need for field data to validate these modeling efforts. One of the first fires to utilize this database was the High Park Fire, which burned approximately 330 km2 in northcentral Colorado in June 2012. Beginning in late summer of 2012 rainfall, site characteristics, and sediment production were measured for 21 unmulched and 8 mulched hillslopes. In 2013 sediment yields from unmulched and mulched hillslopes averaged 7.9 and 3.0 Mg ha-1, respectively, and these values dropped by more than an order of magnitude in 2014. Initial comparisons of measured and predicted hillslope sediment yields were poorly correlated, although they are of a similar order of magnitude. The causes of the poor correlation are being evaluated, and include a discrepancy between the small scale of the measured hillslopes and the much larger scale of the modeled hillslopes, problems with some sediment fences overtopping, an exceptionally large and long duration storm in September 2013, and the lack of suspended sediment data. Additional modeling is being conducted to obtain a better match between the predicted and measured hillslopes; and to model specific subsets of the data to minimize problems with the measured sediment yields and the effect of the September storm. These results will be used to both validate and recalibrate post-fire soil parameters for WEPP based models.

Publisher's Statement

Publiser's version of record: http://adsabs.harvard.edu/abs/2016AGUFM.H43G1546D

Publication Title

2016 AGU Annual Meeting

Link to Full Text

Share

COinS