A51Q-0340
Empirical analysis and hydrological modeling of wildfire impacts on flow regimes in forest watersheds: Eastern vs. western United States

Friday, 18 December 2015
Poster Hall (Moscone South)
Dennis W Hallema1, Ge Sun1, Peter V Caldwell2, Steven P Norman3, Erika Cohen4, Yongqiang Liu5 and Steven McNulty6, (1)USDA Forest Service, Southern Research Station, Raleigh, NC, United States, (2)Coweeta Hydrologic Laboratory, USDA Forest Service, Otto, NC, United States, (3)USDA Forest Service, Asheville, NC, United States, (4)USDA Forest Svc, Eastern Forest Environmental Threat Assessment Center, Raleigh, NC, United States, (5)USDA Forest Service, Athens, GA, United States, (6)USDA Forest Service, Vallejo, CA, United States
Abstract:
Wildfires are natural disturbances vital to many ecosystems, however with longer wildfire seasons, increasing annual area burned, and higher fire severity associated with forest densification, concern has risen about the potential effect of fire on water supplies. The magnitude and timespan of wildfire impacts on hydrology vary considerably with geographic region depending on pre-fire conditions, fire severity, and post-fire precipitation patterns. According to estimates, approximately 50% of freshwater resources in the contiguous United States (CONUS) originate in forest watersheds, and therefore it is essential to understand how wildfires affect the hydrologic cycle, in particular peak flows, base flows, annual water yields and timing of availability of water. We here present an approach in which we combined long stream flow records with a 30-year burn severity dataset and digital elevation models in order to evaluate for 30 watersheds across the CONUS changes in flow regime observed after a major fire. Flow change was positively correlated with the fraction of the watershed burned, which in turn was correlated with the mean slope of the watershed. Predictions based on a monthly empirical rainfall-runoff model and a water balance model (WaSSI) showed that depending on topography, species composition and fire history, post-fire stream flow was higher than predicted in 15 watersheds, lower than predicted in 9 watersheds, and unchanged in 12 watersheds (p<0.05). The outcome of this analysis allows us to identify watersheds that are most sensitive to wildfire impacts.