Loading...
Modeling post-wildfire mountain headwaters hydrology in an experimental watershed in Colorado
Wade, Randall
Wade, Randall
Citations
Altmetric:
Advisor
Editor
Date
Date Issued
2025
Date Submitted
Collections
Files
Research Projects
Organizational Units
Journal Issue
Embargo Expires
Abstract
The number of wildfires and the resulting total burned area has been increasing in Colorado. Besides disturbing vegetation, wildfires have both immediate and long-term effects on hydrologic and biogeochemical cycles of forested mountainous areas. Severe fires generally reduce water infiltration into the soil, increasing surface runoff and causing nutrients and sediments to wash away and subsequently compromise the quality of downstream water resources, in addition to creating landslide and flood hazards. After wildfires, agencies rapidly act to minimize nutrient loss and topsoil erosion using different slope-stabilization methods, such as wood-shred mulching. In these scenarios, hydrologic modeling is a useful tool to understand how perturbations to the water cycle may influence flows, but limited data is available in remote, rugged, burned landscapes for model inputs. This study applied the Agricultural Ecosystems Services (Ages) semi-distributed ecohydrology model using various gridded climatological datasets to simulate post-fire hydrological responses at sites affected by the Cameron Peak wildfire in Colorado.
The study provides insights into vegetation and hydrology interactions across various watershed locations, influenced by factors like crop type, elevation, and aspect. Notably, unburned HRUs displayed consistent interception of precipitation throughout the growing season, contrasting with burned HRUs, which showed increasing interception capability as they revegetated. This difference underpins the significant hydrologic alterations post-fire, with burned areas experiencing higher throughfall and less water retention, leading to increased surface runoff, especially following significant precipitation events. This study focuses on both hourly and sub-hourly resolutions for capturing detailed process dynamics, as well as finer spatial resolution than many hydrologic models, which are not common and represent important first steps toward simulating post-fire hydrologic responses at the appropriate spatiotemporal scale.
Calibration metrics, including the Kling-Gupta Efficiency (KGE), Nash-Sutcliffe Efficiency (NSE), and percent bias (PBIAS) highlighted the model's competent representation of general hydrologic conditions (KGE = 0.7-0.76; NSE =0.12-0.47). However, discrepancies in peak flow predictions emphasized the need for model adjustments, particularly in accurately capturing large runoff events. This aspect was critical during the calibration period, where the model struggled with prediction of peak flows in the calibration period (PBIAS = 36%), but improved significantly for the validation period (PBIAS = 1%).
The model's ability to operate at sub-hourly resolutions opens new avenues for detailing the timing of runoff events—a crucial feature for managing flood risks and monitoring watershed health in post-fire scenarios. This level of temporal resolution is particularly beneficial for capturing the rapid hydrological responses characteristic of smaller or steeply sloped subwatersheds. In this regard, the Bennett Creek Ages model has the advantage over most post-fire modeling studies as the temporal scale is more appropriate for simulating post-fire hydrologic responses. Modeled time of concentration is slightly elevated for mulched subwatershed than unmulched subwatersheds.
This research contributes to advancing knowledge in the areas of mountain and fire-related hydrologic modeling, high-resolution modeling using remote-sensing data, and informs agencies of the impacts of best management practices in critical zone rehabilitation treatments after wildfires.
Associated Publications
Rights
Copyright of the original work is retained by the author.