The High Park fire: Coupled weather‐wildland fire model simulation of a windstorm‐driven wildfire in Colorado's Front Range. Issue 1 (12th January 2015)

• Record Type: Journal Article
• Authors: Coen, Janice L.; Schroeder, Wilfrid
• Is Part Of: Journal of geophysical research. Volume 120:Issue 1(2015:Jan.)
• Page Start: 131
• Page End: 146
• Publication Date: 2015-01-12
• Subjects: Atmospheric physics -- Periodicals; Geophysics -- Periodicals; 551.5
• DOI: 10.1002/2014JD021993
• Languages: English
• ISSNs: 2169-897X

Abstract

<abstract abstract-type="main"> <title>Abstract</title> <p>Weather affects wildland fires at scales from multiseasonal precipitation patterns and anomalies, through synoptic and mesoscale weather patterns, to convective scale motions including fire‐induced winds. This work analyzed the first day's growth of the 2012 High Park fire, which occurred in Colorado's Front Range during widespread drought and an unseasonal June windstorm, assessing to what extent the Coupled Atmosphere‐Wildland Fire Environment coupled numerical weather prediction—wildland fire behavior model could reproduce the event, burn severity patterns, and how the drought impact on fuel moisture impacted the event. Simulated mountaintop wind speeds reaching 47 m s<sup>−1</sup> and gravity wave overturning created strong, gusty surface winds. During the first 9 h, the simulated fire grew underneath the gravity wave's crest and downdraft, sheltered from the windstorm. The simulated fire then climbed a ridge, was exposed to the windstorm, and rapidly traveled east, covering 15 km in 12.3 h. Burning routed up or down drainages caused finger‐like streaks in maximum fire intensity. Reference fire mapping information supported the simulated early growth toward the north, splitting around topographic features, while the simulation's underestimate of extent accrued to 2 km over 21.3 h. While the control simulation employed horizontal grid spacing of 123 m, a simulation refined to 370 m captured some wave motions and overall direction but further underestimated extent and lacked details such as turns in direction, splitting, or fingering at the leading edge. Compared to a simulation with moderately dry fuel conditions, a range of drought‐like fuel moisture conditions produced fires that extended 0–39% farther.</p> </abstract>