The Marshall Fire of 30 December 2021: The Operational Utility of HRRR Forecasts for a Wildfire Resulting from a Perfect Storm of Climate, Mesoscale and Urban/Open Space Factor

The Marshall Fire, which began just before noon local time on 30 Dec 2021, became the most destructive (in cost) wildfire in Colorado history as it evolved into a suburban firestorm in southeastern Boulder County, driven by strong winds and a drought-influenced fuel state. The downslope windstorm, while strong, was not an unusual occurrence in winter along the Front Range of Colorado. However, two important factors increased the potential for the tragic event that occurred. A 6-month unusually dry spell that followed a wet spring, including an extremely late start to the snow season, left the many swaths of open space grassland type vegetation in a very dry state. The second factor were densely populated communities that have been built and are surrounded by large areas of open space grassland. The fire that then began and initially grew in these open areas quickly spread to the communities, driven by hours of winds gusting well over hurricane strength, creating the out-of-control tragic fire.

Our main goal of this presentation is to examine how well the operational HRRR model was able to predict this windstorm and how those predictions were used by the local National Weather Service (NWS) Weather Forecast Office (WFO) in Boulder. One aspect of the HRRR that will be emphasized is the use of the hourly-updating forecasts to indicate consistency in the forecast message. We will also discuss the conditions outlined above that left the Front Range in a particularly vulnerable situation as 2021 came to a close. Recent fires in California and elsewhere have demonstrated that this type of urban/grassland-forest interface is not unique to the area where the Marshall Fire occurred, creating particularly dangerous potential outcomes when extreme conditions come together.