BFRL: 2008 Program Information (V 1.1)

Program Manager:
William Mell
Revised: 10/29/2007

BFRL Goal:
Homeland Security and Disaster Resilience

Relevant Links

Fires at the Wildland-Urban Interface

To develop tools for reducing community losses in wildland-urban interface fires. These tools will assist communities in their fire risk assessment and choice of economically balanced mitigation strategies that limit the ignition of residences and improve firefighter and community safety. A long term close collaboration with NOAA will result in improvements to their national Fire Weather Prediction Forecast System.

What is the problem and why is it hard? Even though isolated fires in residences have dominated fire losses in the US in recent decades, a disturbing trend has emerged in which fires initiating in wildland fuels spread into communities that are adjacent or intermixed with the wildland fuels. These fires are called wildland-urban interface (WUI) fires and can result in large residential losses over community scales. The 2003 Southern California WUI fires resulted in the destruction of thousands of homes with more than $2 B in insured losses. The most destructive of these fires, the Cedar Fire, caused 15 deaths and the loss of 3600 homes. This fire was started by a signal fire from a lost hunter. The Old Fire which destroyed 993 homes was caused by arson. Clearly these fires are a potential threat to WUI communities and homeland security. Several factors, acting in concert, are responsible for the increasing numbers and destructiveness of these events. Among these are a general trend for increased home construction in the WUI (60% of new home construction in the 1990’s), a widespread build up of wildland fuel due to the effective suppression of small wildland fires that would otherwise limit the accumulation of such fuels, and a persistent drought in parts of the US that increases the likelihood and severity of wildland fires. When these conditions are combined with high winds, difficult-to-contain wildfires can result, capable of igniting large numbers of structures in WUI communities (as in the 2003 Southern California fires).

Despite the increasing frequency and losses from WUI fires, there has been very little research on WUI fire behavior and the conditions that can result in community-scale fire spread and destruction. This is due, in part, to the fact that the subject of WUI fire research falls between traditional studies of building fires and forest fires, which are usually the responsibility of different branches of the government. WUI fire behavior is also a difficult problem to model and study experimentally due to the range of physical scales over which the relevant physical processes operate. Destructive WUI fires often occur in extreme environmental conditions, such as high winds, that are difficult to reproduce experimentally. Wildland fires can be large uncontrollable crown fires or relatively low intensity surface fires – both are capable of igniting a vulnerable structure. Firebrands, from both burning vegetation and structures, can be dispersed hundreds of meters downwind and potentially create numerous vegetative and structural fires. To date, only a limited number studies exist on the characterization of firebrands generated from vegetation and structures and their potential to ignite structures. Ambient weather conditions and terrain also can play a major role in wildland and WUI fire behavior.

How is it solved today and by whom? There are two major approaches to mitigate the occurrence and destructiveness of WUI fires. The first approach is to reduce wildland fuel loads in WUI areas. At present funding levels, wildland fuel treatments in less than 5% of the high risk communities are completed yearly. Also, the various fuel treatment methods used are based on very limited scientific studies, leaving their effectiveness largely unproven (from both a risk reduction and an economic point of view). The second approach is to have communities follow a number of risk reduction practices and building codes. Most of these practices follow rule based and empirically determined checklists that lack testing over a range of characteristic conditions and are not the result of a coordinated scientific based effort. Of the 26,000 at risk WUI communities in the US only 12% currently follow and enforce these risk reduction practices.

Research in the US Forest Service has traditionally focused on wildland fire behavior. Currently there are no models for WUI fire behavior that include the role of burning structures across community scales. The Fire Weather Prediction System developed by the National Oceanic and Atmospheric Administration (NOAA) provides wildland and WUI fire incident managers with fire weather forecasts. These forecasts do not provide fire predictions, only atmospheric conditions (e.g., wind speeds and relative humidity) relevant to fire behavior. In addition, the computational grids used in the fire weather predictions have approximately a 8 km spacing which will not resolve many WUI communities. Incident managers can supplement fire weather predictions with local on-site atmospheric measurements from a NOAA incident meteorologist (IMET). There is, however, a clear need for a community scale WUI fire behavior models. For example, NOAA would benefit from a community to sub-regional scale WUI fire model that can be suitably coupled to NOAA’s fire weather prediction system. The forest service would benefit from a WUI fire model capable of aiding their research on effective wildland fuel treatments.

What is the new technical idea and why can we succeed now? The understanding of how fires propagating through the WUI under a variety of weather, fuel, and terrain conditions ignite structures is poor. NIST is in a unique position to develop useful tools for the WUI problem by adapting its established and successful approaches for modeling structural fires and outdoor smoke transport to WUI fires. The Large Fire Laboratory will provide valuable data sets for model validation. BFRL’s knowledge of fire measurements will also be taken to the field, in collaboration with the US Forest Service, to obtain data on real scale fires for full scale model validation. The information gathered from the modeling, experimental, and measurement effort will provide a strong base for the development of applied economics tools to be used by communities when assessing the economic effectiveness of various fire risk mitigation approaches.

Measuring the Demand for Fire Mitigation Services and Technologies in the Wildland-Urban Interface
• D.T. Butry and G.H. Donovan, “Protect Thy Neighbor: Investigating the Spatial Externalities of Community Wildfire Hazard Mitigation,” under peer review, 2007.

Experimental Investigation of Structural Ignition
• S.L. Manzello, T.G. Cleary, J.R. Shields, A. Maranghides, W.E. Mell, J.C. Yang, “An Experimental Investigation of Firebrands: Generation and Ignition of Fuel Beds”, Fire Safety Journal, to appear, 2007.
• S.L. Manzello, A. Maranghides, W.E. Mell, Firebrand Generation from Burning Vegetation, Int’l J. Wildland Fire, to appear 2007.
• A number of conferences presentations (some invited) were made; please see project description for a full listing.

Experimental Investigation of Vegetation Burning
This is a new project in FY08. In FY07 PI was a co-PI of the project listed above.

Development of an Outdoor-Scale Fire Model
• W. Mell, M. A. Jenkins, J. Gould, and P. Cheney, A Physics-Based Approach to Modeling Grassland Fires, International Journal of Wildland Fire, 16:1-22, 2007.
• R. Sun, S. K. Krueger, M. A. Jenkins, W. Mell, J. J. Charney, An Evaluation of the FDS and Clark Coupled Wildfire Model, The Canadian Journal of Forestry Research, 36(11):2894-2908, 2006.
• A number of conference presentations (some invited) were made; please see project description for a full listing.