Over the past ten years, ca. US$ 5.6 billion has been spent on hazardous fuel reduction to treat an average of ca. 2.5 million acres per year across the United States. These expenditures represent one of the primary strategies for the mitigation of catastrophic wildland fire events. At the local scale, the placement and implementation of fuel reduction treatments is complex, involving trade-offs between environmental impacts, threatened and endangered species mitigation, funding, smoke management, parcel ownership, litigation, and weather conditions. Because of the cost and complexity involved, there is a need for implementing treatments in such a way that hazard mitigation, or other management objectives, are optimized.
This research integrates proven, state-of-the-art, remote sensing methodologies with cutting edge numeric modeling of fire spread to test the principals and physics behind fuel reduction treatments. Airborne LiDAR data is used for the spatial quantification and display of three-dimensional fuel loading, expressed as Canopy Bulk Density. The three dimensional fuels data will be coupled with WFDS (the Wildland-urban interface Fire Dynamics Simulator an open-source tool co-developed by NIST and the US Forest Service) that can simulate fire spread and intensity using this data. This synthetic approach to evaluating fuel reduction treatments would allow fire management agencies in optimizing the placement and timing of hazardous fuel reduction treatments. The geographic area within which we are performing this research, The New Jersey Pinelands, is rich with existing data, partnerships, and expertise that will allow us to maximize this investment.
The project covers an integrated approach to the development of fuel treatment tools to support decision-making. Experiments will be conducted by burning large parcels (circa 10 hectares) of pitch pine forest. The two experiments will characterise the different vegetation layers: tree layer by LiDAR and shrub and ground layers by on-site sampling. The fuel characterisation will include the fuel bulk density, the height of the vegetation layer, the fuel moisture content, as well as the geometrical, physical and chemical properties of the vegetation species. The fire behavior will also be studied in details through aerial infrared images of the fire front, measurements of wind velocity at different heights, temperature distribution in the fire plume, as well as heat fluxes ahead of the fire front). The fuel characterisation will be used to implement the model and test its ability to reproduce two different levels of treatment (no treatment and treatment two years before the experiment) and the fire behavior will be used to test the predictive capacities of the model.
The outcome of the project is an evaluation of the ability of the model to define accurately the optimal level of treatment that is necessary to decrease the risk of large scale fires at a level appropriate for management purposes.
