Objective
The main objective of this technology demonstration is to develop a necessary and timely advancement to survey and map understory and canopy fuels that support physics-based models of fire behavior and smoke production. FuelsCraft will provide an approach that avoids costly field surveys and integrates rapid advances in remote sensing technologies. The demonstration will be conducted on two Department of Defense (DoD) installations that have active prescribed burning programs with fuels that are common throughout the southeast U.S. The project will collect additional datasets at these two installations to evaluate the accuracy and application of synthetic fuelbed mapping.
Technology Description
Rapid advancements are being made to inform prescribed fire decision support with next-generation wildland fire behavior and smoke modeling. To date, much of the emphasis on three-dimensional (3D) vegetation characterization has been on improved understanding about how tree boles and crowns influence subcanopy wind fields and fire flow, but additional improvements in understory and surface fuels characterization are still needed to effectively model fire behavior and smoke production. For this demonstration and validation project, the project team will leverage existing wildland fuels datasets to develop synthetic 3D fuel maps of canopy and surface fuels as an interface to the FastFuels framework. The fuel mapping tool created by this project, FuelsCraft, will provide an approach that avoids costly field surveys and integrates rapid advances in remote sensing technologies.
The project will use artificial intelligence, 3D visualization software, and existing field and remotely sensed 3D point cloud data to generate synthetic fuelbed inputs. These inputs will enable a coupled-fire-atmospheric model of fire behavior and smoke to accurately represent wildland fuels common to prescribed burning programs within DoD installations of the southeast U.S. Synthetic fuelbeds are computer-generated 3D gridded mesh objects that realistically represent the physical size, shape, and spatial distribution of actual wildland fuel complexes. By assigning gridded physical characteristics of the synthetic fuel objects, informed by empirical distributions of wildland fuels, fuel properties such as bulk density and surface area-to-volume ratio can be estimated at a particle or object scale, with improved accuracy over remote sensing alone.
Benefits
Current maps of wildland fuels offer coarse approximations of the fuels that are available to burn, and often do not accurately represent the risks that fuels pose risk to installations and operations or the effectiveness that fuel management actions have on reducing risks. The quantitative approach to wildland fuels mapping will greatly improve the accuracy of mapping wildland fuels and modeling risk for decision support. FuelsCraft will enable DoD managers to develop and map site-specific, realistic 3D animations of forests and fuels based on past prescribed fire, harvests, and fuel dynamics over time. FuelsCraft will be codeveloped with the SERDP-supported FastFuels framework to provide realistic representations of understory fuels. This advancement in translating field and remotely sensed datasets to 3D realizations of forest structure has applications outside of prescribed fire decision support (i.e., military training exercises, forest operations). Assessment of the application will include sensitivity analysis of relevant scales and inputs of synthetic fuelbeds using computational fluid dynamics modeling with the Fire Dynamics Simulator to evaluate how the scale and resolution of gridded fuel inputs influences predicted surface and crown fire behavior. The project team plans for an integrated technology transfer with DoD managers through the ESTCP Integrated Research Management Team to ensure that the prototype tool is both usable and provides realistic datasets for operations.