HR: 14:40h
AN: A33F-05    [Abstracts]
TI: Application of Recent Advances in Forward Modeling of Emissions from Boreal and Temperate Wildfires to Real-time Forecasting of Aerosol and Trace Gas Concentrations
AU: * Hyer, E J
EM: ehyer@umd.edu
AF: Naval Research Lab, Marine Meteorology Division, 7 Grace Hopper Avenue, Mail Stop 2, Monterey, CA 93943 United States
AU: Reid, J S
EM: reidj@nrlmry.navy.mil
AF: Naval Research Lab, Marine Meteorology Division, 7 Grace Hopper Avenue, Mail Stop 2, Monterey, CA 93943 United States
AU: Kasischke, E S
EM: ekasisch@umd.edu
AF: University of Maryland Geography, 2181 LeFrak Hall, College Park, MD 20742 United States
AU: Allen, D J
EM: allen@atmos.umd.edu
AF: University of Maryland, Department of Oceanic and Atmospheric Sciences, University of Maryland, College Park, MD 20742-2425 United States
AB: The magnitude of trace gas and aerosol emissions from wildfires is a scientific problem with important implications for atmospheric composition, and is also integral to understanding carbon cycling in terrestrial ecosystems. Recent ecological research on modeling wildfire emissions has integrated theoretical advances derived from ecological fieldwork with improved spatial and temporal databases to produce "post facto" estimates of emissions with high spatial and temporal resolution. These advances have been shown to improve agreement with atmospheric observations at coarse scales, but can in principle be applied to applications, such as forecasting, at finer scales. However, several of the approaches employed in these forward models are incompatible with the requirements of real-time forecasting, requiring modification of data inputs and calculation methods. Because of the differences in data inputs used for real-time and "post-facto" emissions modeling, the key uncertainties in the forward problem are not necessarily the same for these two applications. However, adaptation of these advances in forward modeling to forecasting applications has the potential to improve air quality forecasts, and also to provide a large body of experimental data which can be used to constrain crucial uncertainties in current conceptual models of wildfire emissions. This talk describes a forward modeling method developed at the University of Maryland and its application to the Fire Locating and Modeling of Burning Emissions (FLAMBE) system at the Naval Research Laboratory. Methods for applying the outputs of the NRL aerosol forecasting system to the inverse problem of constraining emissions will also be discussed. The system described can use the feedback supplied by atmospheric observations to improve the emissions source description in the forecasting model, and can also be used for hypothesis testing regarding fire behavior and data inputs.
DE: 0305 Aerosols and particles (0345, 4801, 4906)
DE: 0322 Constituent sources and sinks
DE: 0414 Biogeochemical cycles, processes, and modeling (0412, 0793, 1615, 4805, 4912)
DE: 0428 Carbon cycling (4806)
DE: 0434 Data sets
SC: Atmospheric Sciences [A]
MN: Fall Meeting 2005