A new dispersion model solves the 3-D advection-diffusion equation using a Lagrangian stochastic, Monte Carlo method. This model can simulate the processes of mean wind advection, turbulent diffusion, radioactive decay, first-order chemical reactions, wet deposition, gravitational settling, dry deposition, and buoyant/momentum plume rise. A new atmospheric data assimilation model provides meteorological data fields (including non-divergent mean winds, turbulence and precipitation rates) to the dispersion model. It processes observations (e.g., from surface stations, rawinsondes, profilers) and/or weather forecast model data (global or mesoscale), as well as land-surface data, using a variety of data assimilation methods and atmospheric parameterizations. The primary source of mesoscale model data is the Naval Research Laboratorys COAMPS model. The system supports nested grids, grids with variable resolution in the horizontal and vertical directions, and uses a continuous terrain representation of the surface boundary.
Our new modeling system is intended for use in simple and complex terrain and on multiple space and time scales, from the microscale to mesoscale. Therefore, several levels of validation and evaluation are required. We will show the results of validation tests using analytic solutions to the advection-diffusion equation for inhomogeneous turbulence and mean wind. We will also show representative results of evaluations of the system using field experiments for different scales, such as Project Prairie Grass, the Diablo Canyon Tracer Study, the Savannah River Mesoscale Atmospheric Tracer Experiments, and the European Tracer Experiment.
This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract No. W-7405-ENG-48.