The accurate and timely prediction of the atmospheric dispersion of hazardous materials in densely populated urban areas is a critical homeland security need for emergency preparedness, risk assessment, and vulnerability studies. Numerical simulations represent a unique predictive tool for developing a detailed understanding of three-dimensional flow fields and atmospheric dispersion of hazardous materials released in complex urban settings.

 

Current computational tools for urban dispersion do not include all the necessary elements needed to accurately predict peak concentrations, spatial extent and temporal evolution of hazardous materials in urban environments. To address these limitations we developed an integrated adaptive urban dispersion model using Cartesian-based adaptive mesh refinement (AMR) with a legacy finite element flow solver.  To handle the complex urban topography we use immersed boundary/cut-cell approach for efficient treatment of complex urban geometries. 

 

Our new integrated tool is built by integrating the legacy FEM3MP urban dispersion CFD code with the SAMRAI library, which provides support for parallel structured AMR.  We have developed capabilities to automatically import geometry information from urban shapefiles and construct the immersed-boundary representation in the AMR context.  We will demonstrate our tools for efficient, automated adaptive mesh generation of complex urban environments and show their use for predictive calculations around Manhattan.  We will also demonstrate a fully adaptive simulation whereby the Cartesian grid is automatically refined in parallel during the simulation to resolve important features.  Finally, we will demonstrate how AMR can be effectively used to achieve realistic forcing of urban scale flows through coupling of the farfield coarse-grid urban-scale model with a mesoscale model WRF.

 

This work was performed under the auspices of the U.S. Department of Energy by University of California, Lawrence Livermore National Laboratory under Contract W-7405-Eng-48. UCRL-ABS-213998.