An Overview of the NOAA National Centers for Environmnental Prediction (NCEP) meteorological model products to support atmospheric transport and dispersion studies

This presentation overviews recent numerical weather prediction applications for atmospheric transport and dispersion that are underway at the NOAA National Centers for Environmental Prediction (NCEP). A meteorology team was established at the U.S. Defense Threat Reduction Agency (DTRA) to provide weather support to personnel operating DTRA's Consequence Assessment tools, such as the Hazard Prediction and Assessment Capability (HPAC) tool. NCEP supports these efforts through the provision of high resolution deterministic modeling and probabilistic ensemble model system capabilities. The NCEP WRF Non-hydrostatic Mesoscale Modeling (NMM) system is used along with deterministic modeling predictions provided by Penn State University and the Air Force to predict local and regional flows for any global location when required. Higher resolution nests within the 12 km coarse domain have been relocated for any location where dispersion support is required. The nesting capabilities for NMM were developed to maintain mass across grids, an important feature in complex terrain. The efficient NMM numerics and dynamics allow predictions on very large nested grids in real-time.

The WRF-NMM nested grid system was applied in real-time for the 2006 Torino Italy Olympics to provide high resolution forecasts to drive dispersion models. Examples and evaluation of NMM performance for this application will be described. The NCEP Forecast Verification System (FVS) is regularly used to evaluate most NCEP model performance. New results using FVS to evaluate predictions that are critical to dispersion model applications (eg: planetary boundary layer height, cloud cover, low level winds) will also be summarized.

More detailed evaluation of the WRF-NMM turbulence predictions are also being evaluated with the DTRA team. Special surface flux and atmospheric boundary layer turbulence measurements were deployed at several locations during the 2002 International H20 Project (IHOP). Retrospective simulations using the WRF-NMM for several cases during IHOP were evaluated using these special turbulence measurements. The flux and turbulence evaluations are important to help assess whether meteorological and dispersion models can be coupled more tightly by driving the dispersion predictions with these more basic turbulence fields.

In addition, special ensemble based prediction systems on global (The Medium Range Ensemble Forecast System with horizontal resolutions of around 100 km) to regional scales (The Short Range Ensemble Forecast System with horizontal resolutions of around 32 km) can also be utilized for dispersion applications. The DTRA team supported by NCEP is testing these ensemble predictions to derive velocity variance and cross correlation predictions which in turn are used to estimate large scale uncertainty information in HPAC. Probabilistic products available from these systems and their application to dispersion will be presented. NCEP plans to expand upon these results and experience during upcoming field tests, and to further improve and expand the capability to provide accurate high-resolution and probabilistic weather forecast information to hazard and consequence assessment operations.