Realtime High-Resolution Mesoscale Modeling for the Defense Threat Reduction Agency

Penn State provides operational reachback support to the Defense Threat Reduction Agency (DTRA) for the Penn State - DTRA in-house mesoscale modeling system while also providing operational redundancy by running parallel mesoscale model forecasts on a mirror computer cluster at Penn State for important worldwide events (Beijing Winter Olympics, Democratic National Convention, Republican National Convention, Presidential Inauguration). Mesoscale model forecasts are used to drive the HPAC-SCIPUFF atmospheric transport and dispersion (AT&D) model for hazard prediction and consequence assessment.

DTRA is currently running a high-resolution (to ~1-km horizontal resolution) MM5 modeling system in-house while Penn State runs that system locally in realtime for operational redundancy. Penn State also runs a high-resolution version of the WRF-ARW locally in realtime. This paper presents some examples of realtime high-resolution mesoscale-model and AT&D forecasts, and also compares results from the realtime MM5 and WRF-ARW forecasts for select cases during the 2008 Beijing Winter Olympics.

Results indicate that the MM5 high-resolution realtime forecasts show very good qualitative and statistical agreement with observations, and that WRF has improved since our MM5-WRF comparisons for the 2006 Torino Winter Olympics, as both models produced very similar results for the 2008 Beijing Winter Olympics.