NCAR Annual Report > RAL Annual Report Contents > Strategic Priority > 2. Improving Prediction

National VDRAS Analysis

In 2003, a team from RAL visited the Department of Homeland Security (DHS) Chem/Bio Science and Technology Directorate to discuss the possibility of applying the (NCAR) Crook/Sun Variational Doppler Radar Assimilation System (VDRAS) to homeland security. The message delivered to DHS was that a composite, national-scale, 3-D boundary layer windfield could be retrieved from the nearly 160 Doppler radars situated across the U.S., and moreover, this could be done at a horizontal grid spacing increment on the order of 1 km, and the product could be updated every six minutes, matching the update rate of the volume-scanning Doppler radars. This dense windfield could be used to drive transport and dispersion models at a nearly real-time refresh rate, providing timely and accurate plume information to emergency responders and incident managers. This visit eventually led to a small project awarded by DHS to NCAR/RAL and NOAA/NSSL.  Work in this initial phase has focused on comparing the performance of the NCAR VDRAS and NOAA/NSSL’s competing system.

FY06 Accomplishments


Figure 1: VDRAS wind retrieval (vectors) overlaid on radial wind velocities

A decision was made to use the Joint Urban 2003 dataset from Oklahoma City for the evaluation of the two methods. The test dataset contains 15 days of radar data at a frequency of one volume every 5 to 10 minutes. VDRAS boundary layer wind analyses were produced using the KTLX NEXRAD radar, both with and without the use of additional in-situ surface observations.  An example of a VDRAS retrieval of lowest-level wind speed and direction is shown in Fig. 1. Initial validation results from Battelle-PNL indicate that NCAR's VDRAS winds correlate better with observations from four profiles than those from the NOAA/NSSL wind retrieval method.

FY07 Plans:

Plans for FY2007 are still being determined. Options currently being investigated include performing another set of wind retrievals at another locale and/or a statistical determination of the annual boundary layer wind coverage based upon the availability of NEXRAD radar data. If a full, National-scale deployment is funded, the composite VDRAS windfield product could become one of the most significant and coveted operational, real-time datasets in the U.S.