Assimilation of high-frequency data is best accomplished with 4-dimensional variational systems that constrain the analysis through the use of a modeling system. However, such an assimilation system that is based on a full-physics model is so computationally intensive that it cannot be currently run operationally. To address this need, NCAR has developed a variational assimilation system that is based on a model with simple physics. The Variational Doppler Radar Assimilation System (VDRAS) model assimilates a time history of radial winds from NEXRAD, or any other Doppler-radar, as well as all other available meteorological data. The product is a three-dimensional analysis of the complete wind field and the other standard meteorological variables. A short nowcast of 30 minutes can also be produced by the model. On a dual-processor PC, the system requires only a few minutes to produce an analysis and forecast. With radar-data as input, the horizontal grid increment of the VDRAS analyses and forecasts is about 1 km, the dimension of the analysis areas is typically 100 km, and the update-cycle frequency is approximately 10 minutes. If lidar data are assimilated by the same model, the horizontal grid increment is approximately 75-100 m, and the analysis area has dimensions of 5-10 km.

The VDRAS system with radar input may be run economically to define above-rooftop winds for most metropolitan areas because the National Weather Service's Level-II NEXRAD data are available at no cost on the high-speed CRAFT network. Such a system was run operationally for Oklahoma City in support of the DoD Joint Urban 2003 field experiment. For this purpose, the VDRAS output was coupled to the DoD SCIPUFF transport and dispersion model for prediction of the general rooftop-level movement of tracer material released in the study.

The VDRAS system can be used for homeland security applications, it can be employed in meteorological studies of boundary-layer processes in urban areas, or it can be used for general operational nowcasting of convective or other weather in the urban area.