Over the past several decades the greatest loss of life associated with land-falling tropical cyclones in the United States has been from inland flooding rather than from storm surge or damaging winds. It is well known that current numerical forecasts of precipitation are too uncertain to provide meaningful guidance to forecasters. Improvement of numerical guidance requires refinement of the physical processes in the model and an observational data set to test against. Quantitative validation data for hurricanes at landfall are scarce due to the low number of events and a lack of detailed observations over large domains. This is particularly true for air circulations that help initiate and sustain mesoscale rainbands that produce most of the precipitation that leads to inland flooding.

Recent advent of C-band mobile Doppler radars will help provide the high spatial and temporal resolution observations over large domains that are required for evaluating and improving numerical model performance. In this presentation, we document the first deployment of a mobile C-band dual-Doppler network for hurricane studies. The Shared Mobile Atmospheric Research and Teaching (SMART) radars collected dual-Doppler data for more than 13 hours over a ten-thousand km-squared area of Hurricane Isabel during its landfall on 18 September 2003 in North Carolina. Numerous rainbands, as well as the center of circulation, passed within the prime dual-Doppler zone. Here, the data are used to examine the kinematic and precipitation structure of the rainbands that contributed heavily to the inland flooding.