P2M.6 LAPS assimilation of Radar and Satellite data for Tropical Cyclone NWP

Tuesday, 25 October 2005
Alvarado F and Atria (Hotel Albuquerque at Old Town)
Patrick T. Welsh, Univ. of North Florida, Jacksonville, FL; and P. Bogenschutz

The NOAA Coastal Storms Initiative funded a 5 km resolution explicit cloud version of the WRF (ARW) model in a Linux cluster at WFO Jacksonville, FL. This model uses NOAA FSL LAPS as the “hot start” to do data assimilation and initialize the WRF model. A small team of FSL, WFO JAX and other NOAA participants brought the system online early and under budget, with daily operation in real-time by March 2003.

That WRF (then EMM) version 1.3 system was operated 4 times a day for over a year with verification by FSL's Real Time Verification System (RTVS) in a “frozen” state to allow for good verification statistics. While that version had identified shortcomings, it proved to be competitive with the NCEP Eta model in verification for state variables and rainfall, while clearly superior in kinematics and sea breeze detection.

In July 2004, the model was upgraded to WRF ARW version 2.02 with the inclusion of the NOAH Land-Surface Model and the Yonsei University PBL. This version was far superior to the earlier model and serendipitously timed for the August-September Florida Hurricanes of 2004. While some of the results were lost, the cases to be presented show that the model managed to keep the core hurricane rainband structure through the 24-hour forecast cycle as shown in both hourly and accumulated rainfall when compared to the NCEP Stage IV rainfall results.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner