To generate an experimental rainfall potential, either the operational DMSP SSM/I 16-km (Ferraro et al. 1998) or the NOAA AMSU objective rain rates are used to produce an areal extent of rain and average rain rate through the storm. Short-term projections based on GOES imagery-based storm motion are then combined with the average rain rates to derive a storm's rainfall potential. The Tropical Rainfall Potential, or TRaP product, can be updated as significant changes in storm motion are observed and used to assess the potential for flash flooding conditions once a storm makes landfall.

For this case study, we'll apply the TRaP method to first the AMSU, and then the SSM/I rainfall rates to generate three different rainfall estimates. The first time period uses AMSU rain rates to look at Hurricane Georges about 12 hours before passing over the Florida Keys. Subsequent rainfall estimates are generated for the storm using SSM/I rain rates about 32 and 20 hours preceding the time the eye made landfall along the Gulf Coast. With the latter estimates, we'll focus on the evolution in storm rainfall potential for the period beginning with landfall early on 28 September 1998.