[ Read Responses | Return to the Index ]

[ Previous | Next ]

Message: change (again!) to RUC SST's

Posted by Geoff Manikin on 31 Jan 01, 7:41 MT

The sea-surface temperature analysis used by the RUC has been changed beginning with the 12Z cycle on Wednesday 31 January 2001. The NESDIS analysis based on satellite retrievals was pulled from the RUC (and Eta) in the middle of December due to problems with updating this analysis during periods of significant, persistent cloud cover over the ocean. This was replaced by the Reynolds analysis which combines buoy observations and satellite data.

 The new analysis is the Real Time Global SST (RTG_SST, also referred to as the 2DVAR) which is performed on a 50 km grid and updated daily. Like the Reynolds analysis, the RTG_SST combines in-situ observational data and hi-resolution (4 km) satellite retrievals, but the Reynolds analysis is performed on a lower-resolution grid and combines 7 days worth of data to yield a very smooth analysis. The RTG_SST analysis uses only the most recent 24 hours of data and employs a variable background-error covariance function to produce greater detail.

 The RTG_SST has a much sharper depiction of the Gulf Stream than the other analyses and is the only one to properly depict the colder shelf water along the east and Gulf coasts. Development of the RTG_SST was accelerated after the incorrect depiction of the colder shelf water was found to play a significant role in the Eta model busted snowstorm forecast for the Washington/Baltimore areas on 29-30 December 2000. The RTG_SST was implemented in the Eta model at 12Z 30 January 2001.

 Plots of SST difference fields between the RTG_SST and Reynolds analyses in the RUC for 29 January can be viewed at http://www.emc.ncep.noaa.gov/mmb/ruc2/sst2dvar.html

The FSL RAP/HRRR/RUC Information Forum is maintained with a variant of WebBBS .