I have had several emails asking for comments on the various computer models that the National Hurricane Center (NHC) uses for guidance in producing their forecasts. There is an excellent review of the models on the NHC web page that has been recently updated at http://www.nhc.noaa.gov/aboutmodels.shtml. In addition, each official forecast from the NHC is verified along with most of the model forecasts. If interested in the verification details, I would suggest you look at the annual verifications posted at http://www.nhc.noaa.gov/verification/.
The above graphic shows the variable consensus (TVCN) model which is in between the GHMI and the HWFI models. The TVCN is not a true model but is a combination of up to seven other models (including the GHMI and HWFI). Of the various consensus models used by the NHC, the TVCN was the best in 2008.
The above graphic shows that the official NHC forecast on Erika just issued at 11 am 2 Sep 2009 based, in part, on the guidance models above. The NHC has many other models that are looked at before making their final forecast. Note that the official forecast, in this case, is very close to the consensus model TVCN.
It is also important not to focus on the skinny track lines shown in the first two graphics above. The NHC wisely issues a “cone of uncertainty” with each track forecast. More on that tomorrow.