Derecho wrote:The FSU Superensemble incorporates the previous NHC forecast track, so it's not really a pure model.
Also, NHC notes that they don't get it in time to actually use it for forecasts most of the time, which I thought was interesting.
The accuracy of GUNA (which is a simple average of the GFDL, UKMET, NOGAPS, and GFS) is so close to the FSU Supensemble that the difference doesn't matter much, and the public can get a look at it (with the appropriate software to average the model tracks) unlike the FSU.
People need to stop thinking about having a favorite single "model" and recognize that consensus models are the promised land. Most of the time the NHC forecast tracks are close to or even identical to GUNA.
From what I have heard, NHC does get the Super-Ensemble in a timely fashion now. It is true, however, that last year, that the output did not get to NHC in time to actually incorporate it into their forecasts; it got there just before the advisory package was due out, so all the forecaster could do was look at it and comment in the discussion where his forecast was relative to the SE forecast.
Also, note that the SE produces not just a track forecast, but an intensity forecast as well. Last year it was tops in both categories.