Weatherfreak14 wrote:Is the GFDL pretty accurate?
In terms of track, it's as good as any model - meaning better than climo and better than any met locked in a room without models. It still has a large error, larger than the NHC track. In terms of intensity, it's the best we have but it's not very good. It strongly overestimates storm intensity early in its run (for majors) and size late in the run. I think a skilled met can still beat the GFDL on intensity. I would absolutely take the NHC's Cat 2 estimate over the GFDL's Cat 3. That said, there's a huge error on any intensity forecast.