What's wrong with the GFS?

[cctxhurricanewatcher]

As someone who is not in the profession and just likes weather talk and observation as a hobby, I read on quite a few sources about how bad the GFS is. Then why does the NWS use it as their primary model for forescasting as noted in their discussions? Even I as an amatuer I notice their love affair with that model (no pun attended) and laugh when they have to backtrack from it, expecially when they are about to blow a forecast 24 -48 hous out.

I can't wait till the Houston NWS starts putting up "bears watch" for the upcoming cold cycle.

[yoda]

I don't know why they use the GFS... but sometimes they use the EURO. Anyway, the GFS always overdoes the amount of energy from the SJ stream. It does well with the Northern Jet Stream. However, it does like to paint huge fantasy snowstorms that are like 10+ days out that never materialize... and don't use the GFS for temps in the future, because it is usually overdone.

I'm am also an amateur... so I hope a pro met could make sure my info is correct.

[Aslkahuna]

The GFS, ETA, NGM are the US models which is why you will see NWS using it the most. When it comes to medium range models, the ECMWF (European) is concerned superior to the GFS (or POS as some of us refer to it as). Down here in SE AZ we always have a problem with which model is going to do the best down here in providing a heads up for incoming systems. Last year, the ECMWF and NOGAPS seemed to be the best-so far this year the UKMET and Canadian have been better and this is the key-for whatever reason, Nature likes dealing winning hands to different models each year and it's up to us to find out who has it. Unfortunately, though, it is true that the NWS forecasters tend to over rely upon the GFS-don't know if it's a policy thing or what.

Steve

[Stormsfury]

I don't know why they use the GFS... but sometimes they use the EURO. Anyway, the GFS always overdoes the amount of energy from the SJ stream. It does well with the Northern Jet Stream. However, it does like to paint huge fantasy snowstorms that are like 10+ days out that never materialize... and don't use the GFS for temps in the future, because it is usually overdone.

I'm am also an amateur... so I hope a pro met could make sure my info is correct.

Yoda, the GFS is horrid with the subtropical jet (SBJ), and underplays this ... with northern stream s/w's and such, the GFS will generally do well with ... but the problem with the GFS is ... that it's simply a short range model that's designed to do much better inside of 84 hours due to resolution of the model ... and still it has issues (convective feedback, etc.)

When ANYONE looks at the GFS for MR/LR, it should ONLY be used for overall pattern ... NOT details ...

SF