Why the GFS model is the one every NWS office uses?

[cycloneye]

I ask this question because I notice that every discussion at all the NWS areas in the US and includes Puerto Rico talk about the GFS and not other models.Ok some of the more pros and experts here can give the members some light about why that is the case.

[Ola]

good question, why?

[senorpepr]

Here's my thinking. See, where I work, the typical model of choice for mid-latitude locations is the GFS model. It has a better track record. Normally they (both NWS and where I work) look at other models, but tend to stick with the GFS simply because of it's better record.

It's like a baseball game. You know Bobby is probably going to make a few runs for the teams versus Tommy who'll strike out a few times. You're more likely to put Bobby up simply because of his history. Similar case goes for weather models.

With that being said about GFS, typically NOGAPS is used more for tropical locations. JMA is the typical model of choice for Japan and Korea while the UKMO has a great handle of European weather.

Hopefully this leads toward the answer.....

[Rainband]

I thought it was because GFS is the longest term model that shows all atmospheric variables??

[senorpepr]

I thought it was because GFS is the longest term model that shows all atmospheric variables??

That's also part of it. If you're doing a longterm forecast, you don't want to start off using a model like the MM5, then switch after 72 hours to GFS. It helps in the consistency department.

[MWatkins]

I think there are a few reasons...

1. It's a US model.

2. It runs 4 times a day (2 times with new observations).

3. The reasons senorpepr pointed out.

4. It generally outperforms the other Global models

I remember 10 years ago...the AVN was just the Avaition run of the MRF model...it was used to fill in the gaps for upper-air data...and it's grown from there. If you look at the PSU models page...there are several more available fields to evaluate than in the other models.

The GFS has it's problems...but they are generally well known...and experienced forecasters can account for these and adjust accordingly.

Here's a good...but technical list of observed biases in all of the models...

http://www.hpc.ncep.noaa.gov/mdlbias/biastext.shtml

MW

[Rainband]

Thanks Guys

[Aquawind]

Good question..Great answers..Awesome link MW..

A clip from the GFS portion of the link..

When specific thresholds in the mass fields are met, convective scheme is triggered and then dumps a large amount of QPF over a grid point - releasing so much latent heat over the grid point that the model is forced to adjust the mass fields by producing a local vertical motion max in the mid troposphere (~ 500mb), a corresponding upper level jet max over the vertical motion max - an intense and small scale vort max in the mid levels (MCV).
The model scales up the mesoscale circulation at mid levels and holds onto it as a real feature for as long as 3 days. The model can produce precipitation in association with the feature as it tracks along in the flow.

Clip from the UKMET portion..

When GFS is showing phasing of systems beyond 84 hours, check UKMET to see if solution is consistent..

[Rainband]

Clip from the UKMET portion..

When GFS is showing phasing of systems beyond 84 hours, check UKMET to see if solution is consistent..

I usually do that!! Thanks Aquawind

[Derek Ortt]

GFS is the government model and NWS is also government. The gov't typically uses their own tools before using others

[cycloneye]

Thanks guys for those answers that I wanted to see because I was for a long time asking why that happens but I haven't asked it here before however now I know why it is used by the NWS.

[Anonymous]

Cool info I have also noticed that many NWS offices sometimes go with the ETA as well in short term forecasting

[USAwx1]

Personally, I will use the model which has the BEST handle on the situation I'm faced with. For example the ECMWF will normally out perform the GFS when dealing with significant east coast low pressure systems during the cold season.

The ECMWF put the GFS to shame WRT the New England Nor'easter this past DEC 5-6.

Verification is most important, IF that means using another model BESIDES THE GFS so be it. It's NOT about what country the model is from, and to have a bias toward the GFS just because it's a US generated model is senseless.

A more recent example would be the Ghost system the GFS was developing run after run over the ATL for the early part of this coming week. The remainder of the data (not the CMC, JMA, NOGAPS, ECMWF...NOTHING) was indicating the same thing.

[Pinwheeler]

Didn't you know. The GFS stands for the Girl Friend System. The NWS employees (who have no chance of getting a date otherwise) are taught to kiss up to the GFS and be nice to it.

[USAwx1]

Didn't you know. The GFS stands for the Girl Friend System. The NWS employees (who have no chance of getting a date otherwise) are taught to kiss up to the GFS and be nice to it.

OUCH!! OUCH!! OUCH!!

That was BRUTAL.

BUT just to let you know NOT all Meteorologists whether they are w/ the NWS or not are hard-pressed to find dates. or unattractive.

I like that one though

[OuterBanker]

Thanks MW for the link. Great reading, I now understand why the GFS is used at TPC. After reading all the bias and operational implication I would have to agree that the GFS is the best if you had to choose a single model. Interesting that the UKMET gets better grades for phasing. The way I read it though is the NCEP GFS EPS ensemble is best overall (although strength is subdued). Is this what TPC uses during hurricane season?