GFS resolution to be improved, NOT extended out to 574 hours

[hurricanetrack]

I read at stormtrack.org that the GFS is going to go out to 574 hours and that the hi-res will now be out to 192 hours. Here is the info as posted at stormtrack.org:

EFFECTIVE JUNE 22 2010...BEGINNING WITH THE 1200 COORDINATED UNIVERSAL TIME /UTC/ RUN...THE NATIONAL CENTERS FOR ENVIRONMENTAL PREDICTION /NCEP/ WILL UPGRADE THE GLOBAL FORECAST SYSTEM /GFS/. THE RESOLUTION OF THE GLOBAL FORECAST MODEL WILL BE INCREASED FROM T382 /35 KM/ TO T574 /27 KM/.
THE HIGH RESOLUTION PORTION OF THE FORECAST WILL BE EXTENDED FROM 180 HRS TO 192 HRS. WITH THIS EXTENSION 3 HOURLY OUTPUT WILL BE MADE AVAILABLE OUT TO 192 HOURS.

THIS WILL RESULT IN SIGNIFICANT CHANGES IN THE DEFINITION OF PARAMETERS IN THE 192 HOUR PRESSURE GRIB /PGRB/ AND FLUX FILES. THERE WILL ALSO BE SIGNIFICANT CHANGES IN MODEL PHYSICS ASSOCIATED WITH THIS CHANGE. IN ADDITION MODIFICATIONS WILL BE MADE TO THE CONTENTS OF THE GLOBAL DATA ASSIMILATION SYSTEM /GDAS/ AND GFS PGRB FILES.

CHANGES IN MODEL PHYSICS INCLUDE:

RADIATION AND CLOUD OVERLAP
GRAVITY WAVE DRAG
HURRICANE RELOCATION
NEW PLANETARY BOUNDARY LAYER SCHEME
NEW MASS FLUX SHALLOW CONVECTION
UPDATED DEEP CONVECTION SCHEME
POSITIVE DEFINITE TRACER TRANSPORT SCHEME

THE NEW PARAMETER FOR THE GFS FORECAST PRESSURE GRIB FILES IS:

MAX WIND GUST

Anyone know about this?

[GeneratorPower]

Oh, great. Just wait until we see a monster hurricane hitting New Orleans on the 574 hour GFS. I shouldn't be so cynical, I guess, not until we see if they make some improvements in the modeling.

My prediction: they just added another 192 hours of unusable fantasy forecasts. This isn't a belated April fools joke is it?

[Duddy]

4 weeks out? Seriously?

Certainly this is an April fools joke.

[SETXWXLADY]

Haven't heard of this one. If it's not a joke it's just something to drive hurricane nuts like me crazy even sooner.

OK. Just did some diggin'. Apparently it's legit. Not that I understood what they said but...

http://www.weather.gov/view/national.php?prod=PNS&sid=WSH

[RL3AO]

Well, the resolution increase is nice. Don't care much past 192 hours though.

[HURAKAN]

574 hr or almost 24 days into the future is where dragons and bad dreams reside!!!

[cycloneye]

I would not be surprised if in the future GFS goes out to 1,000 hours.

[RL3AO]

06 GFS shows a big hurricane in the GOM for Mid-July...2011.

[caneflyer]

T574 refers to the resolution of the model, not to how long the forecast runs. T574 refers to how many waves can be represented globally, and is roughly equivalent to a "grid" resolution of 25 km.

[wxman57]

Having just attended the National Hurricane Conferences, I did hear about some changes to the GFS as they pertain to tropical cyclones. The updated GFS resolution was increased from about 40km to a 25km grid. In addition to some of the other changes mentioned in this thread, it will now be initialized with the actual minimum central pressure of any tropical cyclone. This should significantly improve the model output.

When the upgraded GFS model was used to re-run data during Ike, it no longer drove Ike to the lower Texas coast 4-5 days before landfall. Instead, it moved Ike to the upper coast into Galveston Island, very close to its eventual landfall. That's good news, but it doesn't mean we can trust it with every storm now. Let's see how it does with this very active 2010 season.

By the way, I did find the report on the GFS upgrade on the NWS web site:

http://www.weather.gov/view/validProds. ... &node=KWBC

[jinftl]

That's interesting....do mets re-run past storms to test the accuracy of new models? Makes total sense if they do.

As far as a forecast 574 hours out, I can't see that doing anything more than predicting large-scale patterns...not specific landfalls of storms that won't form for a few weeks.

Having just attended the National Hurricane Conferences, I did hear about some changes to the GFS as they pertain to tropical cyclones. The updated GFS resolution was increased from about 40km to a 25km grid. In addition to some of the other changes mentioned in this thread, it will now be initialized with the actual minimum central pressure of any tropical cyclone. This should significantly improve the model output.

When the upgraded GFS model was used to re-run data during Ike, it no longer drove Ike to the lower Texas coast 4-5 days before landfall. Instead, it moved Ike to the upper coast into Galveston Island, very close to its eventual landfall. That's good news, but it doesn't mean we can trust it with every storm now. Let's see how it does with this very active 2010 season.

[AJC3]

Kids, the T382 and T574 DO NOT refer to times. Tnnn refers to triangular truncation of nnn number of waves represented by the model (not to mention the facts that the current output on the NCO website goes out to 384 and not 382 hours, and neither 382 nor 574 are divisible by 6).

You probably want to pass this on over at Stormtrack.

[srainhoutx]

This makes 2 major upgrades to the GFS in almost 6 months. I do hope we are beginning to "fix" the guidance to become more accurate and allow more lead time for prepartions and Advisories. Also with the upgrades to the ECMWF this year, perhaps we are getting just a bit closer to knowing how things will 'play out' concerning longer range guidance.

[cycloneye]

This makes 2 major upgrades to the GFS in almost 6 months. I do hope we are beginning to "fix" the guidance to become more accurate and allow more lead time for prepartions and Advisories. Also with the upgrades to the ECMWF this year, perhaps we are getting just a bit closer to knowing how things will 'play out' concerning longer range guidance.

Agree 100%.Lets see with these upgrades if GFS turns into the stellar model in 2010 and beyond as ECMWF was the best in 2009.

[wxman57]

That's interesting....do mets re-run past storms to test the accuracy of new models? Makes total sense if they do.

Yes, that's actually a standard way of evaluating model "improvements". Initialization data from the past are used to test model changes. Since we already know what happened in the past, we immediately know if the model changes were for the better or not.

[JonathanBelles]

Will this upgrade to the GFS help its horrible intensity forecasts?

[GeneratorPower]

06 GFS shows a big hurricane in the GOM for Mid-July...2011.

You should see the Mayan calendar !!!

[wxman57]

Will this upgrade to the GFS help its horrible intensity forecasts?

The intensity forecasts will improve some, but it's not a TC intensity model, so you won't be able to use the GFS for realistic intensity forecasts.

[hurricanetrack]

Thanks for the clarification!