ATL: IRMA - Models

[KWT]

Unreal deepening again by the 06z GFS, down to 894mbs!

Very powerful hurricanr forecasted right through the Bahamas, still heading NNW 168hrs out.

[xironman]

Hadn't seen it posted yet

[Blown Away]

06z GFS... 174 hours... Just SE/Slower than 00z... Just N of North Bahamas moving slowly NW... Trough retreating quickly...Cat 5 at 887mb, that's amazing!!!... Looks like another CONUS landfall??

WHAT DOES 887mb = MPH???

[xironman]

The GFS is just being ridiculous now, 888 at north Florida latitude.

[TheStormExpert]

Landfall same location as 00z near NC/SC border.

[KWT]

The thing is BA it is going to have a very good set-up aloft with the upper winds to the north of Irma being really supportive of a great outflow so its not impossible that it does dive right down. Overdone, but the general trend for a top end 4, maybe 5 is plausible this time.

[Blown Away]

06z GFS... 198-204 hours... Just SE/Slower than 00z... Landfall Just E of North/South Carolina Border moving NW... Trough gone HP building...Cat 4/5...

Note: Beyond that the HP drives Irma into W North Carolina/Virginia, so really strong HP...

[weathaguyry]

I think the Euro has a slight over-ridging bias, and the GFS has an over-troughing bias, Cindy is a good example of this, when in the longer range, the Euro would keep it buried in the BOC, but the GFS would go into the Florida panhandle. Because of this, its surprising to me that they have switched places. In a couple days we will be able to get a better handle on this, especially in this pattern where there are a lot of variables. Again, this is 1000% a personal opinion, and not to be taken as fact

[xtyphooncyclonex]

The GFS is just being ridiculous now, 888 at north Florida latitude.

Perhaps 175-190 mph

[xironman]

I think the Euro has a slight over-ridging bias, and the GFS has an over-troughing bias, Cindy is a good example of this, when in the longer range, the Euro would keep it buried in the BOC, but the GFS would go into the Florida panhandle. Because of this, its surprising to me that they have switched places. In a couple days we will be able to get a better handle on this, especially in this pattern where there are a lot of variables. Again, this is 1000% a personal opinion, and not to be taken as fact

The GFS and the Euro are within reasonable agreement on the 500mb pattern through day 7. Normally that is a sign that the solution is getting more solid.

[USTropics]

Here is a breakdown of all 51 ECMWF ensemble members from the 00z suite.

17/51 impact Florida


10/51 impact the Carolinas


6/51 impact the Northeast


2/51 impact the GOM


16/51 recurve - https://image.ibb.co/j1vC8a/Webp_net_gifmaker_4.gif

This is for long range guidance only. This also only includes first landfalls (there are numerous runs with multiple landfalls). Recurves are closest decent to CONUS before recurve occurs.

[xtyphooncyclonex]

I think the Euro has a slight over-ridging bias, and the GFS has an over-troughing bias, Cindy is a good example of this, when in the longer range, the Euro would keep it buried in the BOC, but the GFS would go into the Florida panhandle. Because of this, its surprising to me that they have switched places. In a couple days we will be able to get a better handle on this, especially in this pattern where there are a lot of variables. Again, this is 1000% a personal opinion, and not to be taken as fact

I see your point, but ALL of the global models and even the official NHC track have been trending southwest for the past 24 hours or so, hence this is closely watched by enthusiasts, myself included, and meteorologists.

[Kingarabian]

Sept 2 model trends haven't been good for the CONUS. Thankfully it's still nearly 8 days out before any landfall and the situation can change.

[DESTRUCTION5]

06z

00z

Huge scary W shift of the TVCN consensus... NHC typically follows TVCN consensus...

I think this is a good look at what NHC is seeing with being North of some reliable models. Ukmet and HWRF still south of 5am . UKMET isnt paying no mind to trough at all here.

[hurricaneCW]

Worst case scenario would be a direct impact on those islands, Puerto Rico, significant impacts for Hispaniola, Bahamas, and finally a Cat 4+ landfall on the east coast, and those options are definitely on the table.

[WAcyclone]

A short summary of the UKMET forecast:

In the short term, the operational run as well as the UKMET ensembles have shifted significantly closer to the islands:



Last three runs of the ensembles:



The most recent operational seems to indicate a turn to the NW/N somewhere in the Bahamas:

[USTropics]

Below is a chart for average positional error by nautical mile ONLY for Irma (includes all model runs since Irma initialization). ECMWF data only goes out to 72 hours:





data source - http://tropicalatlantic.com/models/models.cgi?basin=al&year=2017&storm=11&display=model_error&error_type=average&latestrun=1&type=table

[forecasterjack]

ECMWF shows a little bit of westerly shear when Irma's in the Bahamas. Based on that, I'd be skeptical of any sub 900mb forecasts. You need a perfect environment to get those pressures. Anything that's not perfect (a hint of westerly shear for example) should limit that potential. Still, we saw what even low end cat 4's can do wind damage wise in Rockport. I sure wouldn't want that headed for my house.

[Blown Away]

00z Euro at 168 hours, moves nearly due N over NW Bahamas and landfalls in N Carolina actually E of the 06z GFS...

[Michele B]

I think some of you never sleep!

Just got up myself and not liking the 5 am models. There is one thing I'm confused about tho. It looks like the models DO have Irmanmaintaining her w to wsw movement for now, but they seem to. E saying when she recurve to the north, it will be quite sharp. In other words the latest models I'm looking at show Irma coming ever closer to tip of S FL yet still turning north and missing us altogether and instead making landfall in S or even N Car.

Does this mean they believe the trough will still be there and will a big player in steering her up the east coast?