ATL: IRMA - Models

[SFLcane]

Holy hell on that GEFS into Florida..

[SouthFLTropics]

HWRF at 126 hours... 929MB NE of Puerto Rico. 118 Knots...Too close for comfort

[caneseddy]

This is a dramatic change from 18z GFS ensemble. Most members agree a landfall.

Yep...most ensembles are now South Carolina and points south..not a good trend if you are in Florida

[ronjon]

The ensembles are showing increasing likelihood of a CONUS hit. We wait and watch.

[Ivanhater]

Most show a continued wnw heading indicating we may see some further SW shifts in the future with a stronger high moving into place

[curtadams]

It's interesting to see the GEFS spread considerably to the southwest in the medium range when compared to the operational run. The GEFS envelope is generally thought to be a little under-dispersive in many cases, but the operational run isn't even in the envelope.

Not quite. There's *one* member landfalling near NYC and a couple more scattered about in odd places. The operational is certainly an atypical member, but it's not entirely out of the spread.

[1900hurricane]

It's interesting to see the GEFS spread considerably to the southwest in the medium range when compared to the operational run. The GEFS envelope is generally thought to be a little under-dispersive in many cases, but the operational run isn't even in the envelope.

Not quite. There's *one* member landfalling near NYC and a couple more scattered about in odd places. The operational is certainly an atypical member, but it's not entirely out of the spread.

Well, kind of. That one member with a similar track is much slower than the operational run. By 12Z September 10 (tau 210), there are no members remotely close to the operational run.

[stormreader]

The one thing that I am concerned about is that no matter how powerful the windshield wiping is, we still have our 3 best models showing landfalls, which is the last thing this country needs.

Not going out to sea. Personally, I'm ruling out the Outer Banks and points north. So don't think it will even be close to going out to sea. Still possible it approaches SE coast and makes a strong NW jog toward the Cape Fear, NC area, but that's about as far north as I think it would go. So broad area for me is Cape Fear, NC down to the coast of Cuba. See how that plays out for a couple of days. If that looks to verify, then further refine it then.

[Dean4Storms]

Anything beyond 5 days is subject to big changes with any model or ensemble. Just an extrapolation of the slightest track or speed change now can make hundreds of miles difference out 6-10 days. A difference of hitting Florida to New England in this situation right now.

[TheStormExpert]

So should we be expecting a westward shift on the 00z GFS run showing a Florida landfall?

[stormreader]

Most show a continued wnw heading indicating we may see some further SW shifts in the future with a stronger high moving into place

Very possible.

[gatorcane]

So should we be expecting a westward shift on the 00z GFS run showing a Florida landfall?

Yes I think we will see a significant west shift. Often times the ensembles gives clues on what the next OP run will do. Yikes that is some ensemble run.

[Langinbang187]

Definitely favoring the Euro setup here. That ridge over Atlantic Canada looks comically overdone.

On the contrary, the trough or amount of cutoffing could be overdone as well.

Keep in mind we've had roughly two weeks of clear, cool weather from the last ridge up here, so the next one could be a monster too.

We haven't had a "monster ridge" in Nova Scotia all Summer. I'm not banking on it.

[Michele B]

It's interesting to see the GEFS spread considerably to the southwest in the medium range when compared to the operational run. The GEFS envelope is generally thought to be a little under-dispersive in many cases, but the operational run isn't even in the envelope.

Not quite. There's *one* member landfalling near NYC and a couple more scattered about in odd places. The operational is certainly an atypical member, but it's not entirely out of the spread.

Well, kind of. That one member with a similar track is much slower than the operational run. By 12Z September 10 (tau 210), there are no members remotely close to the operational run.

Oh, I REALLY don't like this image!

Except that would bring us NE winds....which is the only good part. I wouldn't have to worry about the one tree I'm worried would fall in a storm. NE winds will take it down away from the house!

[Michele B]

The GEFS have a ridiculous signal into Florida at 18z.

Can you post image?

Had to do a double take when I first ran through this on WSI.

Yeah, I saw this. Looks like Charley again.

I don't want to go through that again! But preps are just about done, so...now we wait.

[sma10]

Holy Hannah, did you see the strength of those members striking SE FL? I think they were all cat 5.

[TheStormExpert]

So should we be expecting a westward shift on the 00z GFS run showing a Florida landfall?

Yes I think we will see a significant west shift. Often times the ensembles gives clues on what the next OP run will do. Yikes that is some ensemble run.

Yeah starting to think that MAYBE unfortunately the models are starting to come in-line with Irma getting blocked and forced west into the U.S. likely somewhere along the SE U.S. coastline including Florida.

[otowntiger]

Landfall is is 8 1/2 days from 18z today which means the storm hits NYC Sunday morning at 1am. 8 1/2 days is a long time in the model world but inside of model "fantasy land". And I'd caution everyone that the GFS and European haven't exactly been good in the mid-range which is what we'll be in throughout the labor day weekend. Unless all tracks come into agreement, don't expect to see the real solution, unless one accidentally hits it, before next Tuesday when we get inside of 5 days.

IDK Steve the EURO was pretty good with Harv around 3-4 days. It was the first model to throw out the wacky mid TX coast hit then bounced back to MX then back up the coast. Probably the reason the NHC is weighing so heavily on it this year. By far it has out performed the GFS as we all know...I also like the UKMET model this year...Not bad squirrely has the others. JMO

yeah the Euro did do well with Harvey like you say, only 3-4 days out. How did it do 8-9 days out?

[jlauderdal]

This is a dramatic change from 18z GFS ensemble. Most members agree a landfall.

the recurve idea east of the bahamas sure did die a quick death..

[hurricaneCW]

You can speculate all you want but none of it is legit because it's 8+ days out. I can't even take these runs seriously until Irma is near the islands.

As far as I know all options remain on the table.