GFS shows Mobile Bay/MS border

[hurricanetrack]

I hope I am not repeating another post- but the 0z GFS now shows Mobile Bay area (or even MS/AL border) for a landfall- quite a jog back to the east like Pasch alluded to earlier today. We're also within 5 days of landfall, so these runs are getting to be more and more important.

http://www.nco.ncep.noaa.gov/pmb/nwprod ... p_096m.gif

[ALhurricane]

Major problem with the 00z GFS..

way too far north and east in the first 48 hours. if you believe this is going to be over southern cuba like the GFS is showing then you have my sympathies

once again...if the model cannot figure out the short term right how can it get the later forecast hours right?

[jkt21787]

FWIW (it may not be much), Canadian is coming back with a Mobile landfall as well.

[Houstonia]

DI can't do with another major hurricane. They only had about 6 months to recover from last season and they got hit right away with Cindy.

[AL Chili Pepper]

FWIW (it may not be much), Canadian is coming back with a Mobile landfall as well.

You got a link? That would be a major change from the 12Z.

[jkt21787]

FWIW (it may not be much), Canadian is coming back with a Mobile landfall as well.

You got a link? That would be a major change from the 12Z.

Sorry, Looking at the wrong data.

Canadian still shows FL.

In fact it appears to show a Miami hit!

[dixiebreeze]

They may all start trending more east after tonight. There's a big push swooping down to the GOM from WNW.

[jkt21787]

As ALhurricane said, GFS is not doing well in the short term. We know CMC/Canadian is bad. NAM is bad as well. All of those show the FL landfalls.

The others are showing NO-Mobile landfalls. That's where I'm going right now.

Edit: GFS is NOT showing a FL Landfall. Its showing Mobile as well.

[Stormcenter]

Just keep it away from Houston. We need the rain but NOT that bad.

[AL Chili Pepper]

I agree AL, it's looking more and more like Dennis may even go south of Jamaica. I don't see much of a northerly component right now, at least not enough to have it hugging the southern coast of Cuba.

[Stormcenter]

Wouldn't it be something if never made the NW turn?

[gkrangers]

We gave the GFS too much credit today..it screwed the pooch on this run.

[jkt21787]

Dennis has been consistently going south of recent GFS runs...this could have big implications...

[mobilebay]

Major problem with the 00z GFS..

way too far north and east in the first 48 hours. if you believe this is going to be over southern cuba like the GFS is showing then you have my sympathies

once again...if the model cannot figure out the short term right how can it get the later forecast hours right?

I'm not a Pro-Met and don't claim to be. However, how bad of a run could it be? It is in tightly clustered with other models NOGAPS,UKMET,GFDL, and the NHC projected path. The GFS has been showing this area for days. It had one bad run at 12Z and shifted back East, just like the NHC said it would in their 5PM Discussion.

[mtm4319]

Does anyone know of a static map that shows computer models (like the one on Weather Underground here) but doesn't use the Mercator projection?



I like this one, but I think it moves depending on the storm's location. (I'm trying to make a Google Earth image overlay that will work for most storms.)[/url]

[mobilebay]

also the 00Z UKMET stays exacty the Same= New orleans! I will post the Link in a minute.


http://moe.met.fsu.edu/cgi-bin/ukmtc2.c ... =Animation

[jkt21787]

also the 00Z UKMET stays exacty the Same= New orleans!

https://www.fnmoc.navy.mil/CGI/PUBLIC/w ... rp&tau=096

UKMET has been and continues to remain my favored model, for its consistency in this similar track.

[rtd2]

Major problem with the 00z GFS..

way too far north and east in the first 48 hours. if you believe this is going to be over southern cuba like the GFS is showing then you have my sympathies

once again...if the model cannot figure out the short term right how can it get the later forecast hours right?

I'm not a Pro-Met and don't claim to be. However, how bad of a run could it be? It is in tightly clustered with other models NOGAPS,UKMET,GFDL, and the NHC projected path. The GFS has been showing this area for days. It had one bad run at 12Z and shifted back East, just like the NHC said it would in their 5PM Discussion.

Its NOT a BAD run...It may have Missed the Initial Position slightly but It still Clustered with all the others!

[jkt21787]

Major problem with the 00z GFS..

way too far north and east in the first 48 hours. if you believe this is going to be over southern cuba like the GFS is showing then you have my sympathies

once again...if the model cannot figure out the short term right how can it get the later forecast hours right?

I'm not a Pro-Met and don't claim to be. However, how bad of a run could it be? It is in tightly clustered with other models NOGAPS,UKMET,GFDL, and the NHC projected path. The GFS has been showing this area for days. It had one bad run at 12Z and shifted back East, just like the NHC said it would in their 5PM Discussion.

Its NOT a BAD run...It may have Missed the Initial Position slightly but It still Clustered with all the others!

No, its not nearly as bad as CMC and NAM. However, its missed position and too far N and E could mean its Mobile landfall is a bit east than it should be. Its best to wait for additional runs of course to latch on to the meaning here.

[mobilebay]

NOW!! here is the 00Z UKMET= New orleans. Sorry about the Above.

http://moe.met.fsu.edu/cgi-bin/ukmtc2.c ... =Animation

The point I've been trying to make is there is a pretty tight cluster of models that calls for landfall between New Orleans and Pensacola. These models include The GFS, UKMET, GFDL, NOGAPS, and the NHC model suite. Note- the NHC model suite will probably shift a little East on the next run because of the GFS.