Why Franklin may LOOP after all

[mike18xx]

You know.. I usually don't go with the BAMM... but... *adds a few of her chips along with Mikes*

The BAMM's, IMO, miscasts the strength of the ridge, allowing Franklin to, after looping, then recurve back up the east coast (as if in front of another approaching short-wave) -- IMO, Franklin is going to move S, SW, then W, and accelerate pretty quickly once westward movement commences.

[Sanibel]

Big payoff coming for mike18xx...

[Ivanhater]

You know.. I usually don't go with the BAMM... but... *adds a few of her chips along with Mikes*

The BAMM's, IMO, miscasts the strength of the ridge, allowing Franklin to, after looping, then recurve back up the east coast (as if in front of another approaching short-wave) -- IMO, Franklin is going to move S, SW, then W, and accelerate pretty quickly once westward movement commences.

but it wont mean a thing if thunderstorms dont develop over the swirl, what are the uperlevel winds like for them to develop over it as it moves back west?

[mike18xx]

it wont mean a thing if thunderstorms dont develop over the swirl, what are the uperlevel winds like for them to develop over it as it moves back west?

The new movement southward is now consistant with southerly upper-level winds; the about-face shaves about 20kt off that shear.

It'll look like rags until movement trends from SW to WSW or W, marking arrival at the southern edge of the ridge; the environment is normally very favorable in such position, and intensification can be extremely rapid.

[Matt-hurricanewatcher]

I agree Mike18xx. Wow you seen this days before it happen. Very good job.

[Ivanhater]

it wont mean a thing if thunderstorms dont develop over the swirl, what are the uperlevel winds like for them to develop over it as it moves back west?

The new movement southward is now consistant with southerly upper-level winds; the about-face shaves about 20kt off that shear.

It'll look like rags until movement trends from SW to WSW or W, marking arrival at the southern edge of the ridge; the environment is normally very favorable in such position, and intensification can be extremely rapid.

so what do you think it will become and where do you think it will hit?

[caplan1]

mike18xx If this comes true, then you need to take the first plane to Vegas.

[Steve]

There are a lot of people who felt like a loop was possible (not necessarily on this thread). Frankly, I think it's done 2 start-stop loops already. But it will be a good call if it verifies. Tough to say (if Franklin can come back in a sense) what could happen. There is model support to move him north a few degrees east of the US Coast and being absorbed into a low south of the Maritimes. Of course there are several other solutions as well that could happen.

Steve

[mike18xx]

I agree Mike18xx. Wow you seen this days before it happen. Very good job.

I've seen ridges "abort" east-coast recurvature before, and in every case I could smell it coming save Jeanne (last year), model-majority was caught flat-footed.

IMO, the "rules" (whatever they may be, if any) regarding how LLCs interact with fronts need to be re-written for the models. Here's what happens:

1. Trough approaches sub-20N tropical cyclone (TC), and recurvature begins.
2a. If TC is closed-eye hurricane, steering "consensus" whisks TC E/NE away to N. Atlantic.
2b. If TC is weak, westerly mid-level shear (just ahead of front) shoves convection away, exposing LLC.
3. Surface cool frontal passage OCCLUDES the LLC; mid-level westerlies depart.
4. Behind the front, the LLC is now under the influence of high-pressure steering winds (northerly veering easterly).
5. LLC swirl moves S, then SW, then W, accelerating and rapidly intensifying during W movement.
6. Southeastern states panic; overly model-dependant forecasters red-faced.

=-=-=-=-=-=

Here's the deal: Computer models feature far faster computational power than the human brain, but are subject to the great limitation of data input. Computer models are also not sentient artificial intelligences -- they are unequipped to perform pattern recognition.

Imagine, if you will, the entire continental Unites States, with hourly surface data from ten stations per state and soundings from maybe two to four places per state twice a day (and this is MUCH better than a lot of the world), with all of this data flowing into computer models to predict the weather. -- Seems like a lot of data, right?

Well, it's not, really. Do the math: 50 states x 10 obs x each ob containing about ten disparate pieces of model-incorporated information is about 5,000 "chunks" of data per hour. Now consider this: I'm watching a WV satellite loop, which is updated every 15 minutes, and I am capable of discerning three "layers" of cloud movement (at surface, mid-altitude, and cirrus-level); if I am animating the NOAA GOES loops, my slower-than-a-computer and not-mathematically-inclined but pattern-recognition-capable brain is staring at 532x443x4x3 = 2.8 million "chunks" of data per hour. Even considering that a model has access to all kinds of information that I can't see and that the mathematicians who wrote the code are geniuses, I am still dog-paddling in many orders of magnitude more data than the model.

When a model is written which can ingest a raw satellite feed and perform pixel-by-pixel vector-analysis -- then a model will crush sentient pattern-recognition and extrapolation.

[Ivanhater]

I agree Mike18xx. Wow you seen this days before it happen. Very good job.

I've seen ridges "abort" east-coast recurvature before, and in every case I could smell it coming save Jeanne (last year), model-majority was caught flat-footed.

IMO, the "rules" (whatever they may be, if any) regarding how LLCs interact with fronts need to be re-written for the models. Here's what happens:

1. Trough approaches sub-20N tropical cyclone (TC), and recurvature begins.
2a. If TC is closed-eye hurricane, steering "consensus" wisks TC E/NE away to N. Atlantic.
2b. If TC is weak, westerly mid-level shear (just ahead of front) shoves convection away, exposing LLC.
3. Surface cool frontal passage OCCLUDES the LLC; mid-level westerlies depart.
4. Behind the front, the LLC is now under the influece of high-pressure steering winds (northerly veering easterly).
5. LLC swirl moves S, then SW, then W, accelerating and rapidly intensifying during W movement.
6. Southeastern states panic; overly model-dependant forecasters red-faced.

=-=-=-=-=-=

Here's the deal: Computer models feature far faster computational power than the human brain, but are subject to the great limitation of data input. Computer models are also not sentient artificial intelligences -- they are unequipped to perform pattern recognition.

Imagine, if you will, the entire continental Unites States, with hourly surface data from ten stations per state and soundings from maybe two to four places per state twice a day (and this is MUCH better than a lot of the world), with all of this data flowing into computer models to predict the weather. -- Seems like a lot of data, right?

Well, it's not, really. Do the math: 50 states x 10 obs x each ob containing about ten disparate pieces of model-incorporated information is about 5,000 "chunks" of data per hour. Now consider this: I'm watching a WV satellite loop, which is updated every 15 minutes, and I am capable of discerning three "layers" of cloud movement (at surface, mid-altitude, and cirrus-level); if I am animating the NOAA GOES loops, my slower-than-a-computer and not-mathematically-inclined but pattern-recognition-capable brain is staring at 532x443x4x3 = 2.8 million "chunks" of data per hour. Even considering that a model has access to all kinds of information that I can't see and that the mathematicions who wrote the code are geniuses, I am still dog-paddling in many orders of magnitude more data than the model.

When a model is written which can ingest a raw satellite feed and perform pixel-by-pixel vector-analysis -- then a model will crush sentient pattern-recognition and extrapolation.

good God man, just tell me what you think franklin will do and where he will head?lol....great post though

[cinlfla]

I've seen ridges "abort" east-coast recurvature before, and in every case I could smell it coming save Jeanne (last year), model-majority was caught flat-footed.

IMO, the "rules" (whatever they may be, if any) regarding how LLCs interact with fronts need to be re-written for the models. Here's what happens:

1. Trough approaches sub-20N tropical cyclone (TC), and recurvature begins.
2a. If TC is closed-eye hurricane, steering "consensus" whisks TC E/NE away to N. Atlantic.
2b. If TC is weak, westerly mid-level shear (just ahead of front) shoves convection away, exposing LLC.
3. Surface cool frontal passage OCCLUDES the LLC; mid-level westerlies depart.
4. Behind the front, the LLC is now under the influence of high-pressure steering winds (northerly veering easterly).
5. LLC swirl moves S, then SW, then W, accelerating and rapidly intensifying during W movement.
6. Southeastern states panic; overly model-dependant forecasters red-faced.

=-=-=-=-=-=

Here's the deal: Computer models feature far faster computational power than the human brain, but are subject to the great limitation of data input. Computer models are also not sentient artificial intelligences -- they are unequipped to perform pattern recognition.

Imagine, if you will, the entire continental Unites States, with hourly surface data from ten stations per state and soundings from maybe two to four places per state twice a day (and this is MUCH better than a lot of the world), with all of this data flowing into computer models to predict the weather. -- Seems like a lot of data, right?

Well, it's not, really. Do the math: 50 states x 10 obs x each ob containing about ten disparate pieces of model-incorporated information is about 5,000 "chunks" of data per hour. Now consider this: I'm watching a WV satellite loop, which is updated every 15 minutes, and I am capable of discerning three "layers" of cloud movement (at surface, mid-altitude, and cirrus-level); if I am animating the NOAA GOES loops, my slower-than-a-computer and not-mathematically-inclined but pattern-recognition-capable brain is staring at 532x443x4x3 = 2.8 million "chunks" of data per hour. Even considering that a model has access to all kinds of information that I can't see and that the mathematicians who wrote the code are geniuses, I am still dog-paddling in many orders of magnitude more data than the model.

When a model is written which can ingest a raw satellite feed and perform pixel-by-pixel vector-analysis -- then a model will crush sentient pattern-recognition and extrapolation.

This is just amazing, I mean totaly amazing. This is good info for a novice like myself.

[mike18xx]

so what do you think it will become and where do you think it will hit?

*Shrug*. Too early to tell. I can easily spot a continent-sized high-pressure ridge building; where it'll melon-seed-squirt a TC caught underneath it is more difficult. A lot depends upon "early chaos variables" such as how far south Franklin is shoved -- farther south means longer westward track over probably warmer water later on down the road.

What-the-hey; how 'bout a cat-1 into Savannah?

[Ivanhater]

so what do you think it will become and where do you think it will hit?

*Shrug*. Too early to tell. I can easily spot a continent-sized high-pressure ridge building; where it'll melon-seed-squirt a TC caught underneath it is more difficult. A lot depends upon "early chaos variables" such as how far south Franklin is shoved -- farther south means longer westward track over probably warmer water later on down the road.

What-the-hey; how 'bout a cat-1 into Savannah?

hey of this pans out, you are the best forecaster ive ever meet, keep up the good work mike!

[WeatherEmperor]

Heck if that forecast works out, I will worship and call him my Master for the remainder of this season.

<RICKY>

[mike18xx]

hey of this pans out, you are the best forecaster ive ever meet, keep up the good work mike!

*Shrug*. Just don't forget that Franklin got farther northeast than I predicted earlier in the thread, and that it seperated from the Florida upper-low I thought it would fujiwara with.

[mvtrucking]

Mike,
You have been on to this scenario since at least Saturday @ 3:00am(First post). I am duely impressed with your call so far on this system. My hat is off to ya!.....

[Stormswirl]

Thanks kindly for sharing your knowledge Mike!

[Eyes2theSkies]

mike, if the works out I will write a letter of recommendation to the nhc for you...

[WeatherWiseGuy]

The satellite loop certainly seems to be backing up your scenario. Congrats!

[stormchazer]

I am placing bets where ever mike18xx is!

I have been reading this ongoing post with great interest since its inception. The aurguement seemed easy to accept but since it had not materialized I thought we might hear a retraction from you today, but it appears that you where definately on the right track.