• Hello, please take a minute to check out our awesome content, contributed by the wonderful members of our community. We hope you'll add your own thoughts and opinions by making a free account!

Tropical Major Hurricane Michael

Jimmy Hypocracy said:
I’m gonna go ahead and air this out. Global models did a sh** job with Michael and a sh** job with Florence as far as pressure and strength go. It’s almost like we shouldn’t start paying attention unless models are showing a mid grade Hurricane a few days out
Click to expand...

Models are tools to be used in conjunction with the understanding of what they are best used for. The globals did a fantastic job of picking up on the RI signal that we saw unfold but they aren’t capable of fully forecasting just how strong it would get. This is where experience and other models like the HWRF can be useful.

Having said that the Euro and UK both had multiple runs showing 940s for pressure and a few 930s on the Euro. The FV3 and GFS consistently showed improving shear and quick pressure drops. Putting all that together and having the experience is how forecasting is done. Too many people model hug the verbatim outputs rather than using them as the tools they were designed to be. Just my 2 cents.
 
WSB had a live reporter in Thomasville Ga for their 5 and 6 o’clock news expecting that’s where the storm was headed but it crossed over or close to Marianna Fl with 120plus mph winds. They are about 40 miles west of the expected path.
 
snowlover91 said:
Models are tools to be used in conjunction with the understanding of what they are best used for. The globals did a fantastic job of picking up on the RI signal that we saw unfold but they aren’t capable of fully forecasting just how strong it would get. This is where experience and other models like the HWRF can be useful.

Having said that the Euro and UK both had multiple runs showing 940s for pressure and a few 930s on the Euro. The FV3 and GFS consistently showed improving shear and quick pressure drops. Putting all that together and having the experience is how forecasting is done. Too many people model hug the verbatim outputs rather than using them as the tools they were designed to be. Just my 2 cents.
Click to expand...
Why can’t we give human input before a model runs to let it know the current pressure?
 
Jimmy Hypocracy said:
Why can’t we give human input before a model runs to let it know the current pressure?
Click to expand...

We do. Recon data as it gets closer to land.

The thing here is, while we call these models "bad" and "goofus" and "stupid", they're a lot smarter than we are and crunch a lot of data that we can't possibly do. We just give them algorithms and instruction on what to do.
 
snowlover91 said:
Models are tools to be used in conjunction with the understanding of what they are best used for. The globals did a fantastic job of picking up on the RI signal that we saw unfold but they aren’t capable of fully forecasting just how strong it would get. This is where experience and other models like the HWRF can be useful.

Having said that the Euro and UK both had multiple runs showing 940s for pressure and a few 930s on the Euro. The FV3 and GFS consistently showed improving shear and quick pressure drops. Putting all that together and having the experience is how forecasting is done. Too many people model hug the verbatim outputs rather than using them as the tools they were designed to be. Just my 2 cents.
Click to expand...
Mets were saying high end Cat3 with MAYBE a run at low end Cat4 before landfall. In actuality it was high end Cat4 low end Cat5 at landfall with Cat4 and Cat3 WELL inland. Idk I just refuse to pat the globals on the back when I know drastic improvement is needed
 
NoSnowJoe said:
WSB had a live reporter in Thomasville Ga for their 5 and 6 o’clock news expecting that’s where the storm was headed but it crossed over or close to Marianna Fl with 120plus mph winds. They are about 40 miles west of the expected path.
Click to expand...
Thomasville is east of Marianna...not west.
 
Shawn said:
We do. Recon data as it gets closer to land.

The thing here is, while we call these models "bad" and "goofus" and "stupid", they're a lot smarter than we are and crunch a lot of data that we can't possibly do. We just give them algorithms and instruction on what to do.
Click to expand...
Understandable. But if we know a Hurricane is, say, at 940mb currently and a model runs 15 minutes later and thinks it’s at 960mb how is that helpful in determining future strength and further inland impacts? We have people well inland dealing with a strong Cat3 right now
 
Shawn said:
We do. Recon data as it gets closer to land.

The thing here is, while we call these models "bad" and "goofus" and "stupid", they're a lot smarter than we are and crunch a lot of data that we can't possibly do. We just give them algorithms and instruction on what to do.
Click to expand...
This is what it was like when I was a kid and 1st contracted this unshakeable weather addiction ...

th.jpg
o_O

(George Winterling, Chanel 4, JAX)
 
Jimmy Hypocracy said:
Why can’t we give human input before a model runs to let it know the current pressure?
Click to expand...
I agree on this one. In the past 48 hours I've not seen one global model that had a correct pressure at the current point. Frankly, it left everyone in the dark. Quite a credibility issue. A simple question like what will wind or pressure look like at landfall that is supposed to happen in 6 hours wasn't credible...just my 2 cents.
 
Lanie Pope now calling for 40-60 MPH gusts for the Triad (WS/Greensboro) tomorrow.
 
I would argue as devils advocate on behalf of Models that this particular Hurricane behaved in a manner of which most varirables and trends which may be part of Computer model solutions wouldn't have expected. It was became the 3rd strongest recorded Hurricane in US history and that doesn't happen without unexpected perhaps uncapturable data points.

Applying that logic hopefully this data is captured and used in future models to maybe pull closer to a better solution
 
malak05 said:
I would argue as devils advocate on behalf of Models that this particular Hurricane behaved in a manner of which most varirables and trends which may be part of Computer model solutions wouldn't have expected. It was became the 3rd strongest recorded Hurricane in US history and that doesn't happen without unexpected perhaps uncapturable data points.
Click to expand...
Just toss in a little human common sense and we're good ...
 
Jimmy Hypocracy said:
Mets were saying high end Cat3 with MAYBE a run at low end Cat4 before landfall. In actuality it was high end Cat4 low end Cat5 at landfall with Cat4 and Cat3 WELL inland. Idk I just refuse to pat the globals on the back when I know drastic improvement is needed
Click to expand...

The globals did fine for their intended purpose of the track and RI signal. They’re not intended to be pinpoint accurate for TC intensity. For that the HWRF and HMON are good tools to use and they consistently showed RI and drops into the 930s.

Many times local and tv Mets have their hands tied and just repeat the NHC forecast. The NHC is good at what they do but in cases of RI they are usually too low with strength because of how they determine their strength forecasts. They also got burned by Florence recently so that played into them being a little bit more conservative with their forecasting I’m sure.
 
snowlover91 said:
The globals did fine for their intended purpose of the track and RI signal. They’re not intended to be pinpoint accurate for TC intensity. For that the HWRF and HMON are good tools to use and they consistently showed RI and drops into the 930s.

Many times local and tv Mets have their hands tied and just repeat the NHC forecast. The NHC is good at what they do but in cases of RI they are usually too low with strength because of how they determine their strength forecasts. They also got burned by Florence recently so that played into them being a little bit more conservative with their forecasting I’m sure.
Click to expand...
HWRF did a great job with Michael but like you said it was hard to buy into because it showed Florence as a buzzsaw and we all know how that played out
 
Jimmy Hypocracy said:
Understandable. But if we know a Hurricane is, say, at 940mb currently and a model runs 15 minutes later and thinks it’s at 960mb how is that helpful in determining future strength and further inland impacts? We have people well inland dealing with a strong Cat3 right now
Click to expand...

Globals usually initialize too high when it’s an intense TC and that’s where the hurricane models are more useful. Globals are helpful for track and general signals, hurricane models are the next step to use for wind fields, structure, better initialization strength, etc. It is all in knowing how to use them as the tools they were designed to be used as and knowing their strengths and weaknesses.
 
Jimmy Hypocracy said:
HWRF did a great job with Michael but like you said it was hard to buy into because it showed Florence as a buzzsaw and we all know how that played out
Click to expand...

Just about every model struggled with Florence intensity after the botched EWRC that never really got going again. There will probably be some interesting studies into what happened and hopefully that data used in future model implementations to improve their performance but models still did a good job with the track despite the intensity struggles.
 
You must log in or register to reply here.
Back
Top