NYC/PHL December 24-27 Potential - Part 2

[F5TornadoF5]

48hrs

[psv88]

There is such a thing as a meteorological interpretation of where a model will go in the future-- this is where "science meets art"-- unless youre blind, you see that the GFS just made a HUGE jump, and a much smaller one is necessary to make it a major hit.

And im not disagreeing with your point at all. I was simply asking who they were...to which nobody has responded. And yes it would be interesting to see verification scores of the mets...see who always wishcasts and who is realistic.

[A-L-E-X]

My apologies sir. I am currently writing an Appellate brief for a client and i was typing very fast.

It's ok lol-- I thought for a second I wrote all that haha. I've been up working all night and continuing to do so today so no telling what a lapse of concentration could result in.

[NittanyWx]

Not by eyeballing a static chart. Only after the fact with reassessment or computers to perform bias correction. The "extreme" is correct in this case.

Remind me what your profession is exactly? I only ask this because you speak to having an intricate understanding of why these models should always be right, and that humans cannot spot any error in a solution.

[F5TornadoF5]

60hrs

[A-L-E-X]

And im not disagreeing with your point at all. I was simply asking who they were...to which nobody has responded. And yes it would be interesting to see verification scores of the mets...see who always wishcasts and who is realistic.

DT= Dave Tolleris www.wxrisk.com. He seems to be respectable enough, but you'll have to judge for yourself based on the info you can garner about his thoughts. There are some other mets in the same camp. But he was the loudest most consistent voice. And to be fair, there are others in the more modest camp of 3-6/4-8 inches who are very talented mets also. I havent read anyone who said we totally miss..... so its fair to say that most on this board think the GFS will continue trending at least somewhat.

[eduggs]

By having an extensive knowledge of atmospheric interactions and the behaviors of pressure systems at different levels of the atmosphere.

Does that answer your question?

It's an answer, but an unsatisfactory one. That's exactly what model code represents. And in a much more extensive way than all but a few meteorologists in the world understand. How can an individual mets knowledge compare to the collective knowledge of the field? And then fast forward the flow to days 4, 5, and 6... To eyeball errors at this point is utterly absurd.

[psv88]

I also want to say that i think most people on this board are overall intelligent people...it takes a certain caliber of individual to be curious enough about the weather to watch models and follow a board. We may not all be professional mets, but that does not mean we don't have basic or even advanced reasoning skills.

In reality, having people on a board from professions other than mets is a good thing, it can make the discussion less technical. When mets have to explain concepts to people, it actually increases their own understanding of the models, just like teaching someone helps you improve your own understanding of the subject.

[A-L-E-X]

Remind me what your profession is exactly? I only ask this because you speak to having an intricate understanding of why these models should always be right, and that humans cannot spot any error in a solution.

Would that the models were always right-- we would be living inside a Matrix-like computer sim. Obviously, that isnt the case....

And its more than just about errors, there is an inherent variability that makes it an "art" as much as a science and this is where human insight and intuition come in, as well as experience from past outcomes.

[eduggs]

This an encouraging run, similar set up and phase like the Euro, but not enough to pull it closer, but a Stark contrast to previous runs, way different.

I agree. Run that setup 10 times from 48hour out and I bet you get at least 3 huge hits up and down the coast.

[ParanormalWx]

What the hell happened in this thread? Can we stop the bickering and get back to the models and what they are showing.

[Dsnowx53]

The 12z GEFS means are much more amplified than the 0z GEFS means through 84 hours.

[goombatommy]

DT= Dave Tolleris www.wxrisk.com. He seems to be respectable enough, but you'll have to judge for yourself based on the info you can garner about his thoughts. There are some other mets in the same camp. But he was the loudest most consistent voice. And to be fair, there are others in the more modest camp of 3-6/4-8 inches who are very talented mets also. I havent read anyone who said we totally miss..... so its fair to say that most on this board think the GFS will continue trending at least somewhat.

DT is a good met. Given he can be a tiny bit harsh with his wording, but his logic is usually sound and he definitely knows his stuff.

[NortheastPAWx]

Compared to the GFS it looks the GGEM is further north with the H5 low and slightly further south with the northern stream. (hr60)

[am19psu]

Then why, eduggs, does HPC continually have better error verification than any model? Why does NHC have better track forecast error verification than any model? Why do private forecast companies exist? Think carefully about your answer.

[am19psu]

I also want to say that i think most people on this board are overall intelligent people...it takes a certain caliber of individual to be curious enough about the weather to watch models and follow a board. We may not all be professional mets, but that does not mean we don't have basic or even advanced reasoning skills.

In reality, having people on a board from professions other than mets is a good thing, it can make the discussion less technical. When mets have to explain concepts to people, it actually increases their own understanding of the models, just like teaching someone helps you improve your own understanding of the subject.

For the record, I've got no problem with this. People suggesting that I am doing my job poorly or that I have no scientific integrity after 8 years of meteorological and statistical coursework plus three years as a private sector forecaster really irritate me.

[NJHurricane]

Why? i think its a very fair, relevant question.

People cite Wes, JB, DT, PB and J, as much as the GFS and Euro as if they are their own models...am i wrong?

Does anyone else get annoyed with all the references to these guys? Or at least JB..come on

this is not a thread for discussing individual mets verification scores...feel free to start a separate thread, we are trying to keep this one under 1,000 posts in 12 hours thanks

[A-L-E-X]

Would that the models were always right-- we would be living inside a Matrix-like computer sim. Obviously, that isnt the case....

And its more than just about errors, there is an inherent variability that makes it an "art" as much as a science and this is where human insight and intuition come in, as well as experience from past outcomes.

The atmosphere (and all of reality) is a lot more complex than any supercomputer could ever replicate.

[LVblizzard]

GEFS mean is similar to the GFS at 84, if not slightly east. But that's to be expected with the ensembles.

[ag3]

Much more amped at hour 96 on 12z GFS ensemble mean.

[Dsnowx53]

The 500mb GEFS plots that i get on SV are also much more amplified compared to the previous GEFS runs.

[LVblizzard]

Closed 500 low near the benchmark at 120 on the GEFS.

[ag3]

Hour 108 of the GEFS mean is pretty impressive. Way NW and more amped.

[Blizzardo]

Closed 500 low near the benchmark at 120 on the GEFS.

LV, have to ask.. what storm is the 966 avatar ?

[eduggs]

No, of course the model physics aren't wrong. They solve the Navier-Stokes equations using a set of initial conditions.

However, there are certain parametrizations used to explicitly solve the equations. For example, for clouds, models only produce a certain opacity over a grid point, versus explicitly resolving clouds. Same with updrafts and downdrafts. This can lead unrealistic solutions, particularly when dealing with mesoscale type events that are less than 25 km in extent (or 15 km in the Euro).

Now, will you please drop your condescending line of questioning each time a meteorologist questions a particular model solution? There are things that we look for that clue us in on what the probable final solution will look like, based mainly on experience in the field.

I know models use approximate solutions to certain difeq's. And the errors that can result grow exponentially. Ensemble forecasting helps in this regard, but it's still a challenge. But this is mostly irrelevant to largescale mid-latitude synoptics. At this level it's nearly impossible to tell with any certainty what has propagated from error... and thus whether a solution is viable in its own right.

I use experience to predict the outcome as well. But its mostly patterns, tendencies, and statistics.

[A-L-E-X]

The 500mb GEFS plots that i get on SV are also much more amplified compared to the previous GEFS runs.

Closed 500 low near the benchmark at 120 on the GEFS.

From CoastalWx:

12z ensembles def came west. 1008mb low about 80-100 miles ese of ILM at hr 96. Low moves to about 200+ miles or so east of ORF as a 997 low. at hr 120 it's about 130miles east of the BM as a sub 990 low. Not bad for ensembles.

[ag3]

GGEM is east. Misses our area:

[A-L-E-X]

Def room for a more westward jog on the GFS based on the ensembles.

[eduggs]

Remind me what your profession is exactly? I only ask this because you speak to having an intricate understanding of why these models should always be right, and that humans cannot spot any error in a solution.

Sorry if I gave this impression. Models definitely NOT always right. But humans are unable to make this determination in mid-range without relying on model consistency, trends, and weighted averages.

[MJO812]

GGEM is east. Barely skims our area:

I thought it was going to be more west because the trough looks slightly negative.