NYC/PHL December 24-27 Potential - Part 2

[GD0815]

You have to look at the upper air charts and their trend before you look at the sfc. The GFS definitely trended in the Euro's direction tonight and so did the Nam. And the UKMET always has a few weird runs thrown into any threat for good measure. It's still way too early to make any definitive conclusion one way or the other.

dont waste your time trying to reason with him, it will never stop.

[ag3]

Don't look just at QPF, you have to look at SLP placement.

Of course. Just responding to a post that GGEM was never a hit.

[Midlo Snow Maker]

and i'm using my last one now.

very last thanks for the reports guys

[Isotherm]

and i'm using my last one now. being post limited really doesn't bother me...i mean i've been right the last 2 storms. it's pointless to get excited about a storm that clearly involves phasing in a la nina...most of the time, it's not going to happen. i'm not disappointed at all as i've expected this, but i wish some others would recognize this and not wishcast.

But without any scientific reasoning, the fact that you've been right doesn't really have much value. Anyone can just say "it won't happen" for every single threat.

[Rtd208]

Are you serious? Your biggest problem is your attitude Colin, personally I could do without you posting on this board, I have overall been quiet and never bothered you, but out of all honesty I think your posts are wasteful and take up space more then anything else.

and i'm using my last one now. being post limited really doesn't bother me...i mean i've been right the last 2 storms. it's pointless to get excited about a storm that clearly involves phasing in a la nina...most of the time, it's not going to happen. i'm not disappointed at all as i've expected this, but i wish some others would recognize this and not wishcast.

[eduggs]

I lost confidence in this threat as the initial surface low position went from Cinncinatti, Oh to Panama City, FL on the charts. I want to see decent overrunning precipitation in the lower ohio valley in advance of a coastal low. Without southerly flow on the frontside, the westward extent of QPF is limited except in the best circumstances.

[Zir0b]

while its very different at the surface; at the 500 mb level the GGEM isn't bad at all...and why does it have 2 H5 lows next to each other at 120 hrs?

from dt

the Ggem looks Relatively good

I am NOT worried about the Precip shield at all. 90% of the time

a southern Gulf low coming up the coast passing 50- 100 miles east of HATTERAS Is going to have significant precip with it

[Isotherm]

I think we all need a laugh right now.

http://www.youtube.com/watch?v=BnWDL5bMTgs

[eduggs]

But without any scientific reasoning, the fact that you've been right doesn't really have much value. Anyone can just say "it won't happen" for every single threat.

Scientific reasoning plays almost no roll in forecasting these days. We pretty much wait for model output and copy that into a forecast. If we are smart we use statistical methods to assess probabilities, but that's not how we do things around here.

[eduggs]

from dt

DT is right about moisture conveyance overall, but wrong about westward extent of QPF.

[ARyan]

Scientific reasoning plays almost no roll in forecasting these days. We pretty much wait for model output and copy that into a forecast. If we are smart we use statistical methods to assess probabilities, but that's not how we do things around here.

I hope this is sarcasm...

[yhbrooklyn]

Guys, stop the bickering.

[eduggs]

I hope this is sarcasm...

It should be clear that it's not.

[ARyan]

It should be clear that it's not.

Well...I don't know what you're trying to say but it isn't true about forecasting.

[Isotherm]

It should be clear that it's not.

Well the best forecasters are ones who use scientific reasoning to interpret model output, and judge whether or not they hold any validity. Pattern recognition, analogs, and pattern persistence play a large role in forecasting, which sometimes warrant a prediction against model output. Most of the time a forecast is made based upon the model consensus with an added human touch of his/her scientific read of the pattern (if the model solution makes sense with the pattern in place, or vice versa).

[jm1220]

I lost confidence in this threat as the initial surface low position went from Cinncinatti, Oh to Panama City, FL on the charts. I want to see decent overrunning precipitation in the lower ohio valley in advance of a coastal low. Without southerly flow on the frontside, the westward extent of QPF is limited except in the best circumstances.

I don't believe the Feb 2006 storm that absolutely buried you guys and left me in State College PA with nothing more than some snow showers had much initial overrunning precip in the Ohio Valley or anywhere. And that was in a worse setup than this, just complete coincedence due to good timing. Heavy snow got as far west as the Lehigh Valley in that storm, too.

I think it comes down to the phasing between the streams and where it occurs. If it occurs over the MS valley and the 500mb closed low develops soon after, the low should get captured in time to slam us. If it doesn't, it heads OTS. The discussions about confluence and the rest are likely irrelevant with a massive phase like what the 12z euro showed.

[Dsnowx53]

Just a few thoughts:

The northern stream is in horrible data assimilation areas, which could explain why its degree of diving down is significantly waffling, which obviously affects the timing of a phase. However, I still don't understand why it's not diving down faster, considering that the ridging out west has not seemed to have lost any punch--and as we know, the riding out west is what helps to force the northern stream energy to dive down, which is something we did not have for the last event. There was no forcing mechanism for the entire lobe of northern stream energy to dive southward, and instead we had to rely on the PV being brought well to the west, with only a small chunk of energy trying to capture the southern stream energy last minute...that is not a good scenario. This time, because of the riding, I think it makes sense that it can force a lot more northern stream energy to dive down southward--that combined with a much stronger southern stream vort than last go-around would obviously lead to a more favorable outcome.

However, the northern stream does seem to be trending a bit slower on some of the recent runs, but considering the ridging has not weakened, I'm not sure why the models are doing that. If that is a trend, though, then that obviously would warrant a lot of concern.

In addition, even if the Euro crushes us tonight, that's not necessarily a reason to celebrate. The Euro is obviously by far the slowest model, and while it's clear the the GFS is too fast, I'm not sure that the Euro has the exact timing correct, either...it still does have that bias in the south west. And as we know, the slowest scenario is what helps to give the northern stream time to dive down in time--but if the Euro is just slightly too slow, then the solution might be a no-go.

[SnoSki14]

Well there's still a few days left to see what will happen but I'm not going to get all uptight and upset if we see nothing. To be honest, I wasn't expecting much this season so any snowfall will be a plus.

[Isotherm]

I think it comes down to the phasing between the streams and where it occurs.

Agree, the phase timing is crucial. Ideally we want a slow southern wave and a northern s/w that comes into the picture rather quickly like the 12z euro and phases the storm near AL/MS (GFS phasing occurs closer to the SE US coast).

[Collegestudent11]

One thing to note here is that the model runs are not 300 miles out to sea and weaker. Even the most Out to sea solutions have a powerful storm just off the coast.

[eduggs]

Well...I don't know what you're trying to say but it isn't true about forecasting.

It is. Most weather forecasters are not scientists, that includes degreed Mets. And even if you are a brilliant atmospheric scientist, that doesn't mean you can predict atmospheric flow and dynamics 5 days from now. Humans cannot outforecast supercomputers. Weighted average of multi-model ensembles always beats HPC and everybody else. What passes for "scientific reasoning" on this forum is much closer to guessing or wishcasting than predictive analysis. The only truly useful analysis in the mid-range is statistical (short range is a different animal).

[NittanyWx]

Scientific reasoning plays almost no roll in forecasting these days. We pretty much wait for model output and copy that into a forecast. If we are smart we use statistical methods to assess probabilities, but that's not how we do things around here.

As a meteorologist, I think I have the right to tell you to shut your trap...you have no idea what you're talking about.

[jm1220]

Agree, the phase timing is crucial. Ideally we want a slow southern wave and a northern s/w that comes into the picture rather quickly like the 12z euro and phases the storm near AL/MS (GFS phasing occurs closer to the SE US coast).

Agreed, it needs to phase near the MS valley so that the capture can occur in time. Phasing east of there will likely be too late and result in a capture that likely only benefits Newfoundland.

[eduggs]

Well the best forecasters are ones who use scientific reasoning to interpret model output, and judge whether or not they hold any validity. Pattern recognition, analogs, and pattern persistence play a large role in forecasting, which sometimes warrant a prediction against model output. Most of the time a forecast is made based upon the model consensus with an added human touch of his/her scientific read of the pattern (if the model solution makes sense with the pattern in place, or vice versa).

Pattern recognition is largely a myth. A close analog to a blizzard might ultimately miss. A human might recognize the similarity and predict a blizzard. But a supercomputer is much better suited to predict the likelihood of the storm. The human guesses and the computer performs millions of calculations. How would "science" even play a roll in this scenario? The science is built into the model physics.

[NittanyWx]

It is. Most weather forecasters are not scientists, that includes degreed Mets. And even if you are a brilliant atmospheric scientist, that doesn't mean you can predict atmospheric flow and dynamics 5 days from now. Humans cannot outforecast supercomputers. Weighted average of multi-model ensembles always beats HPC and everybody else. What passes for "scientific reasoning" on this forum is much closer to guessing or wishcasting than predictive analysis. The only truly useful analysis in the mid-range is statistical (short range is a different animal).

Wrong again champ.

[GD0815]

It is. Most weather forecasters are not scientists, that includes degreed Mets. And even if you are a brilliant atmospheric scientist, that doesn't mean you can predict atmospheric flow and dynamics 5 days from now. Humans cannot outforecast supercomputers. Weighted average of multi-model ensembles always beats HPC and everybody else. What passes for "scientific reasoning" on this forum is much closer to guessing or wishcasting than predictive analysis. The only truly useful analysis in the mid-range is statistical (short range is a different animal).

I think if you are going to make such assertions, you should at least give us your background, and what basis you have for making those conclusions? not saying you are wrong (although i have my doubts), but I would like to know where you are coming from with all of this.

[eduggs]

I don't believe the Feb 2006 storm that absolutely buried you guys and left me in State College PA with nothing more than some snow showers had much initial overrunning precip in the Ohio Valley or anywhere. And that was in a worse setup than this, just complete coincedence due to good timing. Heavy snow got as far west as the Lehigh Valley in that storm, too.

For sure there are several great counterexamples. But they are usually hit or miss nail biters.

[ARyan]

It is. Most weather forecasters are not scientists, that includes degreed Mets. And even if you are a brilliant atmospheric scientist, that doesn't mean you can predict atmospheric flow and dynamics 5 days from now. Humans cannot outforecast supercomputers. Weighted average of multi-model ensembles always beats HPC and everybody else. What passes for "scientific reasoning" on this forum is much closer to guessing or wishcasting than predictive analysis. The only truly useful analysis in the mid-range is statistical (short range is a different animal).

Do you know anything about Rossby waves? I'll leave it at that.

[NittanyWx]

It is. Most weather forecasters are not scientists, that includes degreed Mets. And even if you are a brilliant atmospheric scientist, that doesn't mean you can predict atmospheric flow and dynamics 5 days from now. Humans cannot outforecast supercomputers. Weighted average of multi-model ensembles always beats HPC and everybody else. What passes for "scientific reasoning" on this forum is much closer to guessing or wishcasting than predictive analysis. The only truly useful analysis in the mid-range is statistical (short range is a different animal).

Oh, really? I'm not a scientist now? The absurd amounts of thermodynamics, dynamic meteo, atmospheric chemistry, atmospheric microphysics, chemistry, mechanics, electronics and magnetism classes I took are not good enough for you?

Well sh*t, I guess I'll just quit my job now.

[Isotherm]

It is. Most weather forecasters are not scientists, that includes degreed Mets. And even if you are a brilliant atmospheric scientist, that doesn't mean you can predict atmospheric flow and dynamics 5 days from now. Humans cannot outforecast supercomputers. Weighted average of multi-model ensembles always beats HPC and everybody else. What passes for "scientific reasoning" on this forum is much closer to guessing or wishcasting than predictive analysis. The only truly useful analysis in the mid-range is statistical (short range is a different animal).

Strongly disagree, how do you explain long range forecasting? There's definitely been a ton of success via methods of analoging, pattern recognition, and understanding the physics of what pattern(s) result w/ a specific ENSO, NAO,PDO,AMO, etc regime in place. Honestly your comments basically shoot down the meteorology field as a whole. Why have them if we've got these supposed excellent computers. Let me tell you something - computers are fast, and stupid. Humans are slow, but smart. It's a necessary couplet in weather forecasting. Models are only as good as we make them.