Dissecting The Bust

[PB GFI]

I understand, but if you are going to call for a near-certain historic storm, you better have near-unanimous model agreement.

I am not sure if we ever really get that . However Sunday night  I had all the model data QPF at  KNYC and was going off 12 to 1and off the top of my head this is what I wrote

 

GEFS 1.25

NAM 2

EURO 2.5

RGEM 1

UKIE .8

SREF 1.75

 

If 24- 36 does not fall the default forecast is probably 15 to 20 and 20 is a top 5 KNYC storm .

The 12 inch line will start very close to Brooklyn and then it ramps up east .

So I think 24 to 36 was def the ceiling all 3 forecast offices did agree as their maps were perfectly aligned

 

But if 15 to 20 fell KNYC and the N and NW side did not blow up then it would have been inside the envelope of what was a good forecast

GFS .8

[pazzo83]

I am not sure if we ever really get that . However Sunday night  I had all the model data QPF at  KNYC and was going off 12 to 1and off the top of my head this is what I wrote

 

GEFS 1.25

NAM 2

EURO 2.5

RGEM 1

UKIE .8

SREF 1.75

 

If 24- 36 does not fall the default forecast is probably 15 to 20 and 20 is a top 5 KNYC storm .

The 12 inch line will start very close to Brooklyn and then it ramps up east .

So I think 24 to 36 was def the ceiling all 3 forecast offices did agree as their maps were perfectly aligned

 

But if 15 to 20 fell KNYC and the N and NW side did not blow up then it would have been inside the envelope of what was a good forecast

GFS .8

 

NYC doesn't shut down for a 15-20" forecast, that's the difference.

[eyewall]

All anyone had to do was look at the low's location last night and the radar trends and you could see it was time to cut back to the west. They just held on too long.

[Amped]

Replace the nws staff with an automated, gfs, ukmet, cmc, euro average. Eliminates

weenie biases. Also to eliminate stupidity, throw the Nam out of a skyscraper.

[PB GFI]

NYC doesn't shut down for a 15-20" forecast, that's the difference.

Agreed but that was the low end of the guidance 36 hours ago

[pazzo83]

Replace the nws staff with an automated, gfs, ukmet, cmc, euro average. Eliminates

weenie biases. Also to eliminate stupidity, throw the Nam out of a skyscraper.

 

You could use the NAM as like an anti-model - as in, this is most certainly NOT what will happen.

[LocoAko]

Let's call this what it is: reckless forecasting and apparently a wholesale inability to communicate probability and risk (in terms of the forecast) to the general public.  This probably cost the NY metro area upwards of $1bn in terms of lost revenues/wages plus the costs of shutting down a metro area of 20+ million.

 

While I don't disagree, the media had a lot to do with it. I saw plenty of sharing on Twitter of the 90th percentile maps and "worst case scenarios" which were entertaining because of how extreme they were, but totally irresponsible. I can't find them now, but the 5% and 10% maps were quite low and close to what happened. Of course if asked for a single number no one would give them the 10th percentile one, but I saw that shared maybe once while the "worst case scenario" maps were being given out like free candy. 

[FreeRain]

U know what's the most dangerous

Aspect of this whole thing imo? is we get another storm threat of a

Dangerous magnitude sometime soon and people don't take it seriously because of what happened with this storm. This could be a huge problem....

Sent from my iPhone

 

Actually it's a good lesson. Sometimes the weather forecast is wrong. And often times, the media and the politicians are wrong.

[pazzo83]

While I don't disagree, the media had a lot to do with it. I saw plenty of sharing on Twitter of the 90th percentile maps and "worst case scenarios" which were entertaining because of how extreme they were, but totally irresponsible. I can't find them now, but the 5% and 10% maps were quite low and close to what happened. Of course if asked for a single number no one would give them the 10th percentile one, but I saw that shared maybe once while the "worst case scenario" maps were being given out like free candy. 

 

The NWS was issuing statements with 20-30" and maps (not models, forecast maps implying the most likely outcome) of 24-36" less than 24hrs prior to the event commencing.  Obviously I understand that predicting storms of this nature is difficult and inexact (I remember 2001 amongst plenty of other busts).  However, there is a certain amount of responsibility that comes with claiming that something is going to be historic.

 

The only thing this does now is damage the credibility of forecasters.  We can see what that does when you take situations like Irene and then Sandy: the outcome is typically catastrophic.

[NortheastPAWx]

An outsider's take. Feel free to ignore me.

 

The EURO has developed such high regard in forecasting circles and rightfully so, thus when it started dumping 2-3' over places in the 95 corridor, the notion was there to bite and everyone, including Upton and Mt. Holly, took the bait. There comes a time when you have to consider the other models and, imo, that time was 0z Sunday when the EURO was suddenly left an island with those historic amounts (the daytime NAMs yesterday beside that). At that time, it would've been best for Upton and broadcast mets to take a step back with forecasts to a wider range, a compromise - I personally had 14-18" for the city. Instead, folks bought the hype and kept forecasts as they were. The red flags were there. While the eventual final totals for the city proper of ~1' are significant, it obviously wouldn't be enough to shut down the city under normal circumstances.

 

This bust was the product of a lot of hype and over-hugging one model. A conservative voice should have entered the room before things got out of hand. Signals were missed while the hype machine generated by national media went on undisturbed.

 

Just my $0.02. As no meteorologist, I don't have much to comment on in regard to the meteo side of what happened.

 

TL;DR I agree with Pazzo.

[LocoAko]

The NWS was issuing statements with 20-30" and maps (not models, forecast maps implying the most likely outcome) of 24-36" less than 24hrs prior to the event commencing.  Obviously I understand that predicting storms of this nature is difficult and inexact (I remember 2001 amongst plenty of other busts).  However, there is a certain amount of responsibility that comes with claiming that something is going to be historic.

 

Oh, I agree. I just meant that there were probably more opportunities to convey uncertainty than were seized upon in certain circles like the media. 

[TheManWithNoFace]

Loko I totally agree. I don't even know what the purpose of that product is. The worst case should be the top end of the range. Do we need to have a conversation at a met community about what a reasonable range is for a storm like this? I also am not a fan at all of sharing model runs with the public. The ability to interperet guidance and diagnose a situation is what sets meteorologists apart. What is the purpose of sharing the raw materials of forecasting- unless the outlet in question is merely a parrot for computers and wants the viewer to see the bustability in advance for sympathy?

[pazzo83]

Oh, I agree. I just meant that there were probably more opportunities to convey uncertainty than were seized upon in certain circles like the media. 

 

This is kind of the world we live in now though.  Everyone online thinks they are an expert in everything just because they have access to this wealth of information.  Some care needs to be taken in terms of maybe providing context on model snowfall forecast maps etc.  Obviously news media putting these outlandish maps up, like they have done in the past, is just incomprehensible.

[phil882]

I have followed the RGEM ensembles and never did any member produce any excessive QPF for NYC for this storm.

 

Here is the 12Z Monday RGEM ensemble for KNYC:

 

rgem_ensemble_meteogram.jpg

 

Link?

[PB GFI]

The NWS was issuing statements with 20-30" and maps (not models, forecast maps implying the most likely outcome) of 24-36" less than 24hrs prior to the event commencing.  Obviously I understand that predicting storms of this nature is difficult and inexact (I remember 2001 amongst plenty of other busts).  However, there is a certain amount of responsibility that comes with claiming that something is going to be historic.

 

The only thing this does now is damage the credibility of forecasters.  We can see what that does when you take situations like Irene and then Sandy: the outcome is typically catastrophic.

But Chris it comes down to , you have to make a forecast for 20 million people , so do you go with 6 to 30 ?

Because that's what falling  from EWR to the east end .

The 75 mile east jump happened at 6pm . So what do you call for at NOON ?

 

The problem was not the worst case maps the Euro printed out 2.5 at 12 to 1 back to the Poconos

The NAM printed out 2.5 with 3 inch maxes .

The 6z GEFS yesterday had 1.50 at KNYC and 1.75 in western Nassau

The SREFs were 1.75 at KNYC

 

That's 2 to 3 feet at 9 AM . How do ignore that guidance

[PB GFI]

This is all MONDAY MORNING QB SHT . Dumb

[Syrmax]

Let's call this what it is: reckless forecasting and apparently a wholesale inability to communicate probability and risk (in terms of the forecast) to the general public. This probably cost the NY metro area upwards of $1bn in terms of lost revenues/wages plus the costs of shutting down a metro area of 20+ million.

You are right...when NWS BOX and Upton started using histrionic language..."crippling" "historic" " life threatening"...they were essentially Going Big or Going Home, as they say. The moronic (yes, that's actually an accurate description) mass media normally over hype any weather event and this was just throwing red meat to the wolves. A truckload of red meat. How many of these have there been down along the east coast in the past 5 years? At least 3 that I can think of. Maybe more if you throw in the mid Atlantic. How are any of these truly "historic?" People can argue about which model performed better or worse, and why, and that's interesting but the other aspect here is how NWS (and private entities), presented information to the public, knowing how that gets used by an increasingly hype-driven and over-reactionary public/govt management. The way that NWS and the media were hyper ventilating over this Storm, the casual public observer was almost led to believe everyone east of Cleveland was going to die in this storm. I'm obviously exaggerating a bit, but maybe not by much.

[pazzo83]

But Chris it comes down to , you have to make a forecast for 20 million people , so do you go with 6 to 30 ?

Because that's what falling  from EWR to the east end .

The 75 mile east jump happened at 6pm . So what do you call for at NOON ?

 

The problem was not the worst case maps the Euro printed out 2.5 at 12 to 1 back to the Poconos

The NAM printed out 2.5 with 3 inch maxes .

The 6z GEFS yesterday has 1.50 at KNYC and 1.75 in western Nassau

The SREFs were 1.75 at KNYC

 

That's 2 to 3 feet at 9 AM . How do ignore that guidance

 

Well, it should be clear you can toss the NAM right out.  So considering the other models, what's wrong with a 12-24" forecast? The probability of a 30" snowfall regardless of model output, given that it would be > 10% higher than the largest snowfall ever for NYC, is quite low.  12-24" seems like a pretty reasonable approach given the models at the time, and it likely would not have resulted in the unprecedented shutdown of the city.

 

AND, you wouldn't have the now-severely damaged credibility of the local NWS offices.

[wxman]

The truth is this was a bust for a tiny geographic area, the problem is it happens to be one of the most populated places on earth!  As for our CWA, what was one to do, the guidance was 3" to 30"+? Riding the euro made sense, and I bet they'd do it again.  Another point, when people brought up March 2001 before this event, some wrote about how much better the computer models are now versus then.  I do not agree.  The improvements have been marginal over that stretch.  I suppose if we got it perfect, there wouldn't be much fun in the chase.  That's the silver lining here for me.

[TheManWithNoFace]

I'd like to hear some thoughts also from someone who has complete access to the EC, where and how it was wrong, and why the 75-100 mile shift E occurred

[PB GFI]

Well, it should be clear you can toss the NAM right out.  So considering the other models, what's wrong with a 12-24" forecast? The probability of a 30" snowfall regardless of model output, given that it would be > 10% higher than the largest snowfall ever for NYC, is quite low.  12-24" seems like a pretty reasonable approach given the models at the time, and it likely would not have resulted in the unprecedented shutdown of the city.

 

AND, you wouldn't have the now-severely damaged credibility of the local NWS offices.

 

Please don`t give me  F%^y probability .  That is such a  HORRIBLE argument  .They have to go by guidance .

The guidance was 2 feet . Actually it was 25- 30 . It was a cold storm ( ALL SNOW )  with WINDS predicted to gust to 60 back to the city  . There was convection and the risk that banding would increase totals .

And no the Guidance suggested you start at 20 -

This is about a 75 mile jog east at 6pm that no one could control and the poor outcome for snow lovers has them twisted .

Go out to Nassau county where the 20 inch line starts and see how close you were .

[LocoAko]

As a sidenote, from an outsider's (and non-forecaster's) perspective, my Facebook feed of friends back in NJ seems more ticked off with the hype of the storm than the actual lesser snowfall amounts. I think they got ~7-8", which is borderline enough to cause most people to stay home regardless. It is stuff like CNN's "Blizzardmobile" and all of the media dramatics that are being mocked more than the snowfall forecast itself. At least among my friends many know how hard forecasting is and that models aren't perfect. But while I am curious about what went "wrong" from a physics and communications point of view, I think more than blame there's a lot of eye rolling going on (again, from my own limited sphere on social media). 

[pazzo83]

Please don`t give me  F%^y probability .  That is such a  HORRIBLE argument  .They have to go by guidance .

The guidance was 2 feet . Actually it was 25- 30 . It was a cold storm ( ALL SNOW )  with WINDS predicted to gust to 60 back to the city  . There was convection and the risk that banding would increase totals .

And no the Guidance suggested you start at 20 -

This is about a 75 mile jog east at 6pm that no one could control and the poor outcome for snow lovers has them twisted .

Go out to Nassau county where the 20 inch line starts and see how close you were .

 

I would, except I can't because they shut the entire city down (although I guess we have now re-opened?).  I think you have to incorporate a bit of historic reasoning into these forecasts as well.  I think, generally, there was a pretty wide spread in the guidance.  I don't see any other responsible choice than to forecast a wide spread with a low degree of confidence.

[Stormlover74]

This is all MONDAY MORNING QB SHT . Dumb

Isn't that the point of this thread?

[Stormlover74]

I would, except I can't because they shut the entire city down (although I guess we have now re-opened?). I think you have to incorporate a bit of historic reasoning into these forecasts as well. I think, generally, there was a pretty wide spread in the guidance. I don't see any other responsible choice than to forecast a wide spread with a low degree of confidence.

And as you mentioned with Irene and sandy when the next big blizzard is forecast are people going to take it seriously?

[pazzo83]

As a sidenote, from an outsider's (and non-forecaster's) perspective, my Facebook feed of friends back in NJ seems more ticked off with the hype of the storm than the actual lesser snowfall amounts. I think they got ~7-8", which is borderline enough to cause most people to stay home regardless. It is stuff like CNN's "Blizzardmobile" and all of the media dramatics that are being mocked more than the snowfall forecast itself. At least among my friends many know how hard forecasting is and that models aren't perfect. But while I am curious about what went "wrong" from a physics and communications point of view, I think more than blame there's a lot of eye rolling going on (again, from my own limited sphere on social media). 

 

The eye rolling is widespread here.  It's unfortunate because the NWS here is generally really solid.  The next few times there is a forecast for a major storm here, I think a lot of people are going to be hesitant to heed the warnings.

[nycemt123]

U know what's the most dangerous

Aspect of this whole thing imo? is we get another storm threat of a

Dangerous magnitude sometime soon and people don't take it seriously because of what happened with this storm. This could be a huge problem....

Sent from my iPhone

Yep, look how people took Sandy since Irene just wasn't anything special for us...

[Snowshack]

U know what's the most dangerous

Aspect of this whole thing imo? is we get another storm threat of a

Dangerous magnitude sometime soon and people don't take it seriously because of what happened with this storm. This could be a huge problem....

Sent from my iPhone

That gets said after every bust - I think there is some truth to it during tropical season.  But people love to whip themselves into a frenzy over forecasted snow, so if there is another big one forecasted this year im sure there will be another run on shovels/bread/milk/gas.  

[PB GFI]

Isn't that the point of this thread?

Yes but then it evolved into 24 to 36 was unreasonable .

 

The bust was simple , it jumped east at 6pm to the best convection . Happens . Was not really seen that well and most models

printed out 1.5- 1.75 and 2 on the western side and that's where we were at 9am yesterday

 

Happens , we just missed this one .

[kevbo81]

I followed the models from sat on like everyone else. At no time did you have

a unanimous consensus for 2-3 foot totals. if EVERY model was saying

over 1.5 liquid, they would have had more justification for the biblical snowfall predictions on such a grand scale. 

When you get over a foot, or have the potential to, you have to let things play out during 

the storm and adjust accordingly.

 

My idea would be to have a self-imposed "cap" if i was a trained met. 

I would recommend that the NWS or anyone refrain

from putting out forecasts of over 18" unless it's a now-cast situation and you already

have the totals. It's just the way it is; forecasting the highest rates in snow bands is difficult

if not impossible to do with the current state of the models as they are.