Jump to content

*** PLEASE REGISTER AND JOIN OUR DISCUSSION!!! ***

THE STAFF WANT YOU TO JOIN US AT PHILLYWX!!!

Register as a member today, and become part of the Delaware Valley weather community!

Our pro and seasoned amateur meteorologists, and weather enthusiasts from around the PA and NJ area together form a great group discussion, and we're asking folks that read our site today to register as members and post along with us!

Don't be intimidated if you're not an expert, ask questions if you're curious or want to build your knowledge!

Whether it's adding to our local profiles by reporting observations (and maybe becoming a SkyWarn Spotter!), or contributing more on the model interpretation side, we'd like you to join us in a constructive and insightful dialogue around all things Philly Weather!


WeatherOrNot

Irresponsible "computer generated" Now Accu 90 Day Forecast? (2/9)

Recommended Posts

Despite that few of us have a very high opinion of all things Accuweather, I still for some reason have a habit of checking them out.  There's worse.  Sure, their "45 day" is a horrible horrible slap in the face - but their near range, is sometimes ok, and sometimes sees things that haven't really trended in discussion here yet, I dismiss it because you guys haven't talked about it, and sure enough within a day or two you start, and they really did see something - so it's worth looking at in the arsenal.  And their real mets with written forecasts are actually generally quite good - it's the computer generated fluff that's awful. (Especially determining lows.)

 

But today I noticed something that just seems irresponsible, even for what seems to be an entirely computer-generated-from-model-output "45 day" forecast.  The front page for the area discusses the 2/8-2/9 event, adds to it "heavy ice conditions."   The little info graphic and text for Monday just says "some snow" - but on the Monday night segment it says "A major ice storm, with a thick coating of ice" and for Ice accumulation lists 0.9 inches. 

 

At that unprecedented (for the region) accumulation, almost 5x that of last year's ice storm, few buildings and trees would remain standing and power would be out for months.  Business would be gone.  They're effectively calling for a Katrina-level event for the area 5 days out from what is sure to be very gradient driven.  Their own written forecasts don't even discuss the event yet.

 

Obviously you guys are talking nasty ice possibilities as well, so there's the chance they could even be right by sheer dumb luck. But you guys also know that nothing is certain and that those are just possibilities, far from a forecast for something so close this far out.

 

Is this kind of posting sheer irresponsibility on their part to even allow the system to post this sort of thing, or is it a situation of "well that's their reasonable forecast"?
 

Share this post


Link to post
Share on other sites

Despite that few of us have a very high opinion of all things Accuweather, I still for some reason have a habit of checking them out.  There's worse.  Sure, their "45 day" is a horrible horrible slap in the face - but their near range, is sometimes ok, and sometimes sees things that haven't really trended in discussion here yet, I dismiss it because you guys haven't talked about it, and sure enough within a day or two you start, and they really did see something - so it's worth looking at in the arsenal.  And their real mets with written forecasts are actually generally quite good - it's the computer generated fluff that's awful. (Especially determining lows.)

 

But today I noticed something that just seems irresponsible, even for what seems to be an entirely computer-generated-from-model-output "45 day" forecast.  The front page for the area discusses the 2/8-2/9 event, adds to it "heavy ice conditions."   The little info graphic and text for Monday just says "some snow" - but on the Monday night segment it says "A major ice storm, with a thick coating of ice" and for Ice accumulation lists 0.9 inches

 

At that unprecedented (for the region) accumulation, almost 5x that of last year's ice storm, few buildings and trees would remain standing and power would be out for months.  Business would be gone.  They're effectively calling for a Katrina-level event for the area 5 days out from what is sure to be very gradient driven.  Their own written forecasts don't even discuss the event yet.

 

Obviously you guys are talking nasty ice possibilities as well, so there's the chance they could even be right by sheer dumb luck. But you guys also know that nothing is certain and that those are just possibilities, far from a forecast for something so close this far out.

 

Is this kind of posting sheer irresponsibility on their part to even allow the system to post this sort of thing, or is it a situation of "well that's their reasonable forecast"?

It wouldn't be that bad. Ice accumulations that heavy do bring down a lot of trees and power lines though. Many of us would be out of power for a week. A lot of that forecast is automated and from the look of it they just populated the 00z GFS as the forecast. Very easy to see the downfall of that strategy. Anything beyond day 14 in a normal pattern is really a shot in the dark and just a guess.

Share this post


Link to post
Share on other sites

It wouldn't be that bad. Ice accumulations that heavy do bring down a lot of trees and power lines though. Many of us would be out of power for a week. A lot of that forecast is automated and from the look of it they just populated the 00z GFS as the forecast. Very easy to see the downfall of that strategy. Anything beyond day 14 in a normal pattern is really a shot in the dark and just a guess.

 

0.9? Last year was around .20 I believe and the area was without power for a week and few businesses were running - '94 was .25 I thought.  That's quite a lot more they're pointing at. 

 

On the western side of the city the dense forestation between buildings means that much tree damage also means major structure damage - the trees would be on the structures.  .25" ice was taking down some trees and many limbs - .9 would certainly take down far more trees.  That's a pretty high accumulation - more like what you described once about your trip to KY.  But they don't have anywhere near the forestation we have here.  Not sure how many residential rooves could take the weight around here either.  These buildings aren't that well built like mid-western structures.  Nor have they been tested by that much, to my knowledge.  While I am sitting in mortal terror of an ice storm - that terror pertains to .2-.3" accumulations.  That's a pretty dire prediction by any measure - maybe it's warranted - maybe not - but it seems a bit too premature to be coming up with such dire extremes so far in advance.

 

They also love their freezing rain predictions in their medium to long range forecasts. They've frequently had .2+ " frz predictions - roughly 1-3 a week.  But to have such a huge one, with text based cautions about it is odd - usually those numbers are buried in the forecast hourlys and such.   I kind of laughed this one off after comparing it to Tom's FB post looking more at the snow chance until I read the short range board and saw "oh, they really DO see an frz possibility"...

 

Regarding their "45", yeah I had the suspicion it was all automated.  It often conflicts directly with their written forecasts (again, I have no complaints about their real human mets - Sosnowski, Pastelok, both professionals with great analysis writeups, and usually worth reading - it's whoever decided on this computer driven system that stinks.)  And whoever did their PVI & KYW forecasts in the 90's - haven't followed it since then, but man, those were awful. they also tend to change the 45 auto forecast several times a day.  Not sure what their formula is though - sometimes it seems right from GFS, sometimes it seems straight from EC.  It's a shame because overall they're not a TERRIBLE outfit, but the automated from the models system (and the terrible idea of a 45 day) makes them look like clowns more than their real staff deserve.  They usually overbake the lows by 5-8 degrees...but it's predictable and relatively consistent bias. 

 

Related but unrelated question for you personally - that story you described about the huge KY ice storm last year has haunted me since I read it - especially after the ice storm last year.  I lucked out last year, and only lost power for the first day.  Very lucky.  I'd not likely survive a full week or even a few days at low temps without heat.  The fear of it happening AGAIN has literally gnawed at the back of my mind all year - it's become a real, consuming, phobia.  While well outside the meteorology scope, do you have any ideas on how to actually survive that time - meaning, in half a day without power last year, indoor temps were in the mid-40's - that was before it got very cold.  Another day without power indoor temps would easily have been in the 20's and 30's.  A few days - teens and 20's indoors.  How do you keep the plumbing from destroying the whole house?  All I can think of is run every tap, toilet, and shower full blast for the duration.  Can't save the washing machine - No power to run the valves.  Hundreds if not thousands of dollars of water pouring through, just to keep it all from freezing and bursting - which of course means you can't LEAVE the building either - you have to stay permanently in sub-freezing conditions like a Neanderthal.   People turn off the meter during vacations  in winter - but that implies heat in the basement, of which in an ice storm, there is none.  And even if you run kerosene heaters, that means you have to stay in the basement with them and can't leave to get to running water (chamber pot?)

 

Given those fears I've always wondered how ice storms are seen as run of the mill - It seems certain doom to me (and that's assuming that trees don't fall on the house - which is always possible.)

Share this post


Link to post
Share on other sites

I'm pretty sure 1994 had 1/4 to 1/2 inch ice across much of the area with localized spots up to 1" of ice accretion.

Share this post


Link to post
Share on other sites

Despite that few of us have a very high opinion of all things Accuweather, I still for some reason have a habit of checking them out.  There's worse.  Sure, their "45 day" is a horrible horrible slap in the face - but their near range, is sometimes ok, and sometimes sees things that haven't really trended in discussion here yet, I dismiss it because you guys haven't talked about it, and sure enough within a day or two you start, and they really did see something - so it's worth looking at in the arsenal.  And their real mets with written forecasts are actually generally quite good - it's the computer generated fluff that's awful. (Especially determining lows.)

 

But today I noticed something that just seems irresponsible, even for what seems to be an entirely computer-generated-from-model-output "45 day" forecast.  The front page for the area discusses the 2/8-2/9 event, adds to it "heavy ice conditions."   The little info graphic and text for Monday just says "some snow" - but on the Monday night segment it says "A major ice storm, with a thick coating of ice" and for Ice accumulation lists 0.9 inches. 

 

At that unprecedented (for the region) accumulation, almost 5x that of last year's ice storm, few buildings and trees would remain standing and power would be out for months.  Business would be gone.  They're effectively calling for a Katrina-level event for the area 5 days out from what is sure to be very gradient driven.  Their own written forecasts don't even discuss the event yet.

 

Obviously you guys are talking nasty ice possibilities as well, so there's the chance they could even be right by sheer dumb luck. But you guys also know that nothing is certain and that those are just possibilities, far from a forecast for something so close this far out.

 

Is this kind of posting sheer irresponsibility on their part to even allow the system to post this sort of thing, or is it a situation of "well that's their reasonable forecast"?

 

 

I don't know how it is now, since I've been gone for 8 years, but when I was there, little to no attention was given to the night periods of the extended forecast.  This meant that it ended up being raw GFS.  Not sure if that's still the case.

Share this post


Link to post
Share on other sites

0.9? Last year was around .20 I believe and the area was without power for a week and few businesses were running - '94 was .25 I thought.  That's quite a lot more they're pointing at. 

 

On the western side of the city the dense forestation between buildings means that much tree damage also means major structure damage - the trees would be on the structures.  .25" ice was taking down some trees and many limbs - .9 would certainly take down far more trees.  That's a pretty high accumulation - more like what you described once about your trip to KY.  But they don't have anywhere near the forestation we have here.  Not sure how many residential rooves could take the weight around here either.  These buildings aren't that well built like mid-western structures.  Nor have they been tested by that much, to my knowledge.  While I am sitting in mortal terror of an ice storm - that terror pertains to .2-.3" accumulations.  That's a pretty dire prediction by any measure - maybe it's warranted - maybe not - but it seems a bit too premature to be coming up with such dire extremes so far in advance.

 

They also love their freezing rain predictions in their medium to long range forecasts. They've frequently had .2+ " frz predictions - roughly 1-3 a week.  But to have such a huge one, with text based cautions about it is odd - usually those numbers are buried in the forecast hourlys and such.   I kind of laughed this one off after comparing it to Tom's FB post looking more at the snow chance until I read the short range board and saw "oh, they really DO see an frz possibility"...

 

Regarding their "45", yeah I had the suspicion it was all automated.  It often conflicts directly with their written forecasts (again, I have no complaints about their real human mets - Sosnowski, Pastelok, both professionals with great analysis writeups, and usually worth reading - it's whoever decided on this computer driven system that stinks.)  And whoever did their PVI & KYW forecasts in the 90's - haven't followed it since then, but man, those were awful. they also tend to change the 45 auto forecast several times a day.  Not sure what their formula is though - sometimes it seems right from GFS, sometimes it seems straight from EC.  It's a shame because overall they're not a TERRIBLE outfit, but the automated from the models system (and the terrible idea of a 45 day) makes them look like clowns more than their real staff deserve.  They usually overbake the lows by 5-8 degrees...but it's predictable and relatively consistent bias. 

 

Related but unrelated question for you personally - that story you described about the huge KY ice storm last year has haunted me since I read it - especially after the ice storm last year.  I lucked out last year, and only lost power for the first day.  Very lucky.  I'd not likely survive a full week or even a few days at low temps without heat.  The fear of it happening AGAIN has literally gnawed at the back of my mind all year - it's become a real, consuming, phobia.  While well outside the meteorology scope, do you have any ideas on how to actually survive that time - meaning, in half a day without power last year, indoor temps were in the mid-40's - that was before it got very cold.  Another day without power indoor temps would easily have been in the 20's and 30's.  A few days - teens and 20's indoors.  How do you keep the plumbing from destroying the whole house?  All I can think of is run every tap, toilet, and shower full blast for the duration.  Can't save the washing machine - No power to run the valves.  Hundreds if not thousands of dollars of water pouring through, just to keep it all from freezing and bursting - which of course means you can't LEAVE the building either - you have to stay permanently in sub-freezing conditions like a Neanderthal.   People turn off the meter during vacations  in winter - but that implies heat in the basement, of which in an ice storm, there is none.  And even if you run kerosene heaters, that means you have to stay in the basement with them and can't leave to get to running water (chamber pot?)

 

Given those fears I've always wondered how ice storms are seen as run of the mill - It seems certain doom to me (and that's assuming that trees don't fall on the house - which is always possible.)

 

Last year's storm was greatly exacerbated by the heavy snow which was still sitting on many trees.  This increased surface area which greatly increased the ice load.

 

The big ice storm in 94 in Ewing was 1/7/1994, the worst day in PECO history (or has it now been eclipsed?).  That was a good solid inch of ice.  It was both breathtaking and awe-inspiring.  One whole tree came down in my parents back yard (fortunately falling away from the house).  My friends and I also decided to be daredevils and took a walk in an adjacent woodland, and very nearly got crushed by falling branches in the process.  Somehow, some way, my parents never lost power during that (but, the power grid seems especially resilient in their area; the longest I can remember them ever losing power was for 6 hours during Sandy).

Share this post


Link to post
Share on other sites

Weatherornot, ice is bad, but not alocalyptic, but if you really are in such a phobia, you should budget for a generator and an electrician to wire up a plug for it to your major needs. You can keep the major things in the house running and total costs can come under $1500, if you only want the basics. Just a thought...

Share this post


Link to post
Share on other sites

I'm pretty sure 1994 had 1/4 to 1/2 inch ice across much of the area with localized spots up to 1" of ice accretion.

 

'94 - 1/4 I was aware of, but I hadn't heard there was more than that - that might make sense.   The big thing I remember about that one was being unprepared - I had no idea power even could go out in Winter - thought it was a Summer thing.  In hindsight that's kingly stupidity, but at the time, I didn't know.  I remember cooking meat using only a wax candle that night.  Dangerous in many ways, but it was the only food around. That's the fixated memory I have for it, and probably the beginning of winter phobia.

 

 

I don't know how it is now, since I've been gone for 8 years, but when I was there, little to no attention was given to the night periods of the extended forecast.  This meant that it ended up being raw GFS.  Not sure if that's still the case.

 

You worked at Accuweather?  Had no idea - I'll have to strike you off my favorites list and let Tony stand alone ;)    J/k.  Very interesting to hear that actually.  And a little disturbing - night periods are pretty important - especially in winter.  Not sure why they'd prioritize it so low.  Any other interesting "rules of thumb" for those days?  Knowing where the numbers come from on some of that stuff makes it a little more useful for how to use it.

 I wouldn't doubt they still do it that way.

 

 

Last year's storm was greatly exacerbated by the heavy snow which was still sitting on many trees.  This increased surface area which greatly increased the ice load.

 

The big ice storm in 94 in Ewing was 1/7/1994, the worst day in PECO history (or has it now been eclipsed?).  That was a good solid inch of ice.  It was both breathtaking and awe-inspiring.  One whole tree came down in my parents back yard (fortunately falling away from the house).  My friends and I also decided to be daredevils and took a walk in an adjacent woodland, and very nearly got crushed by falling branches in the process.  Somehow, some way, my parents never lost power during that (but, the power grid seems especially resilient in their area; the longest I can remember them ever losing power was for 6 hours during Sandy).

 

I'd forgotten about that - I didn't remember the snow already on the trees - that makes sense.  It had kind of reset my expectations that .25=massive damage and outages. 

 

'94 has twice been eclipsed for PECO - Sandy now ranks #1 and last year's ice storm ranks #2.  Actually I think Irene is now #3, making '94 #4.  You can imagine the paranoia about the electric grid around here with any storms - what you remember from '94 has been surpassed 3 times in the last 5 years.  Economic damage has been pretty bad from it, Sandy and '14 ice storm 18 months apart - that's 2+ weeks of many businesses being out of commission in that time. Lots of smaller businesses were very damaged by it.

 

Yeah, falling trees and breaking pipes are the two fears - the plumbing both from power outages and extreme cold - but the trees every time there's strong wind gusts and ice. Lots of huge trees could easily fall here.  And anyone who doesn't sit in paranoia about that has never seen a tree fall on/through a house - nor heard the sound of it in the middle of the night (Floyd or Henri...can't remember which one, two doors down.)   You never, ever forget that sound ,and wait for it with the sound of every wind gust forever more.  My interest in weather is mostly attributed to these events.  During Sandy I was watching a 150ft pin oak sway a good 20 feet back and forth in the yard. Of course the leaves acted as sails in October...

 

Last year's storm the neighbor's huge poplar hand limbs bent all the way down to the ground (in my yard) - normally they're far above the roof line. And that night I was watching the fireballs and blue arks in the sky under the cloud cover from the exploding transformers pretty much everywhere.

Share this post


Link to post
Share on other sites

Weatherornot, ice is bad, but not alocalyptic, but if you really are in such a phobia, you should budget for a generator and an electrician to wire up a plug for it to your major needs. You can keep the major things in the house running and total costs can come under $1500, if you only want the basics. Just a thought...

 

I've considered that - but where there's no natural gas, generators are a sketchier bet.  Automatic gas based generators would be great.  No gas here, so you still have to fuel it up, and the frequent maintenance of a gasoline engine and the need to store large quantities of gasoline for large generators (big enough to run the major appliances) is problematic, especially since gasoline has a shelf life.  While there are solutions, they're not tremendously economical.  I always look into the idea and come away with the thought that I couldn't actually rely on the solution to be there when I need it.  Fighting in the ice with a gummed up carburetor isn't helpful :)

 

 

Houses aren't going to collapse under 1" of ice.

 

Roof damage maybe - but no.  But houses do collapse under a few tons of trees. Only a problem for wooded areas - but bad in wooded areas.  Though ray's clarification that the snow weight made things worse makes sense.

 

Ray: With the added weight of the snow to the .20 or so fo ice - how much "weight equivalent" would you say there would have been last year in terms of ice accretion if it were an ice-only even without the pre existing snow.

Share this post


Link to post
Share on other sites
You worked at Accuweather?  Had no idea - I'll have to strike you off my favorites list and let Tony stand alone ;)    J/k.  Very interesting to hear that actually.  And a little disturbing - night periods are pretty important - especially in winter.  Not sure why they'd prioritize it so low.  Any other interesting "rules of thumb" for those days?  Knowing where the numbers come from on some of that stuff makes it a little more useful for how to use it.

 I wouldn't doubt they still do it that way.

 

Main reason was that paying clients did not find the night period useful; main focus was newspapers, radio, other media which used the forecast.  Lows had some importance but weather at night was virtually worthless since it simply didn't appear in most products.

 

I'd forgotten about that - I didn't remember the snow already on the trees - that makes sense.  It had kind of reset my expectations that .25=massive damage and outages. 

 

Yeah, the first big February storm on the 3rd was just two days before the ice storm and the intervening period was almost dead calm, so almost all the snow that fell on the trees (heavy and wet as it was) was still on the trees when the ice started.

Share this post


Link to post
Share on other sites

On the original Accuweather topic, it's funny that despite the orange text block citing heavy ice Monday - and the overnight detail box showing "major ice storm" - the info graphic text on the weekly/monthly overview shows a half a cloud and "breezy with a little snow."  Now if a major ice storm were forecast for the evening of a given day, would your forecast be "breezy with a little snow" for your headline?  I don't chalk it up to "bad forecasting" or "sliminess for attention" like some  outlets....just horrible sloppiness in their automatic system, arguably their most popular system,  that allows for incongruous results depending on what part you view. Even if the extreme end of the forecast ends up verifying - the daily caption is just out of step.  And if it doesn't verify, the dire forecast in the details was arbitrary.  If you're going to give a forecast, good or bad, at least stick with it through the whole system! :)  I'm not always a fan but I know they're capable of better than that....

Share this post


Link to post
Share on other sites

Main reason was that paying clients did not find the night period useful; main focus was newspapers, radio, other media which used the forecast.  Lows had some importance but weather at night was virtually worthless since it simply didn't appear in most products.

 

 

Yeah, the first big February storm on the 3rd was just two days before the ice storm and the intervening period was almost dead calm, so almost all the snow that fell on the trees (heavy and wet as it was) was still on the trees when the ice started.

 

Very interesting.  But what newspapers and radio didn't report night time weather - particularly winter night time weather?  Snow at night is still there in the AM rush, of course.  And snow happens more often at night than during the day (or at least it's the night time snows that are more likely to stick, have higher ratios, etc. etc.)  I don't remember the dark days too well, but for TV/Radio, I hadn't realized they didn't really focus on the night weather.  It's probably true I just never noticed - I was always look for both.  Accu's reasoning is sensible from a business perspective, but I can't imagine the outlets that used the forecast not seeing demand for knowing about night time precip and temps - at least in the "modern" 24/7 world.  Maybe back in the 80's not so much when schedules were more predictable for most people. Maybe that's how we ended up with...was it the "boxing day storm" that was the 1ft surprise?

 

I kind of remember that now - last year seemed like the never ending recurring nightmare up until the ice storm - and then after that it was a new level of nightmare that made the previous nightmare seem not so bad. The prior snow became a blur in memory.  I remember the snow AFTER the ice storm was another heavy wet one with winds that risked the power lines already weakened by the ice.

 

The irony of it all is I picked a really bad time to try to learn more about weather.  The idea of it was to understand more about it to fear it less and understand when to not panic.  As it happens, since I started getting into it we've had Sandy, a nightmare winter, and this one shaping up to be possibly another nightmare winter - proving that all my fears come true, over and over, and the more you know about the weather, the more time you have to fear it!  Maybe SOMEDAY we'll get rid of this horrible west ridge and the endless cycle of harsh winters and I'll be able to learn something without the sense of impending doom behind every front!

Share this post


Link to post
Share on other sites

Well, gasoline generators aren't that bad to "maintain", mine has never had an issue, but if that big of a deal, buy a propane generator, they can use same tanks as gas grills. We're talking about rare events here, so refueling a generator for a few days once every several years to have the basics....

Anyway, if you're that phobic, might want to bite the bullet. Good luck I won't gum up.the board with anymore generator comments...unless I'm using mine, lol.

  • Like 1

Share this post


Link to post
Share on other sites

Well, gasoline generators aren't that bad to "maintain", mine has never had an issue, but if that big of a deal, buy a propane generator, they can use same tanks as gas grills. We're talking about rare events here, so refueling a generator for a few days once every several years to have the basics....

Anyway, if you're that phobic, might want to bite the bullet. Good luck I won't gum up.the board with anymore generator comments...unless I'm using mine, lol.

 

Actually I have a small one of those - pricy buggers, but handy to keep around.  But it certainly can't run things like the heater or the water heater.  It does mean I won't have to cook meat on a candle again, however. Even if indoor temps are in the teens :fall-to-pieces:  Kerosene heat works, but it does mean you have to stay there to maintain it, and that's only to keep it barely above freezing (and probably not where the plumbing runs...)

 

 

Last ice storm presented an alternate problem: Everyone was down including the gas stations so even propane, kerosene, and gasolene were unavailable.  If I knew who to bribe at Peco to get them to run gas lines.....

Share this post


Link to post
Share on other sites

Very interesting.  But what newspapers and radio didn't report night time weather - particularly winter night time weather?  Snow at night is still there in the AM rush, of course.  And snow happens more often at night than during the day (or at least it's the night time snows that are more likely to stick, have higher ratios, etc. etc.)  I don't remember the dark days too well, but for TV/Radio, I hadn't realized they didn't really focus on the night weather.  It's probably true I just never noticed - I was always look for both.  Accu's reasoning is sensible from a business perspective, but I can't imagine the outlets that used the forecast not seeing demand for knowing about night time precip and temps - at least in the "modern" 24/7 world.  Maybe back in the 80's not so much when schedules were more predictable for most people. Maybe that's how we ended up with...was it the "boxing day storm" that was the 1ft surprise?

 

The first two nights matter... the rest, not so much.  Most products "loaded" off the automatic database, and after the first night or two, most products just ignored the nights and only loaded the day period.  Just a glance at any newspaper forecast and you'll see this.

Share this post


Link to post
Share on other sites

The first two nights matter... the rest, not so much.  Most products "loaded" off the automatic database, and after the first night or two, most products just ignored the nights and only loaded the day period.  Just a glance at any newspaper forecast and you'll see this.

 

Interesting - extremely interesting actually.

 

I never really looked at newspaper forecasts much - I think largely because they never did provide much information - at all.  In the pre-'net days I think I went mostly on KYW and, TV (before I wizened to how bad TV had become.)

Share this post


Link to post
Share on other sites

Despite that few of us have a very high opinion of all things Accuweather, I still for some reason have a habit of checking them out.  There's worse.  Sure, their "45 day" is a horrible horrible slap in the face - but their near range, is sometimes ok, and sometimes sees things that haven't really trended in discussion here yet, I dismiss it because you guys haven't talked about it, and sure enough within a day or two you start, and they really did see something - so it's worth looking at in the arsenal.  And their real mets with written forecasts are actually generally quite good - it's the computer generated fluff that's awful. (Especially determining lows.)

 

But today I noticed something that just seems irresponsible, even for what seems to be an entirely computer-generated-from-model-output "45 day" forecast.  The front page for the area discusses the 2/8-2/9 event, adds to it "heavy ice conditions."   The little info graphic and text for Monday just says "some snow" - but on the Monday night segment it says "A major ice storm, with a thick coating of ice" and for Ice accumulation lists 0.9 inches. 

 

At that unprecedented (for the region) accumulation, almost 5x that of last year's ice storm, few buildings and trees would remain standing and power would be out for months.  Business would be gone.  They're effectively calling for a Katrina-level event for the area 5 days out from what is sure to be very gradient driven.  Their own written forecasts don't even discuss the event yet.

 

Obviously you guys are talking nasty ice possibilities as well, so there's the chance they could even be right by sheer dumb luck. But you guys also know that nothing is certain and that those are just possibilities, far from a forecast for something so close this far out.

 

Is this kind of posting sheer irresponsibility on their part to even allow the system to post this sort of thing, or is it a situation of "well that's their reasonable forecast"?

 

 

The 00z GFS run was very icy.  Not that anything is etched in stone at this juncture, but the 12z model suite had more sleet and snow to it.

Share this post


Link to post
Share on other sites

The 00z GFS run was very icy.  Not that anything is etched in stone at this juncture, but the 12z model suite had more sleet and snow to it.

 

Aww, you just replied because I took Ray off the "my favorite mets" podium and gave it all to you, didn't you? :bye:  Unless you worked at TWC....or NBC.  Then I might have to just give the podium to Mitch.

 

Still, that's good tentative news, at least for the "I'll listen only to information I want to hear so I don't have heart attack" factor. And while it was a given there would be a lot of waffling from now until Monday, it's always good to watch it drift in a positive direction rather than sticking with knowing you're doomed and trying to reassure yourself it's not a given yet. It'll be interesting to see what Accu updates to after the updated GFS and see how lock-step it is to the model.  When Accu has some big precip/cold/heatwave report I take it with a grain of salt - when I see you guys discussing it - that's when it's time to panic :)

Share this post


Link to post
Share on other sites

The first two nights matter... the rest, not so much.  Most products "loaded" off the automatic database, and after the first night or two, most products just ignored the nights and only loaded the day period.  Just a glance at any newspaper forecast and you'll see this.

 

Do you know how Accu creates their per-locale temps in the auto system?  I mean do they base it on given specific location predictions, or do they basically take the metro and apply a formula to "convert" it for difference or local histories, etc? Might not be something you know, but it's something I've always wondered.

Share this post


Link to post
Share on other sites

Do you know how Accu creates their per-locale temps in the auto system?  I mean do they base it on given specific location predictions, or do they basically take the metro and apply a formula to "convert" it for difference or local histories, etc? Might not be something you know, but it's something I've always wondered.

I don't know if this is how it works now, but when I was there, they took the GFS MOS locations, and for all places they wanted forecasts for which didn't have MOS, they applied a "climo adjustment".  Sometimes it was good, sometimes it wasn't, depending on the weather situation.  No idea if that's how its still done.

Share this post


Link to post
Share on other sites

I don't know if this is how it works now, but when I was there, they took the GFS MOS locations, and for all places they wanted forecasts for which didn't have MOS, they applied a "climo adjustment".  Sometimes it was good, sometimes it wasn't, depending on the weather situation.  No idea if that's how its still done.

 

Thanks! That's one thing I've been curious about since long before I discovered you guys.

 

I'm guessing they still do it that way, based largely on the predictable or quasi-predictable biases I've notcied.  I find if I take their data and apply my own mental "climo adjustment", particularly to lows, I can get fairly close to what ends up verifying.  Usually.  Sometimes though the numbers come from nowhere.  That sounds pretty accurate - and what I've learned here (between fits of panic, terror, and hiding in a hole shaking), I've improved my ability to predict where it'll land a little. 

Share this post


Link to post
Share on other sites

I knew I'd heard that we had widespread amounts of 1/4" to 1/2" ice accretion with localized amounts up to an inch, and here it is:

 

http://www.erh.noaa.gov/phi/hist_phi.html

 

Jan 7

 

...1994...
A great ice storm commenced on the 7th and extended into the 8th. Low pressure
formed east of Wyoming on the 5th, moved to the Central MS Valley on the 6th,
reached Ern KY the morning of the 7th and redeveloped along the Ern VA coast on
the morning of the 8th. A strong high pressure system remained nearly stationary over
Ontario Province, Canada and continued to pump in a fresh supply of cold air near
the ground. This kept temperatures below freezing throughout this event resulting in
one of the worst ice storms on record for SE PA. The freezing rain started lightly the
morning of the 7th, but in spite of this, all roadways were glazed by the start. As rain
increased in intensity the night of the 7th, the accumulation of ice started to down tree
limbs and power lines. Ice accumulations were 1/4" to 1/2" across most of SE PA, but
amounts reached up to 1" in the NW PHL suburbs. While there were 5,000 customers
without power in the Lehigh valley, the hardest hit area was around PHL.
Approximately 590,000 Philadelphia Electric Company (PECO) customers lost
power. This represents about 40% of their customers and was the worst power outage
in PECO's history, surpassing the 400,000 customers who lost power during the
heavy wet snowstorm on Mar 20, 1958. Approximately 150,000 (of 232,000)
customers lost power in Delaware Co, 134,000 in Bucks Co, 112,000 in Chester Co,
105,000 in Montgomery Co and 89,000 in Philadelphia Co. As of Sunday morning
the 9th, 123,000 customers still did not have power. Full power was not restored until
Tuesday morning the 11th. There were literally thousands of ice-related vehicular
accidents and described as "countless" number of personal injuries resulting from
slipping on the ice. Miraculously no one was directly killed as a result of this ice
storm. In addition to damage to electrical lines, the ice accumulation on fruit trees

brought a significant amount of damage. (SD) 

Share this post


Link to post
Share on other sites

I knew I'd heard that we had widespread amounts of 1/4" to 1/2" ice accretion with localized amounts up to an inch, and here it is:

 

http://www.erh.noaa.gov/phi/hist_phi.html

 

Jan 7

 

...1994...

A great ice storm commenced on the 7th and extended into the 8th. Low pressure

formed east of Wyoming on the 5th, moved to the Central MS Valley on the 6th,

reached Ern KY the morning of the 7th and redeveloped along the Ern VA coast on

the morning of the 8th. A strong high pressure system remained nearly stationary over

Ontario Province, Canada and continued to pump in a fresh supply of cold air near

the ground. This kept temperatures below freezing throughout this event resulting in

one of the worst ice storms on record for SE PA. The freezing rain started lightly the

morning of the 7th, but in spite of this, all roadways were glazed by the start. As rain

increased in intensity the night of the 7th, the accumulation of ice started to down tree

limbs and power lines. Ice accumulations were 1/4" to 1/2" across most of SE PA, but

amounts reached up to 1" in the NW PHL suburbs. While there were 5,000 customers

without power in the Lehigh valley, the hardest hit area was around PHL.

Approximately 590,000 Philadelphia Electric Company (PECO) customers lost

power. This represents about 40% of their customers and was the worst power outage

in PECO's history, surpassing the 400,000 customers who lost power during the

heavy wet snowstorm on Mar 20, 1958. Approximately 150,000 (of 232,000)

customers lost power in Delaware Co, 134,000 in Bucks Co, 112,000 in Chester Co,

105,000 in Montgomery Co and 89,000 in Philadelphia Co. As of Sunday morning

the 9th, 123,000 customers still did not have power. Full power was not restored until

Tuesday morning the 11th. There were literally thousands of ice-related vehicular

accidents and described as "countless" number of personal injuries resulting from

slipping on the ice. Miraculously no one was directly killed as a result of this ice

storm. In addition to damage to electrical lines, the ice accumulation on fruit trees

brought a significant amount of damage. (SD)

Tony probably wrote this ;)

Share this post


Link to post
Share on other sites

It's incredible how bad that was - and yet how bad last year's was with so much less ice (now explained by Ray).  Granted Peco's outages are always worse now because there's simply more population, so by the numbers, they'll always have more meters down.  But still, those numbers seem mild compared to the last 3 major outages.  And full power up in 4 days?  After Sandy and last year's ice storm it too 7-14+ days for full power to all locations. It makes the electrical situation of '94 seem tame!

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...