Feb. 23-25th Winter Storm Potential

[SpartyOn]

GFS is at the point, it shouldn't even be given any thought.

I was pondering the same thing...By the looks ( Im only home gamer..lol) I see some feedback issues..I dont claim to be an expert so if Im seeing nothing feel free to educate me!

[baroclinic_instability]

GFS is likely a tick S and I think the op will come a little N--but the effects of convection are clearly rearing its head here as the orientation of the wind fields and the expected mode of eventual convection will help drive surface pressure falls farther S of the position of the synoptic surface low as convection rapidly spreads eastward ahead of the wave. The mass latent heat release will also have an effect--a realistic effect. It can be shown here that there is a fine line in terms of the strength of the upper wave and how it will eventually develop. A stronger upper wave and this storm could theoretically "tank out" with that baro zone--much like the convection during groundhog blizzard helped that storm rapidly deepen and intensify. If the upper wave is a tick too weak--it will actually weaken with time as the mid troposphere warms.

[snowstormcanuck]

18z UKIE looks like it ticked SE. You be the judge.

http://moe.met.fsu.edu/tcgengifs/ukm/2011022218/

[baroclinic_instability]

Although that kind of blanket statement will bring some haters your way, I totally agree. I feel like the Euro kicks its butt at times, and considering this is supposed to be a new and improved model, I'm far from impressed.

Yes the Euro does at times kick the GFS' butt--and it can go the other way as well as the GFS has kicked the Euro's tail a number of times. This isn't a "model war" though. As for the GFS being new and improved--well it is--it is as simple as that. If you used the old GFS operationally you would most definitely attest to it. Also--this storm hasn't happened yet-and as I mentioned just a few mins ago--convection can most certainly alter the synoptic environment--and "convective feedback" issues the HPC folks were mentioning may very well have been real feedbacks to the system. For now--I think a realistic forecast given the data would be farther N than the GFS leaning towards the Euro. Expect continued trends one way or the other for a number of runs.

[wisconsinwx]

Yes the Euro does at times kick the GFS' butt--and it can go the other way as well as the GFS has kicked the Euro's tail a number of times. This isn't a "model war" though. As for the GFS being new and improved--well it is--it is as simple as that. If you used the old GFS operationally you would most definitely attest to it. Also--this storm hasn't happened yet-and as I mentioned just a few mins ago--convection can most certainly alter the synoptic environment--and "convective feedback" issues the HPC folks were mentioning may very well have been real feedbacks to the system. For now--I think a realistic forecast given the data would be farther N than the GFS leaning towards the Euro. Expect continued trends one way or the other for a number of runs.

I'll admit I haven't done a study or looked at both on a regular basis, but the 3 or 4 major systems this winter have been handled very consistently by the Euro compared to the GFS. The December storm, which model was pretty consistent on a decent cutter? The Euro, aside from a fluke run or two. The GFS was suppressed for several runs. Which model was consistent with a track through the Upper Ohio Valley for the Groundhog Day Storm? The Euro again. Which one stood out as being further SE with this last storm through Southern Iowa and Northern Illinois while the rest were further north? The Euro, and it was the closest to being accurate. For my money, I want to rely on a model that will be consistent with the major storm systems. That is definitely the Euro this year, at least in the Midwest.

[SpartyOn]

Looking at the NWS AFDs from many areas..its clear. We have a ways to go before a lock

[baroclinic_instability]

I'll admit I haven't done a study or looked at both on a regular basis, but the 3 or 4 major systems this winter have been handled very consistently by the Euro compared to the GFS. The December storm, which model was pretty consistent on a decent cutter? The Euro, aside from a fluke run or two. The GFS was suppressed for several runs. Which model was consistent with a track through the Upper Ohio Valley for the Groundhog Day Storm? The Euro again. Which one stood out as being further SE with this last storm through Southern Iowa and Northern Illinois while the rest were further north? The Euro, and it was the closest to being accurate. For my money, I want to rely on a model that will be consistent with the major storm systems. That is definitely the Euro this year, at least in the Midwest.

You mention the storms the Euro did well and don't mention any others--if course it sounds like the de facto model by your account.

Also--the groundhog blizzard was handled superb by the GFS even though both the Euro and GFS were too weak. SREF and the amped non-hydro models won that. Using your logic we may as all use the Euro only. In reality it is far more complex than that.

[snowstormcanuck]

18z GEFS have taken a decent sized shift north

[baroclinic_instability]

Looking at the NWS AFDs from many areas..its clear. We have a ways to go before a lock

Yeah I agree--I wouldn't be locking anything in. We clearly have three major considerations. The strength and amplitude of the ejecting wave, the amount and timing of early convection ahead of the main wave, and the type and strength of convection over the MIss. Valley and into the OV. The slight S/SE ticks by the more amped guidance shouldn't be avoided for the reasons laid out previously. For me--the biggest difference I am seeing between the guidance is the strength of the ejecting wave into the plains as the UK and to a degree Euro have a deeper and more organized/amplified S/W into the plains which then amplifies further through baroclinic processes. The weaker solution like the GFS has a much weaker initial S/W and weakens further as convection warms the mid troposphere. We really are on a fine line regarding the strength of the upper wave either amplifying or decaying--hence the rather disparate solutions with increasing time--think of it as an inflection point in mathematics.

[Hoosier]

You mention the storms the Euro did well and don't mention any others--if course it sounds like the de facto model by your account.

Also--the groundhog blizzard was handled superb by the GFS even though both the Euro and GFS were too weak. SREF and the amped non-hydro models won that. Using your logic we may as all use the Euro only. In reality is far more complex that this.

I don't know if I'd agree with your assertion that the Groundhog blizzard was handled superb by the GFS. It made slow, consistent trends but when added up, they were huge.

[baroclinic_instability]

I don't know if I'd agree with your assertion that the Groundhog blizzard was handled superb by the GFS. It made slow, consistent trends but when added up, they were huge.

Well--from a forecasting standpoint--slow consistent changes in the dynamic fields is all a forecaster can truly ask for as no model will be nailing things on the dot 5 days out. I actually don't remember how the Euro performed exactly but I know it did well overall as well.

[Chicago WX]

I like this disagreement. Point and click forecasts for two close locations from two different NWS offices for Thursday night.

Brookston (IWX), about 15 miles north of here.

Thursday Night: Rain likely. Cloudy, with a low around 32. Northeast wind between 10 and 20 mph. Chance of precipitation is 70%. New rainfall amounts between a quarter and half of an inch possible.

Lafayette (IND)

Thursday Night: Snow likely. Cloudy, with a low around 27. Northeast wind around 14 mph. Chance of precipitation is 70%. New precipitation amounts between a quarter and half of an inch possible.

[baroclinic_instability]

I like this disagreement. Point and click forecasts for two close locations from two different NWS offices for Thursday night.

Brookston (IWX), about 15 miles north of here.

Thursday Night: Rain likely. Cloudy, with a low around 32. Northeast wind between 10 and 20 mph. Chance of precipitation is 70%. New rainfall amounts between a quarter and half of an inch possible.

Lafayette (IND)

Thursday Night: Snow likely. Cloudy, with a low around 27. Northeast wind around 14 mph. Chance of precipitation is 70%. New precipitation amounts between a quarter and half of an inch possible.

Do their discussions disagree or is the rain/snow line just that close? I honestly have no idea as I haven't looked that close for the area--I am just asking.

[panhandlehook]

I'll admit I haven't done a study or looked at both on a regular basis, but the 3 or 4 major systems this winter have been handled very consistently by the Euro compared to the GFS. The December storm, which model was pretty consistent on a decent cutter? The Euro, aside from a fluke run or two. The GFS was suppressed for several runs. Which model was consistent with a track through the Upper Ohio Valley for the Groundhog Day Storm? The Euro again. Which one stood out as being further SE with this last storm through Southern Iowa and Northern Illinois while the rest were further north? The Euro, and it was the closest to being accurate. For my money, I want to rely on a model that will be consistent with the major storm systems. That is definitely the Euro this year, at least in the Midwest.

The Euro is KING! Long live the Euro

[Indystorm]

Yes, IWX has been calling for upwards of 6 inches of snow on and north of a line from Winamac to Defiance OH in their Hazardous weather outlook.

[wisconsinwx]

You mention the storms the Euro did well and don't mention any others--if course it sounds like the de facto model by your account.

Also--the groundhog blizzard was handled superb by the GFS even though both the Euro and GFS were too weak. SREF and the amped non-hydro models won that. Using your logic we may as all use the Euro only. In reality it is far more complex than that.

No, I'm not. I mentioned all the major systems that affected the Western Great Lakes (systems of 6 inches or greater in MBY. Of course those are the ones I'm most familiar with model-wise. Regarding the Groundhog Day Storm, the Euro was still the first to shift northwest I believe after both it and the GFS were a tad suppressed to start.

[wisconsinwx]

Well--from a forecasting standpoint--slow consistent changes in the dynamic fields is all a forecaster can truly ask for as no model will be nailing things on the dot 5 days out. I actually don't remember how the Euro performed exactly but I know it did well overall as well.

I agree that in general slow consistent trends from a disparate solution to what ends up being accurate is not a bad performance, but that said, the Euro was definitely better than the GFS with that storm.

[baroclinic_instability]

No, I'm not. I mentioned all the major systems that affected the Western Great Lakes (systems of 6 inches or greater in MBY. Of course those are the ones I'm most familiar with model-wise. Regarding the Groundhog Day Storm, the Euro was still the first to shift northwest I believe after both it and the GFS were a tad suppressed to start.

All this is interesting discussion because you bring up the exact thing that is ruining meteorology today and weather forecasting in general--model casting. I don't expect a non met to understand--but everything you say is what is actually crippling meteorology today--reliance on models verbatim with no understanding of what is going on. For your part it is ok--for mets who use that same logic it is unacceptable. Unfortunately too many mets use the exact same logic. OT and end rant on model-casting.

[Chicago WX]

Do their discussions disagree or is the rain/snow line just that close? I honestly have no idea as I haven't looked that close for the area--I am just asking.

Their discussions differ, plus Brookston is north of here...as such it'd be pretty neat to see snow here in LAF, while a place 15 miles north sees rain. And this is no knock on either office, just goes to show the different thinking on this potential system. Gonna be a nail biter for some.

[baroclinic_instability]

Their discussions differ, plus Brookston is north of here...as such it'd be pretty neat to see snow here in LAF, while a place 15 miles north sees rain. And this is no knock on either office, just goes to show the different thinking on this potential system. Gonna be a nail biter for some.

Ah well that isn't good haha. I didn't realize the snow was S. Yeah that doesn't make much sense.

[SpartyOn]

All this is interesting discussion because you bring up the exact thing that is ruining meteorology today and weather forecasting in general--model casting. I don't expect a non met to understand--but everything you say is what is actually crippling meteorology today--reliance on models verbatim with no understanding of what is going on. For your part it is ok--for mets who use that same logic it is unacceptable. Unfortunately too many mets use the exact same logic. OT and end rant on model-casting.

Baro.

You bring up an interesting point. I have often wondered about model watching becoming a "virus" in modern forecasting. Computers fail...Weather changes..Humans fail & change...To rely on just one seems insane! Man vs Machine is ancient and its blunders are historical.

[baroclinic_instability]

Baro.

You bring up an interesting point. I have often wondered about model watching becoming a "virus" in modern forecasting. Computers fail...Weather changes..Humans fail & change...To rely on just one seems insane! Man vs Machine is ancient and its blunders are historical.

Yeah you said it there. Models have become better and the average forecaster has become worse. Humans as well as models will all eventually fail at times--but if one doesn't even know how to assess and analyze the weather pattern at hand how does one learn? How does one know when a model is busting hard core like the SNE Nor'easter earlier this winter where all models failed miserably but good forecasters were able to NOWcast and understand the storm was about to tank? They don't. We all bust but we only learn from experience and from analysis--model casting alone is not the solution. Hopefully this trend abates soon. There are still plenty of good forecasters and mets out there--I just wonder what it is some mets think their job is if all they can do is glance at models and guess. Models already do that! Of course decision support services and interpretation will always be a part of the job--but even then glancing at the models alone won't be enough to even do those aspects of the job in any type of efficient manner.

[Chicago WX]

All this is interesting discussion because you bring up the exact thing that is ruining meteorology today and weather forecasting in general--model casting. I don't expect a non met to understand--but everything you say is what is actually crippling meteorology today--reliance on models verbatim with no understanding of what is going on. For your part it is ok--for mets who use that same logic it is unacceptable. Unfortunately too many mets use the exact same logic. OT and end rant on model-casting.

Thing is there is no perfect model, which of course you know. That's why it's termed model guidance...not model solutions. I'm certainly just a weenie, but it seems to me that a blend of guidance is usually a good way to go. Granted if there is a wide range of solutions, that's makes it harder...but as we get inside 48 hours the differences usually get less. Sure the Euro verifies better, but it's not like it knocks it out of the park every time. There were a couple of instances back in December where it blew chunks for here.

Your talk of model casting is interesting. I see some TV mets that run the in-house model on the telecast and that's their forecast verbatim. Many times that has blown up in their faces. On the other hand, our local guy here rides the GFS into the ground. I can't think of an instance where any other model has even been mentioned. I guess I'm not trying to completely knock all TV mets, because a lot of them do good work, but I think it's a little "dangerous" to forecast off one model. Or model-cast if you will. But I digress...

[baroclinic_instability]

Thing is there is no perfect model, which of course you know. That's why it's termed model guidance...not model solutions. I'm certainly just a weenie, but it seems to me that a blend of guidance is usually a good way to go. Granted if there is a wide range of solutions, that's makes it harder...but as we get inside 48 hours the differences usually get less. Sure the Euro verifies better, but it's not like it knocks it out of the park every time. There were a couple of instances back in December where it blew chunks for here.

Your talk of model casting is interesting. I see some TV mets that run the in-house model on the telecast and that's their forecast verbatim. Many times that has blown up in their faces. On the other hand, our local guy here rides the GFS into the ground. I can't think of an instance where any other model has even been mentioned. I guess I'm not trying to completely knock all TV mets, because a lot of them do good work, but I think it's a little "dangerous" to forecast off one model. Or model-cast if you will. But I digress...

Last post on my part on the topic--but fully agree. Even here in the cities a lot of TV mets are quick to mention the different "computer models" but without making any call. We have one forecaster in the cities who makes consistent forecast calls using logic that makes use of weather analysis (Chikage Windler--an OU grad), otherwise, TV mets here typically use none. Of course the job requirements and time crunch certainly doesn't help--and the demanding public doesn't help either--but that is part of the job! These guys are making in excess of 100k per year--they should do more than act as a "personality". This is why a lot of markets are now hiring the pretty face and non-met--it is cheaper and they get the same performance from a broadcast major as some met majors. Sad really.

[wisconsinwx]

Thing is there is no perfect model, which of course you know. That's why it's termed model guidance...not model solutions. I'm certainly just a weenie, but it seems to me that a blend of guidance is usually a good way to go. Granted if there is a wide range of solutions, that's makes it harder...but as we get inside 48 hours the differences usually get less. Sure the Euro verifies better, but it's not like it knocks it out of the park every time. There were a couple of instances back in December where it blew chunks for here.

Your talk of model casting is interesting. I see some TV mets that run the in-house model on the telecast and that's their forecast verbatim. Many times that has blown up in their faces. On the other hand, our local guy here rides the GFS into the ground. I can't think of an instance where any other model has even been mentioned. I guess I'm not trying to completely knock all TV mets, because a lot of them do good work, but I think it's a little "dangerous" to forecast off one model. Or model-cast if you will. But I digress...

I understand what you guys are saying, but I'm tired of the GFS making the same mistakes. It seems most storms it starts out much more suppressed than reality. About half the major storms this year the Euro nearly nailed 5 or 6 days out. It would waver once in a while or even have a bad run, but I still trust the Euro much more than the GFS. Does that mean that's the only one I'll follow? No, but if the Euro and GFS are much different, I'll naturally lean toward the Euro because of recent history in my neck of the woods. I think it's important to take a blend of guidance as well, but I'm starting to question whether the GFS is one of the top 2 or 3 models out there.

[wisconsinwx]

Btw, I also agree on hoping that meteorologists use more of their knowledge of weather instead of just spitting out model information. I recognize that the models fail quite often. I think we'd see better results if meteorologists paid little or no attention to models and just look at current information and interpret it themselves.

[Powerball]

but I'm starting to question whether the GFS is one of the top 2 or 3 models out there.

GFS statiscally has the best verification rate behind the EURO.

I think the GFS does better with summertime patterns than the EURO. The EURO has a bias of not being progressive enough.

[snowstormcanuck]

GFS statiscally has the best verification rate behind the EURO.

I think the GFS does better with summertime patterns than the EURO. The EURO has a bias of not being progressive enough.

Wasn't the UKIE 2nd? Or was that just at H5?

[Chicago WX]

Wasn't the UKIE 2nd? Or was that just at H5?

For the past 31 days, the Ukie is #1 at H5 for both day 5 and 6.

http://www.emc.ncep.noaa.gov/gmb/STATS/html/acz5.html

http://www.emc.ncep.noaa.gov/gmb/STATS/html/acz6.html

Here's the past 5 years at H5.

http://www.emc.ncep.noaa.gov/gmb/STATS/html/aczhist6.html

Seasonal scores since 1985.

http://www.emc.ncep.noaa.gov/gmb/STATS/html/seasons.html

[Harry]

Yeah you said it there. Models have become better and the average forecaster has become worse. Humans as well as models will all eventually fail at times--but if one doesn't even know how to assess and analyze the weather pattern at hand how does one learn? How does one know when a model is busting hard core like the SNE Nor'easter earlier this winter where all models failed miserably but good forecasters were able to NOWcast and understand the storm was about to tank? They don't. We all bust but we only learn from experience and from analysis--model casting alone is not the solution. Hopefully this trend abates soon. There are still plenty of good forecasters and mets out there--I just wonder what it is some mets think their job is if all they can do is glance at models and guess. Models already do that! Of course decision support services and interpretation will always be a part of the job--but even then glancing at the models alone won't be enough to even do those aspects of the job in any type of efficient manner.

Hence my use of the indicies. However there is some who *attempt* to do what you are saying/advocating as well and fail miserably. They don't have much use for modeling at all. Pattern recognition is king but knowing the biases of the models ( via use of indices ) makes them a great TOOL. Not the end all but a tool. As mentioned before there is some models that do better with some patterns vs others and vice versa and then you have the issue of eastcoast storms and or any system on/near the coast/in the ocean which models can be way off with as you saw with the New England system. This is where use of SST info etc comes into play with trying to identify the baro zone as i like to call it. Those from that region who know about it and understand it did great and made a great call but others failed.

But anyways my issue with the model humpers are those that don't even take notice of their biases which YES all of them have in some form or another. Ofcourse this takes a ton of following and remembering unless you write it down/save it somewhere. Note my convos about how the models handle blocking/-AO/-NAO etc and which ones do better with it vs those that do better with no blocking/+ao/+nao. I mentioned earlier that the euro tends to struggle more with the northern stuff and thus why it is not as handy as some think during -ao/-nao etc. Why again i mentioned that southern Canada system as that is the sort of thing the euro can mess up with. That ofcourse could send the confluence farther south or keep it farther north and this ofcourse has it's own implications on the track of the main system coming out of the sw into the Plains. This is not counting the stuff you mentioned about the system itself.

I understand what you guys are saying, but I'm tired of the GFS making the same mistakes. It seems most storms it starts out much more suppressed than reality. About half the major storms this year the Euro nearly nailed 5 or 6 days out. It would waver once in a while or even have a bad run, but I still trust the Euro much more than the GFS. Does that mean that's the only one I'll follow? No, but if the Euro and GFS are much different, I'll naturally lean toward the Euro because of recent history in my neck of the woods. I think it's important to take a blend of guidance as well, but I'm starting to question whether the GFS is one of the top 2 or 3 models out there.

There is storms that the euro did not so great with and the GFS did. Again it goes back to what i said above to baro. Not directing this at you but in general this is why i wish a few would not even comment on them like you just did because many don't understand the hows or why's. Model biases/the pattern etc. Saying the GFS sucks or the euro is the end all is not the answer. I'll put the GFS up against the euro any day when it comes to blocks/-ao/-nao. This is where it helps to look beyond your own backyard as well so you can get a feel for the models better and a better feel for the general pattern itself. Basically what baro said.