Good news about the pending hurricane...

[bwifan]

Eastern PA Weather Authority now saying out to sea and never hits the U.S. Still going to have rain from the noreaster this weekend but this is good news. Hoping their call and forecast holds as they are eerily accurate and chief met is a PSU grad...
http://www.lehighvalleylive.com/weather/index.ssf/2015/10/eastern_pa_weather_authority_j.html

 

[NittPicker]

If so then it sounds like the Euro model was more accurate yet again.

 

[Still in State College]

Why isn't the Euro Model known as THE Model?

 

[bwifan]

If so then it sounds like the Euro model was more accurate yet again.

Yes... also Joe Bastardi just changed weatherbell's forecast and it now matches the Euro as well he is on board with the Euro... Pretty tough group to go against those 3 players EPAWA/JB/Euro when they say its going OTS or high percentage its goes out to sea you have believe them

 

[bwifan]

Why isn't the Euro Model known as THE Model?

Most do use it as thee model of choice.

 

[LionJim]

Cool, that's a relief. My daughter is driving from/to college.

Most do use it as thee model of choice.

Yes, it's held in such high esteem that cultured folk refer it to "Thee European Model."

 

[Lane]

Cool, that's a relief. My daughter is driving from/to college.

Yes, it's held in such high esteem that cultured folk refer it to "Thee European Model."

In Columbus, they refer to it as The Ohio State Model.

 

[BobPSU92]

If so then it sounds like the Euro model was more accurate yet again.

Flag! Where is the pic of this Euro model?

 

[BobPSU92]

In Columbus, they refer to it as The Ohio State Model.

Also known as cheating.

 

[bwifan]

Flag! Where is the pic of this Euro model?

Right here is your pic... a real beaut.....

 

[jj410]

Eastern PA Weather Authority now saying out to sea and never hits the U.S. Still going to have rain from the noreaster this weekend but this is good news. Hoping their call and forecast holds as they are eerily accurate and chief met is a PSU grad...
http://www.lehighvalleylive.com/weather/index.ssf/2015/10/eastern_pa_weather_authority_j.html

Actually, NWS still showing considerable cone of uncertainty
http://www.weather.gov/lwx/Joaquin

 

[uppermaclion]

Actually, NWS still showing considerable cone of uncertainty
http://www.weather.gov/lwx/Joaquin

They average the models to get that cone. There is still a group of models that take the storm into the NC/VA area. A of of the other models have started to follow what the Euro is doing. When you average out the divergence, that track is about what you get. It looks like, as of right now, this is headed out to sea, and I wouldn't be surprised to see the NHC change to cone more out to sea in a few minutes at the next update.

 

[bwifan]

They average the models to get that cone. There is still a group of models that take the storm into the NC/VA area. A of of the other models have started to follow what the Euro is doing. When you average out the divergence, that track is about what you get. It looks like, as of right now, this is headed out to sea, and I wouldn't be surprised to see the NHC change to cone more out to sea in a few minutes at the next update.

Correct and we don't look that hard at the op runs of the model we also look much more into the ensembles to see if they support what the op run showed. We are also looking at the 500mb level of the atmosphere and what is going on there not on the surface levels. There is a boatload of physics involved here so in short what we are seeing in the atmosphere has been happening for awhile now and why the forecast is what it is... Could it be wrong? Absolutely but feeling pretty good about that forecast currently...

 

[jj410]

Correct and we don't look that hard at the op runs of the model we also look much more into the ensembles to see if they support what the op run showed. We are also looking at the 500m level of the atmosphere and what is going on there not on the surface levels. There is a boatload of physics involved here so in short what we are seeing in the atmosphere has been happening for awhile now and why the forecast is what it is... Could it be wrong? Absolutely but feeling pretty good about that forecast currently...

The European model has been predicting it to go out to sea pretty much all along, while the US and Canadian models all have been saying otherwise?

 

[mn78psu83]

Actually, NWS still showing considerable cone of uncertainty
http://www.weather.gov/lwx/Joaquin

"cone of uncertainty" - great phrase. Possible alternative use - "As evidenced by his play calling, Donovan's mind was lost in a cone of uncertainty during Saturday's offensive debacle." Don't mean to start anything.

 

[uppermaclion]

The European model has been predicting it to go out to sea pretty much all along, while the US and Canadian models all have been saying otherwise?

Yes. The last several days euro out to sea. Now the other models are starting to agree with the euro. When sandy hit, the euro was the only model that had it hitting NJ several days out. Every other model had it out to sea. Then they all came on board with the euro. Looks like the same thing happening here.

 

[bplionfan]

What a disgrace the US models getting owned by the King EURO once again. All of these upgrades to the GFS and it still can't hold the EURO's jock..

 

[jj410]

Yes. The last several days euro out to sea. Now the other models are starting to agree with the euro. When sandy hit, the euro was the only model that had it hitting NJ several days out. Every other model had it out to sea. Then they all came on board with the euro. Looks like the same thing happening here.

While maybe diversity of models has value, I understand one reason the Europeans are more accurate is that they have more powerful computers? So how is it the most powerful and technologically advanced nation in the world" can't afford bigger computers? Maybe ally with Canada? Penny wise, pound foolish.

 

[uppermaclion]

While maybe diversity of models has value, I understand one reason the Europeans are more accurate is that they have more powerful computers? So how is it the most powerful and technologically advanced nation in the world" can't afford bigger computers? Maybe ally with Canada? Penny wise, pound foolish.

I'm no expert on weather models, but it has nothing to do with the horsepower of the computers it is running on. It has to do with the way the model is written and the physics that the model uses. That is where the Euro wins.

 

[uppermaclion]

What a disgrace the US models getting owned by the King EURO once again. All of these upgrades to the GFS and it still can't hold the EURO's jock..

It is actually quite embarrassing. They should shift some of the money the use for predicting the climate 100 years into the future(lol) and actually spend that on getting a model that works for a few days out. the worst part is the euro is a global model. The US has specialized models just for hurricane models and they are pretty bad. It is sad when a very specialized model loses out to a much more general model but that has been going on for years.

 

[jj410]

I'm no expert on weather models, but it has nothing to do with the horsepower of the computers it is running on. It has to do with the way the model is written and the physics that the model uses. That is where the Euro wins.

From what I read, seems to be some disagreement on that. Three possible interrelated reasons: Inferior models, smaller computers, or diversion of resources among uncoordinated models.
http://blogs.agu.org/wildwildscienc...-satellites-and-more-accurate-weather-models/

 

[mn78psu83]

From what I read, seems to be some disagreement on that. Three possible interrelated reasons: Inferior models, smaller computers, or diversion of resources among uncoordinated models.
http://blogs.agu.org/wildwildscienc...-satellites-and-more-accurate-weather-models/

The interviewee in the below National Geographic interview from 2013 mentions the lack of computing power. I wish the interview were a little longer.

LINK: Why are Europeans Better Predicting Weather?

 

[Howie'81]

Hmph. Are these the same guys who say global warming is "settled science?"

 

[Player2BNamedL8r]

Also known as cheating.

Bob, you're on a roll today!!

 

[mn78psu83]

Hmph. Are these the same guys who say global warming is "settled science?"

Didn't want to go there, but from the interview:



What would it take for the U.S. to have a model as good as the European one?

It would be very easy and it's outrageous it hasn't been done. We need more computer power for the National Weather Service. One way to do that would be to repurpose one of the supercomputers that NOAA uses for climate change research. Another would be to buy a new system. Why not use some of the $50 billion from the Hurricane Sandy relief bill?

TJ?

 

[thecoolestfish]

Yes... also Joe Bastardi just changed weatherbell's forecast and it now matches the Euro as well he is on board with the Euro... Pretty tough group to go against those 3 players EPAWA/JB/Euro when they say its going OTS or high percentage its goes out to sea you have believe them

LOL - Bastardi (and embarrassment to PSU) is always a day late. He talks good though.

 

[uppermaclion]

LOL - Bastardi (and embarrassment to PSU) is always a day late. He talks good though.

Regardless of what you think of his stance on climate chance, he is by far one of the best forecasters there is. His forecasts absolutely destroy the "official" NOAA/NWS(especially the last few years) forecasts and they are done long before they put theirs out.

 

[mn78psu83]

Regardless of what you think of his stance on climate chance, he is by far one of the best forecasters there is. His forecasts absolutely destroy the "official" NOAA/NWS(especially the last few years) forecasts and they are done long before they put theirs out.

What is his take on the El Nino?

 

[uppermaclion]

What is his take on the El Nino?

He does a saturday summary on the weatherbell site. I highly recommend you watch that.

I'm not sure what aspect of his take on the el nino you want. I give my recollections of what he said. Early on when the models were predicting how strong the el nino would be, he said no way and said it would end up pretty close to the intensity it's going to be. He says there is a lag in global temperatures with the el nino, so next year you should see the big spike in temps. he expects the temperatures to fall after that(like they typically do). He is also predicting a return to a cold AMO/PDO in the next several years. That's when he'll know if he's right or wrong about global warming.

 

[hagberg]

The interviewee in the below National Geographic interview from 2013 mentions the lack of computing power. I wish the interview were a little longer.

LINK: Why are Europeans Better Predicting Weather?

It was a lot longer, but they didn't have the computing power to transcribe the whole thing.

 

[psuro]

What a disgrace the US models getting owned by the King EURO once again. All of these upgrades to the GFS and it still can't hold the EURO's jock..

I blame Jay.

 

[mmp121]

Didn't want to go there, but from the interview:



What would it take for the U.S. to have a model as good as the European one?

It would be very easy and it's outrageous it hasn't been done. We need more computer power for the National Weather Service. One way to do that would be to repurpose one of the supercomputers that NOAA uses for climate change research. Another would be to buy a new system. Why not use some of the $50 billion from the Hurricane Sandy relief bill?

TJ?

Seems like NOAA is already getting upgraded super computers.

http://www.noaanews.noaa.gov/stories2015/20150105_supercomputer.html

If things go to plan, the upgraded computing capacity should be coming online this month (10 fold increase) from what was there before. I wonder how long before we start seeing better predictions.

 

[mn78psu83]

Seems like NOAA is already getting upgraded super computers.

http://www.noaanews.noaa.gov/stories2015/20150105_supercomputer.html

If things go to plan, the upgraded computing capacity should be coming online this month (10 fold increase) from what was there before. I wonder how long before we start seeing better predictions.

Would really like to see the correlation between more computing power and better forecasting demonstrated. I wonder if better physics, organization and less political interference are also at work here.

 

[uppermaclion]

Would really like to see the correlation between more computing power and better forecasting demonstrated. I wonder if better physics, organization and less political interference are also at work here.

There's two things I have heard, and I'm only doing as a casual observer of the model(so that means I have no formal education or anything in meteorology). Just stuff I have read/heard etc.

1. The phyics is better with euro. I know alot of people complain about the GFS having feedback issues an stuff pertaining to various weather features. I consider this a "physics" problem more then anything. Now the Euro does have some known issues as well, but overall it's much better.

2. The euro has a much finer resolution, which may be where the compute power comes into play(i'm totally speculating here). I think the weather channel stated they were launching more balloons and stuff this week so that then give the models a better resolution of the area of concern(the Eastern US and Atlantic). so the euro probably starts off with much better/more intial conditions then the other models. Though I would think the same info is fed into all models so I may be totally wrong here.

The real question in my mind is why are the speciality hurricane models so bad? At least the GFS seems to eventually gravitate towards a similar solution then the euro, but the speciality hurricane models should be all over this first and they are not. That is what's really odd. .

 

[bwifan]

Geez I go away to do some research and forecasting and come back to all this...
Without getting all geeky on you yes the euro is a much higher resolution model. Plus especially in this type of pattern the euro's algo's are much better and handle the phasing in the trough so much better. Same reason why for the most part in a traditional zonal flow the gfs does better it's algo's handle that pattern better. The GFS is usually used as a much longer range model meaning 7+ days out to see how the pattern is setting up and possibility of any storms. Once we get inside the 7 day range to the 3-7 day range most of us switch to riding the euro with its high resolution. While still watching the gfs to see what it is saying. Then once we get to 48 hours there are numerous short range models we use as well as nowcast meaning using radar/satellite/water vapor etc to make a forecast. This is a very brief synopsis in a nutshell.... oh and the euro holds serve OTS again.... back to work....

 

[bwifan]

quick additional update hwrf model which is a hurricane based model actually has done ok.... it's been spot on with projecting and guidance on this storm's strength and intensity. It's path was off but just corrected and now has it OTS as well. We are now a Cat 4 hurricane. It is a beast out there in the ocean. The bahamas are going to most likely see some massive destruction from this thing.

 

[mmp121]

I'm going to go out on a limb here with all the talk about compute power leading to more accurate modeling / prediction of the weather.

I didn't sleep at a Holiday Inn or anything, and I don't play a meteorologist on TV. However, being an engineer, I can sympathize with the lack of compute power and being unable to accurately model things in a fixed time window.

Lets say we had a 100% accurate model, which took 48 hours to run on supercomputer A.
Obviously we couldn't get tomorrow's forecast out today as we wouldn't have the 100% accurate data out for 48 hours.

If we could drop the accuracy to 80%, it would take 12 hours to run on same supercomputer A, then we may have a fighting chance of putting something out for tomorrow, but with 20% uncertainty.

I think that in a nutshell that's what's wrong with the GFS (Upgraded) model. They have to run it at a less accurate setting (less 'grid' points) to have it run in a meaningful time period.

I don't know if the Euro model (ECMWF) is more efficient, or they have a larger more powerful supercomputer to be able to run it with more grid points, but it seems like they are able to run their model with more grid points which increases the model's accuracy.

 

[bwifan]

I'm going to go out on a limb here with all the talk about compute power leading to more accurate modeling / prediction of the weather.

I didn't sleep at a Holiday Inn or anything, and I don't play a meteorologist on TV. However, being an engineer, I can sympathize with the lack of compute power and being unable to accurately model things in a fixed time window.

Lets say we had a 100% accurate model, which took 48 hours to run on supercomputer A.
Obviously we couldn't get tomorrow's forecast out today as we wouldn't have the 100% accurate data out for 48 hours.

If we could drop the accuracy to 80%, it would take 12 hours to run on same supercomputer A, then we may have a fighting chance of putting something out for tomorrow, but with 20% uncertainty.

I think that in a nutshell that's what's wrong with the GFS (Upgraded) model. They have to run it at a less accurate setting (less 'grid' points) to have it run in a meaningful time period.

I don't know if the Euro model (ECMWF) is more efficient, or they have a larger more powerful supercomputer to be able to run it with more grid points, but it seems like they are able to run their model with more grid points which increases the model's accuracy.

It's not just computing power... The real fix is in fixing the science/physics behind it and the algo's.... They have enough computing power it's more the behind the scenes that needs to be fixed and made better.

 

[Woodpecker]

The sad truth is that it's NOT the model that prevented the hurricane from heading this way - it's ME! Yes, I am responsible since I just fired up my generator and it's all set for a disaster. The Fates noted this and promptly ordered a right turn for Joaquin.

You're welcome!

 

[LafayetteBear]

"cone of uncertainty" - great phrase.

"Cone of Uncertainty." Awesome. BTW, this thread is nirvana. If you're a weather geek.