Here’s what the world’s most accurate weather model predicts for Irma

A potentially catastrophic hurricane will approach southern Florida this weekend.

Read the whole story

 

[niwax]

I don't want to respond to specific comments, because this is more of a general thing I've been seeing on the internet, but regarding climate change:

Climate change doesn't really cause hurricanes, by they can certainly amplify them. That still doesn't necessarily mean we know that a big storm is climate change related or not, either. CC could amp a small storm to a slightly larger storm, while a huge storm happens outside of CC influence. Individual storms are not clear indicators, and CC normally involves looking at a large sample over time.

That doesn't mean it's impossible to understand CC's impact on an individual storm, either. There's already some data to suggest that Harvey was significantly amplified by CC, but it's still wise to wait before making bold statements on the internet.

Plus, there's another issue of climate change/negative human impact. The flooding in Houston was certainly a man made problem, regardless of CC's impact on the storm. They knew it could happen and the issues was ignored by the powers that be. IIRC, Florida is not much better in how cities are designed.

What we do know is that the entire region is warmer than usual right now. Which means weather like this will keep occurring if that temperature becomes normal. The current hurricanes are not a direct definite result of long-term climate change, but we are getting a preview of what hurricane season will look like with oceans at +1°:

 

[LetterRip]

... and to think that all I wanted to know is if Irma will wipe Mar A Largo off the map.
"Global warming is a Chinese misinformation."
"Oh yeah? How does that misinformation FEELS like now?"

Except that if you think that Global Warming has played a significant role in anything to do with these two hurricanes, you would be wrong.

I read it, he argues that .5-1 degree of SST (sea surface temperature anomaly) resulted in about a 3.5% increase in rain amount - so about 1 inch out of 30 inches of rain due to the Clausius Clperyron relationship (aka CC scaling). Of course if you go with the upper range - that is 2.1 inches.

This ignores SuperCC/2xCC scaling that can happen in extreme weather events.

http://journals.ametsoc.org/doi/pdf/10. ... -12-0317.1

So now we are at 4.2 inches.

He isn't convinced that that stalling can be attributable to AGW. I think he ignores that the westerlies have shifted northward as has the ITCZ

https://www.gfdl.noaa.gov/bibliography/ ... rt0901.pdf

https://www.livescience.com/5514-tropic ... north.html

So, yes I've read it, I just think he ignores facts that are inconvenient to his narrative.

 

[Random John Smith Guy]

And here I am in South Carolina hoping this doesn't smash here.

In Maryland, know I will get rain but hopefully this isn't going to be a repeat of Labor Day 1935.

 

[mainmac]

... and to think that all I wanted to know is if Irma will wipe Mar A Largo off the map.
"Global warming is a Chinese misinformation."
"Oh yeah? How does that misinformation FEELS like now?"

Except that if you think that Global Warming has played a significant role in anything to do with these two hurricanes, you would be wrong.

Not to say Global warming isn't happening, or human caused, but rather in this case it's not anywhere near a major contributor to the intensity, timing, track, etc of these particular storms

Edit: have to wonder if any of those downvoting bothered to click on, much less actually read the linked info from a well respected expert in atmospheric sciences, weather forecasting, and climate.

Yes and no. While global warming didn't cause the 'stalling' over Houston, the increase in the number of high intensity hurricanes means Houston (or some other major population) was at better odds of having a hurricane stall over it.

 

[chudan]

If you want to see what's going on with I-95 and the Florida Turnpike in regards to evacuation, just search 'I-95 traffic cam miami' or 'florida turnpike traffic cam miami' (also, can use palm beach, ft lauderdale, orlando etc as city).

 

[nedscott]

I don't want to respond to specific comments, because this is more of a general thing I've been seeing on the internet, but regarding climate change:

Climate change doesn't really cause hurricanes, by they can certainly amplify them. That still doesn't necessarily mean we know that a big storm is climate change related or not, either. CC could amp a small storm to a slightly larger storm, while a huge storm happens outside of CC influence. Individual storms are not clear indicators, and CC normally involves looking at a large sample over time.

That doesn't mean it's impossible to understand CC's impact on an individual storm, either. There's already some data to suggest that Harvey was significantly amplified by CC, but it's still wise to wait before making bold statements on the internet.

Plus, there's another issue of climate change/negative human impact. The flooding in Houston was certainly a man made problem, regardless of CC's impact on the storm. They knew it could happen and the issues was ignored by the powers that be. IIRC, Florida is not much better in how cities are designed.

What we do know is that the entire region is warmer than usual right now. Which means weather like this will keep occurring if that temperature becomes normal. The current hurricanes are not a direct definite result of long-term climate change, but we are getting a preview of what hurricane season will look like with oceans at +1°:

Indeed, and I think Azethoth666's post explains it far better than I did.

 

[azazel1024]

I wouldn't depend on being able to get out of South Florida on Friday either. I'm sure I-95 will be (already is?) a parking lot going north.

Take the Turnpike...

At this time the safest way to get out of South Florida is by submarine....

Depends on who the captain is.

https://meincmagazine.com/information-tec ... ters-head/

 

[uhuznaa]

... and to think that all I wanted to know is if Irma will wipe Mar A Largo off the map.
"Global warming is a Chinese misinformation."
"Oh yeah? How does that misinformation FEELS like now?"

Except that if you think that Global Warming has played a significant role in anything to do with these two hurricanes, you would be wrong.

Not to say Global warming isn't happening, or human caused, but rather in this case it's not anywhere near a major contributor to the intensity, timing, track, etc of these particular storms

Edit: have to wonder if any of those downvoting bothered to click on, much less actually read the linked info from a well respected expert in atmospheric sciences, weather forecasting, and climate.

Yes and no. While global warming didn't cause the 'stalling' over Houston, the increase in the number of high intensity hurricanes means Houston (or some other major population) was at better odds of having a hurricane stall over it.

Arguing about if a specific weather event was caused by the climate changing is like arguing about a specific cigarette (of a lot) someone smoked at some point having caused his lung cancer. There's nothing you can prove here, but this does not mean that there is no connection. You have to look at the bigger picture and at probabilities. Use numbers, not specific events.

 

[Faanchou]

I don't want to respond to specific comments, because this is more of a general thing I've been seeing on the internet, but regarding climate change:

Climate change doesn't really cause hurricanes, by they can certainly amplify them. That still doesn't necessarily mean we know that a big storm is climate change related or not, either. CC could amp a small storm to a slightly larger storm, while a huge storm happens outside of CC influence. Individual storms are not clear indicators, and CC normally involves looking at a large sample over time.

That doesn't mean it's impossible to understand CC's impact on an individual storm, either. There's already some data to suggest that Harvey was significantly amplified by CC, but it's still wise to wait before making bold statements on the internet.

Plus, there's another issue of climate change/negative human impact. The flooding in Houston was certainly a man made problem, regardless of CC's impact on the storm. They knew it could happen and the issues was ignored by the powers that be. IIRC, Florida is not much better in how cities are designed.

What we do know is that the entire region is warmer than usual right now. Which means weather like this will keep occurring if that temperature becomes normal. The current hurricanes are not a direct definite result of long-term climate change, but we are getting a preview of what hurricane season will look like with oceans at +1°:

That is true. While the temperature difference may not rise enough to produce predictable changes, the increased surface temperatures mean that the hurricanes that do form can now carry significantly more moisture. +1°C means up to +6.2% in water content, +6°C takes it over +40%.

Source: http://www.engineeringtoolbox.com/moist ... d_281.html

 

[raxx7]

Therefore the big question facing forecasters heading into the afternoon hours was this: What would the European model do with the 12z run?

That question seems premature before explaining what "12z" means.....?

12z means the model run was initiated at 12:00 UTC or 8am ET.

IIRC 12z stands for Zulu time. Which is UTC, and if you're inbetween when Zulu Time and UTC were used commonly, you probably know it as GMT or Greenwich time.

That would be a time line of WWII , GMT being sometime in the 50's and UTC in the 2000's? I am probably wrong on those time frames though, been a long time since I read on this.

GMT (mean astronomical time as observed in the Greenwich Observatory) was used by the UK as reference since 1801 at least and was universally adopted as reference in 1884.

UTC (atomic time plus an integer number of leap seconds which such that |UTC-GMT|<0.9) was invented in 1961.

"Zulu time" has switched from GMT to UTC as the later replaced the former.

 

[fatgeek]

Here's a related good story from NBC: https://www.nbcnews.com/storyline/hurri ... ge-n798961.

I'd love to read a technical article on Ars about the hardware and software running in these supercomputers, and what kinds of mathematical equations require the use of such supercomputers in hurricane prediction.

At some point I'll get around to writing about this in detail, and what NOAA is doing to try and catch up.

am I wrong, always thought it was a computational issue? they are throwing way more power at it?

It is a computational issue, but that is only a small subset of the problem as I understand it. But I need to dig into the issue more deeply before writing authoritatively.

Cliff Mass at U of Washington has some definite opinions on the subject and has written about it more than once: http://cliffmass.blogspot.com/2016/07/to-become-world-class-us-operational_13.html

 

[Veritas super omens]

Therefore the big question facing forecasters heading into the afternoon hours was this: What would the European model do with the 12z run?

That question seems premature before explaining what "12z" means.....?

12z means the model run was initiated at 12:00 UTC or 8am ET.

IIRC 12z stands for Zulu time. Which is UTC, and if you're inbetween when Zulu Time and UTC were used commonly, you probably know it as GMT or Greenwich time.

That would be a time line of WWII , GMT being sometime in the 50's and UTC in the 2000's? I am probably wrong on those time frames though, been a long time since I read on this.

GMT (mean astronomical time as observed in the Greenwich Observatory) was used by the UK as reference since 1801 at least and was universally adopted as reference in 1884.

UTC (atomic time plus an integer number of leap seconds which such that |UTC-GMT|<0.9) was invented in 1961.

"Zulu time" has switched from GMT to UTC as the later replaced the former.

All of which are soon to be replace by RET if the orange tabby grabber has his way....


RET= Rapture End Times

 

[CraigJ ✅]

... and to think that all I wanted to know is if Irma will wipe Mar A Largo off the map.
"Global warming is a Chinese misinformation."
"Oh yeah? How does that misinformation FEELS like now?"

Except that if you think that Global Warming has played a significant role in anything to do with these two hurricanes, you would be wrong.

Not to say Global warming isn't happening, or human caused, but rather in this case it's not anywhere near a major contributor to the intensity, timing, track, etc of these particular storms

Edit: have to wonder if any of those downvoting bothered to click on, much less actually read the linked info from a well respected expert in atmospheric sciences, weather forecasting, and climate.

Global warming = more energy in the ocean and atmosphere. In this case more energy = more heat.
More energy means potentially more frequent, and more intense storms.

It's a fact that warmer water = more intense hurricanes

NASA have a surprisingly entertaining article Recipe for a Hurricane (their stuff is usually pretty dry) that looks at the conditions necessary for a hurricane to form. http://www.nasa.gov/vision/earth/enviro ... ECIPE.html

However, NASA are outdone by the BBC who've produced an attractive animated guide to hurricane formation. http://news.bbc.co.uk/2/hi/science/nature/4588149.stm

An instructive summary of the science is Hurricanes and Global Warming - Is There a Connection? by Real Climate. http://www.realclimate.org/index.php/ar ... l-warming/

Climate Progress discusses the physical reasons why global warming leads to more intense hurricanes. http://climateprogress.org/2008/08/31/w ... av-part-1/

Now to accept this you must first accept the fact that Global Warming is real. You don't need to accept AGW to accept that GW is causing more intense storms. (but you should).

 

[redbike]

Why is the European model proprietary

Good question, to which I have no direct answer.

Explore ECMWF's website, and you'll find this page

https://www.ecmwf.int/en/forecasts/accessing-forecasts

Yes, there's a level of public access, but as compared to NWS / NOAA, it's meager.

As a mere mortal, to view to ECMWF's Caribbean Model, I use

http://www.myfoxhurricane.com/custom/mo ... bbean.html

This is a Fox Television Stations, Inc. website. Presumably, Fox has paid the ECMWF fee.

In addition to the ECMWF graphic toward the top of the page, there are more neat graphics toward the bottom.

 

[real mikeb_60]

If you want to see what's going on with I-95 and the Florida Turnpike in regards to evacuation, just search 'I-95 traffic cam miami' or 'florida turnpike traffic cam miami' (also, can use palm beach, ft lauderdale, orlando etc as city).

Or just pull up Google map of FL and turn on the traffic layer.

 

[Azethoth666]

Here's a related good story from NBC: https://www.nbcnews.com/storyline/hurri ... ge-n798961.

I'd love to read a technical article on Ars about the hardware and software running in these supercomputers, and what kinds of mathematical equations require the use of such supercomputers in hurricane prediction.

At some point I'll get around to writing about this in detail, and what NOAA is doing to try and catch up.

am I wrong, always thought it was a computational issue? they are throwing way more power at it?

Excuse me, no : the explanation for this is clear. It is well known that European academics value math and science much more than USAnians do ; and their students excel in it far better.

We are seeing the results of lazy American disregard for the sciences at work here.

I like your jingoism. It is so adorably parochial! It may even be true for K12.

However, you will need to explain why we statistically get all the Nobel prizes, host all the top universities, have the biggest economy, etc.

If you want to point to financial neglect as the culprit I can agree with you. All else is bullshit.

 

[chanfle98]

Hey genius, Hillary won by 30 points in Miami. We are not all idiots down here.

Hey Miami!

You can meditate on your vote for anti climate change Donald Trump while your homes are being destroyed. And you can look forward to more of the same in the future.

 

[ihatewinter]

Since every 'prediction' shows uncertainty on Sunday (72 hours) about the 'turn' (or lack of one or the severity of that 'turn') from northwest to north track, I'll watch with interest on this specific
'prediction'. Because that 'turn' could be crucial to future events that will transpire right quick depending on which side of the Florida peninsula one chooses to evacuate and to where.

 

[ihatewinter]

Just saw on the weather channel that the European model which has been the most accurate up to date shows it going into the gulf and up the west coast of Florida.

You should not base articles on information that is in flux.

Agree wholeheartedly

 

[uhuznaa]

I think this will be the most watched storm in history... all over the planet.

 

[ihatewinter]

I think this will be the most watched storm in history... all over the planet.

That's not exactly a brainstorm. More eyes, more means of enabling eyes to watch, more watching.

 

[Azethoth666]

... and to think that all I wanted to know is if Irma will wipe Mar A Largo off the map.
"Global warming is a Chinese misinformation."
"Oh yeah? How does that misinformation FEELS like now?"

Except that if you think that Global Warming has played a significant role in anything to do with these two hurricanes, you would be wrong.

I read it, he argues that .5-1 degree of SST (sea surface temperature anomaly) resulted in about a 3.5% increase in rain amount - so about 1 inch out of 30 inches of rain due to the Clausius Clperyron relationship (aka CC scaling). Of course if you go with the upper range - that is 2.1 inches.

This ignores SuperCC/2xCC scaling that can happen in extreme weather events.

http://journals.ametsoc.org/doi/pdf/10. ... -12-0317.1

So now we are at 4.2 inches.

He isn't convinced that that stalling can be attributable to AGW. I think he ignores that the westerlies have shifted northward as has the ITCZ

https://www.gfdl.noaa.gov/bibliography/ ... rt0901.pdf

https://www.livescience.com/5514-tropic ... north.html

So, yes I've read it, I just think he ignores facts that are inconvenient to his narrative.

You seem like an actual climate scientist. Is this correct or are you just googling?

You say we can go with the upper range, but we can not. He is already assuming 100% of the increase is man made. The IPCC consensus is around 50% us, 50% natural. I know they are conservative but I have no feel for the real breakdown. Certainly it is not 100%.

So properly we would halve and then double due to the 2CC? Same result, but we are no longer looking at an overestimation but something statistically closer to reality?

Finally, is this guy a climate denier? I did not get that sense from the post, but I did not bother to dig deeper.

PS I forgot to address the stalling. Based on what you posted he really dropped the ball on that part.

 

[siliconaddict]

I'm pissed as fuck. I've been telling my friends to GTFO who are down there. Most have already packed the car and plan to head out Friday at 2AM. But I have a couple friends who can't leave. Their boss's have threatened to fire them if they leave before an official evacuation occurs. Right. So lets wait until the roads are congested to get out. Dock their pay if anything. But firing just so you can make sure your ass is safe. sounds like whoever they work for they need to get a new job anyways.

 

[normally butters]

Palm Beach county has warned people to stay put and shelter in place, as gasoline supplies are insufficient to support evacuation.

 

[Router66]

... and to think that all I wanted to know is if Irma will wipe Mar A Largo off the map.
"Global warming is a Chinese misinformation."
"Oh yeah? How does that misinformation FEELS like now?"

Except that if you think that Global Warming has played a significant role in anything to do with these two hurricanes, you would be wrong.

I read it, he argues that .5-1 degree of SST (sea surface temperature anomaly) resulted in about a 3.5% increase in rain amount - so about 1 inch out of 30 inches of rain due to the Clausius Clperyron relationship (aka CC scaling). Of course if you go with the upper range - that is 2.1 inches.

This ignores SuperCC/2xCC scaling that can happen in extreme weather events.

http://journals.ametsoc.org/doi/pdf/10. ... -12-0317.1

So now we are at 4.2 inches.

He isn't convinced that that stalling can be attributable to AGW. I think he ignores that the westerlies have shifted northward as has the ITCZ

https://www.gfdl.noaa.gov/bibliography/ ... rt0901.pdf

https://www.livescience.com/5514-tropic ... north.html

So, yes I've read it, I just think he ignores facts that are inconvenient to his narrative.

You seem like an actual climate scientist. Is this correct or are you just googling?

You say we can go with the upper range, but we can not. He is already assuming 100% of the increase is man made. The IPCC consensus is around 50% us, 50% natural. I know they are conservative but I have no feel for the real breakdown. Certainly it is not 100%.

So properly we would halve and then double due to the 2CC? Same result, but we are no longer looking at an overestimation but something statistically closer to reality?

Finally, is this guy a climate denier? I did not get that sense from the post, but I did not bother to dig deeper.

PS I forgot to address the stalling. Based on what you posted he really dropped the ball on that part.

Of course he can use the upper range. According to cliffmass, the .5C temperature increase corresponds to an increase of 1 inch of rain. It follows that 1C --> 2 inches. Nothing to do with anthropogenic causes or lack of them.

Also... where are the maths behind cliffmass' claim? Honest question.

 

[wyrmhole]

You seem like an actual climate scientist. Is this correct or are you just googling?

You say we can go with the upper range, but we can not. He is already assuming 100% of the increase is man made. The IPCC consensus is around 50% us, 50% natural. I know they are conservative but I have no feel for the real breakdown. Certainly it is not 100%.

50% is the very low end of likely human attributions. It certainly could be 100%, as that is also within the range of likely human attributions. Greater than 100% is as well -- implying that the natural signal is cooling.

 

[Vimon]

the European Center for Medium-Range Weather Forecasts has the best forecast model in the world

The EU might have the best model but when it comes to naming stuff we are way behind the US. The ECMRWF... what kind of name is that? Try to say it out loud... you can't!
I'm sure that if it was a US based thing it would be named something like All Weather Encompassing System Online, Micdrop, Exitstage or for short AWESOME!

Also good luck to you in FL, sorry the EU model says you need it.

 

[jdale]

Just saw on the weather channel that the European model which has been the most accurate up to date shows it going into the gulf and up the west coast of Florida.

You should not base articles on information that is in flux.

Based on your last sentence, no one should write about any storm until it is over. That's not very useful. Predictions may be imperfect but our current forecasts are far better than nothing, and knowing their strengths and limitations is useful too. Not to mention interesting.

Certainly no one should read one single article to decide whether they are or are not going to be affected, and then ignore all subsequent news.

 

[JustUsul]

Just saw on the weather channel that the European model which has been the most accurate up to date shows it going into the gulf and up the west coast of Florida.

You should not base articles on information that is in flux.

Odd. Looking at tropicaltidbits.com, the ECMWF Wednesday 12z run appears to show it going up Florida's east coast. This is the same run that Eric discusses in this article. And given that the next 00z run won't be released until 2:30AM EST, I'm not sure how the Weather Channel has such advanced knowledge?

 

[JustUsul]

the European Center for Medium-Range Weather Forecasts has the best forecast model in the world

The EU might have the best model but when it comes to naming stuff we are way behind the US. The ECMRWF... what kind of name is that? Try to say it out loud... you can't!
I'm sure that if it was a US based thing it would be named something like All Weather Encompassing System Online, Micdrop, Exitstage or for short AWESOME!

Also good luck to you in FL, sorry the EU model says you need it.

This has been my biggest complaint about weather and climate models; I can never keep them straight, because the acronyms never spell out anything meaningful! At least astronomers trying to keep their acronyms memorable. For instance, see
Dumb Or Overly Forced Astronomical Acronyms Site (or DOOFAAS).

 

[ihatewinter]

Just saw on the weather channel that the European model which has been the most accurate up to date shows it going into the gulf and up the west coast of Florida.

You should not base articles on information that is in flux.

Based on your last sentence, no one should write about any storm until it is over. That's not very useful. Predictions may be imperfect but our current forecasts are far better than nothing, and knowing their strengths and limitations is useful too. Not to mention interesting.

Certainly no one should read one single article to decide whether they are or are not going to be affected, and then ignore all subsequent news.

The point is, that every 'prediction' had an uncertainty cone at the Sunday (72 hour mark at the time of prediction) and that no 'prediction' can be made with any certainty until the conditions come clearer the nearer we get to that mark. Since Florida is a peninsula that uncertainty is huge.
Downvotes don't change that reality.

 

[drtrout]

ECMWF used to be better, but now GFS has caught up. See Figure 6 in the latest National Hurricane Center validation report at: http://www.nhc.noaa.gov/verification/pd ... n_2016.pdf .

As a meteorologist, the way I use both models is if one is an outlier from other models, you have to weight your forecast towards the outlier. Either GFS or ECMWF may have picked up something the others missed. In this case, they weighed towards ECMWF, but they probably would have done the same for GFS it is was an outlier.

HWRF is very close in skill, too, which had many upgrades the last two years. Model improvements evolve quickly.

On another topic, I wouldn't give ECMWF too much credit. As you mentioned, they only do forecasts twice a day while GFS does it four times per day. And, ECMWF does not share much of their data and charges for the rest, which I think really stinks when the US is very generous with all its weather products. It gives the ECMWF a cost-edge.

 

[rex86]

Why is the European model proprietary, and exactly what does that mean in this context? Does it mean they only provide data how the model predict a system will move and not the entire solution?

I think that in this case it means the results are made available to subscribers, and not made public domain as with of lot of the other models.

although it might also refer to the algorithms themselves and who owns them, in this context however I believe he was referring to the data being produced.

The data are freely available to the member states of the project, ECMWF is a joint European project funded by close to 40 European states, and so the bestest and newest results are available to them. Some archival data sets are available to the public but if you're not a member state then you have to pay to get the newest data.

I don't know if that's also applicable to the code, but it's quite possible that the code is not open to the public because of competition, but then maybe it's because of that massive supercomputing cluster they're using to run the model on. I'm sure the power bill is very hot.

 

[Vimon]

ECMWF used to be better, but now GFS has caught up. See Figure 6 in the latest National Hurricane Center validation report at: http://www.nhc.noaa.gov/verification/pd ... n_2016.pdf . As a meteorologist, the way I use both models is if one is an outlier, you have to weight your forecast towards the outlier because these are the best two and one may have picked up something the other missed. HWRF is very close in skill, too, which had many upgrades the last two years.

Model improvements evolve quickly.

The graph you point out doesnt have a ECMWF or GFS marker. Do you mind pointing out which points you are refering to? Because Fig 6 in your link doesn't include any of your shorthands which means we are left to assumptions. Given that you are a scientist you wouldn't want that. Therefor please point out the "lines" you are refering to.

OFCL,GFSI,GHMI,HWFI,EMXI,...

 

[Upsidedownjustice]

Punkin head says "fake news"