Toggle sidebar

OT: upgrade in weather forecasting software

M

MorrisAsh

All-Conference
Dec 5, 2015
2,635
4,428
113
R

RUScrew85

Heisman
Nov 7, 2003
30,054
16,939
0
 
  • Like
Reactions: T2Kplus10, WhiteBus and rurichdog
R

RU848789

Heisman
Jul 27, 2001
65,390
44,413
113
MorrisAsh said:
www.nytimes.com/2019/06/12/climate/noaa-weather-forecast-gfs.html?action=click&module=MoreInSection&pgtype=Article&region=Footer&contentCollection=Climate%20and%20Environment

Aiming to reduce errors like the one it made in 2012, when it wrongly forecast the track of Hurricane Sandy into the New York area, the National Oceanic and Atmospheric Administration on Wednesday introduced a major upgrade to the software at the heart of its weather prediction capability.
Click to expand...

The "new" GFS-FV3 model has been running for the past year or so and yes, it's overall scores are better than the old GFS, by a little, but strong anecdotal evidence in the meteorological community is that it did worse than the old GFS and other models this past winter in winter storm situations, regularly overpredicting snowfall. Will try to dig some of that up later (one guy on the weather boards did a bit of an analysis, iirc). They supposedly made some more "tweaks" since this winter - we'll see, I guess...
 
R

RU848789

Heisman
Jul 27, 2001
65,390
44,413
113
By the way, if you like heavy rain, you're going to love the next 3 days. Lots of tropical moisture and a stalled front will combine for quite a bit of rain. Flash flood warning just issued for NE Mercer, SE Somerset and Central Middlesex up to just about Metuchen. Up to 1-2" more rain to come in the next few hours.

https://forecast.weather.gov/wwamap/wwatxtget.php…
 
R

RU848789

Heisman
Jul 27, 2001
65,390
44,413
113
RU848789 said:
By the way, if you like heavy rain, you're going to love the next 3 days. Lots of tropical moisture and a stalled front will combine for quite a bit of rain. Flash flood warning just issued for NE Mercer, SE Somerset and Central Middlesex up to just about Metuchen. Up to 1-2" more rain to come in the next few hours.

https://forecast.weather.gov/wwamap/wwatxtget.php…
Click to expand...
And a severe t-storm warning for much of the same areas, extending into Monmouth, with a watch for counties essentially south of the Raritan...

https://forecast.weather.gov/showsi...m+Watch&lat=40.3766&lon=-74.4244#.XQkrD7xKjDc
 
R

RU848789

Heisman
Jul 27, 2001
65,390
44,413
113
Posting from our huge work BBQ (1500 people). Rain miraculously stayed just north of Rahway from 9-11:30 am during setup and from 11:30-1:30 for the first 2 hours of the 3 hour event; most of that time it was pouring 2-5 miles N of here. We did have tents for 500+, but rain would've killed it. Had a couple of sprinkles from 1:00-1:30 pm and then it has rained pretty heavily since then.

Kind of funny to see everyone hanging out under the tents and then those who forgot umbrellas walking away under the plastic tablecloths being used as rain gear. Luckily I'm under a gazebo with the equipment and everything is dry. Listening to my new playlist is still way better than working, lol. Could've been way, way worse.
 
A

anon_ivydyf0amkzay

Guest
I’m glad we are finally stepping up to Europe circa 2005! An old USAF buddy of mine who works in numerical modeling always laments how far the 8 ball we are compared to the euros...it becomes especially clear to him when he travels to The Hadley Center...
 
R

Retired711

Heisman
Nov 20, 2001
19,971
10,151
58
RU848789 said:
Rumor has it we're bringing back the ETA, lol...
Click to expand...

I distinctly remember that you posted last winter that the new GFS wasn't much better than the old one, so I wasn't excited by the Times story. Why do you think the GFS wasn't improved every much? Is it a funding problem?
 
R

RU848789

Heisman
Jul 27, 2001
65,390
44,413
113
hrdcorhays said:
@RU848789 Did you read Michael Lewis's book The Fifth Risk? The parts about Barry Myers and NOAA are so disgusting.
Click to expand...
No, but it's on my list. Love Lewis's books. But am very familiar with Myers and he's just one more example of a horrific appointment of a major government agency by Trump. The man is not even a scientist and has advocated limiting NOAA to data gathering for private companies like AccuWeather, so they can sell their forecast products, which is the total antithesis of what is in the public's best interest. Not to mention he oversaw a company with one of the worst records of sexual harrassment anywhere. The opinion piece below, by a former NOAA Administrator, is scathing - and several other former NOAA Administrators are also against his appointment.

https://www.nytimes.com/2019/05/01/opinion/the-senate-should-reject-trumps-noaa-nominee.html
 
  • Like
Reactions: hrdcorhays
T

T2Kplus10

Heisman
Feb 24, 2010
28,180
17,566
0
MorrisAsh said:
www.nytimes.com/2019/06/12/climate/noaa-weather-forecast-gfs.html?action=click&module=MoreInSection&pgtype=Article&region=Footer&contentCollection=Climate%20and%20Environment

Aiming to reduce errors like the one it made in 2012, when it wrongly forecast the track of Hurricane Sandy into the New York area, the National Oceanic and Atmospheric Administration on Wednesday introduced a major upgrade to the software at the heart of its weather prediction capability.
Click to expand...
Weather forecasting is abysmal, so it can't get any worse.
 
T

truth_KB

Redshirt
Jul 7, 2005
12
5
0
Lewis never lets anyone get in the way of whatever story he is writing/peddling to his readers. Definition of no-fact checkers he should have all of his books listed as FICTION, as he never lets the truth or facts get in the way of his current mission.
 
R

RU848789

Heisman
Jul 27, 2001
65,390
44,413
113
T2Kplus10 said:
Weather forecasting is abysmal, so it can't get any worse.
Click to expand...
Trolls gonna troll. I let many more of your troll posts go than I used to, but I will point out the idiocy of your latest effort, just so others know that you're trolling. Below is an excerpt from an Atlantic article from earlier this year and the link to that article, as well as a link to the source article from a paper in Science.

Meteorologists have never gotten a shiny magazine cover or a brooding Aaron Sorkin film, and the weather-research hub of Norman, Oklahoma, is rarely mentioned in the same breath as Palo Alto. But over the past few decades, scientists have gotten significantly—even staggeringly—better at predicting the weather.

How much better? “A modern five-day forecast is as accurate as a one-day forecast was in 1980,” says a new paper, published last week in the journal Science. “Useful forecasts now reach nine to 10 days into the future.”

The paper is a birthday present from meteorology to itself: The American Meteorological Society turns 100 this year. But it also acts as a good report card on how far weather prediction has come.

“Modern 72-hour predictions of hurricane tracks are more accurate than 24-hour forecasts were 40 years ago,” the authors write. The federal government now predicts storm surge, stream level, and the likelihood of drought. It has also gotten better at talking about its forecasts: As I wrote in 2017, the National Weather Service has dropped professional jargon in favor of clear, direct, and everyday language.

[URL]https://www.theatlantic.com/science/archive/2019/01/polar-vortex-weather-forecasting-good-now/581605/[/URL]

https://science.sciencemag.org/content/363/6425/342
 
R

RU848789

Heisman
Jul 27, 2001
65,390
44,413
113
camdenlawprof said:
I distinctly remember that you posted last winter that the new GFS wasn't much better than the old one, so I wasn't excited by the Times story. Why do you think the GFS wasn't improved every much? Is it a funding problem?
Click to expand...

Low US funding support certainly was an issue for a long time, as per the first linked article from 2015. However, I don't know the details of how much $$/effort was put into the new GFS-FV3 model to know if that gap was closed over the past couple of years. Certainly, the performance gap was only closed a bit, as the new GFS is now 3rd best instead of 4th best, as per the 2nd link, still behind the Euro and UK models, which is not where NOAA hoped it would be, although I haven't seen any updated performance data since the "tweaked" version went live. I have my doubts that tweaks will bring it to the performance level of the Euro.

https://www.scientificamerican.com/...better-than-americans-at-forecasting-storms1/

https://www.forbes.com/sites/marsha...l-wars-take-a-new-turn-in-march/#3e5af9266c2b
 
R

RU848789

Heisman
Jul 27, 2001
65,390
44,413
113
RU4Real said:
Click to expand...
Turns out you didn't Photoshop the cow in, lol - I actually checked and it's in the original trailer ad, which is funny, as no tornado strong enough to launch a cow into flight would have that cow clearly visible in the debris field. Hollywood.
 
R

RU4Real

Heisman
Jul 25, 2001
50,955
30,734
0
RU848789 said:
Turns out you didn't Photoshop the cow in, lol - I actually checked and it's in the original trailer ad, which is funny, as no tornado strong enough to launch a cow into flight would have that cow clearly visible in the debris field. Hollywood.
Click to expand...

I'm concerned that that's all you got out of that.
 
R

RU848789

Heisman
Jul 27, 2001
65,390
44,413
113
RU4Real said:
I'm concerned that that's all you got out of that.
Click to expand...
Don't worry - I got the obvious part. Now we need a winter weather weenie film - maybe some storm chasers stranded in a biblical lake effect snowstorm (The Day After Tomorrow covered that territory a little bit but that was more of a disaster film).
 
R

RU4Real

Heisman
Jul 25, 2001
50,955
30,734
0
RU848789 said:
Don't worry - I got the obvious part. Now we need a winter weather weenie film - maybe some storm chasers stranded in a biblical lake effect snowstorm (The Day After Tomorrow covered that territory a little bit but that was more of a disaster film).
Click to expand...

I was in Buffalo in '14. You would have been REALLY jealous.
 
T

T2Kplus10

Heisman
Feb 24, 2010
28,180
17,566
0
RU848789 said:
Trolls gonna troll. I let many more of your troll posts go than I used to, but I will point out the idiocy of your latest effort, just so others know that you're trolling. Below is an excerpt from an Atlantic article from earlier this year and the link to that article, as well as a link to the source article from a paper in Science.

Meteorologists have never gotten a shiny magazine cover or a brooding Aaron Sorkin film, and the weather-research hub of Norman, Oklahoma, is rarely mentioned in the same breath as Palo Alto. But over the past few decades, scientists have gotten significantly—even staggeringly—better at predicting the weather.

How much better? “A modern five-day forecast is as accurate as a one-day forecast was in 1980,” says a new paper, published last week in the journal Science. “Useful forecasts now reach nine to 10 days into the future.”

The paper is a birthday present from meteorology to itself: The American Meteorological Society turns 100 this year. But it also acts as a good report card on how far weather prediction has come.

“Modern 72-hour predictions of hurricane tracks are more accurate than 24-hour forecasts were 40 years ago,” the authors write. The federal government now predicts storm surge, stream level, and the likelihood of drought. It has also gotten better at talking about its forecasts: As I wrote in 2017, the National Weather Service has dropped professional jargon in favor of clear, direct, and everyday language.

https://www.theatlantic.com/science...r-vortex-weather-forecasting-good-now/581605/

https://science.sciencemag.org/content/363/6425/342
Click to expand...
Nobody read any of this post.
 
R

RU848789

Heisman
Jul 27, 2001
65,390
44,413
113
RU848789 said:
Trolls gonna troll. I let many more of your troll posts go than I used to, but I will point out the idiocy of your latest effort, just so others know that you're trolling. Below is an excerpt from an Atlantic article from earlier this year and the link to that article, as well as a link to the source article from a paper in Science.

Meteorologists have never gotten a shiny magazine cover or a brooding Aaron Sorkin film, and the weather-research hub of Norman, Oklahoma, is rarely mentioned in the same breath as Palo Alto. But over the past few decades, scientists have gotten significantly—even staggeringly—better at predicting the weather.

How much better? “A modern five-day forecast is as accurate as a one-day forecast was in 1980,” says a new paper, published last week in the journal Science. “Useful forecasts now reach nine to 10 days into the future.”

The paper is a birthday present from meteorology to itself: The American Meteorological Society turns 100 this year. But it also acts as a good report card on how far weather prediction has come.

“Modern 72-hour predictions of hurricane tracks are more accurate than 24-hour forecasts were 40 years ago,” the authors write. The federal government now predicts storm surge, stream level, and the likelihood of drought. It has also gotten better at talking about its forecasts: As I wrote in 2017, the National Weather Service has dropped professional jargon in favor of clear, direct, and everyday language.

https://www.theatlantic.com/science...r-vortex-weather-forecasting-good-now/581605/

https://science.sciencemag.org/content/363/6425/342
Click to expand...

T2Kplus10 said:
Nobody read any of this post, other than me, obviously, since I replied to it. Turns out, I'm actually obsessed with your weather posts, as you can probably tell, since I reply to every one of them within minutes of them being posted, usually. I just wanted to come clean, as my therapist said I should stop projecting my inadequacies onto others by trolling them on internet message boards. So consider this a start. I'm sorry.
Click to expand...

FIFY.
 
You must log in or register to reply here.
Top Bottom