The Long-Awaited Upgrade To the US Weather Forecast Model Is Here (arstechnica.com) 59
An anonymous reader quotes a report from Ars Technica: For the first time in about 40 years, the guts of the U.S. model got swapped out for something new today. The upgrade brings us a new "Finite-Volume Cubed-Sphere" (or FV3) dynamical core, which simulates the basic atmospheric physics at the heart of this endeavor, a change that has been in the works for a while. The new core had its origins in simulating atmospheric chemistry but ended up being adapted into other models. A few years ago, it was selected to replace the old core in the U.S. Global Forecast System model. And for more than a year now, the new version of the model has been running in parallel so its results could be compared to the operational model.
The results have been a little mixed. The new core improves computational efficiency and allows some processes to be simulated at a higher resolution -- unequivocal improvements. It also simulates the physics of water vapor more realistically. In a press conference today, NOAA scientists cited a number of areas where forecast improvements have been seen. Forecast tracks of hurricanes and the mid-latitude storms that frequently sweep across the U.S. have both improved, they said, along with forecasts of hurricane strength. Forecast precipitation amounts were also cited as a key area of progress. But there have also been grumblings in the weather community over the past year about results that didn't seem so hot. For example, surface temperatures have been biased low in some situations, throwing off forecasts.
The results have been a little mixed. The new core improves computational efficiency and allows some processes to be simulated at a higher resolution -- unequivocal improvements. It also simulates the physics of water vapor more realistically. In a press conference today, NOAA scientists cited a number of areas where forecast improvements have been seen. Forecast tracks of hurricanes and the mid-latitude storms that frequently sweep across the U.S. have both improved, they said, along with forecasts of hurricane strength. Forecast precipitation amounts were also cited as a key area of progress. But there have also been grumblings in the weather community over the past year about results that didn't seem so hot. For example, surface temperatures have been biased low in some situations, throwing off forecasts.
Re: If they are predictably off ... (Score:2, Insightful)
Maybe figure out why the model is off instead of creating potentially more errors.
Re: (Score:2)
They did adjust it. The thing about the low temps is from February.
FV3 simulation (Score:1)
https://www.youtube.com/watch?... [youtube.com]
Re: (Score:1)
Nancy's last name starts with a 'P' not a 'T'.
Re: (Score:3, Informative)
"Weather forecast model"
>How will the Climate Change activists react to real data on climate?
There's an extrupulous phrase I never heard before, dug up recently while I excavated old usenet archives. It's wisdom, the wiseness I wish to be engraved on my tombstone when I die. Here friend, I'll share with you the wisdom, so you can attain it yourself:
"Weather's not climate."
Re: (Score:1)
>How will the Climate Change activists react to real data on climate?
The same way they have been, by "correcting" historical data and throwing out conflicting modern data to keep their agenda alive.
No doubt the climate is changing - it's always been changing.
Show me the data that the current climate is the best for us to prosper. Oh, that's right it doesn't exist and cannot.
CO2 is plant food, not pollution.
Please keep other emitted shit out of the air please.
Re: (Score:2)
And there's a word that Google's never heard before. I've embiggened both my knowledge and vocabulary today!
Great (Score:2)
Pity it'll be fucked when 5G comes out and ruins the data going in to the model.
Re: (Score:2)
Weather forecasting *is* better today, but there are two main reasons:
1) larger faster computers can handle more detailed models.
2) more satellites give better on-going data correction for the predictions of the models.
The story indicates this new model isn't that much better than the old one. They'd probably do better to run them in parallel and average the predictions. Use a weighted average when figuring the announced results, with the weights adjusted by a neural network that you trained on last year'
They've Cubed a sphere?? (Score:2)
And I thought that squaring a circle was hard enough.
https://en.wikipedia.org/wiki/... [wikipedia.org]
Re: (Score:2)
Seriously, though, the thing about squaring the circle is that it has to preserve area. The issue in this case is that you need to come up with a well-behaved grid that covers the planet, with as little distortion as possible and with no singularities. Especially without any ill-conditioned volumes close to any place where important weather-generating activity happens, like the poles.
At any rate, you can't use a single 2D grid because of hairy ball theorem [wikipedia.org].
Yes, that joke you're thinking of right now has alr
Obviously ,,, (Score:2)
The upgrade brings us a new "Finite-Volume Cubed-Sphere" (or FV3) dynamical core, ...
Because the old "Infinite-Density Squared-Torus" (or IDT^2) static shell took too long to compute -- and no one knew what *it* did either.
Will this model... (Score:1)
Useless comments (Score:2)
Moderation idea (Score:3, Insightful)
Adult discussion-Total fail.
Slashdot should post one story like this a month, where every social reject is drawn in and as you say, nothing is said though a lot of words are posted.
After several hours have passed, Slashdot removes the story and IP bans all users who have posted.
Ahh, bliss.
What they really need to do... (Score:1)
...is crowdsource the data a bit more. Sucks when the forecast says "20% chance of rain" and it's pouring outside, and the actual weather is 5 degrees off the forecast prediction.
Re: (Score:1)
Forecast: high of 85, currently 92 (Score:2)
I like the NOAA site, but from mid-June through mid-September, on any given afternoon you can look at the charts and they will say "Forecast: high of 85" but then you dig further and see "Currently 92." I laugh and shake my head, but it happens nearly every day. You'd think they'd make adjustments or at least update the daily forecast to realize that if it's overshot by 7 degrees, that should be the new, amended forecast.
This is probably unrelated to the new spherical cube thingy, but this cranky aging man
Gee, I might finally get some useful reports (Score:2)
The new core improves computational efficiency and allows some processes to be simulated at a higher resolution -- unequivocal improvements. It also simulates the physics of water vapor more realistically.
The weather report was fairly useful where I lived last, in Kelseyville. Now I'm in Albion, and it's completely worthless. I mean that literally, I get more useful information by just going outside. Or hell, looking outside. But this is a complex environment with hills and a lot of marine influence, so it's not surprising if nobody can predict my weather accurately. My guesses would be that the resolution and the modeling of water vapor would in fact be the specific big problems with modeling it.
But can it beat... (Score:1)
...the ECMWF?
Accuracy is important - try it (Score:2)
It's not "the" US model, it's "A" US model, specifically the GFS. There are multiple models; the GFS, HRRR, HREF, RAP, NAM....
It's comical that people are complaining about model accuracy but can't be accurate about which one they're complaining about.