-
Welcome to TalkWeather! We see you lurking around TalkWeather! Take the extra step and join us today to view attachments, see less ads and maybe even join the discussion. CLICK TO JOIN TALKWEATHER
Severe Weather 3/15 - 3/16
- Thread starter CheeselandSkies
- Start date
Aaron Rider
Member
I know, I know, I've mentioned this twice already in the thread, but I'd really like to see an in depth analysis of the 5/2/1929 outbreak in Virginia and Maryland! There's just nothing like it since! I have not found anything satisfactory online
WeathermanLeprechaun
Member
It's poorly documented unfortunately but outbreaks in that area are anomalous. Anything that involves beating the so called magnet of the Appalachian Mountains is intriguing. I wonder if some sort of simple thermodynamic combination can help overcome the usual blocking regime over here. 5/2/1929 and June 1944 were the most potent to do this. Not much like them since.
It really intrigues me also. Do certain areas just ramp up recurrance intervals when a long term pattern is favorable? Michigan has woke up the past few years though it isn't the state for consistent, major activity that it used to be. The Plains hasn't really been consistently active with violent events. Utah used to regularly record tornadoes in January and February in the 1800s, but i don't believe they've recorded much in that period ever since. Some areas just seem to wake up occasionally and sometimes it's long term.
If you look at the some of the ERA5 or 20CR data for 5/2/1929, the 500 MB looks similar to yesterday. Except the jet ejection was timed just right, and actually punched, then overlapped the warm sector. Essentially firing the storms as it nosed into the area, then with the energy directly overhead, allowing them to sustain. None of those tornados were very long tracked, but notable for that region, and seem to have fired on the cold front (which didn’t have as extreme gradient across it as yesterday did) so it was probably semi-discrete for a while but then lined out.
Your surface low on that day actually was very close to the outbreak area, moving northeast from Cincinnati and was in the Pittsburgh/Cleveland area as the F3s touched down in northern Virginia. That day, like most anomalous events, just had all the right ingredients at the right place and time.
WeathermanLeprechaun
Member
The reanalysis data on Tornado Archive doesn't work on mobile unfortunately. Is there any other areas i could view it?
Grand Poo Bah
Member
Each new AI model costs exponentially more to train, and requires exponentially more processing power, but the actual intelligence part really seems to be hitting a plateau. $10 trillion is slated to be invested into AI by 2030. If the results aren't everything that has been promised (and more) it'll set us back a decade easily. All these data centers are going to account for 25% of all US power consumptions, and will already be outdated within 18 months. In my personal experience it seems the latest models are actually dumber and more prone to hallucinations than their previous generation. There's no doubt AI will have tremendous value, but it's not going to match the absurd amount of money that has been invested, and growth will grind to a halt.
At least they are evolving in the same way as us humans
Grand Poo Bah
Member
Something I really wish we could see after events like this is archived CAMs that we can compare side-by-side with actual observed results. I want to see which CAM gets trough orientation correct most often, which shows the OWS best, which nails SRH, MSL pressure and surface wind, CAPE, and which gets shear. It'd be much easier to spot flies in the ointment if we knew, which models were most reliable for each element.
I think it was you who mentioned the fact that the Mid-Atlantic is simply, well, at the far east of CONUS and as a consequence, the evolution of troughs makes it very rare for major tornado events to happen there. I also believe it was @bckhd2 who queried about the Impenetrable Storm Wall™ at the Georgia/Alabama border, and I must reckon these factors are likely related to both of these phenomena. Watching and looking back on a lot of outbreaks in the Deep South, Georgia's placement in the CONUS almost always has it receiving basically a glancing blow from the best environmental parameters in events, even major ones. We obviously have way more exceptions to the rule than, say, Maryland, but it seems the same general idea stands. In our particular case, the combination of the trusty Appalachian Wedge, being a bit farther removed from best dynamics than Mississippi and Alabama, having systems traversing our area at late hours of the night with notable consistency, and likely additional factors all seem to contribute.
Unfortunately, no. It’s the best public recreation of that data, but you have to use desktop of course.
I’ve been able to use python to access ERA5 but that’s a non-public hobbyist process I’ve been working on.
The idea that AI is gonna get smarter than the actual meteorologists is bullsh*t especially with analog data/past weather events. That AI doesn't know what that meteorologist has seen or been thru weather wise plus if that the meteorologist has been in the same area for many years, he has a general idea of the landscape. The AI doesn't know that. It's just another tool for the meteorologist to use just like the forecast models, etc. are just tools.
Last edited:
TH2002
Member
- Messages
- 4,966
- Reaction score
- 11,086
- Location
- California, United States
- Special Affiliations
- SKYWARN® Volunteer
I don't know if I should be doing this, but I did some quick searches over on a certain site, and well...
Older one, but I'm throwing it in as a free bonus:
Older one, but I'm throwing it in as a free bonus:
I see wxtwitter is still the same as always.
WeathermanLeprechaun
Member
WeathermanLeprechaun
Member
Nitter unfortunately is experiencing certain issues because now you can't check anyone's post history without restarting to their most recent about two pages in. It's a shame because I usually check a lot of people's old posts out to learn new things and figure out new events. "the general public aren't complaining" oh yes they are.
bckhd2
Member
Yes! It's so interesting. Would love to see a study done on it someday, but I think you've already identified all of the major factors! It's fascinating to me how such relatively short geographic distances can lead to such drastic differences in storm impacts.
The Nashville, TN, NWS office has posted their most recent update to their damage survey of the 3/15 storms:
As of the time of writing, they've confirmed three tornadoes (one EF0 and two EF1s), none of which produced any injuries or fatalities.
As of the time of writing, they've confirmed three tornadoes (one EF0 and two EF1s), none of which produced any injuries or fatalities.
akt1985
Member
Here is a video from NWS Columbia, South Carolina about what went wrong with the forecast on Monday.
Central Ohio Wx
Member
An RSI value of 26.967 has been retroactively applied to the blizzard this storm system caused by the NCEI; this makes it the first Category 5 blizzard since January 2016 and the highest RSI value retroactively applied to a storm since 1996.
