• 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

Southeast Svr Wx threat April 8-11, 2024

Over 5800 Lightning strikes
 

Attachments

  • GOES16_1km_ir_202404100305_26.50_37.00_-100.75_-85.00_ir1_ltngge_hgwy_warn_weathernerds.png
    GOES16_1km_ir_202404100305_26.50_37.00_-100.75_-85.00_ir1_ltngge_hgwy_warn_weathernerds.png
    236.7 KB · Views: 0
Storms along the TX/LA line in the 10 hatch have really struggled to look like anything other than rainy blobs despite strong shear. Some rotation south of Kirbyville, TX but it's not currently warned; even SVR.

Edit: Looking a bit more impressive, they might actually want to warn it.
 

Attachments

  • LCH_0322.png
    LCH_0322.png
    903.4 KB · Views: 0
Heaviest rainfall corridor (4+ inches) according to latest 00z RRFS-A thru 01z Thursday
 

Attachments

  • qpf_acc-imp.us_se.png
    qpf_acc-imp.us_se.png
    578.2 KB · Views: 0
So a coding challenge I worked on this evening: What could a "verified" forecast look like?

I made a program that takes in the latest SPC convective risks and simulates a tornado outbreak. (I don't have it for hail or high winds atm.) An orange dot represents a tornado touching down at that spot. A white dot represents an EF2+ tornado touching down at that spot.

I'll show a few random runs -- each run is different, but the probabilities are based on the SPC's probabilities.

I have 7 shown here, and these are NOT cherry-picked: They are literally the 7 runs in a row I did just for this post. I think it's a helpful visual to see that even "busts" can be valid probability-wise, and the probabilities allow for a wide range of scenarios.

Now to wait for the new Day 1 to hit lol

1712723484397.png

1712723518495.png
1712723542795.png
1712723570305.png
1712723597801.png

1712723623203.png
1712723644870.png
 
warneagle said:
WHAT TIME IS THE ECLIPSE WEDGE GOING TO HIT MY HOUSE??????
Click to expand...
Posted on Facebook that the threat will say along and south of I-20 tonight. Still had people commenting us about towns in North MS.
 
wx_guy said:
So a coding challenge I worked on this evening: What could a "verified" forecast look like?

I made a program that takes in the latest SPC convective risks and simulates a tornado outbreak. (I don't have it for hail or high winds atm.) An orange dot represents a tornado touching down at that spot. A white dot represents an EF2+ tornado touching down at that spot.

I'll show a few random runs -- each run is different, but the probabilities are based on the SPC's probabilities.

I have 7 shown here, and these are NOT cherry-picked: They are literally the 7 runs in a row I did just for this post. I think it's a helpful visual to see that even "busts" can be valid probability-wise, and the probabilities allow for a wide range of scenarios.

Now to wait for the new Day 1 to hit lol

View attachment 25274

View attachment 25275
View attachment 25276
View attachment 25277
View attachment 25278

View attachment 25279
View attachment 25280
Click to expand...
Awesome work! The natural statistical variation is really neat, and is a rather good example of the breadth of possible outcomes you can get with any given setup.
 
1712748713281.png~

These are probably the strongest QLCS associated radar velocities I can recall seeing... 130mph. Not only that, there are at least 4 tornadic circulations in the single tornado warning alone. Another reminder to never underestimate QLCS events - especially considering we will most likely see similar stuff to this continuing across LA, MS and AL for the next 12 hours. Still think the 15% was the right call from the SPC.
 
First benefit of KHDC realized - it got a much better look at the earlier tornado-warned supercell near Rosedale, LA than KLIX would have, and the storm was about the same distance from KLCH.

It's also getting a better look at the newly warned QLCS circulation approaching Saint Francisville than would have been possible before.
 
Just released the new D1 MDT for 13z.

Here's the tornado threat, with a couple of simulations (see previous post) of the tornadoes. (Please note these are probabilistically-generated points, not based on any meteorology other than the SPC's numbers.)

1712753691354.png

1712755592802.png
 
CheeselandSkies said:
Ugh. IMO they should NEVER pair "large and extremely dangerous tornado"/PDS verbiage with "radar indicated rotation." To use the former it should be "radar confirmed tornado" via a large, deep and persistent TDS; or preferably confirmed via spotter reports.
Click to expand...
Yeah, that's weird and probably a bad idea for them to do.
 
CheeselandSkies said:
Ugh. IMO they should NEVER pair "large and extremely dangerous tornado"/PDS verbiage with "radar indicated rotation." To use the former it should be "radar confirmed tornado" via a large, deep and persistent TDS; or preferably confirmed via spotter reports.
Click to expand...

completely agree. i will say, WAFB [baton rouge] just reported on air that the nws said they used the wrong product in issuing that TOR.
 
Really don’t like the look of that storm coming into New Orleans. It’s got some pretty strong rotation, albeit mostly broad at the moment. But it looks like it could really tighten up any minute now.
 
This is the NSSL CAM experimental model using the MPAS-RT, but many of the CAMs are starting to show this same pattern -- a dangerous bow echo, not dissimilar from a derecho, speeding across south AL/north FL later this evening. If this verifies, strong winds over 65-70 mph appears very plausible, and the SPC might need to extend the hatched wind risk further east into AL/FL.

1712756903206.png
 
wx_guy said:
Just released the new D1 MDT for 13z.

Here's the tornado threat, with a couple of simulations (see previous post) of the tornadoes. (Please note these are probabilistically-generated points, not based on any meteorology other than the SPC's numbers.)
Click to expand...

Do you think you could post the source code to Github or whatever, so that we can try our own hands at running some simulations ourselves? Because this looks quite fascinating to me, enough so that I want to try my hand at generating some simulations myself just to see what kinds of results that would produce!:cool:
 
You must log in or register to reply here.
Back
Top