Joplin, Missouri Tornado Assessment - recommendations

[BeauDodson]

Interesting research there Beau.

It appears SGF has a pretty large FAR. I know they issue a lot of tornado warnings based on squall line events.

I can speak for a lot of people that I've talked with and they typically don't take shelter during severe thunderstorm events. They will move their car under somewhere if they think there is going to be hail though.

This study is also of interest

http://ams.confex.com/ams/39BROADCAST/flvgateway.cgi/id/18179?recordingid=18179

Gabriel S. Garfield, University of Oklahoma, Norman, OK; and K. E. Klockow, S. Hoekstra, and S. Cobb

[JoMo]

Out of those 3 statements, I've got to say, the "Dread" statement would make me take the situation more seriously. The "Memorable" statement is too long. The "Current" does not provide enough description.

And I like the "High risk" vs % of risk or numerical models on the other scenario.

I wish they could pinpoint locations better though.

[B-Rent]

I just wonder how many of Springfield's warnings are not verified because of rural locations. Hard to say. They defin have a reputation for issuing a lot of tornado warnings (as mentioned on this and other forums). Whether that is right or wrong - good or bad - I don't know. Each office is different.

I don't think that SGF make up is much diffrent than PAH. About 3-5 "large" (by rural standards) cities, decently populated county seat in each county. Five to ten small towns per county. Maybe the only diffrence is road network, but doesn't PAH use air resourses too?

[Fred Gossage]

If there was a legitimate couplet associated with a comma head or an embedded supercell (e.g. many of the 4/27 morning couplets) a TOR would still be warranted, of course. The warning would be used for instances when QLCS spinups are impossible to detect, and would serve the same purpose as the current blanket SVRs/TORs.

I used to be in the more conservative camp concerning tornado warnings with QLCS events also, until the morning QLCS on 4/27. What I think you guys fail to realize is... while yes, most of the morning tornadoes did have couplets of some intensity.... the two strongest tornadoes from the morning round in Alabama (southern Tuscaloosa County and the Parrish-Cordova morning tornadoes) were both almost rated EF4 but BOTH had weak signatures (reflectivity notches and velocity signatures), at best. In fact, the EF3 that tracked across southern Tuscaloosa County had such an unidentifiable radar signature... that it went without warning until a good 15 miles further down the track, when late damage reports started trickling in. Both of these tornadoes, by the way, were close enough to have been identified easily by GWX and/or BMX had an established mesocyclone been present. For what it's worth.... both of these borderline violent tornadoes.... were not from embedded supercellular structures, but rather along the leading edge of the QLCS, not close to any identifiable kinks in the line or breaks in the line. Given that the environment during that pre-dawn time period wasn't too unlike most QLCS tornado warning cases we see (high shear, coupled with low but surface-based instability), there's no easy way to differentiate situations like this from all the false alarm scenarios we deal with in almost the exact same environmental setups.... at least until we're at a point where maybe our radar science is better and/or we better understand topographical and low-level boundary influences in situations like that. Until then, there is no way possible to differentiate these false alarm scenarios that we're all too quick to armchair quarterback style criticize.... from a situation like the morning of 4/27.... where you have barely identifiable radar signatures (if any at all), but strong to almost violent tornadoes ongoing... in a meteorological setup not unlike the ones that we commonly see associated with the high number of false alarm events.

[isohume]

I used to be in the more conservative camp concerning tornado warnings with QLCS events also, until the morning QLCS on 4/27. What I think you guys fail to realize is... while yes, most of the morning tornadoes did have couplets of some intensity.... the two strongest tornadoes from the morning round in Alabama (southern Tuscaloosa County and the Parrish-Cordova morning tornadoes) were both almost rated EF4 but BOTH had weak signatures (reflectivity notches and velocity signatures), at best. In fact, the EF3 that tracked across southern Tuscaloosa County had such an unidentifiable radar signature... that it went without warning until a good 15 miles further down the track, when late damage reports started trickling in. Both of these tornadoes, by the way, were close enough to have been identified easily by GWX and/or BMX had an established mesocyclone been present. For what it's worth.... both of these borderline violent tornadoes.... were not from embedded supercellular structures, but rather along the leading edge of the QLCS, not close to any identifiable kinks in the line or breaks in the line. Given that the environment during that pre-dawn time period wasn't too unlike most QLCS tornado warning cases we see (high shear, coupled with low but surface-based instability), there's no easy way to differentiate situations like this from all the false alarm scenarios we deal with in almost the exact same environmental setups.... at least until we're at a point where maybe our radar science is better and/or we better understand topographical and low-level boundary influences in situations like that. Until then, there is no way possible to differentiate these false alarm scenarios that we're all too quick to armchair quarterback style criticize.... from a situation like the morning of 4/27.... where you have barely identifiable radar signatures (if any at all), but strong to almost violent tornadoes ongoing... in a meteorological setup not unlike the ones that we commonly see associated with the high number of false alarm events.

Good post, I agree. Issuing tor warnings based on QLCS signatures is never a satisfying situation to be in. Many times you are going on theory, such as the probable best sheared location within a a bow echo or trying to time a possible broken-S line. These types of tornadoes are hard to verify as they can be very short lived or not touch down at all. On the other hand...lower end tors can and do develop in these situations...and just because a tor is only an EF0/EF1 doesn't mean it cant cause significant damage, injuries and deaths. They certainly can and they should be warned for. Of course these types of warnings help to increase the FAR...which all goes back to the current state and theory of operational radar science.