Killer Tornado in Evansville, Indiana

Based on the radar/radial velocity presentation it looks as if this was a quasi-linear HP supercell. The storm most definately had a well defined hook echo with a "flying eagle" presentation within the band of storms. Can someone post some SFC obs from around this time? I'm sure the winds were backed ahead of the SFC trof and low, but I want to see the data. Thanks.
 
Obviously, the NWS did a pretty good job in issuing the warnings. It seems even the people at the point of touchdown had at least 10 minutes warning, with those farther down the path getting much more.

However, there were a number of key factors at work that played a significant role in the number of deaths we saw:

1. It hit at 2 am when most people were sleeping and visibility was poor.

2. There wasn't a great deal of concern yesterday within the media and certainly little in the way of pre-storm hype to grab people's attention. Most people who were even aware of a threat were thinking MO and IL would be the place where most of the activity would take place if something did happen. Afterall, that was the area mentioned on TWC and other media outlets when the severe weather threat was discussed. But by midnight, not a great deal had taken place and many probably wrote off the threat and paid little attention from that point forward. Even many of the local news channels seemed caught off guard by this event. In addition, when things did start to develop, it developed quickly and much farther to the E of where most people had been anticipating the tornado threat to be.

3. The forward speed of the storms was very fast, in some cases, 60 mph. With such great forward speed, these tornadoes were really rolling, giving those people who were caught unaware little time to react.

4. The time of year was also bad. While we know better and often expect severe weather this time of year the same as during Spring, your average person doesn't associate November with severe weather outbreaks.

So what we have here was a worst case scenario. A large, powerful, long track tornado, moving through a populated area (and a trailer park), late at night and at a time and in an area where the people weren't really prepared for nor expecting strong tornadic activity. We've seen similar situations play out many times before. Camilla, GA in 2000 and the Blue Ash section of the Cincy area in 1999 are just a couple of recent examples.
 
At around 9:00 in the evening, the RUC had the LLJ progged at 60+ mph up through central Indiana in it's 6 hour forecast model. Had the surface low deepening to 997 mb, and had the upper trough clearly going to a negative tilt. I guess anticipated de-coupling of the boundary layer moisture with sunset was somewhat of an offsetting factor, but given these dynamics coming together, I was a little surprised at the lack of emphasis. Also, some of the storm motions were much higher than 60mph. On Lvl 3, I was seeing clockings of over 80 mph, with one cell at the nose of the bow at 122 mph. Realize it probably wasn't a derecho in the pure meteorological sense, but the effects were comparable, and I'm a little surprised the SPC didn't go ahead and invoke a special PDS. Hard to say if that would have made the locals even more on their toes and provided more lead time, but - at least over the short-term - this looked like one of those "synoptically evident" events to me.
 
Angie, you said you hear reports of tornado damage, in North Central TN, right? I think that is incorrect, from whatever source you obtained that information from. I believe the damage IF ANY, came from damaging winds, rather than a tornado in that vicinity. I thought the tornado was isolated in the KY/IN area, yes? I am fairly positive, that there was not an actual tornado in TN, however there were reports of Funnel Clouds in western TN.
 
The upper-level dynamics were strong, but there was very little in the way of surface-based instability. Typically, as the sun sets and the surface layer cools, substantial CINH develops, which is largely the reason why we typically see a transitition to elevated convection overnight. The strong surface low created a strong pressure gradient and resultant strong gradient wind. This likely kept the boundary layer semi-well mixed, prohibiting decoupling and a transition to a parcel updraft above the surface ('elevated convection'). If we had 2000-3000 j/kg sbCAPE, then I think the ugrade to a MDT risk (or higher) would have been warranted. However, with VERY little surface-based instability, it's very difficult to say that conditions are very favorable for tornadoes. The storm in particular likely ingested a localized area of higher surface-based instability, which caused it to "flare-up" and do what it did.

We do see a situation like this several times in the cool-season. A lot of times, there are many folks who wonder why the SPC goes MDT or HIGH risk on days with very little instability progged on a NOV-FEB day. I remember a HIGH risk day in LA/eastern TX a few years ago -- some folks critized the SPC for overblowing the situation given the lack of instability. Well, it goes both ways. This certainly does go to show that tornadoes are an inherent risk with any supercell.
 
Andrew, here are the counties in TN that were under tornado warnings during the overnight:

Benton, Stewart, Montgomery, Cheatham, Davidson, Robertson, Sumner, Macon and Trousdale.

All I can confirm is that there was some wind damage in some of those areas, but I haven't heard any of it has been confirmed as tornado damage at this point.

I suppose we'll have to wait until tomorrow until any suspect damage is looked at by the NWS before we can say ye or nay as to whether there was any tornado damage in TN.
 
"SEVERE THUNDERSTORMS CAN AND OCCASIONALLY DO PRODUCE TORNADOES."

I tire of people saying that the public needs to be alert for an F3 tornado in a SVR warning... Remember, SVR warnings are supposed to mean "you could be hurt or killed." Do they? Nope. Will they in the future? Probably not. So since we all know that a SVR means "this is a bad storm, but likely won't do you any harm" we can't use that as an excuse for residents not taking cover.

In this case, nobody expected a major tornado to form near EVV. Nobody.
 
Originally posted by Jeff Snyder
The upper-level dynamics were strong, but there was very little in the way of surface-based instability.

Jeff, something I didn't quite understand last night in the SPC's forecasting, though. It seems like, in prior situations where convection has already initiated at night, and the storm motions will take the storms into a very, very strong LLJ, it has been almost automatic that the forecasters note this, with the realization that strengthening will occur regardless of the subsiding instability. Once initiated, doesn't CIN become less of a factor? I just remember several situations last spring where a strengthening nocturnal LLJ seemed to be such a key factor in the SPC's discussions, and - frankly - just sat there last night staring at what the RUC was saying, staring at the incredible radar echoes and wondering, is anyone else seeing this? As usual, your feedback is appreciated from one still learning.
 
I watched this unfold on the wire and on local television. There was PLENTY of warning out of Paducah. Tornado warnings were issued even EARLIER for the counties west and southwest of Evansville. Had people been paying attention then they would have known a tornadic supercell was heading their way.

First tornado warning went out at 1:30 a.m.

http://kamala.cod.edu/offs/KPAH/0511060732.wfus53.html


It was late at night...so that hurt the warning process. My weather radio went off for these tornadoes.

This was a bad tornado...no doubt about it. People need to have those NOAA Weather Radios. They do save lives.
 
To be honest, I am not very surprised to see big-time tornadic supercells occured last night... And as I mentioned earlier in the FCST thread, tornadoes were likely (however, didn't expect them over IN/KY - but rather mainly over southern/central IL).

Originally posted by 11/05 FCST thread - Nick Grillo+--><div class='quotetop'>QUOTE(11/05 FCST thread - Nick Grillo)</div>
Given forecast strong low-level shear (0-3km SRH currently >250m2/s2 across the area) and favorable deep-layer shear, the potential for a few violent tornadic supercells exists. [/b]

<!--QuoteBegin-Mike Johnston

Jeff, something I didn't quite understand last night in the SPC's forecasting, though. It seems like, in prior situations where convection has already initiated at night, and the storm motions will take the storms into a very, very strong LLJ, it has been almost automatic that the forecasters note this, with the realization that strengthening will occur regardless of the subsiding instability. Once initiated, doesn't CIN become less of a factor? I just remember several situations last spring where a strengthening nocturnal LLJ seemed to be such a key factor in the SPC's discussions, and - frankly - just sat there last night staring at what the RUC was saying, staring at the incredible radar echoes and wondering, is anyone else seeing this? As usual, your feedback is appreciated from one still learning.

Often, as the boundary layer cools at night - most storms will become elevated and will feed in parcels from above the surface layer... So, the storm updraft is drawing in it's air from above the surface (say their is a inversion at 850mb, then say the storm will be drawing in parcels from 700mb, etc.). This will often happen, and elevated convection will form due to the nocturnally-strengthening CINH. Of course, this is not always the matter (like the case last night, when surface-based supercells formed in a region with enhanced surface-based CAPE and localized strong low-level shear) and last night, as Jeff mentioned, the strong surface low / associated strong wind gradient kept the boundary layer somewhat well-mixed (allowing for storms to draw in parcels from the surface layer). Therefore, when CINH increases at night - storms will become "elevated" and will draw in air from above the CAP (thus, little of a tornado threat, but rather large hail and damaging winds).
 
Originally posted by Mike Johnston+--><div class='quotetop'>QUOTE(Mike Johnston)</div>
<!--QuoteBegin-Jeff Snyder
The upper-level dynamics were strong, but there was very little in the way of surface-based instability.

Jeff, something I didn't quite understand last night in the SPC's forecasting, though. It seems like, in prior situations where convection has already initiated at night, and the storm motions will take the storms into a very, very strong LLJ, it has been almost automatic that the forecasters note this, with the realization that strengthening will occur regardless of the subsiding instability. Once initiated, doesn't CIN become less of a factor? I just remember several situations last spring where a strengthening nocturnal LLJ seemed to be such a key factor in the SPC's discussions, and - frankly - just sat there last night staring at what the RUC was saying, staring at the incredible radar echoes and wondering, is anyone else seeing this? As usual, your feedback is appreciated from one still learning.[/b]

Mike,

It's easier for a supercell to sustain itself in a given environment with moderate or high CINH than it is to initiate. We've all seen this in action -- many storms develop in the late afternoon or evening, but only a few sustain themselves into the night. Once a storm is going, there is increased convergence at the base of the updraft, so it usually takes stronger environmental CINH to 'cut off' the updraft given the stronger convergence. Now, the rotation of the updraft (which defines a mesocyclone in a supercell) induces vertical pressure perturbation gradients, which further enhances the updraft velocity and the convergence at the base of the updraft.

In regards to the LLJ, the nocturnal strengthening of the LLJ is often a result of boundary layer decoupling (which results in a transition from subgeostrophic flow common in the low-levels to supergeostrophic flow). Now, in times of rapid surface cyclogenesis, the strengthening of the LLJ may not be enhanced as much by boundary layer decoupling as by the increase in the pressure gradient as the cyclone develops.

So, IF a supercell can persist into the night, as remain surface-based, the increased shear resulting from the stregntehnign of the LLJ can create a very favorable environment for tornadogenesis. However, surface-based instability wasn't all that strong during the daylight hours, much less after dark. As I said before, the strong gradient wind may have prevented boundary-layer decoupling to a degree, so the surface-based instability probably didn't decrease as much in this case as is typically experienced near and afters sunset. These high-shear cases are relatively common during the cool season (late fall through early spring), and very few yield a strong tornadic supercell. I haven't looked at a whole lot of data from last night, but I assume that the supercell ingested a rather localized area of strong instability that allowed it to either strengthen more for a while, or allowed it to become entirely surface-based.

EDIT: Just saw Nick's answer above, which explains this well. Also, I remember several cases this past spring which were high-shear, low instability in the warm sector. Unfortunately (or fortunately), we saw a very different result in these cases this spring, with the only tornadoes in the cool sector beneath the upper-level/mid-level low.
 
Thanks, Nick - I understand a little better now.

Re: the event last night, there did persist just a plume of instability at the surface, and I guess it was just enough given the extraordinary dynamics. I guess this gives renewed testament to the old EHI; some of those helicity measures were absolutely off the charts last night.

Congrats on your early eye for on this system. It did seem a bit marginal Saturday morning, and you had some naysayers posting you down on this board, but when that cap and clear slot were over SW MO in the afternoon, we saw that 62td, that upper trough started digging down hard and then CIN eroding from the west...well, it was really something to watch unfold. Not exactly a great photo op, but your instincts and analysis were right on!
 
Back
Top