How Could This Happen?

Blog reader Jenny asked (see comments) how the tornado in Charlotte could be missed. We’ve been having a similar controversy here in Kansas about the missed tornado warning for Harveyville

Charlotte Observer photo

Here is a video that shows the meteorologist saying there is “no indication” of the tornado while it was destroying homes and injuring four.  

Coverage of the destruction in Charlotte is here

In both cases, the tornadoes were F-2 intensity and did significant damage. While I cannot speak for the National Weather Service, let me make a couple of generalized comments that might help.

These were not the classic supercell-type tornado (think Greensburg, Tuscaloosa, Joplin). We almost always know where they are and where they are going early enough to provide an effective warning.

Harveyville and Charlotte are a much smaller and short-lived type of tornado known as a “squall line” or QLCS tornado. While weaker, they are still dangerous as the tragic death in Harveyville demonstrates. From where I sit, if your home is destroyed, you don’t care what type of tornado struck it. 

For reasons that are not clear to me, the NWS is really struggling with the issue of different types of tornadoes (note: highly recommended link if you are interested in the topic). So, as previously discussed, beginning April 1, their offices in St. Louis, Kansas City, Springfield (MO), Topeka and Wichita will start issuing four levels of tornado warnings:

  • Severe thunderstorm warning with the sentence “a tornado is possible.”
  • The “ordinary” tornado warning.
  • The “particularly dangerous situation” tornado warning (what should have been issued given the type of tornadoes that struck the Charlotte and Harveyville areas)
  • The “tornado emergency” tornado warning where they forecast “catastrophic” damage.
There are two major problems with this approach. 
  • We do not advise people to prepare differently, so what is the point?
  • The very real risk of confusion. What is a school principal supposed to do with a “severe thunderstorm” warning that says a “tornado is possible”? Will anyone be able to keep straight whether a particularly dangerous situation is more or less dangerous than a tornado emergency? 

My unsolicited advice to the NWS is to forecast trying to forecast what type of damage the storm is going to do and redouble efforts to get the warning scientifically correct. That includes obtaining more frequent radar scans of the storms, wider distribution of the FAA’s Terminal Doppler Weather Radar (and, it looks like the NWS is making progress on this one!), and ways of getting the warnings distributed more quickly.  

So, Jenny, this is my analysis of the situation. Hope it is helpful. Thank you again for the question.  

This entry was posted in Uncategorized by Mike. Bookmark the permalink.

6 thoughts on “How Could This Happen?

  1. Mike, here is a really good summary by local TV met. @wxbrad about this event. This is a link to his post if you haven’t seen it yet. (I hope this lets me post the URL) I do not live in that metro area, but I do live in the state. It is interesting to read about the issue that he addresses of the large metro area being broken up in to three different CWAs.

    http://wxbrad.com/mecklenburg-cabarrus-county-tornado

  2. Thanks for the explanation, Mike. I really appreciate you taking the time to answer my question. I just came across the article that Ryan posted in his comment, which is also a very informative explanation. Kind of crazy that a city the size of Charlotte doesn’t have its own Doppler radar! In light of this, is there anything the average person can do to prepare for situations like this, short of staying up all night to keep an eye on the weather/twitter/tv/etc.?

  3. Ryan, thanks for passing it along. It is a good explanation by a very good meteorologist. I agree that it makes no sense to split the warning responsibility among three offices.

    That said, I do not share his caution toward the TDWR. They are excellent radars that update every minute which is vital when trying to anticipate fast-moving and quick developing situations.

    The good news is that the NWS — finally — is going to make wider distribution of the TDWR data starting in April.

  4. Mike, The issue I have with this particular TDWR is a large chunk of the 340° is blocked by a water tower. Then you factor in the attenuation problem and the absorption issues of the smaller beam width it does have disadvantages. I would prefer a full Dual-Pol Nexrad site with the 10cm beam width and power that the other NWS sites have. The other issue for us is that GSP is primarily looking at the 88-D in these situation and has to switch to the TDWR closer to Charlotte.

  5. Hi Brad,

    In the near future, perhaps as early as April, the NWS is going to provide the TDWR data at one-minute intervals, which is a huge improvement over the 4.1 min. (at best) for the -88D. Because the TDWR velocity is so much higher resolution than the WSR-88D, it is far easier to see small tornadoes, downbursts, and similar features. You’ll find the velocity data survives the attenuation quite well. Because of their bandwidth issues, they are only going to transmit the lowest two tilts, but we’re happy to have it to serve our clients in the area.

    Thanks for the great write-up. I passed it along to our staff.

    Mike

    • That’s awesome to hear, I agree we see a lot more microbursts for sure with the TDWR than we did before. It’s hard to tell if it’s better than the 88-D here because the 88-d it’s frankly useless over a large part of my DMA. I mean the beam hits Uptown Charlotte at over 7,000′ Luckily we have a 250,000 watt c-band radar that helps fill the gap as well. I’ll take any data I can get my hands on. :-)