Joplin Tornado Warning Redux

A tornado occurred in the Joplin area early yesterday afternoon and, unfortunately, there were issues with the National Weather Service's warning just as there were on that terrible day in May, 2011, when 161 people lost their lives in that horribly warned tornado

According to KSNF-TV, which could view the tornado from its cloud camera (just as it did with the May 22, 2011, tornado), the tornado touched down at 1:35pm. Unfortunately, the tornado warning wasn't issued until 1:42pm, which made for a "lead-time" of an awful -7 minutes. 
Below is a screen capture of a video of the tornado (the video is at the red link, above). There's a tornado siren in the video and it is not rotating -- meaning it is not sounding!
Even worse, per a meteorologist in the region, when the tornado was first reported, the NWS called it -- you guessed it! -- a "landspout."

Increasingly, I must ask: what training are NWS meteorologists receiving?

While warning of this tornado was not easy, the Springfield NWS office (which warns Joplin) made it much more difficult for itself by running their radar at six-minute (360 seconds) when they could have run it as fast as 80 second intervals. Had they run it at a faster interval, they might have seen the tornado's rotation on Doppler in time to get some warning out. 

Below is the Dopper wind data from the Springfield radar. There was a significant area of rotation at 1:05pm southwest of the city. I have highlighted "Airport Drive" because that is the area in which the damage occurred.

At 1:30pm, the mesocyclone is weak. This would not have prompted me to issue the warning. But, you can see the problem when you compare it to the image below. The time is 1:30 (arrow). 

Six minutes later the mesocyclone is a bit more organized and the tornado is on the ground. Whether there would have been a stronger rotation signature in between the 1:30 and 1:36pm images is unknown. But, running the radar at six-minute intervals when a significant mesocyclone (top image) had presented itself is inexcusable.
These radars are getting very old and there are, at present, no plans to replace them. So, NWS offices are told to run the radar more slowly to minimize wear. That's fine. But, the whole point of having radars is to warn of dangerous storms. When there is the potential for tornadoes, they should be run at 80-second intervals. 

These issues urgently need fixing.


  1. The Tornado Outbreak in SW OK a couple days ago... They were not going to issue a watch as storms began firing. They gave it a 20% chance for watch issuance. Then they ultimately issued a severe thunderstorm watch. There were numerous tornado warnings with confirmed tornadoes - watch went until 9:00 pm without upgrading. No modifications to the SPC day 1 Outlook either. They do not issue a tornado watch until about 10:30 pm that evening. With a large confirmed tornado on the ground with two circulations, the watch text gave only a 20% chance for a strong tornado. Again, no modifications to the outlook either. My faith in the NWS/SPC as a whole to be able to adjust on the fly is at an all-time low. -Wes

    1. I've noticed that SPC has had some difficult days this year. That said, I don't believe anyone could have anticipated a violent tornado in southwest Oklahoma.


Post a Comment

Popular posts from this blog

[1:10am Update] Tornado Forecast for Rest of the Night

First Tornado Watch of the Day Issued

Hilary's Forecast Path Shifts West; Updated 9:20am PDT