Today is the 15th Commemoration of the Joplin Tornado -- The Deadliest Tornado of the Storm Warning Era (With 11am Update)

In an unspeakable tragedy, the government's warning system failed in Joplin, MO 

the evening of May 22, 2011 -- 15 years ago today. 

The Result? 161 Perished.

Ruskin Heights Tornado about the time it was damaging my future bride's home.  
The tornadohad a 71-mile path and was F-5 intensity. Photo from Wikipedia.

May 20, 1957, was the date of the Ruskin Heights Tornado and the date the federal government (finally) got into the tornado warning business. Those warnings have saved tens of thousands of lives. On that Monday evening, the Weather Bureau's tornado warning was accurate and their tornado watch -- which had been issued hours earlier -- was right on the money.* 

Over time, both the science (our knowledge of how tornadoes function and how best to warn of them) and the technology (radar, weather satellites, then Doppler radar) helped weather science cut the tornado death rate (deaths per million population) by a stunning 95%.

Then, shockingly, it all collapsed in Joplin, MO the evening of May 22, 2011... 15 years ago today.

I wrote a short book that details the tornado, its effects on people, and what went wrong that horrible evening. The book was thoroughly researched including three trips to the city to interview everyone from hospital management to members of the Chambers of Commerce to ordinary people who were in the path of the storm. 

KSNF TV

When the Sirens Were Silent explains how it was a straightforward tornado warning situation and that subsequent statements from National Weather Service (NWS) employees that the tornado "formed over Joplin" and that a warning could not have been issued until the tornado was in west Joplin are false.  

That Sunday afternoon, AccuWeather's began issuing 70+ mph wind warnings more than an hour in advance. Then, we saw a (based on radar and other meteorological data) tornado was likely to form. Our tornado warning is below:

AccuWeather's Tornado Warning issued for RailAmerica

This warning provided more than 20 minutes of advance notice. The client's feedback is below.

The NWS issued their tornado warning a minute later. But it wasn't the timing that was critical. It was a fatal flaw in the wording.

At 5:14pm CDT...National Weather Service Doppler Radar indicated a tornado near Rivertown...or 4 miles north of Baxter Springs...moving northeast at 40 mph. 

Do you see the word "Joplin" in that description of the developing tornado's location? No!

But even worse -- and this is the fatal flaw -- the NWS said the tornado was moving northeast rather than east. If you look at the satellite image below, the developing tornado "near Riverton" (which is just west of the town of Galena) moves "northeast," it will completely miss Joplin!

Things worsened at 5:38pm. The NWS published, Trained spotters near Joplin or 6 miles northeast of Galena moving northeast at 45 mph." That location is plotted by the northern circle below. 

The upper circle is "6 miles northeast of Galena." The arrow depicts the "northeast" movement
stated in the National Weather Service's tornado warning. The tornado completely misses Joplin
The lower circle is the actual location of the tornado. It is moving due east right into the heart of populated Joplin. 

When I interviewed KSNF-TV (NBC) meteorologist Caitlin McArdle, she said "I was shocked the tornado was in that location [southwest Joplin, rather than north of the city as the NWS described at 5:38]!" She had thought the tornado was going to go to the northwest of the city. Their tower camera had been pointed north hoping to get a glimpse of the storm. The video of her coverage is here. Her entire tone and urgency change when she realizes the tower camera is showing the tornado moving into Joplin. The tornado continued to move east and cut through the heart of the city. 

If all of that wasn't bad enough, the sirens weren't sounded for the Joplin tornado warning until the tornado was entering the city. 

The 161 deaths (which, based on a confidential analysis shared with me, was at least 100 more than would have occurred if the NWS warnings had been correct) would have been bad enough. But the issues that occurred that evening were compounded by an after-storm analysis that was little more than a coverup.

Many believed the report and so concentrated on the wrong things and did not fix what actually went wrong. Since Joplin, the NWS's tornado warning program has collapsed to an extent I would never have thought possible. 

We've already documented the St. Louis NWS office issuing 27 consecutive tornado warnings on April 17 -- without a tornado occurring in those warnings -- due to many examples of a false radar signature known as a "side lobe." 

Below is a Twitter/X post from an amateur meteorologist who correctly identified a side lobe Monday. Yet this false echo evidently was the basis of a "tornado emergency warning" (TORE). TORE is the NWS's highest and most urgent category of tornado warning -- but little happened in this "emergency."

His comments include, "NWS tanked their reputation" and "what a TORE [tornado emergency] stands for (an extreme tornado threat) is in jeopardy," e.g., these false alarms are destroying the credibility of the warning program. 

From a note to me from a politician (received earlier this week):

The spate of bad tornado warnings this spring has been phenomenal. They constantly put out tornado warnings along gust fronts. 

[NWS issues] severe thunderstorm warnings like Halloween candy...every storm gets a piece of the action. At one point the other night there was a solid line of [severe thunderstorm] warnings from Lamoni, IA to NE Oklahome, yet a simple investigation along the line showed only a few spots where warnings were warranted. We were under a warning [at his home] and got a gust to about 30 mph for 10 seconds or so. It's become a joke. 

For years after Joplin, I and many other meteorologists have said, "If things aren't fixed, 'another Joplin' is inevitable.'" We had it this past July 4 when 135 people died in pre-dawn flash floods along the Guadalupe River in Texas. In this case, the National Weather Service's warnings were adequate (there was room for improvement) but local emergency management failed badly. 

When you add all of this into the problems cited here and here, plus many more that I haven't even had time to write about, one seriously wonders whether the system has deteriorated to the point where it is too late for a National Disaster Review Board to be effective. Can the NWS be saved?

Repairing the National Weather Service is an urgent task for Congress. 

*Ironically, 69 years later, the NWS missed two EF-2's and two other lesser tornadoes in the same geographic area. 

11am Update - The above incident (pick link) is referred to in the last paragraph of this press release from Representative Sharice Davids of Kansas: