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 collaps...

2022 tornado, image by Ben Holcomb The NWS did not  provide an advance warning of this Kansas tornado as the tornado was in progress when the warning was issued. [If you aren't a meteorologist or meteorology student, feel free to skip this post.] On December 1, we published a piece outlining our recommendations pertaining to the future of the National Weather Service . There are serious  issues pertaining to the way meteorologists are educated and trained in 2024 that desperately need to be addressed in 2025.  Today, I want to focus on one of the recommendations pertaining to meteorologists' training:   There is an expectation that meteorologists should be able to intervene and change a forecast when the computer forecasts are obviously wrong. But, as aviation safety has learned, there is no way for people to overrule defective automation output when they have little or no experience doing so.  An analogy: Why did Asiana Airlines Flight 214 crash in 2013 in...

I've written up fewer of the National Weather Service's missed tornado warnings this summer because I fear they have become repetitive reading. However, because yesterday evening's situation in northern South Dakota was straightforward, it is quite concerning no tornado warning was issued.   I am publishing the Twitter/X advisories that I provided to my followers not to boast but to disprove the National Weather Service's (NWS) past contentions of "Monday morning quarterbacking" which are silly. That said, if I was able to provide advance warnings from Wichita, there is no reason the NWS local office serving northern South Dakota couldn't, also.  6:14pm Advisory: While there were issues with the North Dakota tornado situation ("T" above), let's focus on South Dakota in this posting. The southern circle (orange arrow) showed an area of rapidly increasing rotation (right panel, below) in the Doppler wind data.  If the NWS or other organization ...

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 torna...

WLNS TV, Lansing This time it was in central Lower Michigan. After dark. The time when the National Weather Service (NWS) says tornadoes are most dangerous.  Everything pointed to a tornado warning being issued -- but it wasn't -- until the tornado was on the ground, doing damage.  I am going to review this case as an illustration of why the USA desperately needs a National Disaster Review Board to determine why the National Weather Service's tornado warning program has gone so far off course as compared to its high quality from about 2000 to 2010.  Atmospheric Conditions  Conditions were ripe for tornadoes yesterday evening in central Lower Michigan. The "Significant" (meaning EF-2 intensity or greater) Tornado Parameter was at a high  value of 5 (2 is generally thought of as the threshold).  The NWS Storm Prediction Center, in spite of its forecasts earlier in the day (see: here   ) that there was a significant risk of tornadoes, chose to go with a s...