There's no rational business reason to keep a factory operating during extreme weather. All those candles they manufactured? Destroyed by the storm. And now an entire shift of your workforce is wiped out.
That's easy to say in hindsight, but it depends on a few factors IMO.
> How extreme was the expected weather that day?
> How much of a lead time did the area have that actual tornadoes were coming?
> Was there any government advisory telling people that it was highly recommended to head home that the factory ignored?
> If they acted immediately, would there have been enough time for all people to get home?
> Were the workers informed of the situation, and did they prefer to head home or stay there?
That you are asking this question suggests to me that you have no experience with tornadoes whatsoever.
Vehicles are probably the worst place you can be in a tornado: they offer poor protection from the flying debris of a tornado, and are liable to themselves become part of the debris [1]. Most tornado warnings, especially the more extreme warnings, include messages that say something along the lines of "if you're in a car, GET OUT NOW and get to shelter."
The government would not recommend that people head home in event of a tornado warning. Tornado warnings are issued only on evidence of a tornado, which is way too short notice (think 15 minutes if you're lucky) for anyone to attempt to change their location to somewhere else.
[1] It's worth noting that one of the pictures I've seen of this tornado is a box truck impaled by a tree. Or maybe a tree impaled by a box truck--they are both in places that they shouldn't be, so it's hard to tell who struck whom.
1. Quite extreme. NWS had been highlighting the potential for a significant outbreak of severe weather for at least a day at that point. My local WFO (KIND) referred to the system’s potential as “exceptionally dangerous,” which is not language that NWS forecasters use haphazardly.
2. Mayfield had over 2.5 hours of warning from KPAH.[0] I’ve noted elsewhere that this tweet wasn’t a formal warning, of course, but it was a strong heads-up that something big was in the offing. At the very least, local emergency managers should have been aware of this and monitoring the situation very closely.
3. The tweets from KPAH regarding the storm and naming Mayfield specifically were pretty strong. Add to this the actual warnings themselves, which were Particularly Dangerous Situation warnings. A Tornado Emergency warning was also issued as it became clear Mayfield was in the crosshairs. Those are the strongest warnings in the NWS’s arsenal.
The first tornado warning on that cell was issued around 6pm CST, and the tornado almost immediately became destructive. Forecasters knew that the cell was entering a convective environment that was highly conducive for the cell to persist and even strengthen. (The various Mesoscale Discussions that the SPC issued for this cell make that clear.) So, between that knowledge and the track that the cell was on, KPAH was able to determine that Mayfield was in serious danger hours beforehand.
So, the possibility at least existed that plant managers could have known what was coming. One question might be what the local emergency management agency did to disseminate these warnings to the community. Most people, unless they’re weather nerds, probably aren’t reading mesoscale discussions from SPC or forecast discussions and tweets from their local WFO. But I’d expect emergency management to be doing so and then making determinations about how to respond to that information.
So, it’s hard to say precisely how much plant managers knew, but it’s possible that they could have known well ahead of time. It’s also possible that they believed it to be more prudent to keep people in the plant rather than sending them home to structures that may well have been much less sturdy.
[0] See https://cliffmass.blogspot.com/2021/12/a-critical-gap-in-tor...
> [0] See https://cliffmass.blogspot.com/2021/12/a-critical-gap-in-tor...
I think the author of that blog post made a time zone error. He is referencing this tweet in support of "2.5 hours warning": https://twitter.com/NWSPaducah/status/1469503102218760194 but Twitter tells me that tweet was sent at 9:03pm CT, not 7:03pm (Twitter would show him 7:03pm if he's on the west coast which I believe he is). So that would be more like 20-25 minutes warning (which is still impressive for a tornado), not 2.5 hours.
You’re right that Cliff made a time zone error, which consequently undermines at least some of my points. The lead time was still fairly significant (and there were other indications that enhanced monitoring of the situation was warranted), but that certainly isn’t the same as 2.5 hours’ worth of advanced notice!
Thanks again for catching that.
They had 20 minutes per the article, in that time frame you would have a traffic jam in the parking lot as everyone tried to leave down the 2 lane road that services the factory.
They had 2 days warning. not 20 minutes.
Comment say that Mayfield had about 2.5 hours of warning based on this blog: https://cliffmass.blogspot.com/2021/12/a-critical-gap-in-tor...
If you look at the area encompassed by the tornado watch, should they have evacuated it in it's entirety? https://blogger.googleusercontent.com/img/a/AVvXsEji0sqQJqWo...
Look about 300mi x 170mi area. If you take Jones burro as the center, its 120mi to the edge, technically if everyone could evacuate at 60mph you could get everyone out in 2 hours.... but with NWS tornado warnings have a False Alarm Rate (FAR) of roughly 70% you probably won't get that many people to evacuate at all unless you make it mandatory. Then it will take much longer than 2 hours to get people out.
You are talking about massive areas with small amounts of time: https://blogger.googleusercontent.com/img/a/AVvXsEhcPxy7gbyr... is roughly 430 square miles for something that is small and fast moving.
I'm talking about not forcing them to come in to work for a 3rd shift overnight run that can most definitely be delayed by a shift for a weather pattern known to create high numbers of tornadoes.