Comment by Tagbert | Hacker News Reader

Comment by Tagbert | Hacker News Reader

Parent Full thread

Tornados can be predicted in broad terms but are highly variable on the ground.

Weather forecasts can usually identify that the conditions favorable for tornados are happening. That is announced as a Tornado Watch. If someone sees a tornado, they will issue a Tornado Warning, usually for the whole county. They will sound sirens if available. People are urged to take shelter, ideally underground. Otherwise they are told to shelter in the most substantial part of a building. Many people do take shelter but others decide to tough it out.

Tornados are relatively small compared to hurricanes and earthquakes but they are very intense within a few hundred foot wide area where they tear everything apart and blow it around with 300mph winds doing more destruction with that debris. They move in erratic directions and can pickup and hop around. There can be multiple tornados at one time.

Any one person is not a great risk of harm but anyone in the path of the tornado is at great risk. People living in the area have been through tornado warnings before without harm. This can lead to people assuming that they will be fine and not taking precautions. Chances are they are right but if they are wrong… it goes very wrong.

> Tornados can be predicted in broad terms

There are occasions where tornados can be predicted accurately with radar data, yet with only about a 4 minute lead time. The hook echo has been recognized for decades with recognition and detection further refined with Next Gen Doppler radar.[1] But not all tornadic storms display a hook echo and not all hook echoes produce tornadoes.

[1] https://en.wikipedia.org/wiki/Hook_echo

This is accurate. Particularly the “alarm fatigue”.

I’ve lived in the Midwest my whole life. I’ve heard many tornado sirens, been through many watches and warnings, and never once actually been close to a tornado that did any damage.

A few months ago I was woken in the middle of the night by my wife who heard sirens. I was groggy, and before I was convinced to get out of bed, the sirens stopped. This has happened dozens of times in my life, and felt entirely routine - there was no panic. A few minutes later my phone gave an emergency alert announcing a Tornado Watch. I decided it was time to move to the basement, so I went downstairs and set up candles and blankets, and then began the process of gently waking my 3 young children in an attempt to move them without terrifying them.

As I moved the last of them, I glanced at a clock and saw it was 11:53. We moved to the basement and all was quiet. It seemed like another false alarm.

The next morning I discovered that the tornado hit two blocks away from me, and leveled entire street’s worth of houses. Radar confirmed the tornado passed through my neighborhood at 11:50. If I had been two blocks south, at least me and my son would be dead. Potentially more of us.

Alarm fatigue is real, but don’t let it happen with tornado sirens. They come quick and hit hard.

While I can really *relate and understand how a "alarm fatigue" can build up, especially when you have to choose between sleep and taking shelter. However, I'm guessing that since these workers were currently at work, at an Amazon facility, have the reputation of being infamous for monitoring and clocking workers, these alarm fatigue considerations would not apply in the same way.

What might apply creating similar alarm fatigue effect, and this is really just me guessing based on Amazon facilities reputation, is the incentive (or its lack thereof) Amazon may provide for workers to evacuate vs continue their work.

* I have never experienced tornado alarm, but have experienced multiple rockets alarms, some of which, I've (foolishly) decided to ignore in favor of getting another hour of sleep.

Reply on news.ycombinator.com