Tornado Warnings/Sirens

[TimeForGuinness]

Did you watch the video I posted? Those guys were pinpointing it. They were predicting where it would go. They were predicting which places wouldn't get it and would only get hail. Eitehr they are the worst meteorologists ever just out spewing malarkey or the paths are predicted.

They were pinpointing it because the hook had already manifested, they didn't predict where the hook would occur 30 minutes beforehand. You can also see from your video that their prediction of the storm moving forward had some variance associated with it...it wasn't an exact line/path of where the storm was going, it was a wedge of where the storm could go.

I will agree that once the hook/tornado is formed, they have a decent prediction of how it's going to move, but even that has a certain variance associated with it. Also, pop-ups do occur and catch meteorologists off guard.

Meteorologists know when the conditions can allow for severe storms, they can't predict down to a street corner until the hook/tornado is formed, but waiting to sound the alarm until that happens might be too late. People may not be watching the news to judge the severity of their own personal situation, therefore the sirens are necessary to alert the general public that conditions have deteriorated to the point where tornadoes can occur, and to take shelter. I'd rather people have enough of a heads-up so that they can calmly move to safer areas, not a last minute warning which could cause panic and chaos.

If you're watching the news, good for you, you can assess your own danger...but don't get upset about them blowing the sirens for the common good who might not know that the weather situation has changed for the worse.

[AWvsCBsteeeerike3]

Did you watch the video I posted? Those guys were pinpointing it. They were predicting where it would go. They were predicting which places wouldn't get it and would only get hail. Eitehr they are the worst meteorologists ever just out spewing malarkey or the paths are predicted.

They were pinpointing it because the hook had already manifested, they didn't predict where the hook would occur 30 minutes beforehand. You can also see from your video that their prediction of the storm moving forward had some variance associated with it...it wasn't an exact line/path of where the storm was going, it was a wedge of where the storm could go.

I will agree that once the hook/tornado is formed, they have a decent prediction of how it's going to move, but even that has a certain variance associated with it. Also, pop-ups do occur and catch meteorologists off guard.

Meteorologists know when the conditions can allow for severe storms, they can't predict down to a street corner until the hook/tornado is formed, but waiting to sound the alarm until that happens might be too late. People may not be watching the news to judge the severity of their own personal situation, therefore the sirens are necessary to alert the general public that conditions have deteriorated to the point where tornadoes can occur, and to take shelter. I'd rather people have enough of a heads-up so that they can calmly move to safer areas, not a last minute warning which could cause panic and chaos.

If you're watching the news, good for you, you can assess your own danger...but don't get upset about them blowing the sirens for the common good who might not know that the weather situation has changed for the worse.

But, that is precisely when they blow it.

They issue a watch if conditions are present for a tornado to form. They issue a warning when they see it forming and they blow the sirens seconds after that. Sure, the tornado might not have formed, but the hook has before they issue the warning.

So, they can predict it down to the precise intersection at the time they issue the warning and therefore blow the sirens.

They don't sound the sirens for a watch, like you're suggesting. Thank goodness for that. They'd go off 15 times a month!

[IMADreamer]

Sorry for my absence. I dug through some radar archives I saved over the years to find a couple examples and explain what's going on most of the time.

What AWvs is saying is true for the most part. With discrete super cells. Tornado track is evident 100% of the time. There is no reason for big blanket warnings. I've been out of meteorology for years and once an event is happening I can follow along and tell you whose going to get hit and who won't with certainty. So the guys on the cutting edge with the latest tools (dual pol radar, etc) can certainly do it. Take this cell for example from the St. Louis tornado a few years back that hit Lambert.


STL5 by EXORIMADreamer, on Flickr

As you can see in the above screen capture the tornado is obvious. It's one of the most classic examples of a tornado on radar I've ever seen. The left frame is base reflective, the right frame is radial velocity. There is obviously a tornado there and in this case you just calculate storm motion and speed to make your warning box, which I have done in blue. Now the NWS box will be bigger, probably twice as wide because of two things. Our system of warning people is extremely antiquated. This is due to the NWS being horribly underfunded. The second is if they do get it wrong and people are killed those in congress who want to do away with the NWS will have more fire power to do so. So the thought is it is better to warn more people and error on the safe side, then to provide the most accurate warning and take a chance. The NWS especially after the Alabama super outbreak, Joplin, and the tea party congress face huge pressure to perform at high levels. I don't mean to bring politics into it but it's 100% true. There are people in govt that want the NWS gone this is what the NWS has to deal with. While I haven't been a working Meteorologist in a long time I have lots of friends in various NWS jobs and they say the political pressure has been ratcheted up hugely in the last two years.

Ok example two, which is much harder to forecast and results in much less accurate Tor warnings. The quasi linear convective system. (squall line)

squal-line by EXORIMADreamer, on Flickr
This is storm system from several years back that hit my town and blew down literally hundreds of trees and did major damage to several structures. Three tornadoes were reported with this line. Where are they?

Nope can't see em on radar for two reasons. The first are these squall line situations result in very quick spin ups that may only last a few seconds or minutes. The radar only updates about every 5-7 minutes depending on setting, tilt, etc. In these situations there will not be the classic hook signature on radar. In this situation the NWS will do "blanket" warnings. Usually they will try and stick with a severe storm warning on these but once the radars algorithm starts detecting tornado vortex signatures they will sometimes issues a huge tornado warning box. This is a hotly debated tactic in the weather community and we are seeing lots of pressure to stop doing this.

I still contend that most of what people complain about with "false warnings" is due to local emergency managers sounding sirens when they shouldn't be. For example any time there is a tornado warning for any part of my county the town I live in sounds the siren. This means there could be a tornado on the ground 50 miles away from me because I live in a big county. Yet the sirens sound here. What we have to do is create a localized warning system in this country that is standardized across the whole nation in every community. Then we have to fund the NWS correctly, which could lead to advanced programs so that when the Met in charge of warnings draws the warning box automatic warnings are sent to people in those boxes only. Not the whole county. I think we are moving to that eventually, but it's years away, if not decades because the funding just isn't there.

[TimeForGuinness]

Did you watch the video I posted? Those guys were pinpointing it. They were predicting where it would go. They were predicting which places wouldn't get it and would only get hail. Eitehr they are the worst meteorologists ever just out spewing malarkey or the paths are predicted.

They were pinpointing it because the hook had already manifested, they didn't predict where the hook would occur 30 minutes beforehand. You can also see from your video that their prediction of the storm moving forward had some variance associated with it...it wasn't an exact line/path of where the storm was going, it was a wedge of where the storm could go.

I will agree that once the hook/tornado is formed, they have a decent prediction of how it's going to move, but even that has a certain variance associated with it. Also, pop-ups do occur and catch meteorologists off guard.

Meteorologists know when the conditions can allow for severe storms, they can't predict down to a street corner until the hook/tornado is formed, but waiting to sound the alarm until that happens might be too late. People may not be watching the news to judge the severity of their own personal situation, therefore the sirens are necessary to alert the general public that conditions have deteriorated to the point where tornadoes can occur, and to take shelter. I'd rather people have enough of a heads-up so that they can calmly move to safer areas, not a last minute warning which could cause panic and chaos.

If you're watching the news, good for you, you can assess your own danger...but don't get upset about them blowing the sirens for the common good who might not know that the weather situation has changed for the worse.

But, that is precisely when they blow it.

They issue a watch if conditions are present for a tornado to form. They issue a warning when they see it forming and they blow the sirens seconds after that. Sure, the tornado might not have formed, but the hook has before they issue the warning.

So, they can predict it down to the precise intersection at the time they issue the warning and therefore blow the sirens.

They don't sound the sirens for a watch, like you're suggesting. Thank goodness for that. They'd go off 15 times a month!

I'm not suggesting that they sound the sirens for a watch...but I want them to sound the sirens to the surrounding area if conditions have deteriorated to the point where a tornado has a high probability of happening, not just certain streets. My point is, meteorologists aren't 100% right, the weather system is dynamic and not 100% predictable...a lot of this data analysis is storm dependent, so creating a more complex way to warn people doesn't provide a huge benefit...other than quieting the people who find it necessary to complain about hearing a tornado siren that doesn't seem to immediately affect them.

You know what I do when I hear a tornado siren, and it's sunny out? I don't ignore it, but I think either that it's a test or the other side of the county must be getting pounded...I don't think about emailing the local emergency manager to complain about why they sounded the siren.

[AWvsCBsteeeerike3]

I'm confused....

People may not be watching the news to judge the severity of their own personal situation, therefore the sirens are necessary to alert the general public that conditions have deteriorated to the point where tornadoes can occur, and to take shelter. I'd rather people have enough of a heads-up so that they can calmly move to safer areas, not a last minute warning which could cause panic and chaos.

I'm not suggesting that they sound the sirens for a watch...

In the above quote that is almost exactly what you are suggesting...

but I want them to sound the sirens to the surrounding area if conditions have deteriorated to the point where a tornado has a high probability of happening, not just certain streets. My point is, meteorologists aren't 100% right, the weather system is dynamic and not 100% predictable...a lot of this data analysis is storm dependent, so creating a more complex way to warn people doesn't provide a huge benefit...other than quieting the people who find it necessary to complain about hearing a tornado siren that doesn't seem to immediately affect them.

Well, sure. But, that's not what was said earlier.

Second, as I said before, I think we're selling meteorologists a bit short here. If you don't sell them short and believe in their abilities, you can come up with a better system than what is out there. I'd venture to say you could come up with a much, much better system.

Third...You look for the signs to issue a warning. And do it. Also, you don't need to blow a tornado siren to get everyone's attention. A severe thunderstorm warning should work fine. If people are outside playing in the woods when there is a massive storm approaching, then they're idiots and have their head up their ass. Sure, you could blow a siren or fly a helicopter directly overhead and drop pamplets saying "MASSIVE STORM APPROACHING - TAKE COVER" but that's not going to guarantee they do it. If a big red blob on the weather radar doesn't prevent them from going out somewhere they are 30 minutes from a shelter, I'm not sure what will. If you're in your back yard, you don't need a siren to notice deteriorating conditions where you can run and check real quick.

Anyway, after the warning is issued, obviously those directly impacted get the “Oh [expletive] siren” We’ve all heard it. It’s the one that we hear all the time. Do it for a couple mile radius. Then, the towns in the path get a different warning. A couple short beeps (or whatever) for towns 5-10 miles away. Three short beeps for towns 10-20 miles away. Obviously the path expands as you go further out. But at least you are telling them a tornado is COMING as opposed to the way they do it now and sound the siren AFTER. If it dissapates, sweet. No more warnings. If it continues, boom, hit the oh [expletive] alarm in one of the cities that have already gotten the warning minutes ahead of time.

If people can't figure it out, maybe they should call their emergency manager to figure it out....which leads me to my last point....

You know what I do when I hear a tornado siren, and it's sunny out? I don't ignore it, but I think either that it's a test or the other side of the county must be getting pounded...I don't think about emailing the local emergency manager to complain about why they sounded the siren.

First off, this is a gross mischarecterization of the situation. I spoke to the guy, cordially, and didn't even have to insinuate that they blow the sirens too much. He offered it. And just explained they do, in fact, error on the side of safety. Is it an [expletive] move to call him to understand how it works, albeit not being an [expletive]? I dunno. I'd suspect if everyone knew exactly how it worked, they might be better off. [expletive] me for being curious, I guess. Second, it's not a one time thing as you suggest, it's a recurring theme.

[AWvsCBsteeeerike3]

Sorry for my absence. I dug through some radar archives I saved over the years to find a couple examples and explain what's going on most of the time.

What AWvs is saying is true for the most part. With discrete super cells. Tornado track is evident 100% of the time. There is no reason for big blanket warnings. I've been out of meteorology for years and once an event is happening I can follow along and tell you whose going to get hit and who won't with certainty. So the guys on the cutting edge with the latest tools (dual pol radar, etc) can certainly do it. Take this cell for example from the St. Louis tornado a few years back that hit Lambert.


STL5 by EXORIMADreamer, on Flickr

As you can see in the above screen capture the tornado is obvious. It's one of the most classic examples of a tornado on radar I've ever seen. The left frame is base reflective, the right frame is radial velocity. There is obviously a tornado there and in this case you just calculate storm motion and speed to make your warning box, which I have done in blue. Now the NWS box will be bigger, probably twice as wide because of two things. Our system of warning people is extremely antiquated. This is due to the NWS being horribly underfunded. The second is if they do get it wrong and people are killed those in congress who want to do away with the NWS will have more fire power to do so. So the thought is it is better to warn more people and error on the safe side, then to provide the most accurate warning and take a chance. The NWS especially after the Alabama super outbreak, Joplin, and the tea party congress face huge pressure to perform at high levels. I don't mean to bring politics into it but it's 100% true. There are people in govt that want the NWS gone this is what the NWS has to deal with. While I haven't been a working Meteorologist in a long time I have lots of friends in various NWS jobs and they say the political pressure has been ratcheted up hugely in the last two years.

Ok example two, which is much harder to forecast and results in much less accurate Tor warnings. The quasi linear convective system. (squall line)

squal-line by EXORIMADreamer, on Flickr
This is storm system from several years back that hit my town and blew down literally hundreds of trees and did major damage to several structures. Three tornadoes were reported with this line. Where are they?

Nope can't see em on radar for two reasons. The first are these squall line situations result in very quick spin ups that may only last a few seconds or minutes. The radar only updates about every 5-7 minutes depending on setting, tilt, etc. In these situations there will not be the classic hook signature on radar. In this situation the NWS will do "blanket" warnings. Usually they will try and stick with a severe storm warning on these but once the radars algorithm starts detecting tornado vortex signatures they will sometimes issues a huge tornado warning box. This is a hotly debated tactic in the weather community and we are seeing lots of pressure to stop doing this.

I still contend that most of what people complain about with "false warnings" is due to local emergency managers sounding sirens when they shouldn't be. For example any time there is a tornado warning for any part of my county the town I live in sounds the siren. This means there could be a tornado on the ground 50 miles away from me because I live in a big county. Yet the sirens sound here. What we have to do is create a localized warning system in this country that is standardized across the whole nation in every community. Then we have to fund the NWS correctly, which could lead to advanced programs so that when the Met in charge of warnings draws the warning box automatic warnings are sent to people in those boxes only. Not the whole county. I think we are moving to that eventually, but it's years away, if not decades because the funding just isn't there.

This is awesome.

Thanks for the info.

It really is a shame the NWS is under scrutiny. I talked to one of their meteorologists this AM and he was so scared to talk it was weird. It was like he was afraid to give out any inforamtion because it could be used against him. I finally got him talking by asking a couple different questions and what not. And, then he opened up. But, yeah, at first he was like.....uh, who is this, why are you asking, where do you live. I told him I was curious how they did Tornado warnings. He was like...uh, who is this, why are you asking, where do you live. So, I told him I was curious what the process was for sounding the alarms in Little Rock, where I lived, as I had just moved there. Just a little white lie, but whatever. So, he asked if it was for a newspaper article. I assured him it was just an inquisition as to rather or not they could issue a warning based off radar and models. That I was an engineer, gave him my name and company and assured it was just because I was curious and found the whole process interesting. Followed them on twitter and thought they were doing a great job.... He kind of opened up then. Said he was one of the meteorologists and yes they predict the tornadoes based on more information that just the 'hooks' but that those are in fact a huge thing. They're all different, some easier than others to predict and interpet, yada yada. He said they could in fact pinpoint pretty closely what was going on with most of them....but didn't go into why they don't as you just did.

Really, he was a pretty interesting guy. I don't know if part of his job is public relations, the phone number was on the internet. But he was knowledgeable and friendly once he opened up.

As a matter of fact, I think I'm going to write ole Timmy Griffin a letter on their behalf and tell him that they're awesome and need more money I wonder how to hint at a better improved warning system without making it seem like they're doing something incorrectly though...

[TimeForGuinness]

Sorry maybe I'm getting the definitions & terminology of watches and warnings mixed up, and nitpicking where I don't need to be. Let me start over, since I'm having trouble remembering what I've written...maybe I need to clarify it, maybe we're arguing the same point. Maybe I'm confusing myself.

If the conditions have deteriorated to the point where severe storms are occurring or about to occur (ie: tornadoes, straight line winds, etc), I think the siren should be sounded. I guess that's more learning towards a tornado warning where "hooks" are more defined...but I wouldn't mind if it's more preemptive if they can accurately predict it. The problem like I said is that they might not be able to predict where a hook will develop...currently, they just know that the conditions are right for a hook to develop. It's really storm dependent, like Dreamer pointed out.

I'm not saying I don't trust meteorologists, or that I'm selling them short, I believe in what they do and admire them for it. I'm one of the biggest science advocates on this board. My point is that it isn't a 100% known, it's a probability assessment (even if it's a really high probability assessment, it's still isn't 100%). There is variance (even if it's small). Certain storms/tornadoes they might be right 99% of the time...other storms (like Dreamer posted) is a crap shoot. They do their best with the tools and science that they have available. I do agree that with the technology that they have, they can assess the situation better and offer emergency guidance to the more effected areas...but once again that is storm dependent. I believe and am confident that they can isolate storms down to the gnat's ass, but do we need to set up a neighborhood local emergency warning system? Is that worth the money? Is that worth the risk? What happens if they warn one neighborhood but conditions change too rapidly to warn a neighboring community?

That being said, depending on how the siren system is set up (local/city/county), it could be overkill for certain areas..especially if you're on the eastern part of a county 20-30 minutes away from the actual storm. Another point is that maybe you're just hearing the neighboring area's siren instead of your own area...I've experienced that before.

I guess my point is, even with the technology to pin-point dangerous storms, a broad warning system is still not a bad idea...from a risk, financial, and legal point of view.

This is a good discussion, and I'm not picking on you (except for the part about calling the local emergency manager), you set up the rant

[AWvsCBsteeeerike3]

I think we're talking about different things and the same thing along the way.

first, I disagree that they should sound the sirens for anything but a tornado warning. There needs to be something above and beyond that warns citizens of an imminent threat that a tornado brings, not to downplay other storms. But, that's where I stand. And, that's the thinking now...they are just doing a poor job of implementing it...or are doing an implementation that is way beyond safe.

Here's an example that happened Tuesday:

http://www.littlerock.org/%21userfiles/ ... em_Map.pdf



A tornado occurs on the very east side of town and the storm is headed east. The sirens up to 15 miles west are blowing. This is like 15 minutes or more after the storm that caused this tornado passed this area. (It was a super thin line of strong storms maybe a mile wide). At the time, the weather was calm. Every meteorologist was talking about other areas. There was no danger. But, I understand why they did it. I just disagree with it.

Imo, what needs to happen is they (the City of Little Rock) need to localize the sirens in a case like that. Blow the ones within a couple miles (that they control). Or, an even better option is like IMA is saying and give the NWS control. Blow the sirens in the area, blow the sirens to the east as well in the path they predict. Sure, if they completely screw up the path, the people affected don't get warned. Of course, under the current system they're not going to anyway.

If you want to have different alerts for future tornadoes vs imminent hits, that's irrelevant, really.

That would alert anyone that is affected now and anyone that has the possibility of being affected in the future. Further, if the alarm disappears, the sirens cease. And, at worst, it acts as a storm warning. No, I wouldn't blow it with the point of it being a storm warning. But, it would likely act as one many of the times with the legit threat of a tornado. Also, it prevents those that aren't affected from being innundated with the noise....that yes is annoying!

[TimeForGuinness]

Yeah, that seems weird that the sirens went off after the storm passed, never heard of that before. The only thing I can think of is that all the sirens are tied together with one button push? Now if the sirens went off as the storm was coming into the area, and the storm could drop a tornado at any time, then I don't think that's a problem.

I still probably wouldn't care as much as you do, I just would have thought the eastern part of the city was getting hammered.

[AWvsCBsteeeerike3]

Yes. I talked to the guy. They set them all off as soon as there is a warning issued. Not a big deal, in the overall scheme of things. I shouldn't get that irritated by it and won't now that I talked to him and he seemed to understand that it is overkill sometime but they decided to let everyone know that bad weather was in the area....even if it wasn't

But, there is room for improvement. I do stand by that argument.