SN Reports had 15 min lead time 3/8/10

[Kiel Ortega]

I struggle to believe that whatever they were doing for that 15 minutes between ground truth and issuing the warning - was a better choice than clicking the link and seeing the tornado live.

This wasn't the biggest event (staffing) and OUN has to launch a sounding at 23Z for 00Z reporting. Further, these weren't exactly the most exciting storms as far as radar presentation. So it's easy to criticize (as I see here and as I even did following the event), but if all that matters is slapping a warning on a map without regard to accuracy, what's the point to storm-based warnings?

 

[rdale]

Personal opinion is that you widen out the polygon and cover the blobs in question versus doing nothing... Very few use the polygons at this time anyways.

It appears the initial warning was not based on any radar interpretation (likely due to the lack of features) but on the spotter reports (and an estimated distance.)

 

[Jason Boggs]

Not quite as bad as the late tornado warning, but I relayed info about quarter size hail just south of Roll, and the cell never went severe warned. Not the same storm as the one that produced the tornado, but it still never got warned on.

 

[rdale]

I don't see that in the SN logs.

 

[Jeff Snyder]

I don't see that in the SN logs.

Jason - is this your report?

[FONT=lucida sans typewriter, lucida console, courier][COLOR=#003366][B]PRELIMINARY LOCAL STORM REPORT[/B][/COLOR]  
[COLOR=#003366][B]NATIONAL WEATHER SERVICE NORMAN OK[/B][/COLOR]  
[COLOR=#003366][B]548 PM CST MON MAR 08 2010[/B][/COLOR]  
[/FONT]
[FONT=lucida sans typewriter, lucida console, courier]0504 PM     HAIL             2 S ROLL                35.75N 99.72W   
03/08/2010  E1.00 INCH       ROGER MILLS        OK   UNKNOWN           
  
            RELAYED FROM WFO AMARILLO    [/FONT]

-> http://kamala.cod.edu/offs/KOUN/1003082348.nwus54.html

 

[Tyler Allison]

Let's all remember that there's a human on the end of the chain of information here. It's up to him/her to use their best judgment. Just because the worlds greatest storm chaser calls them and says a tornado is on the ground...doesn't mean they _have_ to fire the warning button. Not everyone knows everyone else, and not all NWS employees know or even understand the Spotter Network. I know we'd all like it to work that way..but I don't think any amount of complaining and finger pointing is going to make it happen.

If the NWS wanted tornado warnings to be published solely based on an SN report (or any reporting mechanism) we wouldn't need a human. They could hard wire the SN system.

Sometimes humans make mistakes (we don't even know if this is the case in this instance), but more often than not it's the right thing to have a human reviewing the input.

Looking back from the comfort of my arm chair....sure...the tornado warning should have gone out 10 minutes earlier. But I wasn't there and I don't know all the reasons/requirements/issues involved.

Also, claiming the report wasn't precise enough to issue a warning is a red herring. You either believe the report or you don't. You know _EXACTLY_ where the spotter is. The tornado can't be more than 10-20 miles away at most (line of sight limitations). Open up the warning polygon like rdale said and let it fly. Warning for a 40 mile square area on a confirmed tornado in the area and more folks than necessary take cover is _FAR_ superior to not warning anyone and someone dying simply because you wanted more precision. You can refine the area later once you see the rotation on radar.

-Tyler

 

[Ryan Wichman]

I have talked about this with several chasers over the past few hours and questioned whether or not to say anything...but why didn't someone just call the OUN NWS to report the tornado?
Perhaps a NWS employee that is on ST can shed some insight but it would make sense to me that calling would be faster and more effective (as opposed to a SN report that comes across as 'unverified').
From the prespective of the NWS, who knows who the person to post the SN report is. I realize that it is a hassle to to have all the NWS number handy, but cell phones make it easy to save numbers well in advance these days.
Just my view on it.

 

[Joey Ketcham]

I have talked about this with several chasers over the past few hours and questioned whether or not to say anything...but why didn't someone just call the OUN NWS to report the tornado?
Perhaps a NWS employee that is on ST can shed some insight but it would make sense to me that calling would be faster and more effective (as opposed to a SN report that comes across as 'unverified').
From the prespective of the NWS, who knows who the person to post the SN report is. I realize that it is a hassle to to have all the NWS number handy, but cell phones make it easy to save numbers well in advance these days.
Just my view on it.

I agree. I think SN is a great tool, but in my opinion the best method of reporting is calling it in direct to the NWS. Since SN first came out, I think in total I only made one or two reports over it and that was just to report flooding. It is my policy that if I'm observing a tornado, than I am on the phone calling it in to the NWS. I then at least know that, without a doubt, they have the information. I use SN mainly so that the NWS will know my exact location.

And as Tyler said, I am sure there are some within the NWS that doesn't know much, if anything at all, about SN. I know that links and messages were sent to them, but that is no guarantee that they are going to see that. Speaking for myself, I work in IT and I am often so busy during the day - even while sitting at my desk - that someone will message me over MSN and I may not get to it until 20+ minutes later. And as I've told the guys under me, if it's something that needs my immediate attention come to me directly or call me. That's how I look at it.

Now if these same chasers had called directly into the NWS with their tornado report and the NWS just sat on it without issuing a warning for 15 minutes, then I would more likely be on the other side pointing my finger at the NWS shaking my head. But that not being the case, I don't think we should be so quick to blame the NWS.

It's just my opinion that SN reporting should only be done for non-tornadic events such as flooding, damaging winds, hail, etc... and for tornadic situations call that in direct.

 

[Tyler Allison]

(as opposed to a SN report that comes across as 'unverified').

Actually...a random call from a random person claiming to know what they are talking about is more "unverified" than an SN report. At least anyone using the SN has been required to take a training module and pass a test.

This comes down to trust, not communication method. You can default trust a tornado report...or you can default not trust. I'm guessing the person running the "desk" has a default not trust personal preference. Nothing wrong with that...you'll have a lower false positive, but a higher false negative. IMO, when dealing with tornadoes a higher false positive is preferred.

This also goes to show that just because _ONE_ person has dropped a tornado report into the SN or called the office...._EVERYONE_ who can see the tornado should be reporting it. Don't assume someone else is/has reported it. If 10 people with visual on the tornado all put in a tornado report into the SN, that's hard to ignore...no mater how much you don't trust.

Also, when you call the NWS that information only goes to the NWS...not everyone else that wants to know about it. I'm not saying don't call...I'm saying we shouldn't accept the argument that calling has a higher default "trust" value.

-Tyler

 

[Jeff Snyder]

I have talked about this with several chasers over the past few hours and questioned whether or not to say anything...but why didn't someone just call the OUN NWS to report the tornado?

I tried to call several times (both the office and 911), but I couldn't make a single call (or get data) for most of the afternoon/evening. I'm not sure if other chasers were experiencing the same problems, but at least one chaser tried to call.

I do agree that voice calls MAY be better than a SN report in an emergency situation since the person on the other end of the line can seek further clarification, ask questions, etc.. In other words, a telephone call is a 2-way convo, which isn't something that's currently a feature of SN.

I respect OUN partially because they TEND to have a greater threshold for tornado warning issuance. Whereas I've seen many cases of rather questionable tornado warnings from other NWSFOs, I'm very pleased that OUN seems to place a bit more focus on keeping the False Alarm Rate in check. Almost necessarily, raising the treshold for tor warnings to reduce false alarms can also reduce the probability of detection (at least with a positive lead time). Were any TV stations live from the field before the tornado developed? If the staffing were relatively limited (given an expectation of a marginal severe event), I could see how they may not have the personnel available to carefully watch streaming video from chasers. I was surprised to see how many minutes passed between the time I saw the tornado and the time I heard the warning come across on AHWR/NWR.

 

[rdale]

If the NWS wanted tornado warnings to be published solely based on an SN report (or any reporting mechanism) we wouldn't need a human. They could hard wire the SN system.

The big thing here though is that this is more than just a SN report. The chatroom logs show that at 5:19pm the info was manually posted in the Norman room along with a link to the video so the forecasters could see for themselves.

At 5:27, another note was sent in the chatroom, and referred to the live video once again.

I don't think anyone is suggesting that a SN tornado automatically generates a Tornado Warning. I think the question is - the office was notified of a tornado via spotternet, and was notified that live video of the tornado was available via their chatroom. What happened to have a 10+ minute delay between that "confirmed report" and the warning? Is there something in the system that is broke and can be fixed? Was it just inattention?

Nobody is out to "get" anyone, but I doubt anyone thinks it is a good idea to just shrug your shoulders and sweep it under the rug either.

 

[Tyler Allison]

the office was notified of a tornado via spotternet, and was notified that live video of the tornado was available via their chatroom. What happened to have a 10+ minute delay between that "confirmed report" and the warning? Is there something in the system that is broke and can be fixed? Was it just inattention?

True. I assumed inattention was not possible and nothing technologically was broken. Which means the person heard/saw and decided to wait until they could verify with radar. Could be a bad assumption on my part.

-Tyler

 

[Joey Ketcham]

The chatroom logs show that at 5:19pm the info was manually posted in the Norman room along with a link to the video so the forecasters could see for themselves.

At 5:27, another note was sent in the chatroom, and referred to the live video once again.

It could been easily missed. Was the NWS employee actively chatting at the time? As I said in my earlier post, I get so busy even at my desk that someone will message me and it'll be 20+ minutes before I even look to see what they said.

 

[Tyler Allison]

As I said in my earlier post, I get so busy even at my desk that someone will message me and it'll be 20+ minutes before I even look to see what they said.

Equating not patching someones box within 10 minutes of an IM and not hitting the panic button when you see an IM are not the same. If it turns out that the information was "missed", we may have a systematic failure...not a person failure. If the NWS employee is suppose to be doing 48 things _and_ watch NWSChat...maybe he/she can't...and the US government needs to appropriately staff the NWS to do the job the public has asked them to do.

I suspect we'll never know what actually happened....we can only speculate. Most of the time they do a good job, I know I wouldn't want their job. I got enough trouble watching out for myself

 

[Jason Boggs]

Jason - is this your report?

[FONT=lucida sans typewriter, lucida console, courier][COLOR=#003366][B]PRELIMINARY LOCAL STORM REPORT[/B][/COLOR]  
[COLOR=#003366][B]NATIONAL WEATHER SERVICE NORMAN OK[/B][/COLOR]  
[COLOR=#003366][B]548 PM CST MON MAR 08 2010[/B][/COLOR]  
[/FONT]
[FONT=lucida sans typewriter, lucida console, courier]0504 PM     HAIL             2 S ROLL                35.75N 99.72W   
03/08/2010  E1.00 INCH       ROGER MILLS        OK   UNKNOWN           
  
            RELAYED FROM WFO AMARILLO    [/FONT]

-> http://kamala.cod.edu/offs/KOUN/1003082348.nwus54.html

Yea, that's it. Correct me if I'm wrong, but to my knowledge the cell never went t-storm warned.