NWS Norman Local Storm Report Question

[mikegeukes]

Here is a question, I notice this year, that the NWS Norman office
in the local storm reports have not reported any tornadoes, while
tornadoes have occurred in the CWA. Is this a new policy change.

Also I notice in the Public Information Statements to see if any
of tornadoes were surveyed, and not seen anything.

Mike

 

[Jeff Snyder]

There was a bit of related discussion on this over in the 3/30 DISC thread (starting at Shane's post, continuing on from there). Shane said that the folks at WFO OUN told him not to look for an LSR for the tornado(es) on that day. I've expressed my thoughts in that thread, so I'll leave it for someone who works at OUN (Rick probably) to chime in here if he wishes.

On a somewhat related note, I was impressed with the St. Louis NWSFO, which had a PNS out about a weak tornado that occurring Sunday evening (4/2)... The PNS was issued at 4am on 4/3 -- that's some rapid assessment or at least dissemination of information.

 

[Rick Smith]

Our policy on issuing LSR's for tornadoes has not changed.

First of all, the LSR is usually most effective during, or perhaps immediately after an event. It is used mainly to relay near realtime storm reports to downstream NWS offices, the local broadcast media and the SPC.

We do not report tornadoes in an LSR unless we have high confidence that 1) the tornado actually occurred and 2) we know the location and time of the tornado. Even though LSR's are called PRELIMINARY storm reports, many seem to ignore the preliminary part and take what is sent in an LSR as the final truth (witness the national media listing 60+ tornadoes from Sunday). If we are confident in the basic facts about a tornado, we will report it in an LSR.

A tornado report does not become "official" until it is entered into Storm Data. By that time, the WFO has had time to research the event, gathering all available reports, images and videos, and comparing that information with radar data. Then and only then does the event become part of the official database.

Concerning surveys...we do not survey every single tornado report we receive. Due to resource limitations and other factors, we survey what we believe to be "significant tornadoes." So far this year, we have not had any significant tornadoes in our CWA.

I hope this helps. You should note that I speak only for WFO OUN, and not for any other offices.

Rick

 

[Evan Bookbinder]

I would tend to agree, however argue that in this case, there were several trained spotters (including a very well respected chaser) reporting these tornadoes, with concrete photographs/video.

While the LSRs are preliminary, SPC puts together a nice graphic depicting the coverage of severe weather over the course of the day. I'm certainly not arguing that preliminary reports are final. We all agree that yesterday's 63 tornado reports will be significantly widdled down as tracks are combined, or that some damage is determined to be downburst/straight line winds. Sure some reports of tornadoes are bogus or misinterpreted by those reporting them, but with even modest certainty (multiple reports, known chaser, photos, video, damage, etc...), I'd rather provide the preliminary information and adjust it accordingly with followup LSRs the next day, public info statements, web articles, etc... Unfortunately, Storm Data is a great resource, but is grossly underutilized, generally because of the many month delay before these final stats are published.

Of equal importance is the following example from a perspective of uniformity and consistency among the meteorological community. National news media station WXYZ calls NWS office ABC requesting commentary on the scope of an event, and in addition to requesting information about the local event, poses a question about the regional/national coverage of tornadoes (states affected, numbers, etc...).

In the Mar 30th case, SPCs page showed no tornadoes in OK, and yet already posted on this forum were clear cut tornadic photographs with commentary provided about calling the NWS. These same photos may even have already been provided to TV stations or other media outlets, as is typical of many chasers in this forum. Tough scenario to be in for the person handling this interview.

I surely hope that Shane and others will continue to call their reports into NWS offices. Mike Peregrine's testimonial from that same day in NW MO undoubtedly resulted in lives saved and advance warning.

These statements reflect my opinion only and not necessarily those of my employer.

Respectfully,
Evan

 

[Jeff Snyder]

While the LSRs are preliminary, SPC puts together a nice graphic depicting the coverage of severe weather over the course of the day. I'm certainly not arguing that preliminary reports are final. We all agree that yesterday's 63 tornado reports will be significantly widdled down as tracks are combined, or that some damage is determined to be downburst/straight line winds. Sure some reports of tornadoes are bogus or misinterpreted by those reporting them, but with even modest certainty (multiple reports, known chaser, photos, video, damage, etc...), I'd rather provide the preliminary information and adjust it accordingly with followup LSRs the next day, public info statements, web articles, etc... Unfortunately, Storm Data is a great resource, but is grossly underutilized, generally because of the many month delay before these final stats are published.
[/b]

I'm not sure I'm entirely qualified to comment a lot on this, but I do agree with Evan. Storm Data, while "official", is not very user friendly when someone wants to reference severe reports from a particular day. The way it is now, you have to search for the date, state, and type on the NCDC site. Then, you're presented a list of all COUNTIES affected. Well, this doesn't help when determining the number of tornadoes on a particular day, since a tornado that spans several counties will receive several Storm Data entries, and piecing together these pieces can be tedious and difficult at times. Sure, we can wait for Hart (I believe) to put together all the tracks for us for use in SvrPlot v.2.0, but that only occurs annually. I'd think that iif you were to ask most chasers or media folks, 95% of them would say they reference the SPC Prelim logs much more often than Storm Data entries. I know that some NWSFOs release Storm Data info in a more organized/clear/concise format than is found on the NCDC site (or at least it appears to be so), but again, I think far fewer folks will reference said database relative to SPC Prelim logs.

With this in mind, what does it hurt to issue an LSR after an event? That way, it'll be in the SPC Prelim Reports database. IMO, some info is better than no info. My question is this -- why NOT issue an LSR after an event, particularly when presented with solid and reliable information that the event actually occurred?

 

[rdale]

NWS policy is that as long as it's within 7 days of the event, offices WILL issue a LSR. I don't see how this can be an option that local offices can avoid:

5. Preliminary Local Storm Report (product category LSR).

5.1 Mission Connection. Preliminary Local Storm Reports provide the Storm Prediction
Center (SPC), adjacent WFOs, the public, media and emergency managers with reported
observations of hazardous weather events. Preliminary Local Storm Reports also serve as the
primary basis for the official monthly publication “Storm Data.â€￾

5.2 Issuance Guidelines.

5.2.1 Creation Software. WFOs should use the AWIPS LSR generation software for reports.

5.2.2 Issuance Criteria. WFOs will issue LSRs for severe weather events such as tornadoes,
waterspouts, large hail, thunderstorm/marine wind gusts and flash floods. WFOs should issue
LSRs for other events listed in Appendix B. LSRs should be issued for events that meet or
exceed applicable warning criteria. Other events may be included. LSRs should be issued as
close to real time as possible. WFOs should issue LSRs to “summarizeâ€￾ a list of reports during
and/or at the end of an event (e.g. severe weather outbreak, winter storm). Events reported more
than seven days after occurrence will be included in monthly Storm Data reports instead of
LSRs.

 

[Donald Van Dyke]

Our policy on issuing LSR's for tornadoes has not changed.

First of all, the LSR is usually most effective during, or perhaps immediately after an event. It is used mainly to relay near realtime storm reports to downstream NWS offices, the local broadcast media and the SPC.

We do not report tornadoes in an LSR unless we have high confidence that 1) the tornado actually occurred and 2) we know the location and time of the tornado. Even though LSR's are called PRELIMINARY storm reports, many seem to ignore the preliminary part and take what is sent in an LSR as the final truth (witness the national media listing 60+ tornadoes from Sunday). If we are confident in the basic facts about a tornado, we will report it in an LSR.

A tornado report does not become "official" until it is entered into Storm Data. By that time, the WFO has had time to research the event, gathering all available reports, images and videos, and comparing that information with radar data. Then and only then does the event become part of the official database.

Concerning surveys...we do not survey every single tornado report we receive. Due to resource limitations and other factors, we survey what we believe to be "significant tornadoes." So far this year, we have not had any significant tornadoes in our CWA.

I hope this helps. You should note that I speak only for WFO OUN, and not for any other offices.

Rick
[/b]

Respectfully, the following part of this is flawed reasoning IMO.

Even though LSR's are called PRELIMINARY storm reports, many seem to ignore the preliminary part and take what is sent in an LSR as the final truth (witness the national media listing 60+ tornadoes from Sunday)[/b]

I agree totally with the point that many people may ignore the prelim part, however, that is THEIR problem. It states "preliminary" on the product. IMO what is happening now is that WFO OUN is trying to create another product out of the LSR from what it really is. In other words, I don't think they're using the LSR for its true purpose because of the mis-use of it that they feel is coming from other people (public, media, etc). I guess I just feel that it's circular reasoning to not issue a product that's technically sound because a few people are using it wrong. I think it's better fundamentally to educate the people who are mis-using it rather than alter the product to compensate for improper behavior when there's really nothing technically wrong with the product the way it is.

I hope that has made a little sense and doesn't sound too confusing.

 

[Donald Van Dyke]

Shane said that the folks at WFO OUN told him not to look for an LSR for the tornado(es) on that day. [/b]

NWS policy is that as long as it's within 7 days of the event, offices WILL issue a LSR. I don't see how this can be an option that local offices can avoid:

5. Preliminary Local Storm Report (product category LSR).

5.1 Mission Connection. Preliminary Local Storm Reports provide the Storm Prediction
Center (SPC), adjacent WFOs, the public, media and emergency managers with reported
observations of hazardous weather events. Preliminary Local Storm Reports also serve as the
primary basis for the official monthly publication “Storm Data.â€￾

5.2 Issuance Guidelines.

5.2.1 Creation Software. WFOs should use the AWIPS LSR generation software for reports.

5.2.2 Issuance Criteria. WFOs will issue LSRs for severe weather events such as tornadoes,
waterspouts, large hail, thunderstorm/marine wind gusts and flash floods. WFOs should issue
LSRs for other events listed in Appendix B. LSRs should be issued for events that meet or
exceed applicable warning criteria. Other events may be included. LSRs should be issued as
close to real time as possible. WFOs should issue LSRs to “summarizeâ€￾ a list of reports during
and/or at the end of an event (e.g. severe weather outbreak, winter storm). Events reported more
than seven days after occurrence will be included in monthly Storm Data reports instead of
LSRs.
[/b]

Another point about issuing LSRs......don't they automatically go to the proper day on the SPC prelim report page when they're issued, even if it is a couple of days after the event? If so, then why not go ahead and issue an LSR statement at that time after the tornado is confirmed so that at least some accuracy is added to the "prelim" reports page that SPC maintains? If the complaint about the LSR product and prelim SPC storm report page is lack of accuracy, then that could be improved upon IMO by following the above suggestion.

 

[beaudodson]

Our policy on issuing LSR's for tornadoes has not changed.

First of all, the LSR is usually most effective during, or perhaps immediately after an event. It is used mainly to relay near realtime storm reports to downstream NWS offices, the local broadcast media and the SPC.

We do not report tornadoes in an LSR unless we have high confidence that 1) the tornado actually occurred and 2) we know the location and time of the tornado. Even though LSR's are called PRELIMINARY storm reports, many seem to ignore the preliminary part and take what is sent in an LSR as the final truth (witness the national media listing 60+ tornadoes from Sunday). If we are confident in the basic facts about a tornado, we will report it in an LSR.

A tornado report does not become "official" until it is entered into Storm Data. By that time, the WFO has had time to research the event, gathering all available reports, images and videos, and comparing that information with radar data. Then and only then does the event become part of the official database.

Concerning surveys...we do not survey every single tornado report we receive. Due to resource limitations and other factors, we survey what we believe to be "significant tornadoes." So far this year, we have not had any significant tornadoes in our CWA.

I hope this helps. You should note that I speak only for WFO OUN, and not for any other offices.

Rick

Wow that is unfortunate. Sounds like we will, for years to come, have incomplete storm data from Oklahoma. It is difficult to rate tornadoes if one doesn't do a storm survey. Nearly impossible (if not impossible).

How would you know that you have not had any significant tornadoes...if you don't do a full survey? Other offices have the staff to do surveys. How is that? Did the federal government cut back on the office out there? This could be the case, I am not familiar with the office in question. I am sure someone on the forum could answer that question. Obviously if the staff has been cut then that would be reason to not do storm surveys. If that is not the case then how is it that other offices seem to find the resources to do surveys? I am curious.

It also sounds like, even with high confidence, that the offices isn't admitting there was or wasn't tornadoes in the local storm reports. See Evans remarks above. His remarks seem to indicate that chasers and spotters called in tornadoes and they still didn't show up in local storm reports - even though the people calling them in were credible individuals - more so than most offices could ever dream of! Yet even with that fact we did not see local storm reports. Is the policy such that only if a NWS Employee calls in and reports a tornado that then and only then will a local storm report be issued? I don't believe it is relevant that the reports are preliminary - of course they are preliminary but that doesn't mean they are not useful.

I can understand that there are always going to be some reports that seem fishy enough to not warrant a local storm report. That does not appear to be the case here though. Obviously when TRAINED spotters and chasers are taking photos and video of tornadoes then it is the real deal. I think most NWS Employees can tell the difference - not to mention that the NWS has radar and all their other equipment to help determine whether or not there is a possible tornado in the area where one is being reported. It sounds like the office in question just doesn't want to put out an LSR. For their own reaons (whatever that might be).

On another note most offices issue local storm reports immediately after the report is received. This is beneficial to emergency management, media, and the public. Why is this subject treated differently out there in Oklahoma? Does the media, public, and emergency management have some other mechanism for receiving storm reports from that area? If so then what is it? Do tell! I am sure most in here would be curious how that works. Has the staff been cut so much that not only can they not do storm surveys but they also don't have enough personnel on duty to put out local storm reports? If this is the case then that is a good excuse. Nothing you can do about that.

It would seem to me, and this is just my opinion of course, that it is a disservice to the public and the safety of the public to withhold local storm reports. I can only speak for my own area but we find them extremely crucial in determining what one county to the next can expect from a storm system. Obviously if the county next to me has baseball size hail and several spotter/chaser reports of a tornado on the ground then I would like to know about that - in as much detail as possible. KPAH normally will issue a local storm report as soon as they receive the information...many times within a minute or two. I have noticed most offices seem to operate in that manner. So what makes Oklahoma an exception to the rule? I am curious. Are there different policies at each office? I see that Rdale posted some information on that. If there are not different policies then perhaps the locals out in that region should contact someone higher up, within the National Weather Service, concerning this matter.

 

[Jay McCoy]

OUN doesnt do surveys unless the tornado is signifigant?? What is the criteria? bodies?? I know on 5/5/07 there was a large long lasting tornado that crossed I-40 and hit Sweetwater with major damage to a number of houses and the high school. This event was video taped and shown on OKC tv stations so obviously OUN knows about it yet no LSR or rating. That same night we had the Vici wedge that numerous chasers saw and it also caused damage but again no survey or rating. Arnett the day before damaged some buildings but again no survey. I can understand when they are like DDC and have so many tornado paths they cant cover them all but since OUN hasnt had hardly any tornados in the past few years you figure they would have the time to check the couple tornados they did have. Sounds to me like OUN is dropping the ball big time or being lazy. Either way they are not serving their tax funded purpose. I bet DDC would love to be able to sit back and relax. Maybe its been so long since OUN has seen a tornado they forgot how to do surverys.

 

[Kiel Ortega]

OUN doesnt do surveys unless the tornado is signifigant?? What is the criteria? bodies?? I know on 5/5/07 there was a large long lasting tornado that crossed I-40 and hit Sweetwater with major damage to a number of houses and the high school. This event was video taped and shown on OKC tv stations so obviously OUN knows about it yet no LSR or rating. That same night we had the Vici wedge that numerous chasers saw and it also caused damage but again no survey or rating. Arnett the day before damaged some buildings but again no survey. I can understand when they are like DDC and have so many tornado paths they cant cover them all but since OUN hasnt had hardly any tornados in the past few years you figure they would have the time to check the couple tornados they did have. Sounds to me like OUN is dropping the ball big time or being lazy. Either way they are not serving their tax funded purpose. I bet DDC would love to be able to sit back and relax. Maybe its been so long since OUN has seen a tornado they forgot how to do surverys.

Ooops...someone needs to learn about google http://www.srh.noaa.gov/oun/wxevents/20070505/
In fact, go to http://www.srh.noaa.gov/oun/wxevents

Can't comment on the "wedge", except I know Rick surveyed Sweetwater and from what I hear, that took awhile and so I bet he didn't have the daylight to get up to Vici. Further, it didn't sound like the Vici "wedge" was really that significant (as in, all I heard was minor house damage). Remeber too, that the NWS office gets information that you don't know about. I didn't know that some counties will survey damage and then send the reports to the NWS. With the new EF scale, what's the point to surveying unless the damage sounds significant?

And I'm gonna get a little pissed here, because I've surveyed 5 tornadoes this year for OUN; one had a PNS for it, the others didn't. All were EF0-1 (though I still debate the El Reno rating with those with a higher pay grade )---none a significant tornado. So, thanks...I guess I'm lazy for taking the time to do these surveys. In fact, I know my group at NSSL has surveyed 10-12 tornadoes for OUN this year in support of other projects held here in Norman. So, thanks, I'll tell everyone they're lazy just because you didn't get a precious public info statement.


Arnette has been discussed elsehwere so I won't touch it again; I will note that several of us from NSSL were out there and in contact with OUN about the situation. Again, though, what's the point of surveying nothing? Other than chasers' having a pretty map to point at (and this crap about warning skill and what not, read my thesis when I finish it and you'll see why reporting is worthless anyways for scoring warning skill....errr, that's another rant for another time). I think what's more pathetic is how a tornado was reported on ST before it was reported to OUN. Wow...where was the uproar for that?!

 

[Jeff Snyder]

For the record, Rick Smith is the Warning Coordination Meteorologist at the OUN NWSFO. One can better argue or present a point if one does so tactfully and respectfully. I'm not necessarily speaking of any posts in this thread yet, but please keep this in mind.

See http://www.srh.noaa.gov/oun/wxevents/ for some of the detailed damage surveys from this spring (3/29 OKC tornado, 5/4 Sweetwater, 5/5 Oklahoma and Seminole Co. tornadoes, and 5/8 El Reno tornado). I expressed my opinion on this recently in a different thread, though I'll say that I'd be nice to use the new Damage Indicators for trees (which can be used to rate as high as EF3) to accurately rate tornadoes that may have, in the past, been given only an EF0 rating owing to the lack of significant structures hit. People worrying about the effect of the new EF-scale on tornado climatology should worry, IMO, equally about "presumed EF0" tornadoes that are assigned that rating because the tornadoes weren't "significant". If more surveys have been completed, then I think many would just like to see the results. Just a request for more disclosure

 

[Shane Adams]

I know it's not a priority for any NWSFO, but I'd be willing to wait as long as it took to do a survey, without complaining, if I knew a Public Information Statement would be released about it. I don't understand why those aren't mandatory. It's nice to know what happened once the NWS figures it out. I know about "Storm Data", but so much time passes between the official confirmation and the release of that event's particular month of Storm Data, the point is kinda moot (as far as knowing what happened while the event's still fresh).

 

[Kevin Scharfenberg]

Personally, I think it's much smarter for a forecast office to spend their limited funds and staff hours on training and research & development, and not chasing down every last F0-F1 spinup in a wheat field. Even if a barn or two is hit, or there is some significant tree damage, this can be handled with relatively high precision by looking at digital photos taken by the local media or emergency manager. Nor do I think it's smart to waste time and energy typing up a public info statement for every last one of these brief spin-ups just to satisfy the curiousity of a handful of weather weenies who can't wait another month or two for Storm Data (and Jeff, there is disclosure...it's in this publication).

 

[Glen Romine]

Might have been shown elsewhere and I missed it - but for those looking for tornado ratings from OUN that don't want to wait for the Storm Data publication, the following web page appears to have farily up-to-date summary information:

http://www.srh.noaa.gov/oun/tornadodata/ok/tornado2007.php


In terms of the politics of to survey or not to survey, I think it is very unrealistic to think more than a few tornadoes will ever have a detailed survey done. Perhaps even fewer than before with the new EF scale, as it now requires a survey team to look in much greater detail (even more time consuming). I doubt there is much hope of better surveying until private industry brings money to get it. Certainly insurance companies would benefit from a better knowledge of tornadic wind climatology - but may not be willing to pay enough for it. Surveys are also time critical for best accuracy, and when the weather pattern is active it is unrealistic to think the NWS can spare the manpower to survey any but the most likely to expose failure in the NWS warning system.