AccuWeather Strikes Again

[PSUBlizzicane2007]

http://www.washingtonpost.com/blogs/capital-weather-gang/post/yesterdays-storms-ground-shaking-thunder-plus-controversy-in-college-park/2011/10/13/gIQA4IVfjL_blog.html

Just had this sent to me in an email.

I think this is getting into pretty dangerous territory, and someone needs to prevent AccuWeather from pulling stunts like this, especially since there was no corresponding Tornado Warning from the NWS, nor was there any actual tornadoes reported near UMD.

On the flip side, UMD should have also paid more attention to the NWS instead of brazenly warning the entire campus based on a questionable alert system from a private company.

Thoughts on this?

[nrgjeff]

Private commercial weather forecasting firms have been issuing their own customized alerts for years now (15-20+ years). They are not the only one. The alerts are business specific, and not intended for the general public. I personally issued a tornado alert for a client, which was struck by a damaging tornado, but the storm had been declared outflow dominant. I saw it differently, and correctly. On the other hand I have talked Facilities/Security through tornado warnings in their county in which the storm would miss them by a safe distance.

That being said, the NWS is absolutely the best source for the public to receive warnings. I believe hundreds or thousands of lives are saved each year by NWS warnings. Keep hurricanes and winter storms in mind too. The false alarm rate is dropping for tornadoes, even as we still err on the side of safety. The public needs a standardized official information source. Viva NWS!

[Parsley]

http://www.washingto...IVfjL_blog.html

Just had this sent to me in an email.

I think this is getting into pretty dangerous territory, and someone needs to prevent AccuWeather from pulling stunts like this, especially since there was no corresponding Tornado Warning from the NWS, nor was there any actual tornadoes reported near UMD.

On the flip side, UMD should have also paid more attention to the NWS instead of brazenly warning the entire campus based on a questionable alert system from a private company.

Thoughts on this?

If someone wants to throw $$$$ Accuweather's way.....they (the clients) are one ones to blame. UMD should know better and follow the NWS.

[Ed Lizard]

Was this the storm in question by campus? That lowering does look suspicious. (Disclaimer- not an expert, never chased, but I watch lots of chaser shows on TV)

I don't suppose anybody has a radial velocity capture of the cell about the time AccuWeather sent an alert...

[Ellinwood]

Was this the storm in question by campus? That lowering does look suspicious. (Disclaimer- not an expert, never chased, but I watch lots of chaser shows on TV)

I don't suppose anybody has a radial velocity capture of the cell about the time AccuWeather sent an alert...

That lowering is a shelf cloud and does not spawn tornadoes.

As far as the radar, there wasn't really any indication of rotation... just some strong outflow winds. NWS did a fairly good job with warnings in the region that day (though one or both of the initial tornado warnings had public/police reported stuff before they got issued, so that's kind of a strike against).

[The Dude]

That lowering is a shelf cloud and does not spawn tornadoes.

As far as the radar, there wasn't really any indication of rotation... just some strong outflow winds. NWS did a fairly good job with warnings in the region that day (though one or both of the initial tornado warnings had public/police reported stuff before they got issued, so that's kind of a strike against).

Yup. There were some decent couplets between Richmond and Fairfax, but the rotation became broad very quickly once the storms got North of 66. There was never anything but an area of broadening rotation heading towards College Park.

[tornadotony]

That lowering is a shelf cloud and does not spawn tornadoes.

As far as the radar, there wasn't really any indication of rotation... just some strong outflow winds. NWS did a fairly good job with warnings in the region that day (though one or both of the initial tornado warnings had public/police reported stuff before they got issued, so that's kind of a strike against).

Um, that thing in the right background doesn't exactly look like a shelf cloud. Agreed that there is one in the foreground, but that looks like maybe RFD around the feature above the treeline.

[Ellinwood]

Um, that thing in the right background doesn't exactly look like a shelf cloud. Agreed that there is one in the foreground, but that looks like maybe RFD around the feature above the treeline.

It looks like it's all one feature to me... just going around the main downdraft+rain shaft (not rotating around it).

[Jesse Ferrell - AccuWx]

Our official reply is this:

On October 13 at 6:38 p.m., AccuWeather issued a warning to the University of Maryland that the College Park campus was at risk of tornadic activity. Though no tornadoes subsequently touched down on the College Park campus, we had verified the conditions that create tornadoes were present. The National Weather Service confirmed that multiple tornadoes touched down that evening to the south of campus.

Unofficially, I would add:

1. It's not unusual for private weather companies to issue warnings to their clients. In this case we were probably trying to give them a longer lead time to a potentially dangerous situation.

2. We certainly don't tell them what to do with them or how to issue them.

3. Here is a plot of the NWS Warnings. College Park is shown in the upper right. Not counting County Warnings, measuring in Google Earth it appears the closet NWS polygon tornado warning was about 12 miles away and the nearest NWS tornado report was 38 miles away from College Park.

[k***]

lol @ any alert that says a tornado will hit in 12 minutes

[Sunny and Warm]

lol @ any alert that says a tornado will hit in 12 minutes

Can you show me the alert that says that? thanks

[AtticaFanatica]

The alert that was sent out to students said "within 10 minutes." I think that was their (EM) interpretation of the information they received however, so it's not clear to me Accu-Weather forecasted a tornado in some amount of time. I hope not anyway.

[Ellinwood]

Can you show me the alert that says that? thanks

Wasn't 12 minutes specifically, but it still illustrates his point (the time estimates were indeed lolworthy).

http://www.foxnews.com/us/2011/10/16/university-maryland-police-defend-tornado-warnings/

Thunderstorms rolled over the campus Thursday night when university police sent out an email at 6:28 p.m. indicating that a "tornado is forecast to strike campus within the next 13 minutes." A follow-up at 6:52 p.m. said a tornado could touch down within 10 minutes.

A third message at 6:59 p.m. said a tornado could strike within three to four minutes.

[Dalfy]

The alert that was sent out to students said "within 10 minutes." I think that was their (EM) interpretation of the information they received however, so it's not clear to me Accu-Weather forecasted a tornado in some amount of time. I hope not anyway.

this is the text that I got that day:

University of Maryland a tornado is forecast to strike the campus within the next 10 minutes. Seek shelter immediately.

Then we got another text saying:

University of maryland a tornado is forecast to strike the campus within the next 3 to 4 minutes. Remain in shelter until advised.

I was skeptical the whole time, especially after I checked ncep and didn't see any warning at all.

[PSUBlizzicane2007]

Our official reply is this:

That is an extremely irresponsible reply to the event. The tornadoes that touched down were SW and well into Virginia... not "to the south of campus" which makes it sound like the tornadoes were extremely close. It was clear to most meteorologists that the storm was weakening as it approached CP.

Unofficially, I would add:

1. It's not unusual for private weather companies to issue warnings to their clients. In this case we were probably trying to give them a longer lead time to a potentially dangerous situation.

This is going to also be in response to nrgjeff's post above. While I appreciate the market that weather risk companies are filling, it is extremely irresponsible to put out individual tornado alerts like this. I completely agree that you should put out generalized tornado alerts (a version of a tornado watch or notifying clients that tornadic storms are possible that day)... but all localized, specific warnings should be from the NWS. I would say that you could act as a relay of that information.

2. We certainly don't tell them what to do with them or how to issue them.

Of course, and this is where UMD is at fault. They should have known better to corroborate this information with the NWS. Unfortunately, instances like this can result in weather risk companies losing clients because they acted irresponsibly, however.

3. Here is a plot of the NWS Warnings. College Park is shown in the upper right. Not counting County Warnings, measuring in Google Earth it appears the closet NWS polygon tornado warning was about 12 miles away and the nearest NWS tornado report was 38 miles away from College Park.

This is not saying much considering how much tornado warning locations have improved over the past few years. Regardless of how far away they were (which, by the way, 12... and more specifically 38 miles... are not distances that any forecaster should consider remotely accurate.

If AccuWeather meant for this to be a more generalized statement, then they should say that up front. Causing clients like UMD to get confused about the differences between Tornado Watch, Tornado Warning, and whatever Tornado Alert AccuWeather issues is the prime reason why private forecasters should not issue such alerts. Confusing the public in emergency situations is not acceptable.

[isohume]

"Though no tornadoes subsequently touched down on the College Park campus, we had verified the conditions that create tornadoes were present."

What does this mean..."verified the conditions"? And why was it used as a rationalization?

[Ed Lizard]

"Though no tornadoes subsequently touched down on the College Park campus, we had verified the conditions that create tornadoes were present."

What does this mean..."verified the conditions"? And why was it used as a rationalization?

Is that like a watch? Conditions were favorable? Or did they have strong rotation indicated by radar, ie, like a warning?

[isohume]

Is that like a watch? Conditions were favorable? Or did they have strong rotation indicated by radar, ie, like a warning?

Not sure...but either way it makes it sound like it was a hit. I'd like to examine the NEXRAD data just for kicks.

[PSUBlizzicane2007]

"Though no tornadoes subsequently touched down on the College Park campus, we had verified the conditions that create tornadoes were present."

What does this mean..."verified the conditions"? And why was it used as a rationalization?

I'm more concerned with the "tornadoes occurred south of campus" statement.

[Ian]

Um, that thing in the right background doesn't exactly look like a shelf cloud. Agreed that there is one in the foreground, but that looks like maybe RFD around the feature above the treeline.

Storm about 2 hours earlier. Based on the description from the photog I think anything in that cell would have been rain wrapped from his location.

re: the warning etc. I have a comment in that article in the comment section which covers my thoughts rather than retype them here. I was covering CWG at the time and we were getting bombarded with people wondering why the sirens were blaring in College Park.

There was weak remnant rotation that was continnually weakening as the storm became outflow dominant.

I wonder if the college made the wording more threatening than they were passed. I sorta hope so. The idea that they can say "a tornado will hit campus in 13 minutes" is kinda silly.

[Ian]

The alert that was sent out to students said "within 10 minutes." I think that was their (EM) interpretation of the information they received however, so it's not clear to me Accu-Weather forecasted a tornado in some amount of time. I hope not anyway.

----- Original Message -----

From: "Emergency Alert" [[email protected]]

Sent: 10/13/2011 06:28 PM AST

To: "UMD Alert Users" <[email protected]>

Subject: Tornado Warning

A tornado is forecast to strike the campus within the next 13 minutes. Seek shelter immediately.

Sent by UMD Alerts to All users (E-mail, Pagers, Cell phones) through UMD Alerts

... powered by the Roam Secure Alert Network

[Ian]

Our official reply is this:

Unofficially, I would add:

1. It's not unusual for private weather companies to issue warnings to their clients. In this case we were probably trying to give them a longer lead time to a potentially dangerous situation.

2. We certainly don't tell them what to do with them or how to issue them.

3. Here is a plot of the NWS Warnings. College Park is shown in the upper right. Not counting County Warnings, measuring in Google Earth it appears the closet NWS polygon tornado warning was about 12 miles away and the nearest NWS tornado report was 38 miles away from College Park.

There were no active NWS tornado warnings for about 45 minutes to an hour prior to it getting to College Park.

Here is an SRV loop, maybe the first frame or two were still warned: http://t.co/Vy7JjifL

It's not necessarily that there was no rotation etc. But the warning was odd nonetheless. Especially when you factor int he fact that you guys repeatedly bash the NWS for overwarning people and it being unhelpful to do so. Why does the same not apply when you have your client sound sirens for a half hour followed by a later statement to disregard all future warnings?

[Sunny and Warm]

----- Original Message -----

From: "Emergency Alert" [[email protected]]

Sent: 10/13/2011 06:28 PM AST

To: "UMD Alert Users" <[email protected]>

Subject: Tornado Warning

A tornado is forecast to strike the campus within the next 13 minutes. Seek shelter immediately.

Sent by UMD Alerts to All users (E-mail, Pagers, Cell phones) through UMD Alerts

... powered by the Roam Secure Alert Network

It would be of interest to see the exact warning Accwx gave to UMD. I'd like to see whether the wording was upgraded by UMD on their own or if that is verbatim what Accwx gave to them. If verbatim, then that IS irresponsible of Accwx.

[BethesdaWX]

That lowering is a shelf cloud and does not spawn tornadoes.

As far as the radar, there wasn't really any indication of rotation... just some strong outflow winds. NWS did a fairly good job with warnings in the region that day (though one or both of the initial tornado warnings had public/police reported stuff before they got issued, so that's kind of a strike against).

Can't those gust fronts produce spin-up twisters on the outer edge?

[Ian]

Can't those gust fronts produce spin-up twisters on the outer edge?

even less so than a weak ef0 gustnadoes are not long-track. there was a tornado confirmation in fairfax from a different cell but it was characterized as sporadic damage. dc and e/w was basically the dividing line of good ingredients v poor ingredients. there was seemingly very little likelihood that a tornado would be on the ground for tens of miles at that pt if ever.

[BethesdaWX]

even less so than a weak ef0 gustnadoes are not long-track. there was a tornado confirmation in fairfax from a different cell but it was characterized as sporadic damage. dc and e/w was basically the dividing line of good ingredients v poor ingredients. there was seemingly very little likelihood that a tornado would be on the ground for tens of miles at that pt if ever.

Oh accuwx had the thing warned all the way through? Not surprised, Accuwx always overkillz. I was under the impression it was the heavy straight-line winds bordering an area of more stagnant air that sparked the warning but regardless I'm not sure why anyone would choose accuwx over the NWS, one is free, one is not..

[Ian]

Oh accuwx had the thing warned all the way through? Not surprised, Accuwx always overkillz. I was under the impression it was the heavy straight-line winds bordering an area of more stagnant air that sparked the warning but regardless I'm not sure why anyone would choose accuwx over the NWS, one is free, one is not..

It was two supercells becoming embedded in a line and weakening. They were warning the remnant couplet.

[AtticaFanatica]

I wonder if the college made the wording more threatening than they were passed. I sorta hope so. The idea that they can say "a tornado will hit campus in 13 minutes" is kinda silly.

Yeah, that's what I was getting at.

[MGorse]

There were no active NWS tornado warnings for about 45 minutes to an hour prior to it getting to College Park.

Here is an SRV loop, maybe the first frame or two were still warned: http://t.co/Vy7JjifL

It's not necessarily that there was no rotation etc. But the warning was odd nonetheless. Especially when you factor in the fact that you guys repeatedly bash the NWS for overwarning people and it being unhelpful to do so. Why does the same not apply when you have your client sound sirens for a half hour followed by a later statement to disregard all future warnings?

This.

[CheesyPoofs]

I'm sure part of the elevated wording in the UMD texts/the fact that UMD took it so seriously was because of their past history. As many remember, the school was hit by an F3 in 2001 that unfortunately killed 2 people.

The general public isn't as knowledgeable about weather as your typical weenie. Given past history, even if there is no NWS warning, if any met tells them that a tornado "is coming" (disregarding the irresponsibility of that warning), the university is probably going to want to act.

That excuses the university for their reaction. That doesn't excuse Accuweather for their "warning," however.