WARNING: TEXT IN RED WAS ADDED AFTER THE ORIGINAL ARTICLE WAS PUBLISHED, IN RESPONSE TO COMMENTS.
UPDATE 9/15/09: The Associated Press did not respond to my request to interview Mr. Lammers.
DISCLAIMER: AccuWeather sells the AP Photo Library for the Associated Press.
UPDATE 8/30: It would seem that the NWS office in question is as uninformed about this issue as we all are. They said today in an email "We don't know what the term "alert" is referring to, "who" made the call or "who" called off this alert or when this occurred. We suspect it might have been a private forecasting firm. We just don't know. I believe all this information originated with AP so unfortunately we don't have much to go on. There has never been a Tornado Warning issued by this office based on a TVS generated by algorithms detected near the wind farms."
So it appears it was NOT a Tornado Warning (as I had assumed - and hereby apologize for that assumption - based on the Associated Press article), at least not one issued by the NWS in Dodge City, but no one seems to know who or what it was. I will email the Associated Press to see if there is some way I can get in contact with the writer of the article, Dirk Lammers.
UPDATE 8/30: The naysayers continue to leave Comments, but none can explain what type of "tornado alert" is not a Tornado Warning but must be "called off" by the NWS. If you know what really went down, let me know. I am continuing to research this, including asking the NWS and other local meteorologists what happened.
UPDATE 8/29: Blog reader Andy has left a Comment below that it may not have been a Tornado Warning, but some other kind of "tornado alert" that did not go out to the public. To be fair, the article I quoted does not say "Tornado Warning," though this article quoting a local
meteorologist oops - "reporter" -- does.
ORIGINAL POST: There's an Associated Press story today (and others on Google News) talking about wind turbine farms showing up on NEXRAD Radar. I've been following this story for quite a while, after I visited our local wind farm in 2007.
RANDOM PHOTOS FROM MY TRIP TO THE LOCAL WIND FARM (RELOAD FOR MORE | SEE ALL)
The article says that both amateur and professional weathermen have been fooled by the signatures, which may look like a thunderstorm. In fact the government even issued a false Tornado Warning in Kansas. Here's an example of the problem of wind farms appearing on radar, courtesy the NWS:
This is not new news; the NWS has an entire website dedicated to this subject that's been online for several years. What is news is the affect its having on professional and amateur weather forecasters as wind farms become more popular. Here is a related article about a wind farm in the NWS-Milwaukee's area (thanks Scott L.).
NOTE: I'm not going to rehash Wind Power Myths or pros & cons here, because I did that last year in this article.
Here are some important points:
1. This should not be fooling trained professionals into issuing Tornado Warnings
2. For amateur radar viewers, this is not the only kind of false echo that you have to worry about
3. The effect is limited to a small distance from the radar, but you may think you saw them further out
4. The NWS is unable to remove wind farms like they would mountains or other clutter.
Here is some more information on each point:
1. I was surprised to hear that trained radar professionals in our government issued a false Tornado Warning based on wind farms. Although the radar signatures can be impressive(see example below), it should be easy to use other tools such as Visible Satellite radar looping, or, heck, webcams. (Disclaimers: Visible satellite or webcams might not update quick enough or be viable in certain situations; in regards to radar loops, the wind farms should always be there and won't move over time -- unless you're in a situation where storms are also not moving, this should be a big clue).
2. The idea of seeing something on radar that looks like a storm, but isn't, is not a new phenomenon. There are several situations where the radar reports back false data. For more on this, see my blog published in 2007 entitled "Radar, Lies & Videotape, Part 2"
3. The NWS site includes line-of-sight maps for each radar site (showing where wind farms could show up on radar, something they generalize as "11-35 miles" in the article). The point is that the radar beam starts at several hundred feet up, often on a hill where hills are available, and points upward from there. The illustration below attempts to show this.
Now, you may swear you've seen your local wind farms out to 100 miles on the radar. I've seen them too. But I have to assume (here in the East anyway) what we're really seeing is actually the mountains that the farms are erected on. The NWS should be able to remove mountains from the radar images at your local office with software masks, but they may not always do that (I assume because it could leave holes in the rain echoes covering them). Possible exceptions include situations in which there is an inversion causing the radar to hit the ground, which can also show interstate traffic which has a similar look, although it's more obviously NOT storms because it's a long thin line.
4. So why doesn't the NWS just remove the wind farms with that same software? As the article points out (this finally "clicked" for me today) the software assumes the objects aren't moving when it removes them. As you can see on the image above, turbine blades (like the interstate traffic) do appear to have a velocity signature (probably because, duh, they are moving!).
P.S.: Wind farms can also affect air traffic radars (at least the ones the Royal Air Force is using in the U.K..
Comments (7): Jeff:
In your update you state that a Meteorologist said it was a warning. You are mispoken, the columnist of that article, Nichole Teich, is a reporter for a Kansas City radio station not a Meteorologist. The actual meteorologist listed in that article stated nothing about a tornado warning, but stated the difference between a wind turbine on radar and a thunderstorm.
FROM JESSE: You are correct, I will fix that. I meant to say "local reporter". However, this still does not change the fact that's what the reporter said. If you can get the reporter to retract the story, let me know.
Posted by Jeff | August 30, 2009 5:14 AM Bill:
I find flaw with your statement above,
"1. I was surprised to hear that trained radar professionals in our government issued a false Tornado Warning based on wind farms. Although the radar signatures can be impressive, it should be easy to use other tools such as Visible Satellite radar looping, or, heck, webcams."
It is pretty bold to lay full responsibility on the forecaster which "MAY" have had this happen to them. Granted you know nothing of the atmospheric conditions of that day (there may have been additional thunderstorm activity in the area, which given an active radar screen the wind turbine may have blended into the scope). The image you posted is misleading as you are inferring to readers that these were the conditions at that time.
FROM JESSE: Good point, I should have made is clearer that it was an example. I have corrected that above in red.
No where in your posting is a date of this occurrence or a capture of the image at that time or even verification from the office of this occurrence, only MEDIA speculation which we all know is 100% accurate!
FROM JESSE: But I can't censor the facts as reported by the media (accuracy nonwithstanding), and I won't let you do that either. If it wasn't a Tornado Warning, you need to prove your case too and you need to correct the original reporter. If it was a Tornado Warning then it IS the forecasters responsibility to issue it and know what he's doing. In the end, he is responsible.
Secondly, lord knows if you've ever been to Dodge City Kansas, but the probability of finding a webcam where this object was located is close to ZERO. In addition, should the radar screen have been active that day, visible imagery would have been of no valuable use due to cloud contamination.
FROM JESSE: You are correct, however, on saying that I should have expanded on how visible satellite or webcams could not be of use. I have added "or viable" in the original post above in red.
The value of TVS algorithms in warning operations is very low and likely hardly utilized by warning operators.
FROM JESSE: I'm glad you agree with my statement above that "errant TVSs are a dime a dozen."
To correct a statement, the WSR-88D does not indicate a TVS, the algorithm which process that data is the responsible party. This algorithm can vary from software vendor to software vendor and is often a point of confusion with amateur users of radar visualization software (usually several settings in the software which can be adjusted for detection or announcement of a TVS as based on processed velocity data).
FROM JESSE: This is a red herring - you guys still haven't answered my question - how can a NWS meteorologist "call off" a TVS, or any other software/hardware alert? Until that question is answered, I have to stick by my assumption that it was a Warning.
Furthermore, without actual proof (a warning text or other) proving this event (claim of a false tornado warning), I would find it very suspicious that a NWS trained warning operator would warn without questioning the Emergency Manager (or point of contact) as to visible confirmation. Should this wind turbine field have been present for some time, I highly doubt a forecaster would fall victim to it.
Then again this was not brought up in your post.
FROM JESSE: Go back and actually read my article. That's why the headline said "Tornado Warning?" (note the question mark) and I said "I was surprised" which you also said above, so I think we are on the same page. Everybody's surprised. End of story.
Posted by Bill | August 30, 2009 5:08 AM Bill:
Jesse, Andy is 100% correct. There is a big difference between the WSR-88D detecting a TVS, and an actual tornado warning being issued. Also, the "article's" author was clearly not well trained. She should have known better than to call a TVS radar signature a tornado warning.
FROM JESSE: Bill, what I still haven't heard is how a NWS Met can "remove" a TVS from the PUP. Please respond.
Posted by Bill | August 29, 2009 12:33 PM Dennis:
Just put radar absorbent paint on the turbine blades. The govt must have a lot of that stuff sitting around with the cancelling of the F22 fighter program...
FROM JESSE: Well... I guess that's a thought!
Posted by Dennis | August 28, 2009 10:57 PM Andy:
Your blog is misleading. The article says:
"In Kansas, it was a computer program that picked up on the pattern and issued the alert. A meteorologist who was aware of the phenomenon quickly called off the alert."
Computers don't issue warnings, their algorithms detect and alert a variety of users to things like tornadic vortex signatures (TVS). It says the meteorologist went on to analyze and dismiss the erroneous TVS. Happens all the time for a variety of reasons. Nowhere did it say the meteorologist was fooled and issued a warning.
FROM JESSE: This article says that it was a Tornado Warning. I understand what you are saying, but you can't "call off" a TVS, there is no way to alter the NIDS data coming out of the PUP (to my knowledge). So I assumed that they were referring to a Tornado Warning. You are correct, errant TVS's are a dime a dozen. Perhaps there is some sort of pre-warning alert in-between NIDS and the Warning creation software that we're talking about, something that would "create an alert" to prompt a NWS meteorolgist to issue a warning?
Posted by Andy | August 28, 2009 1:01 PM Will:
Hey Jesse, This is off the subject, but could we be getting any outer bands from Danny here in Richmond, VA? It has been very hot and dry this month and we need the rain badly. Thanks, Will
FROM JESSE: Check AccuWeather.com for details but much of the Northeast quadrant of the country will get some rain this weekend as a result of Danny and the other systems.
Posted by Will | August 27, 2009 8:56 PM Joe:
Jessie your surprised that the government has failed in another area HA.Also i live here in WNC just west of Ashville and i was wondering if you know what year was the earliest peak of the leaf season?Got a few trees in my yard have shed there leaves already and some have changed.Its only August i have never seen the fall colors come this early and when you go up on the Blue ridge park way there is plenty of color up there.Whats going on?
FROM JESSE: My guess would be we will have an early fall season due to the cool summer but we are researching fall-related info for an upcoming story on AccuWeather.com.Posted by Joe | August 27, 2009 12:15 PM
Thank you for your patience during our recent Comments outage. Comments have returned, including comments on previous stories & blogs before the outage. As before, Comments that don't add to the conversation may be automatically or manually removed by Facebook or AccuWeather. Profanity, personal attacks, and spam will not be tolerated.
Extreme weather hit the West this week, with the heaviest snow and rain in years.
Massive winter storms in the Southeast and California start out the new year.
A large storm in the heart of the U.S. spawned wind gusts over 100 mph and blizzard conditions.
Like there were in past years, there could be tornadoes on Christmas day.