UAL967 encounters severe turbulence


UAL Flight 967 enroute from Washington Dulles to Los Angeles encountered severe turbulence over the central US, causing multiple injuries to passengers and a diversion to Denver:

Flying over rapidly-developing thunderstorms can sometimes be a dangerous thing!


Yeah, I just heard the story on NBC 7/39. Flight number corrected to UA 967. Apparently no deaths, but a few of the injuries were serious. Diverted to DIA. … 15966.html


Grrr… Why is their no flight track available? … /KIAD/KDEN


It’s back now.


As a lowly pax, and maybe we don’t know the details yet, how does this happen? Is it a matter of unforeseen clear air turbulence? Some storm cell that quickly developed and no leading plane experienced? Is there any way for doppler to spot these things?




Flight Attendant: You know, if you do try and get some sleep, the flight will go a lot faster.
Jack Ryan: I can never sleep on a plane. Turbulence.
Flight Attendant: Pardon?
Jack Ryan: Turbulence. Solar radiation heats the Earth’s crust, warm air rises, cold air descends - turbulence. I, I don’t like that.
Flight Attendant: Oh. Well, try to get some sleep anyway.

Outside of PIREPS turbulence is pretty hard to pinpoint usually.


I think doppler on the ground detects changes in wind direction (I.E wind shear) but I don’t think airplanes have doppler?

As Pika already said, CAT (Clear air turbulence) is identified by PIREPS (Pilot Reports). It just may so happen this airplane was the “first report”.


This was certainly not CAT – as the satellite images show, the plane was flying right through a very vigorous thunderstorm. I’m guessing they were using radar to try to avoid what appeared to be the most intense precipitation cores, but they might have encountered a strong updraft (or downdraft) that did not have a strong radar reflectivity signal associated with it.


Where do you see this to be so sure it wasn’t CAT???

Looking at the flight track posted above in Flight Aware, looks to me they were well south of the thunderstorm complex radar returns over IA and KS.

Looking at the satellite images from your own post, and extrapolating the filight path Flight Aware shows, the track was well south of the animated complex moving south.

Your article is very clear that the encounter location is unclear since it was plotted over two states.

Taken from your link
"It is unclear of the exact location of the turbulence encounter, since AWIPS plotted the SEVERE TURB report from UAL967 at Flight Level 34,000 feet over both Missouri and Kansas."


That’s another thing I was going to ask you experts here: how do I know which was the “real” location of the turbulence report? Could they have sent it twice: once at the location of the actual event (within the thunderstorm complex), and again at the waypoint where they diverted to Denver?

I also looked at GOES “water vapor channel” imagery:

which (in non-cloudy air) shows middle-tropospheric features such as gravity waves that might have been generated by such a rapidly-intensifying thunderstorm complex. I didn’t see any evidence of such a CAT-inducing gravity wave on the water vapor imagery south of the storms (which isn’t to say that one didn’t exist…this is only 4-km resolution satellite imagery, after all).


While I can’t answer your question with regards to real location of the turbulence report, PIREPS going back 12 hours can be found at

Again I respectfully ask, what makes you think the actual location of the event was within the thunderstorm complex? What data are you using to determine this that would conflict with the Flight Aware flight path? BTW, the actual flight path was what was filed so it does not appear the pilot asked for any deviations. (blue path vs green line) for weather.

One other thing I should add the radar image provided by Flight Aware was midstream of the flight (as I recall in a previous post) and the path would have been clearly outside any thunderstorm complex.


This is why you should always remain seated with you belt securely fastened! Odds are the injured where those people who ignore all the safety briefings and signs.


Disclaimer: I’m not a pilot…just a satellite meteorologist. And looking at the location of the PIREP over MO, I’d say that they were flying through the cloud mass that was associated with rapidly-growing thunderstorms. Looking at the visible satellite imagery toward the bottom of my blog post, you can see that there were some vigorous overshooting tops (producing small shadows on the cloud anvil top) – and these overshooting tops could have been intense updrafts that were not associated with any radar echoes. Plus, updrafts can have a significant amount of tilt in the vertical in high wind shear environments – so where radar echoes are and where storm-top satellite features are do not (if ever) line up.

So in my non-pilot mind, the real question still remains: where exactly was the SVR TB encounter? Right where they made the turn to DIA? Or back in MO, but they still flew some distance while the crew was assessing the state of the injuries and the pilots were coordinating with ATC for a diversion to DIA?

EDIT: Why didn’t I think of turning to Fox News for the definitive answer? The turbulence encounter was over southwest Missouri. :slight_smile:


Heh, can’t wait to hear from others on your thoughts of using Fox news as a definitive answer :open_mouth:

Also taken from your very Fox source while bopping in there…

“The website, which uses information from the FAA to track the path of aircraft, shows the United jet flew south of a storm in Missouri and Kansas.

It’s just not standard practice for a pilot to fly near a thunderstorm. Severe storms, FAA recommends 20 NM clearance and the flight track on Flight Aware seems to verify the flight complied with that recommendation. The fact that they did not deviate also makes me think they felt they were far enough away from the activity.

For a pilots point of view, … m0701.html may help you what is suggested for course of action.

Scroll down to 7-1-14 for weather avoidance assistance in the above reference…

AOPA thunderstorm flying can be found at Note in this reference, pilots are to avoid flying under the anvil. (as I understand it to avoid possible hail being tossed out)

I fly a small GA plane and it’s ingrained in our training T’storms are to be avoided at all costs.

I just don’t see objective data between the satellite images in your original post and Flight Aware supporting that this flight was inside the thunderstorms.:

I see exactly what you are talking about on the images with the overshooting anvil tops but that seems to come down on the flight path after the flight has flown by.

I learned more about “tilts” in updrafts in my glider lessons these past couple of months then in my 8 years of flying a small GA plane :wink: Pilots become weather weanies in various degrees.


We have been exchanging several emails with folks out at NCAR in Boulder CO on this particular case…and here is a plot I just received in one of the emails. The time of the PIREP was 00:25, and this is the radar reflectivity at that time, along with the position of United Flight 967:

This plot supports the location of the first PIREP in Missouri, as seen on the satellite images I posted on the CIMSS Satellite Blog:

Why the second PIREP shows up in Kansas (at the same 00:25 time) remains a mystery (to me, anyway).


Interesting… Thanks for this insight.

I wonder if the pirep update at 725 is related to the descent beginning at 7:00? The descent looks like it started 60 miles west of St Louis per Google Earth.


Now wait, I’m getting even more confused now. Assuming this flight log is accurate:

there was a significant drop in altitude (from FL380 to FL340) from 00:00 to 00:03 UTC. Or is that fairly representative of a typical altitude correction along such a flight?

Anyway, if that was indeed the severe turbulence encounter at 00:00 to 00:03 UTC, why does it show up as a time of 00:25 UTC on the PIREP? Are such events reported automatically, or is it still a manual process that is done after things “calm down a bit”?

Sorry for asking all the dumb questions here…just trying to get a better understanding of how all this works. We are striving to gather good satellite cases with signatures of potential turbulence, but chasing down the exact time and location of the encounter often seems to be a difficult piece of the puzzle.


No dumb questions!

The descent could be for many reasons, one being my conjecture that may be where the severe turbulence happened to maybe it was ATC working traffic. I would find it unusual myself only because it would not be part of a normal descent profile (strictly my opinion and conjecture)

In my experiences, filing a PIREP is a completely manual process

To get a feel how a pirep is filed in flight and how much a manual process it is, I have a video at If you want to skip to the actual PIREP, my PIREP begins at 3 minutes into the video where I file a PIREP with Flight Service.

As you can see it took quite some time to file the report. Not sure how long it takes to get from FSS to the websites that host PIREPS.

As you can see in the video, it took me 2 minutes to complete an in flight PIREP. 2 minutes flight time in a jet can represent a significant amout of distance like nearly 10 miles in 2 minutes as round figures. In my lil plane I covered 2 miles a minute on a good day :slight_smile:

So for your purposes of exact time, I’d suspect it would be virtually impossible to get the exact time and place because the PIREP as you indicated can be filed later then the event. Person receiving the PIREP may annotate the event time as time received. Conjecture on my part but a possibility.

Urgent PIREPS (UUAs) of icing and severe turbulence are reported directly to ATC (centers and approaches) and as I understand it, they input into the computer. How long it takes for it to update again would be a mystery to me.

Below is a link to the standardize PIREP form.



This might be what the AIM recommends but this is not always what happens.
I follow other flagged carriers threw, around and close to active thunderstorms all the time.

If I remember, the weather image is not alway a match to the flight path on FA?

My guess is that they flew threw the anvil (overhang) of a big ass storm. If I remember right that storm from was spewing 1" hail and tornadoes.


Yep. … 5843#85843

WAG would be the weather depicted on the map would be when the flight was over Indiana crossing into MO (1/2 way point) and probably not far off what would have been depicted when the initial descent was from 38K down to 34K.

EDIT - Oh Hell, the timestamp of the weather is on the map DUH. 7:07 was the time for the radar.