AA 106 @ JFK (13 Jan 23)

[wd-15717]

Apparent near-miss at JFK last night crossing active runway in front of a departing Delta.

[DIBO]

Based on FR24 replay, could have been this:
https://cimg2.ibsrv.net/gimg/pprune.org-vbulletin/725x601/aa106_27435f36ca07f2c42ea24d49db59d7d2d49465f8.jpg

[UnreliableSource]

https://twitter.com/xJonNYC/status/1614370783899303938

[palomar92000]

Apparent near-miss at JFK last night crossing active runway in front of a departing Delta.

Looks like AAL106 crossed 4L at Juliet, about 4000' down the runway. Seems like a very close call.

[dixi188]

I don't know JFK, but, don't they have red stop bar lights at all runways?

[BFSGrad]

I don't know JFK, but, don't they have red stop bar lights at all runways?
This FAA diagram shows REL at J crossing 4L.

KJFK RSL (https://www.faa.gov/air_traffic/technology/rwsl/media/JFK.pdf)

[finfly1]

This has surprisingly few comments. It could have easily been an award winning disaster.

Listening to the voices of the two pilots and two controllers was enlightening. It will be interesting to see the experience levels of some of the players if a report IS made public..

[skydler]

ATC audio:

5AbpxqtXcPk

[DaveReidUK]

NTSB investigation launched.

Comparison of (A) presumably intended ground track (as taken by many of the preceding 4L departures) and (B) actual ground track:

(A) follows ring taxiway Bravo; as Bravo starts to curve round to the northeast, takes the 3rd right exit and crosses the runway ahead

(B) follows ring taxiway Bravo; as Bravo starts to curve round to the northeast, takes the 4th right exit and crosses the runway ahead


A:
https://cimg7.ibsrv.net/gimg/pprune.org-vbulletin/453x555/intended_track_65ca0f0dd2f7a2235b873b48b7a1faf4a618ca7f.jpg


B:
https://cimg8.ibsrv.net/gimg/pprune.org-vbulletin/450x555/actual_track_82bf0f8d7797d264a995f16d36affcc6511b9277.jpg

[Airmann]

My thoughts

This was a threat waiting to be realized. Miss a right turn, take the next one, sounds pretty harmless in most situations, but with the layout at JFK could have ended in serious disaster. Something needs to be done about the taxiway configuration at that area of the airport and/or standard taxi routings. I suspect something will be done to the latter.

What was the status of the automated runway warning status lights? It seems likely that the departing aircraft was far enough along that they didn't see it. What about the aircraft that crossed?

I didn't really like the way the ATC informed the crew to stop. I know that ATC are trained not to be alarmed, but being a bit firmer and using words like 'stop', 'abort', 'reject' would be more appropriate in my opinion, and would do a better job of communicating the urgent need to the the crew of the need to reject.

[Mogwi]

Who says Friday 13th is unlucky?

Mog

[draglift]

So the Delta went back to the gate. Passengers and crew deplaned and were put up in hotels overnight.

The American 106 continued to London.

Interesting decision to continue. Knowing you have made a serious error that caused another aircraft to reject its takeoff and was not just a runway incursion but a near collision must have had quite an effect on the crew. It would have been hard to put that behind them and concentrate on doing a good job of getting the aeroplane to London without going over and over things in their minds. Quite a distraction knowing you are going to be the subject of an investigation and it would not have been a happy 6 hour flight.

They had a telephone conversation with tower after the event but I wonder if they had a conversation with AA too or if they just decided to continue the flight?

The American was told to cross runway 31L but instead crossed runway 04L.

[Timmy Tomkins]

Ground radar?

[oversteer]

given a <$200 consumer phone can do over-the-ground gps routing at driving speeds there is no excuse for the same technology not to be fitted to all passenger aircraft

[ATC Watcher]

It was hot but well handled by ATC, regarding the Phraseology " cancel take off plans" repeated 3 times sounds like a (new?) FAA procedure and yes, is not ICAO phraseology, but so many things are not ICAO in US phraseology , but it worked well.
If the red bars were on (they should have been ) then the AA dif not only violate its clearance but also crossed red bars.. ( if they were on ) .The investigation report will be interesting to read .

[Andrewgr2]

So the Delta went back to the gate. Passengers and crew deplaned and were put up in hotels overnight.

The American 106 continued to London.

Interesting decision to continue. Knowing you have made a serious error that caused another aircraft to reject its takeoff and was not just a runway incursion but a near collision must have had quite an effect on the crew. It would have been hard to put that behind them and concentrate on doing a good job of getting the aeroplane to London without going over and over things in their minds. Quite a distraction knowing you are going to be the subject of an investigation and it would not have been a happy 6 hour flight.

They had a telephone conversation with tower after the event but I wonder if they had a conversation with AA too or if they just decided to continue the flight?

The American was told to cross runway 31L but instead crossed runway 04L.

Wouldn’t the AA106 crew normally have been suspended pending investigation following an apparent error of this magnitude?

[deltahotel]

There may be red stop bars, but my experience at JFK over the last 13 years is that they are not lit. However they also have flashing amber lights all the way across the rwy entrances/crossing points which are hard to miss.

[Timmy Tomkins]

There may be red stop bars, but my experience at JFK over the last 13 years is that they are not lit. However they also have flashing amber lights all the way across the rwy entrances/crossing points which are hard to miss.
They may be hard to miss but they don't stop you. If JFK is not using lights correctly then it is a big issue (Milan) and shows how useful the "follow the greens" are at Heathrow.

[MLHeliwrench]

There may be red stop bars, but my experience at JFK over the last 13 years is that they are not lit. However they also have flashing amber lights all the way across the rwy entrances/crossing points which are hard to miss.

Yes, but the AA crew thought they were cleared to cross the runway, so they would be expecting the flashing lights. The AA crew got disorientated with where they were on the field.

As for why they continued on to London, I wonder what was said during the dressing down phone call to tower? Was it clearly communicated that it was a serious near collision, or just a failure to follow taxi instructions? There is no way for pilot flying to be switched out I wonder (In order to continue the flight to London)? Even switched with possible cruise or training pilot on board already?

SPECULATION: It seems the female crewmember was running radio calls while male was taxiing. Then after the "please call the towner" ATC command, the male crew member comes on, kind of wound up, reading back what he thought they were cleared for (his tone is kind of implying an ATC read out error). I wonder who was sitting in the right seat and possibly saw the aircraft taking off right at them as they crossed the runway?

[BFSGrad]

I didn't really like the way the ATC informed the crew to stop. I know that ATC are trained not to be alarmed, but being a bit firmer and using words like 'stop', 'abort', 'reject' would be more appropriate in my opinion, and would do a better job of communicating the urgent need to the the crew of the need to reject.
"Cancel Takeoff Clearance" is verbatim phraseology from 7110.65.

[Talkdownman]

"Cancel Takeoff Clearance" is verbatim phraseology from 7110.65.
The UK has:

"Hold position, cancel take-off, I say again, cancel take-off, acknowledge"

...but also, if it is necessary that the aircraft should abandon take-off, the option of:

"DAL1943, stop immediately I say again, DAL1943, stop immediately, acknowledge"

The difference being: 'Cancel' a clearance, or 'Stop' a movement.

[Airmann]

"Cancel Takeoff Clearance" is verbatim phraseology from 7110.65. Yeah I'm not saying what he said was wrong. I'm saying that whatever he said doesn't seem to imply the kind of urgency that is required in those kinds of situations. I must admit, once the takeoff roll has started I've seen/experienced two things: one, engine noise increases so whatever volume setting existed during taxi may now be too low. Secondly, let's be honest we aren't as focused on ATC during take off as our attention may be focuses else where.

There needs to be some sort of change that ensures an urgent need to reject from ATC cannot in anyway be missed by pilots, and simply saying 'cancel take-off clearance' doesn't seem to imply that you'de better start stopping now or your going to kill hundreds of people, including yourself.

Firstly, there needs to be international standardization and the phrase used must not in anyway contain words in a format that could be interpreted in any other way (by any crew from any background, level of experience or nationality) other than what is intended.

Secpndly, ATC needed to repeat the phrase twice. So did Delta reject because of ATCs call or because they saw the American ahead of them? Obviously this is just conjecture/speculation but If it was the latter then there has been a COMPELTE FAILURE in one of the key layers of safety in the system.

[BFSGrad]

Secpndly, ATC needed to repeat the phrase twice. So did Delta reject because of ATCs call or because they saw the American ahead of them? Obviously this is just conjecture/speculation but If it was the latter then there has been a COMPELTE FAILURE in one of the key layers of safety in the system.Based on every video I’ve watched of the incident, JFK tower does repeat “cancel takeoff clearance.”

The video linked at post #8 is the first I’ve watched that included ATC instructions to AA106 to “hold position.”

Note that all of this audio is subject to the limitations of the Live ATC source and the many assumptions of the video creators.

[ATC Watcher]

ICAO Phraseology Reference guide :
RTF Cancelling Take-off Clearance Aircraft has not commenced take-off roll:
Big Jet 345 hold position, Cancel take-off, I say again cancel take-off due to ...(vehicle on the runway)
answer: Holding, Big Jet 345

Aircraft has commenced take-off roll:
Big Jet 345 stop immediately, (Big Jet 345 stop immediately)!
answer : Stopping, Big Jet 345

That is what the rest of the world pilots and controllers learn, . Not saying ATC phraseology used in the US (or in the UK is wrong) , it will work , as shown here however the 2 parties were US trained ,
But there is a major difference between "hold position cancel take off " , and "Stop immediately" depending if the aircrfat has started his take off roll or not. .

[LandIT]

Cross runway 31L at Kilo

Takeoff runway should have been briefed as 4L, as should the usual route to it that is to turn down Kilo.
How did AA106 crew think they were going to get to 4L by crossing it (not even at Kilo) and continuing on Juliet towards 4R.
This seems to have been far more than spatial disorientation.
Next issue for the investigators: is the CVR of this taxiing time going to be available to them? (which should be the reason why their take-off should also have been cancelled).

[RatherBeFlying]

Will the AA106 CVR during the incident be preserved after a trans Atlantic flight?

[DaveReidUK]

Next issue for the investigators: is the CVR of this taxiing time going to be available to them? (which should be the reason why their take-off should also have been cancelled).

The chances of that are zero.

[hans brinker]

This has surprisingly few comments. It could have easily been an award winning disaster.

Listening to the voices of the two pilots and two controllers was enlightening. It will be interesting to see the experience levels of some of the players if a report IS made public..

Don't know about ATC. AA pilots, flying LH likely 10K+ for everyone on the FD (probably 3). DL pilots harder to guess but likely 10K+ for both too, although they have had some younger people get hired, and junior people upgrade on NB.

[Chiefttp]

From friends at both American and Delta Airlines. The American crew was replaced and a reserve crew flew the flight.

Delta’s policy concerning rejected takeoffs,
“After an aborted takeoff the crew is required to call a duty pilot. I believe from there a joint decision is made on whether to return to gate or continue the flight.”

They obviously decided to return to the gate.

[Docbert]

From friends at both American and Delta Airlines. The American crew was replaced and a reserve crew flew the flight.

The FR24 track shows the American aircraft does not return to the gate - it stops for about 10 minutes, taxis back around to 31L, and is airborne 34 minutes after the incident occurred.

[stilton]

From friends at both American and Delta Airlines. The American crew was replaced and a reserve crew flew the flight.

Delta’s policy concerning rejected takeoffs,
“After an aborted takeoff the crew is required to call a duty pilot. I believe from there a joint decision is made on whether to return to gate or continue the flight.”

They obviously decided to return to the gate.




The Delta crew returned to the gate but just to clarify the AA crew continued to London, no indication they returned


A reserve crew operated the return from there ?

[Chiefttp]

Stilton,
here’s a direct quote from another aviation website,

“From person I know at AA on 777

The operating crew was pulled off and did DH.
Another crew operated the AA flight.”

To be fair, other posters are saying the incident crew did fly to London, but were deadheaded back by the company.

[draglift]

As the AA plane departed the gate at JFK 4 minutes early and arrived at LHR 13 minutes early there is clearly no way it could have gone back onto the gate and changed crews.



https://cimg3.ibsrv.net/gimg/pprune.org-vbulletin/1646x761/566ff26b_8858_419c_b89a_e5282ff0c92c_8607f325825e70ad29500f0 3835cea44bcba893c.jpeg

[BFSGrad]

Following timeline pulled from ADS-B data:

01:44:29Z DL1943 start TO roll
01:44:30Z AA106 at J-K-B intersection
01:44:39Z DL1943 40 kts
01:44:43Z AA106 entering 4L at J
01:44:47Z DL1943 79 kts
01:44:54Z AA106 at middle of 4L (interpolated between data points assuming steady taxi speed)
01:44:55Z DL1943 104 kts, max speed, at K3/K4, approx 2000 ft from J
01:44:58Z DL1943 85 kts
01:45:00Z DL1943 62 kts
01:45:05Z AA106 clear of 4L on J
01:45:05Z DL1943 33 kts at 31L/4L intersection

Curious to know if the DL crew rejected takeoff based on observed conflict or ATC direction. Or perhaps both events were simultaneous. Even at night, seems a 777 broadside at 2000 ft would be hard to miss as a takeoff hazard. I suspect there was the expected human brain delay as the DL crew convinced themselves that they were seeing the unexpected. Also curious how AA crew verified runway was clear to cross. At that point, landing lights from DL1943 should have been clearly visible (4500 ft away) as takeoff roll was in progress.

[Airmann]

The Delta crew returned to the gate but just to clarify the AA crew continued to London, no indication they returned A reserve crew operated the return from there ? I wonder if you really want a crew that has just messed up in the way they had to be flying? They might have a bit too much on their minds thinking about what just happened, a bit distracted.

[MarkerInbound]

Yeah I'm not saying what he said was wrong. I'm saying that whatever he said doesn't seem to imply the kind of urgency that is required in those kinds of situations.

I think prefacing the call with a reference to body excretion probably alerted the crew to the importance of the message.

[KRviator]

As the AA plane departed the gate at JFK 4 minutes early and arrived at LHR 13 minutes early there is clearly no way it could have gone back onto the gate and changed crews.They didn't do a crew change at all. They did a figure 8 around Z,H,Y & J and promptly departed from 31L. Guess that's one way to hide the evidence - piss off on a transatlantic flight to overwrite the CVR?

Doesn't the CFR's have something about preserving evidence in the event of an incident? I can't imagine they'll be able to explain that decision away...

EDIT: Yep - they do. Granted I'm not particularly fluent with 'Muricanese, but the following seems pretty clear-cut.
830.10 Preservation of aircraft wreckage, mail, cargo, and records.
(a) The operator of an aircraft involved in an accident or incident for which notification must be given is responsible for preserving to the extent possible any aircraft wreckage, cargo, and mail aboard the aircraft, and all records, including all recording mediums of flight, maintenance, and voice recorders, pertaining to the operation and maintenance of the aircraft and to the airmen until the Board takes custody thereof or a release is granted...Note - "Incident" includes a runway incursion where an aircraft or vehicle had to take corrective action to avoid a collision..."Any event in which an operator, when operating an airplane as an air carrier at a public-use airport on land:
(i) Lands or departs on a taxiway, incorrect runway, or other area not designed as a runway; or
(ii) Experiences a runway incursion that requires the operator or the crew of another aircraft or vehicle to take immediate corrective action to avoid a collision.
https://cimg1.ibsrv.net/gimg/pprune.org-vbulletin/1117x620/aa106_ade6b61c10069fba5f27fd2894fc8ab449b90724.jpg

[DaveReidUK]

They did a figure 8 around Z,H,Y & J and promptly departed from 31L.

Not quite. Once on the move again after the incident, they stopped for a second time, for just over 6 minutes on Yankee, before proceeding left onto Juliet for 31L.

[remi]

Given modern CVRs having a generally enabled push to erase that is part of SOP after parking, is there a "preserve" button? I'm unaware. Seems like it would be a lose your certificate problem if you did not preserve the relevant CVR after an incident like this even if that required heading back to the gate. From the crew response it's clear there was something interesting going on in the AA cockpit.

[AmarokGTI]

It continues to amaze me that in the US at a point like this when the AA crew appear to have reduced situational awareness, that ATC load them up with the “possible pilot deviation + phone number” business. It’s AA at JFK, it’s not like they can pretend it didn’t happen. No harm in dealing with that stuff in the background eg on return to the gate or, if the flight departs, to send it via the company. Loading people up with extra tasks when they have had (or may have had) and incident isn’t helpful. Of course, neither is clearing a flight to land when there are multiple departures and arrivals ahead, but that’s also ops normal there.

[Dunhovrin]

Loading people up with extra tasks when they have had (or may have had) and incident isn’t helpful.

Like the Delta - barely coming to rest after a medium-speed abort and probably still to run the post-RTO checks, ATC chip in with "what are your intentions?".

"Chuffing wait will you!"

[punkalouver]

One note concerning the controllers. I do seem to remember from reading past incident reports, that when a controller clears an aircraft to cross a runway, they have certain responsibilities, such as looking to confirm a runway is clear and possibly looking at the aircraft being cleared to cross to confirm its position for where it will cross. I do wonder how they do all this at busy times in complicated situations when the chatter is non-stop. In addition, the crossing clearance may have been given well before the aircraft reached Kilo taxiway making such a task more difficult in a busy environment.

[Torquetalk]

Like the Delta - barely coming to rest after a medium-speed abort and probably still to run the post-RTO checks, ATC chip in with "what are your intentions?".

"Chuffing wait will you!"

True. But then the crew can make that space with the internationally recognised capacity maker “stand by”

After all, the immediate threat was resolved; the next one is departing from procedures and missing steps

[draglift]

Having operated out of JFK two days before the incident with the same runways in use I can see exactly how the Swiss cheese holes lined up. The incident actually started at the gate.

The JFK ATIS gives runway 04L and runway 31L for departure. You do not know which one you will get so you guess one and plan for that although you get the figures for the other runway too.This is the first Swiss cheese hole.

The datalink clearance comes through with a SID. However the SID, i.e. a Kennedy 5 does not specify which runway. At other airports around the world the SID name defines which runway you use. This is the second Swiss cheese hole.

So you brief for departure and push back from the gate still without knowing which runway you are departing from. This is not desirable.

When you come out of the ramp at TA you do not know if you are going to go clockwise or anteclockwise, it can be either, and so you are listening hard to whether it is Alpha or Bravo taxiway and where your holding point limit is.

Crucially the ground controller in the initial taxi clearance to AA106 did say runway 04L at the start of the transmission but this was not read back by the female pilot who just read back the taxi instructions. The taxi instructions they were given were relevant from their taxi position to both runway 31L and runway 04L. The next Swiss cheese hole lines up.

I think they believed they were taxiing to runway 31L for departure. They taxied as instructed on Bravo and held at Kilo. The next instruction they were expecting to receive was to cross the runway in front of them and continue to the hold for 31L. As a bit of confirmation bias they could see a plane at the hold for 31L.

As expected they then received clearance to "Cross runway...." and taxied forward but did not take the first right and went ahead believing they were cleared to cross that runway in their route to 31L. They did not realise they were cleared to cross runway 31L and not 04L. The final cheese hole lined up for the incident to happen.

I have sympathy for the crew and had they been given a departure runway in their original clearance and briefed the taxi route in advance before pushback I do not think this would have happened.

[JanetFlight]

Something missing in my mind...

https://www.youtube.com/watch?v=nsFIHKkN2oU

Acc this YT including full Taxi clrs from the very beginning issued by the TWR and ack by the lady pilot, all the clrs and taxi instructions were given regarding RWY 4L for dep.
The readbacks either, BUT...missing the RWY at 0:26 by the lady (only twys), wich ATCer included few seconds before as 04L(0:20).
However they were heading for 31L, wich acc some ppruners here it was in facto the rwy they used for departure minutes later...so, here is my question:
I presume they had loading all the T/O perf // data // SID's, etc, acc RWY31L, and also having such in their minds.
But all the way they were readbacking it to the ATC as acting for a 4L dep, including when the lady said in readback "cross 31 at kilo"
So...maybe some biased since the start to use 31L insted of 4L??
Sorry for this humble question...anyway, kudos to those who saved the day ;)

EDIT: Ppruner colleague Draglift posting at the same time of me, said much of what i think too...my apologies (not copying at all his/her way of thinking).

[Uplinker]

I don't know if this was a factor, but having operated in and out of both JFK and LHR, the quality of ATC communications is very challenging at JFK.

I am sorry, but the JFK controllers all talk as if their hair is on fire, like a machine gun, and it can be very challenging to understand the instructions, whereas at LHR, controllers are calm and speak calmly.

I can't help feeling that if the JFK controllers were to speak slowly and calmly, aircraft would have a much less confusing and less stressful time at JFK. There may be valid reasons why they don't speak slowly, such as there not enough enough controllers, and if so this should be addressed.

[Bergerie1]

draglift, A very succinct analysis of what we know so far. It is a very long time since I last operated out of JFK and, without being derogatory to my American friends, I do think that the lack of R/T discipline by the controllers (and the pilots) plus the rapid-fire way in which American ATCOs like to give their instructions all contribute to rendering the safety nets less effective in plugging the holes in this Swiss cheese.

[MLHeliwrench]

They didn't do a crew change at all. They did a figure 8 around Z,H,Y & J and promptly departed from 31L. Guess that's one way to hide the evidence - piss off on a transatlantic flight to overwrite the CVR?

IS it possible to swap out crew to one already onboard as deadhead/pax? six minutes would not be enough to brief flight plan/weather etc....

[MightyGem]

Similar to this:

https://www.youtube.com/watch?v=equVF3ULVw8

[wiggy]

IS it possible to swap out crew to one already onboard as deadhead/pax? six minutes would not be enough to brief flight plan/weather etc....

Starter for ten would be was there such a crew onboard?

Such wasn't exactly run of the mill or routine where I worked (not US) though not completely unknown and even when there was such a crew onboard they weren't required to be on stand-by to jump into the operating seats at any time in the flight.

[ATC Watcher]

The Swiss cheese model described by Draglift is a good description of what is likely to have happenned. Taxying in the US major airports is not easy and is not helped by the rapid local phraseology/slang used by the ramp guys. I have not been inside the TWR in JFK for decades, but before the guys doing ramp control were not among the best trained .. Do not know if this is stil the case.
But to take the defence of the TWR controllers here, well, first the runway crossing clearance was clear , and confirmed/acknoledged by the AAL lady : " Cross 31 L at K " it was not to cross 4L at J..
Next the tlephone number; this is the standard procedure , you do not discuss errors on the ffrequency , and you have to tell the pilots if you are going to make a report on them. So it is not arrogance or showing who is the boss, , it is just following the procedure.
As to the speed at which they speak, well it is the same everywhere , if, as a controller, you are alone on a position and it is very busy you do not have much other choice. Getting more staff to open another frequency/ postion would definitivelyhelp but it is not possible with the chronic satff shortage in so many places, and all the 3 New York
I would still like to know if there were red stop bars light on around 4L though., to eliminate one more possible cheese hole. .

[AmarokGTI]

Next the tlephone number; this is the standard procedure , you do not discuss errors on the ffrequency , and you have to tell the pilots if you are going to make a report on them. So it is not arrogance or showing who is the boss, , it is just following the procedure.

personally I’m not questioning whether they followed the procedure. Rather, I’m suggesting the procedure is flawed. (Good application of a bad rule)

[ATC Watcher]

personally I’m not questioning whether they followed the procedure. Rather, I’m suggesting the procedure is flawed. (Good application of a bad rule)Not ideal I agree but what alternative proposal would you make ? Saying nothing and let the aircraft go and inform the crew when it returns 2 days later is not really an option is it?

Personally, if it was me making the rules, I would ask ATC to simply cancel the departure slot, taxi the aircraft back to the ramp, and get someone debrief they crew and secure the CVR . I can see the airlines agreeing to those procedures already

[KRviator]

Not ideal I agree but what alternative proposal would you make ? Saying nothing and let the aircraft go and inform the crew when it returns 2 days later is not really an option is it?

Personally, if it was me making the rules, I would ask ATC to simply cancel the departure slot, taxi the aircraft back to the ramp, and get someone debrief they crew and secure the CVR . I can see the airlines agreeing to those procedures alreadyI'm amazed "someone" at American, the CP/Duty Pilot/Ops/Whoever this crew would have contacted actually thought having the crew continue the flight was a reasonable course of action - and that's not even considering away the loss of evidence with the overwriting of the CVR.

No one wants to delay a widebody full of passengers, but after a serious incident the judgement and performance of the crew is likely to be impaired and to have them stewing on such an event when they could have returned to the gate, downloaded the recorders and called out a standby crew in a couple hours to continue the flight would appear to have been preferable - and I think the NTSB will concur when they address the lack of availability of the CVR data on what is a (very) serious near miss.

Legitimate question - could ATC have refused AA106 clearance to depart until any evidence was secured? I know that's supposed to be the responsibility of the Operator, but where it is clear they are not going to do so as in this case, is there a precedent for Ground directing them to return to the gate?

[stilton]

I'm amazed "someone" at American, the CP/Duty Pilot/Ops/Whoever this crew would have contacted actually thought having the crew continue the flight was a reasonable course of action - and that's not even considering away the loss of evidence with the overwriting of the CVR.

No one wants to delay a widebody full of passengers, but after a serious incident the judgement and performance of the crew is likely to be impaired and to have them stewing on such an event when they could have returned to the gate, downloaded the recorders and called out a standby crew in a couple hours to continue the flight would appear to have been preferable - and I think the NTSB will concur when they address the lack of availability of the CVR data on what is a (very) serious near miss.

Legitimate question - could ATC have refused AA106 clearance to depart until any evidence was secured? I know that's supposed to be the responsibility of the Operator, but where it is clear they are not going to do so as in this case, is there a precedent for Ground directing them to return to the gate?


It’s possible the AA crew did not contact or discuss the incident with any company management entity, or if they did the full gravity of the incident might not have been understood


If they had decided to return to the gate and request a replacement crew then that discussion would most certainly have taken place although initially contact could well be with a union safety rep


In the moment the decision to continue might have seemed logical bearing in mind no actual collision occurred and how well did they see the approaching Delta Aircraft if at all considering the geometry of the two aircraft ?


Of course it’s not known how detailed the telephone conversation was between the AA captain and the tower, that may or may not have impressed on him what a close call it was


I don’t think there’s any question the AA crew should have returned to the gate

[Check Airman]

This has surprisingly few comments. It could have easily been an award winning disaster.

Listening to the voices of the two pilots and two controllers was enlightening. It will be interesting to see the experience levels of some of the players if a report IS made public..

I think you’ll find that there was certainly not a lack of experience. The crew would have spent years flying narrowbody schedules before being senior enough to bid the 777.

[Check Airman]

Cross runway 31L at Kilo

Takeoff runway should have been briefed as 4L, as should the usual route to it that is to turn down Kilo.
How did AA106 crew think they were going to get to 4L by crossing it (not even at Kilo) and continuing on Juliet towards 4R.
This seems to have been far more than spatial disorientation.
Next issue for the investigators: is the CVR of this taxiing time going to be available to them? (which should be the reason why their take-off should also have been cancelled).

I don’t brief taxi routes for this exact reason. I’m betting the report will include an element of expectation bias.

[Sailvi767]

That particular runway configuration is almost the standard at JFK in the winter. The crew was JFK based and I find it hard to believe they did not expect to depart 4L as that is the norm for all international flights. In fact while taxiing out they passed intersection KE on 31L where another AA flight was cleared into position and hold at KE. The investigation will certainly find out both the expected runway briefed and what was loaded in the FMS.

[daelight]

Surely the crew would have glanced right / left before crossing the runway and noticed the blinding lights of an airliner taking off on that runway? All other holes of the cheese withstanding, surely the whole 'airmanship' mantra the chaps upfront pride themsleves on would call for such an obvious check?!

[MechEngr]

I don't know if this was a factor, but having operated in and out of both JFK and LHR, the quality of ATC communications is very challenging at JFK.

I am sorry, but the JFK controllers all talk as if their hair is on fire, like a machine gun, and it can be very challenging to understand the instructions, whereas at LHR, controllers are calm and speak calmly.

I can't help feeling that if the JFK controllers were to speak slowly and calmly, aircraft would have a much less confusing and less stressful time at JFK. There may be valid reasons why they don't speak slowly, such as there not enough enough controllers, and if so this should be addressed.

This is New York. For them that is slow. You want drawl, go to EL Paso.

Who designed that rabbit warren of a taxiway system?

[springbok449]

Cross runway 31L at Kilo

Takeoff runway should have been briefed as 4L, as should the usual route to it that is to turn down Kilo.
How did AA106 crew think they were going to get to 4L by crossing it (not even at Kilo) and continuing on Juliet towards 4R.
This seems to have been far more than spatial disorientation.
Next issue for the investigators: is the CVR of this taxiing time going to be available to them? (which should be the reason why their take-off should also have been cancelled).

Unfortunately not that simple, JFK often operates mixed mode RWYs for take off with both 04L and 31L in use.

However when this is the case 31L take offs are normally from KE intersection rather than full length to avoid crossing 04L on the take off roll.

[slast]

Haven't seen this elsewhere in the thread: there's an interesting piece here commenting that cockpit procedures changes may have been a factor.... .
Christine NegroniNew AA Pilot Procedures Could be Part of Investigation into Near Collision at JFK (https://christinenegroni.com/new-aa-pilot-procedures-could-be-part-of-investigation-into-near-collision-at-jfk/)
and a related one here from January 3: Pilots Balk As American Airlines Enforces New Cockpit Procedures (forbes.com) (https://www.forbes.com/sites/tedreed/2023/01/03/pilots-balk-as-american-airlines-imposes-new-cockpit-procedures/)
That one refers to "Allied Pilots Association said the implementation involves an attempt to alter critical procedures through a 35-page bulletin and changes in a 65- page manual rather than through in person training." Anyyone here have a copy of the bulletin and/or manual ?

[Doors to Automatic]

As a layman (albeit a fairly knowledgeable one) I always find these sorts of situation almost unbelievable. Having received taxi clearance for a 4L departure and a further clearance to cross 31L how can such a fundamental mistake be made by an experienced crew in good visibility and at their home airport?

I also agree with the other posters who have questioned ATC terminology as being too long-winded. As pilots simply say "STOP" during the take-off run to abort, I am surprised that ATC don't use the same terminology. "Delta 1943, STOP STOP STOP" would take less time and project far more urgency.

[DaveReidUK]

Haven't seen this elsewhere in the thread: there's an interesting piece here commenting that cockpit procedures changes may have been a factor.... .
Christine NegroniNew AA Pilot Procedures Could be Part of Investigation into Near Collision at JFK (https://christinenegroni.com/new-aa-pilot-procedures-could-be-part-of-investigation-into-near-collision-at-jfk/)
and a related one here from January 3: Pilots Balk As American Airlines Enforces New Cockpit Procedures (forbes.com) (https://www.forbes.com/sites/tedreed/2023/01/03/pilots-balk-as-american-airlines-imposes-new-cockpit-procedures/)
That one refers to "Allied Pilots Association said the implementation involves an attempt to alter critical procedures through a 35-page bulletin and changes in a 65- page manual rather than through in person training." Anyyone here have a copy of the bulletin and/or manual ?

APA Press Release re changes: Unwise and Unsafe (https://www.alliedpilots.org/News/ID/10941/Unwise-and-Unsafe)

[Uplinker]

This is New York. For them that is slow. You want drawl, go to EL Paso.

Who designed that rabbit warren of a taxiway system?

I realise you are joking, but I don't want 'drawl' I just want safe, clear instructions that I can understand and write down (in short-hand). But the speed of vocal delivery at JFK can prevent this, and is, I think, a safety issue; being at least twice the speed at LHR, AMS, even CDG !

Every time we flew into JFK, all three of the heavy crew were nervously ears-peeled to hopefully hear and understand the machine gun taxi instructions first time. Having to ask for a repeat was pointless, as the controller then got really pissed off and would speak even more quickly, as if s/he thought we were stupid.

By speaking so quickly, ATC cause themselves problems and extra work. Slooooooow Dooooown a bit and give yourselves and everyone a break ! :ok:

The taxiways are complex but not particularly worse than LHR or CDG, but the way ATC is set up leaves a lot to be desired - a separate ramp controller frequency doesn't help, (aircrews), with having to contact ramp on box 2 as you taxi in, to find out your stand to tell the controller. Can't the controller themselves just look over at the ramp screen - we have too much to do trying to understand the machine gun instructions and get them right without one of us going away to box 2.

Something needs to be done though. A better traffic system could be developed at JFK, such as SIDs linked to specific runways, as has been mentioned, and perhaps a standard 'one way' circulatory taxi system?

[ATC Watcher]

Something needs to be done though. A better traffic system could be developed at JFK, such as SIDs linked to specific runways, as has been mentioned, and perhaps a standard 'one way' circulatory taxi system?


No need to reinvent the wheel, aan excellent system exsist already since a long time : it is called SMGCS , exepensive though , and I think that it is always the same issue : money, How much you are willing to spend on safety. that has no direct return on investment.
From ICAO :
SMGCS :
At airports where Category II and III operations take place or where ground movement requirements are complex, a Surface Movement Guidance and Control System (SMGCS (https://www.skybrary.aero/index.php/A-SMGCS)) may be installed in order to simplify the ATC GND task. Such a system has selective switching of taxiway centreline lighting so that individually defined blocks of taxiway, each beginning and ending at a lit Stop Bar, can be individually illuminated ahead by ATC when a valid clearance has been issued. The associated stop bars are extinguished as an aircraft with a valid clearance approaches the next section of taxiway and the centreline lighting for that block is selected.

[Mike6567]

Retired now and have't been to JFK for about 30 years. My wife said don't they have red stop lights to prevent aircraft entering the active runway? I am sure these modern airports have this but presumably JFK does not.

[BFSGrad]

Retired now and have't been to JFK for about 30 years. My wife said don't they have red stop lights to prevent aircraft entering the active runway? I am sure these modern airports have this but presumably JFK does not.
In post #6, I noted that JFK has Runway Status Lights, which includes Runway Entrance Lights at J crossing 4L.

JFK also has ASDE-X.

But we don’t know (do we?) at this point if either or both systems were operational at the time of the incident.

[Mike6567]

In post #6, I noted that JFK has Runway Status Lights, which includes Runway Entrance Lights at J crossing 4L.

JFK also has ASDE-X.

But we don’t know (do we?) at this point if either or both systems were operational at the time of the incident.

Thanks. Missed your info on Runway Status Lights.

[Uplinker]

Hi ATC Watcher.

Yeah, but really is a hugely complex electronic system really needed?

I visited the old control tower at (London) Heathrow, where the taxiway lights were controlled by the controllers via a big board with the airport runways and taxiways drawn out and manual switches - "follow the greens" - which worked really well.
And as a pilot, the old block number system at LHR - " follow the greens and hold at block 202" - worked very well, but was outlawed by ?Europe.

There are probably many more short interconnecting taxiways at JFK, which complicate the issue, but surely a general traffic flow pattern could be devised that would reduce the amount of instructions required ?

[wiggy]

There are probably many more short interconnecting taxiways at JFK, which complicate the issue, but surely a general traffic flow pattern could be devised that would reduce the amount of instructions required ?

I agree the interconnectors can be a problem but of course as I recall it there is a general flow - around the inner, A, clockwise and the outer, B, counter clockwise, so on taxing from the ramp instructions such as "Taxi on Bravo, hold short of Kilo Delta" would get you a long way..

Always found taxiing in was potentially a bit more difficult but once you were onto A or B it was fairly straight forward.

[itsnotthatbloodyhard]

I realise you are joking, but I don't want 'drawl' I just want safe, clear instructions that I can understand and write down (in short-hand). But the speed of vocal delivery at JFK can prevent this, and is, I think, a safety issue; being at least twice the speed at LHR, AMS, even CDG !

Every time we flew into JFK, all three of the heavy crew were nervously ears-peeled to hopefully hear and understand the machine gun taxi instructions first time. Having to ask for a repeat was pointless, as the controller then got really pissed off and would speak even more quickly, as if s/he thought we were stupid.

By speaking so quickly, ATC cause themselves problems and extra work. Slooooooow Dooooown a bit and give yourselves and everyone a break ! :ok:



Agreed. The interesting thing for me was that in the immediate aftermath of the incident, the controller’s delivery was very measured and much more like what you get from ATC elsewhere. So it can be done.

[Gne]

Agree - the quick fire delivery at JFK has been oft criticized but never defended. Is that because it is undefendable or because the controllers at JFK are too busy to be involved in PPRuNe discussions?

Gne

[BFSGrad]

From CNN:A source familiar with the near-collision of two airliners on a John F. Kennedy International Airport runway tells CNN that special flashing lights which warn pilots against taxiing across a runway were in fact working. On Wednesday, the source told CNN that following the incident, airport staff “went out immediately” to confirm Runway Status Lights were working properly.

[sandos]

From CNN:A source familiar with the near-collision of two airliners on a John F. Kennedy International Airport runway tells CNN that special flashing lights which warn pilots against taxiing across a runway were in fact working. On Wednesday, the source told CNN that following the incident, airport staff “went out immediately” to confirm Runway Status Lights were working properly.

But were they also correctly configured? This is different from working.

[737pilotguy]

Complacency during taxi is too common. A three person flight deck and still things get omitted. If unsure, they should've held position, runway warning lights or not...

[xetroV]

If unsure, they should've held position, runway warning lights or not...
That's an important if. Judging from the ATC recording they were not unsure at all; they were just wrong, which is a big difference.

[Captain Fishy]

Having operated out of JFK two days before the incident with the same runways in use I can see exactly how the Swiss cheese holes lined up. The incident actually started at the gate.

The JFK ATIS gives runway 04L and runway 31L for departure. You do not know which one you will get so you guess one and plan for that although you get the figures for the other runway too.This is the first Swiss cheese hole.

The datalink clearance comes through with a SID. However the SID, i.e. a Kennedy 5 does not specify which runway. At other airports around the world the SID name defines which runway you use. This is the second Swiss cheese hole.

So you brief for departure and push back from the gate still without knowing which runway you are departing from. This is not desirable.

When you come out of the ramp at TA you do not know if you are going to go clockwise or anteclockwise, it can be either, and so you are listening hard to whether it is Alpha or Bravo taxiway and where your holding point limit is.

Crucially the ground controller in the initial taxi clearance to AA106 did say runway 04L at the start of the transmission but this was not read back by the female pilot who just read back the taxi instructions. The taxi instructions they were given were relevant from their taxi position to both runway 31L and runway 04L. The next Swiss cheese hole lines up.

I think they believed they were taxiing to runway 31L for departure. They taxied as instructed on Bravo and held at Kilo. The next instruction they were expecting to receive was to cross the runway in front of them and continue to the hold for 31L. As a bit of confirmation bias they could see a plane at the hold for 31L.

As expected they then received clearance to "Cross runway...." and taxied forward but did not take the first right and went ahead believing they were cleared to cross that runway in their route to 31L. They did not realise they were cleared to cross runway 31L and not 04L. The final cheese hole lined up for the incident to happen.

I have sympathy for the crew and had they been given a departure runway in their original clearance and briefed the taxi route in advance before pushback I do not think this would have happened.


This is the root of the problem. The departure runway is not specified on the ACARS departure clearance, only the SID, initial altitude, transition and squawk. The only mention of the departure runway is on the ATIS and, as in this case, is sometimes more than one runway. I operate to JFK frequently and am only too familiar with this 'clearance by osmosis' problem. Sometimes you can tell which runway it is or isn't from the SID designator but not always. Checking FR24 for traffic flows can help but it really shouldn't be necessary. Another hole they have awaiting in the cheese is then to change the departure runway after taxi has commenced, which if you're not there often can cause problems with SA too. We all know what assumptions are... assume nothing at JFK!

[draglift]

Another Swiss cheese hole is when the AA crew were holding and expecting to cross the runway ahead of them ATC said "Cross runway 31L at Kilo"

If ATC had said to them, "Turn right on Kilo and cross 31L" I do not think the incident would have happened as the instruction to turn right would have been different to what they were expecting and would have made them realise.

[bean]

Another Swiss cheese hole is when the AA crew were holding and expecting to cross the runway ahead of them ATC said "Cross runway 31L at Kilo"

If ATC had said to them, "Turn right on Kilo and cross 31L" I do not think the incident would have happened as the instruction to turn right would have been different to what they were expecting and would have made them realise.
Good point

[SWBKCB]

Another Swiss cheese hole is when the AA crew were holding and expecting to cross the runway ahead of them ATC said "Cross runway 31L at Kilo"

If ATC had said to them, "Turn right on Kilo and cross 31L" I do not think the incident would have happened as the instruction to turn right would have been different to what they were expecting and would have made them realise.

Isn't the key bit here "Cross runway 31L"? Even at the little puddle jumper airports I fly into there are big red signs telling you what the ahead runway is and the holding point.

Surely you would want to visually confirm the identity of any active runway you are about to enter?

[ATC Watcher]

If ATC had said to them, "Turn right on Kilo and cross 31L" I do not think the incident would have happened as the instruction to turn right would have been different to what they were expecting and would have made them realise.
Do not try to shift or dilute the errors made by the AAL crew . There are long established taxi procedures, and they work 99.999% of the time . Instructing pilots to turn right or left from 3 Km away will create far more possibilities for confusion and errors than anything else.
Whether the stop bars were on or not is the thing I would like to know though.
I know from an IFALPA/IFATCA survey many years ago asking which airports pilots were routinely instructed by ATC to cross illuminated red bars that there were quite a few at the time . I thought the problem was solved. Or does any of you here still experience this somewhere today ?

[punkalouver]

A good reason why I write down the clearances as they are given. In this case, 04L would have been right there in front of them as a reminder in case they had accidentally gotten 31L in their mind for departure.

It is not always realistic but for those not intimately familiar with the airport, it can be useful to stop after getting the clearance(or already be stopped prior to getting it) and review it with map to get the picture in mind(much more effective than doing this while taxiing). One can also review the names of the cross-taxiways that precede kilo so that they are better prepared to be aware when they are getting close to Kilo.

Obviously, if very familiar with the route/airport, one will likely be comfortable with taxiing while getting the clearance, like many of us do at a home base complicated airport(no guarantee that these two pilots had operated out of JFK for a long time).

But having that runway written down may pay off 5 minutes later when you quickly look at your piece of paper and discover that you have the wrong runway in mind. It may sound unnecessary to write it down but the runway to be used is usually at the beginning of the clearance and if a lot is added on to the transmission or something else is distracting, you may have forgotten what runway was given and then made an incorrect assumption, but pen mark on paper will not erase and should trigger one to question, if it is noticed.

Of course, the concept of not crossing an illuminated stop bar even if proper clearance is given is important as is a good look down the runway and perhaps questioning takeoff lights illuminated instead of assuming that the other aircraft is holding in position.

In addition, read the runway ID signs before crossing the runway. When you finish you acknowledgement to ATC to cross 31L, say out loud ‘cleared to cross 31L” to the other pilot and look for the sign or runway marking with that number and say it out loud. If you find yourself saying ‘cleared to cross 31L’ then discover yourself saying ‘04L’, you know it is time to stop.

[250 kts]

ATC audio:

5AbpxqtXcPk
Are we all comfortable with the quality of the "slurred" ATC as AAL106 is cleared to taxi? I had to listen to it 3 times to actually decipher that 4Left is mentioned, and even then it's only knowing he said 4L that it is a bit clearer.

[Uplinker]

+1 punkalouver

I am sure most write down taxi instructions? I do in short-hand, so I can read them back correctly and to have the clearance in front of me if I should forget some of the details.

If it is a "simple airfield" with few taxiways, I might use the MCDU scratch-pad

The quality and delivery of speech from ATC at JFK has been noted - always my least favourite part of a long-haul trip !

[100Series]

Departing JFK many years ago the cleance could include almost a full route as far as the ocean,read at lightning speed. Smart FOs would write down the flight plan in advance so they could read it back as fast as it was given. Extra points were earnt by replying for him, if the captain was doing the rt, didn't expect the torrent of words, and would need a repeat.
My favourite though was the southern U.S. captain who, after have had two repeats, drawlled "You'all hear how fast I speak, well that's how fast I write. Say again,"

(and don't get me started on "Climb to and maintain". Extra verbiage and a superfluous, possibly confusing, to)

[WillowRun 6-3]

Sorry about having to bump into the thread here, but noticed a news article this afternoon which, to my merely SLF/attorney knowledge, seems quite pertinent.

Forbes reporting that the 106 FO, on first flight on 777 type (after experience on 737 and type training), was busy with new procedures, including making a p.a. announcement about imminent departure (article says "takeoff", but departure is the correct term, isn't it?). Article refers to input from an unnamed AA source; quoting, "she had just read the 35-page bulletin that changes procedures." Some content too about whether the changes in procedures were quite alright or not; per article, Allied Pilots Ass'n had initiated an "appeal" of their implementation.

Article notes also there was a third pilot on the 106 flight deck; that neither the PIC nor 3rd aviator saw the "stop bar lights on the runway"; and that AA 106 had just switched from company frequency to Tower frequency. AA 106 did not know the seriousness of the incursion until they arrived in London, per Forbes article. "Delta was cleared for takeoff before they switched over" to the tower frequency, the article quotes the source as saying.

An aviator active on Twitter, Kelly Lepley, posted the Forbes article. The American source it claims to rely on is identified as a "pilot" but not otherwise. [@kclepley - iirc, a UPS 747 captain]

Apologies twice for barging in on the aviator discussion.

[wiggy]

Are we all comfortable with the quality of the "slurred" ATC as AAL106 is cleared to taxi? I had to listen to it 3 times to actually decipher that 4Left is mentioned, and even then it's only knowing he said 4L that it is a bit clearer.

TBF we'd need to hear the original to judge if that was a factor but it sounds fairly normal for JFK (not that I go there anymore)

I think the one issue with the controller that appears to be a foul was that it appears (from the recording) that he didn't challenge the lack of runway assignment in the crew readback.

Other than that I'm afraid I am agree to some degree with ATC watcher's opening comment a few posts upthread and the comment very early on by a poster about the crew and expectation...

[BFSGrad]

An aviator active on Twitter, Kelly Lepley, posted the Forbes article. The American source it claims to rely on is identified as a "pilot" but not otherwise. [@kclepley - iirc, a UPS 747 captain]
Here's the referenced Forbes article:

First Officer On American JFK Runway Incursion Flight Had Added Task At Departure, Source Says (https://www.forbes.com/sites/tedreed/2023/01/19/first-officer-on-american-jfk-runway-incursion-flight-had-added-task-at-departure-source-says/?sh=6928617767ac)

Not convinced this narrative reflects actual events.

[DaveReidUK]

Forbes reporting that the 106 FO, on first flight on 777 type (after experience on 737 and type training), was busy with new procedures, including making a p.a. announcement about imminent departure (article says "takeoff", but departure is the correct term, isn't it?).

No, the terminology in Forbes is correct.

"Departure" is the process that starts with pushback (hence an on-time pushback = an on-time departure). A PA announcement about imminent pushback would be unlikely to be a factor in a runway incursion, for obvious reasons.

[blind pew]

Reading the Forbes article it would appear that some soul decided to interfere with what should be a sterile phase of the flight in that the monitoring pilot had a series of tasks which could and should have been carried out when the aircraft was stationary. Especially relevant as the copilot was new on type.
It isn’t the first time that I’ve seen ill thought procedure changes which left no one monitoring.
Since the tasks involved checking take off performance then Pm wouldn’t have been looking outside.
Perhaps it’s also about time that the 30 mins CVR rule is changed - after all it was introduced 50 years ago.
(I was on BALPA tech committee in the early 70s tasked with airfield lighting amongst other things - operated a heavy in 1978 into JFK and know how difficult RT and taxying can be without unnecessary distractions).

[ATC Watcher]

A sentence in the article is intriguing : it says both the captain , who was taxying ,and the supernumerary crew on the jumpset COULD not see the stopbars . Using "could" instead of "did" would indicate that either they physically from their seats could not see them perhaps because of the turns they were performing , or that they were not on.
But that could be just a journalist interpretation , who also apparently was wrong when he says that the crew did not know they were involved an incident until they reached London. , the ATC audio clearly shows otherwise. That they did not realise the seriousness of it, that I can believe, but they were informed they did not follow the taxi instructions.

[MarkerInbound]

Perhaps it’s also about time that the 30 mins CVR rule is changed - after all it was introduced 50 years ago.


It’s now 2 hours in the FAA world. I’ve heard EASA has gone to 25 hours. I know the NTSB is pushing for 25 hours and have a list of investigations that have been hindered by lack of CVR records. I’m sure this event will be added to that list.

[Lake1952]

That could be just a journalist interpretation , who also apparently was wrong when he says that the crew did not know they were involved an incident until they reached London. , the ATC audio clearly shows otherwise. That they did not realise the seriousness of it, that I can believe, but they were informed they did not follow the taxi instructions.

The AA crew was informed there was a deviation and told to make a phone call. The male pilot, presumably the Captain, intimated they were cleared to cross 4L when their instructions were to cross 31L at Kilo. Tower controller was remarkably calm, said we will listen to the tapes, but that they were using 4L for departures. I haven't seen any report on the length of the "phone call " or what was said. But AA 106 sat at the end of 31L for about 30 minutes. While they may not have realized how close they were to disaster, or how much notoriety was in store, they had to know they screwed up.

[tdracer]

It’s now 2 hours in the FAA world. I’ve heard EASA has gone to 25 hours. I know the NTSB is pushing for 25 hours and have a list of investigations that have been hindered by lack of CVR records. I’m sure this event will be added to that list.
Problem is, the rule would probably only apply to 'new build' aircraft. Although it's gotten cheaper and easier to retrofit CVR boxes since everything's gone digital, it's still expensive.
There are still lots of aircraft out there flying around with tape based CVR/FDR equipment.

[Chiefttp]

Unquestionably the AA crew knew the implications of their incursion. It’s bad enough to have a runway incursion, it’s really bad to cause another aircraft to reject due to said incursion. As far as JFK ATC high speed delivery, I’m a native New Yorker and that’s the way we talk..I do agree it doesn’t lend itself well to ATC and aviation ops in general and I predict one of the big outcomes from this incident will be a slower, paced delivery by ATC at JFK. I will always warn any new F/O’s to write down everything and listen carefully when we’re at JFK.
Willow 6-3,
The fact that the F/O was new to the jet has little bearing in my opinion. You said you were an Attorney, so If you decide to trade in your Mercedes for a BMW, would that be a valid excuse to run a red light in your new BMW? Also as far as the jumpseater, 3rd crewmember. If he was a jumpseater, (hitching a ride) they will usually try to not get in the way, outside of checking for anything major (Flap settings etc), unless the Captain briefs him to do otherwise. In my Air Force days, we had a taxi incident in Africa. The AF decided that at all times while taxiing, all crew members would scan outside and clear our routing. So we normally had at least 3 or 4 crewmember scanning outside and backing us up.I predict another takeaway from this incident will be sterile cockpit while taxiing.

[Bergerie1]

Chiefttp, I know you all talk like that but my question is, with so many foreign pilots flying into JFK, would it not be wiser, for safety's sake, to try to accomodate them? Even though this incident did not involve foreign pilots, neverthless, a slightly slower delivery would benefit everyone.

[Seaking74]

I tuned into JFK ATC last night, as I do most nights, and my favourite controller was on, he has a broad, thick NY accent (think "cwoffee"). It was noticeable that he was speaking more clearly last night, especially to foreign flight crew, as was the female controller who took over from him. Did a memo go around already perhaps?

[Chiefttp]

Chiefttp, I know you all talk like that but my question is, with so many foreign pilots flying into JFK, would it not be wiser, for safety's sake, to try to accomodate them? Even though this incident did not involve foreign pilots, neverthless, a slightly slower delivery would benefit everyone.

I agree with you 100%, what I meant by my comment is sometimes New Yorkers don’t realize how fast they speak naturally. It’s not necessarily intentional. Trust me I’m in awe of foreign crews who have to communicate in rapid fire English when it’s not their native tongue.

[Uplinker]

Yep. And nothing is gained by speaking so quickly. It's one thing doing so in a shop or a bar in New York city, but at a large, complex, busy airfield; where safety is critical, and many foreign crews operate; speaking slowly and clearly is very much a good idea.

You want everyone to clearly hear and understand, (and write down) clearances. There is no need to speak so fast, (and risk misunderstandings or having to repeat). It sometimes seems to me to be the oral equivalent of driving with one hand on the wheel and one arm leaning on the window ledge - it might look flash but it is less safe. :sad:

[AerocatS2A]

Also as far as the jumpseater, 3rd crewmember. If he was a jumpseater, (hitching a ride) they will usually try to not get in the way, outside of checking for anything major (Flap settings etc), unless the Captain briefs him to do otherwise.

3rd crew member would've been part of the crew on a LAX - LHR flight, not just a jumpseater.

[Chiefttp]

Was the flight LAX-LHR or JFK-LHR? I personally have met the F/O on the crew, She lives in the NY area, based in JFK but I guess the flight could have originated in LAX. Don’t know

[blind pew]

In 1978 I operated into JFK for the first time during training, BOAC had excellent briefing documentation and I had written down the expected clearance..I said say again twice and on the third go I thought I had read back the clearance correctly..at Deer Park the captain did something I didn’t expect and challenged him..I had got it wrong.
The following year I started working for the Swiss and was impressed by they linguistics until I had a rollicking from a Maastrict controller because the captain hadn’t understood me. I then drastically changed the way I spoke; annunciated in extremis, slowed my delivery, used simple words and always asked a question or relayed information that needed a response which couldn’t be answered with a yes or no. Such was the change that when I visited my old BEA crew room one of my mates thought I had a stroke.
It saved further internal cockpit confusion but we blocked a runway - probably Chicago- where we caused a missed approach..the captain never realised although the controllers instructions were clear.
Around 15 years ago I used to operate in Montpellier controlled airspace for 4 or 5 hours at a time..they had a female controller who obviously didn’t speak adequate English.
Maybe one day a system will be used which doesn’t rely on spoken clearances, writing them down and conflict warning such as on my car. The technology is there.

[WillowRun 6-3]

Chiefttp,
I was just relating here what was reported in the Forbes article and not trying to assign relevance to any particular fact, including the F/O's time on the type.
(And you'd be quite disappointed by the mundane motor vehicle I operate, quite unlike high-end German imports.)

[remi]

Reading the Forbes article it would appear that some soul decided to interfere with what should be a sterile phase of the flight in that the monitoring pilot had a series of tasks which could and should have been carried out when the aircraft was stationary. Especially relevant as the copilot was new on type.
It isn’t the first time that I’ve seen ill thought procedure changes which left no one monitoring.
Since the tasks involved checking take off performance then Pm wouldn’t have been looking outside.
Perhaps it’s also about time that the 30 mins CVR rule is changed - after all it was introduced 50 years ago.
(I was on BALPA tech committee in the early 70s tasked with airfield lighting amongst other things - operated a heavy in 1978 into JFK and know how difficult RT and taxying can be without unnecessary distractions).
Push to erase at the gate should not be a thing. (TWA 841) Cockpit cameras should be a thing. Erase at pushback or before maintenance makes so much more sense. There are accidents where CVR and FDR on a preceding leg would have been helpful to investigators.

My understanding is that leaks of US and European CVR have been extraordinarily infrequent, with most due to legal proceedings. Incidents of cockpit audio that have embarrassed crew and carrier have generally been from hot mics with ATC or intercom; in fact those are the only ones that come to light publicly as far as I recall.

I understand the privacy issues and no one likes people looking over their shoulder, but frankly we are all surveilled in the workplace now, and transportation seems like a role where it is truly warranted. There are strong legal protections where CVR is concerned and they seem to work.

[AerocatS2A]

Was the flight LAX-LHR or JFK-LHR? I personally have met the F/O on the crew, She lives in the NY area, based in JFK but I guess the flight could have originated in LAX. Don’t know
Oops. JFK obviously.

[FlightlessParrot]

Not an aviation professional, but have professional experience with speaking so as to be understood. Speed of oral delivery varies with language, region, individual, mood, context (including the size of the room, and presence or absence of microphone); in some languages, the unit of expression is the breath-group, not the word (in my inexpert experience, that makes a big difference between Spanish and Italian in their intellgibility to someone with a rudimentary knowledge of the language). But I am surprised that in the extensive training given to ATC people, there seems to be no training, and no standardisation, in speed and style of delivery. Can this really be so?

Has anyone done any research on speed of delivery and the consequences for efficiency in ATC (I imagine a curve, bell-like but with quirks). There are Departments of Applied Linguistics constantly looking for research projects to justify their existence; otherwise, one could assume that a speed that reduces "Say agains" to a very low level would be a good point to aim for.

[FlightDetent]

No, the terminology in Forbes is correct.

"Departure" is the process that starts with pushback (hence an on-time pushback = an on-time departure). A PA announcement Wrong, that's dual use actually.

Departure for pilots is the term for takeoff when they are not allowed to say the word 'take off' itself. It has been declared an exclusive and restricted use as a key lesson learned from the Tenerife tragedy.

Willow is mimicking R/T phraseology here, albeit perhaps unnecessarily. Simple does it, takeoff works unless you are talking about takeoff over the radio. :8

[WillowRun 6-3]

FlightDetente
..... actually that comment was just trying to catch a published news article in a mistake. As usual what an SLF/attorney doesn't know can . . . hurt their post.

[MarkerInbound]

Problem is, the rule would probably only apply to 'new build' aircraft. Although it's gotten cheaper and easier to retrofit CVR boxes since everything's gone digital, it's still expensive.
There are still lots of aircraft out there flying around with tape based CVR/FDR equipment.

At least every plane in the 121 world has a two hour CVR. FAR 121.359 was changed in March of 2008 to require any plane manufactured on or after April 7th 2010 to have a 2 hour CVR. If the plane was manufactured prior to April 7th 2010 it had to have a 2 hour CVR by April 7th of 2012. And EASA does requires any plane built in the last two years with a Max TO weight above 27 tonnes to have a 25 hour CVR so there have to be some out there. So if the NTSB can push the FAA hard enough we might have them in a few years.

[Uplinker]

.........changed the way I spoke; annunciated in extremis, slowed my delivery, used simple words..........they had a.......controller who obviously didn’t speak adequate English.
Maybe one day a system will be used which doesn’t rely on spoken clearances, writing them down and conflict warning such as on my car. The technology is there.

The UK CAA CAP 413 publication is a very good guide to spoken radio communications. Might seem a little anal but it is clear and it works.

[blind pew]

The UK CAA CAP 413 publication is a very good guide to spoken radio communications. Might seem a little anal but it is clear and it works.
yes but we aren’t dealing with English English spoken at normal rates.
Many moons ago I did a semi wet lease on the MD80 with a Swiss German skipper (limited english ability) for Air Afrique into Conakry where the recent occupants of the palace had been left hanging from a bridge. The runway was notamed as reduced length by 1/3rd due to a trench across it. ATC had been replaced by Chinese whose french and english was look it up in a phrase book without phonetic pronunciation.
So it was a 1’000ft pass which only revealed the orange strip (soil colour), elect to land from opposite direction avoiding restricted airspace around presidential palace (plate stated we would be shot down) and a short landing..fortunately the trench had been back filled.
WRT to the incident..the controller did an excellent job..sh@t happens.
to put it in perspectives one year recently there were three near accidents in ireland all down to ATC ..one aircraft landed over a gang mower on the active runway, à dubious cleared to cross after the rolling aircraft which saw an executive jet fly under a chopper just after rotation and a near CFIT iirc..and they don’t speak like New Yorkers.
I can also recall a mate in a 747 missing an aircraft by less than 10ft in Melbourne when another aircraft was towed across the runway in use.

[B2N2]

ATC audio:

5AbpxqtXcPk

I hate to say this but there are aviators that have and aviators that will…..matter of time.
Despite my best sympathy to the crew:

Everyone involved a Native American English speaker.
ATIS must have mentioned both departure runways in use, 4L and 31L/KE.
Even though it sounds like “Fowl left” or even “fohleft” that’s two syllables instead of 3 for tree-one-left or any variation thereof. 31L/KE is 5 syllables even if you don’t speak English.
Taxi clearance more clearly states hold short 31L.
Three people misidentified their location.
Ground controller spotted the problem first followed by “TWR”.


* you would obviously never interpret a taxi clearance by counting syllables but the brain has a habit of putting patterns together that it recognizes

[Jwscud]

.
to put it in perspectives one year recently there were three near accidents in ireland all down to ATC ..one aircraft landed over a gang mower on the active runway, à dubious cleared to cross after the rolling aircraft which saw an executive jet fly under a chopper just after rotation and a near CFIT iirc..and they don’t speak like New Yorkers.
I can also recall a mate in a 747 missing an aircraft by less than 10ft in Melbourne when another aircraft was towed across the runway in use.

I would respectfully disagree. Dublin ATC are as close to New York as it gets in Europe. Not exactly a surprise that they have some shared verbal heritage either!

[patrickal]

Article notes also there was a third pilot on the 106 flight deck; that neither the PIC nor 3rd aviator saw the "stop bar lights on the runway"; and that AA 106 had just switched from company frequency to Tower frequency. AA 106 did not know the seriousness of the incursion until they arrived in London, per Forbes article. "Delta was cleared for takeoff before they switched over" to the tower frequency, the article quotes the source as saying.


Perhaps a dumb question here, but after vacating the American apron, shouldn't they switch to and at least monitor Kennedy Ground as they taxi out? Why were they on company frequency after taxing for more than a mile from the gate?

[Chiefttp]

Patrickal,
Normally, one radio, the primary, is tuned to Ground, the other tuned to a company frequency during taxi. We have more than one radio.

[ethicalconundrum]

Patrickal,
Normally, one radio, the primary, is tuned to Ground, the other tuned to a company frequency during taxi. We have more than one radio.

Contributing factor? Split attention between grd and co freq? The CVR would have given information if the crew was talk/listening to co freq during taxi.

Long long ago I taxied across a runway and ground admonished me that was clear to(but not across) 31. I sat while they played it back, and I was very lucky that grd cleared me to a point on the opposing side of rwy 31, but it was not explicit that I was cleared to cross. However, the clearance was for a location that could only be reached by crossing 31. There definitely was no mention of 31 in the readback. I think the grd controller was unsure of which side of 31 I was on, and thought I was on east side, when I was on west side. I got away with it.

[Uplinker]

yes but we aren’t dealing with English English spoken at normal rates.....
Yes, I know. That was kind of my point ! :)

CAP 413 might seem stilted and silly, (and is UK), but I use its methodology, and have rarely had a problem being understood. I am sure some US ATC probably think I am being a ****, or am trying to be posh, but neither is true - I am just trying to be understood clearly, because clear communication is a real safety issue. :ok:

[22/04]

I suspect there is a danger of starting a war here but I agree abut CAP 413 - and the way even at my lowly level as an a/g operator my examiner does listen to me occasionally and point out deviations.

There are lots of differences and I would be interested to understand how they came about. Taxi instructions tend to be to a defined point here e.g "taxi E5 via taxiways Juliet and Echo" and there then shouldn't be a need for "hold short" which is ambiguous. The point is defined and unambiguous.

Similarly the use of "clear to land" and then a description of traffic ahead rather than "continue approach" with landing clearance only given when the runway is vacant and nothing happening ahead of the aircraft other than a clear runway.

I get the impression that in the U.S commanders are left more in control of their aircraft and ATC is less instructional.

[Chiefttp]

I don’t know what CAP 413 is, but I’m deducing it’s a reg on how to communicate clearly. Undoubtably, one of the “takeaways” from this incident will be a less frantic, slower paced, communication flow from controllers at JFK. As I’ve stated before, as a native New Yorker I’m used to the pace and speed of New Yorker’s speech, but it’s not appropriate for aviation communications at JFK.

[DaveReidUK]

I don’t know what CAP 413 is, but I’m deducing it’s a reg on how to communicate clearly.

https://cimg6.ibsrv.net/gimg/pprune.org-vbulletin/500x500/cap_413_171af849ab9a7f9f74041b0e3368a680d2986e9b.jpg

[22/04]

CAP 413: Radiotelephony Manual (caa.co.uk) (https://publicapps.caa.co.uk/modalapplication.aspx?catid=1&appid=11&mode=detail&id=9857)

[BFSGrad]

I don’t know what CAP 413 is, but I’m deducing it’s a reg on how to communicate clearly. Undoubtably, one of the “takeaways” from this incident will be a less frantic, slower paced, communication flow from controllers at JFK. As I’ve stated before, as a native New Yorker I’m used to the pace and speed of New Yorker’s speech, but it’s not appropriate for aviation communications at JFK.Are we saying that the AIM and 7110.65 are inadequate to instruct pilots and controllers in proper radio technique and phraseology? I think the fast-talking JFK ATC as a causal factor is a distraction. Listening to the public ATC audio of the incident, I didn’t find anything unusual compared to any large, busy ATC facility in the U.S. IMO, the JFK ground/tower controllers spoke more clearly than the AA106 FO. As I noted in my post #23, the public ATC audio is from a secondary source and is likely of inferior quality compared to what is on the ATC tapes and the CVR (was, RIP). Where the public audio/video shows “blocked” or “inaudible,” those portions may have been perfectly audible to ATC and the AA106 crew.

If JFK (and LGA, TEB, EWR) ATC comms are really a safety issue, is anyone in the profession noting this on an official basis? APA/ALPA? Airlines? ATC comms are a dialogue, not a monologue. Pilots always have the ability and responsibility to ask ATC to repeat any unclear comms as well as control the pace and clarity of their own comms to ensure safety.

[MJA Chaser]

fast-talking JFK ATC as a causal factor is a distraction.

My assesment is that they were speaking way faster than the recommended 100 WPM.

[Uplinker]

.......... Listening to the public ATC audio of the incident, I didn’t find anything unusual compared to any large, busy ATC facility in the U.S...........
........If JFK (and LGA, TEB, EWR) ATC comms are really a safety issue, is anyone in the profession noting this on an official basis? APA/ALPA? Airlines? ATC comms are a dialogue, not a monologue. Pilots always have the ability and responsibility to ask ATC to repeat any unclear comms as well as control the pace and clarity of their own comms to ensure safety.

Umm yeah. Try reporting bad USA ATC comms as an English pilot working into JFK and see how far you get ........

[ATC Watcher]

While I basically agree with the commenets made about the speed of delivery in of some US controllers in busy airports and I would add even on some en route sectors, I do not think it is the main issue here. We could also debate here on how the ICAO ELPAC ( the language proficeinecy tests) were embraced by the US and the FAA when they became mandatory in 2011 but, it is another story , but the fact here that I believe this had very little to do in this incident . As correctly pointed out by BFSGrad , the quality of the internet audio recording most likely comig from a hand held cheap scanner in aparking lot, and would not the same quality as what you hear on the cockpit VHF,,. In addition the AA crew did not question the taxi clearance and everybody involved was a native english speaking American.

From what I read so far, the new briefing and checks to be performed while taxing, perhaps having to deal with another issue or having a discussion during that time made them misinterpret the crossing part of the clearance.. Unfortunately the CVR will not give us this info and most importantly the timig of things, so the incident report will have to rely on memory of the involved persons days afterwards .not really ideal to draw conclusions.

[megan]

one of the “takeaways” from this incident will be a less frantic, slower paced, communication flow from controllers at JFK. As I’ve stated before, as a native New Yorker I’m used to the pace and speed of New Yorker’s speech, but it’s not appropriate for aviation communications at JFKCommentary about the pace of JFK comms has been a source of much discussion for years and years, all in the negative, so I doubt anything will become less frantic and slower paced as you suggest Chiefttp, despite you acknowledging it’s not appropriate for aviation communications

[BFSGrad]

Toward the end of the 2/7 testimony before the House Transportation & Infrastructure Committee, the NTSB chairman provided the following answer in response to a question about the JFK incident:

‘It was the ASDE system—airport surface detection equipment—system that notified the air traffic controller that there was an impending collision.’

[Seat4A]

"Preliminary report is now out." Posted Feb 10 2023, 3:25pm ET

https://twitter.com/RossFeinstein/status/1624142574209404930?cxt=HHwWhMC4hZ7EjootAAAA

[waito]

"Preliminary report is now out." Posted Feb 10 2023, 3:25pm ET

https://twitter.com/RossFeinstein/status/1624142574209404930?cxt=HHwWhMC4hZ7EjootAAAA

A short one. Can I copy?

On January 13, 2023, about 2044 local time, American Airlines (AA) flight 106, a Boeing 777- 200, crossed runway 4L, without air traffic control (ATC) clearance, at John F. Kennedy International Airport (JFK), Queens, New York causing Delta Airlines (DAL) flight 1943, a Boeing 737-900ER, to abort its takeoff on runway 4L. Of the 6 crew and 153 passengers on DAL 1943, and 12 crew and 137 passengers on AA106, there were no injuries. There was no damage to either aircraft. AA 106 was a 14 Code of Federal Regulations (CFR) Part 121 scheduled international passenger flight from JFK to London Heathrow International Airport, London, United Kingdom (LHR). DAL 1943 was a CFR Part 121 scheduled international passenger flight from JFK to Santa Domingo, Dominican Republic (SDQ). AA 106 was instructed to taxi from the ramp to Runway 4L via taxiway B and hold short of taxiway K. As the aircraft left the ramp, the ground controller cleared AA 106 to cross runway 31L at taxiway K. Upon reaching the Taxiway B/Taxiway K intersection, AA 106 continued straight to taxiway J crossing runway 4L without ATC clearance. At the time AA106 entered on runway 4L at taxiway J, DAL 1943 had begun its takeoff roll increasing speed through 80 knots, and was abeam taxiway K3, about 2700 feet from the taxiway J intersection.

The ASDE-X alerted the JFK ATC tower to the conflict and the tower controller issued a takeoff cancellation to DAL 1943. The DAL 1943 crew aborted takeoff at about 100 knots and came to stop about 500 feet short of taxiway J. The closest point between the two aircraft was about 1400 feet and occurred as DAL 1943 decelerated past taxiway K4 and AA106 exited the runway at taxiway J. NTSB groups formed to investigate this accident: operational factors, human performance, air traffic control, and flight data recorder. Parties to the investigation include the Federal Aviation Administration (FAA), the Boeing Company, Delta Airlines, American Airlines, and the Allied Pilots Association (APA). On February 1, 2023, NTSB conducted ATC interviews, which were audio recorded for transcription. Recorder data from both aircraft were obtained. Cockpit voice recorder data were both overwritten. Flight crew statements were received. In addition to the crew statements, NTSB attempted to interview the American Airlines flight crew three different times. American Airlines cleared the flight crew’s schedule to ensure their availability; however, the flight crew refused to be interviewed on the basis that their statements would be audio recorded for transcription. On behalf of the crew, the APA party representative informed the NTSB that the crew would not consent to participate in audio recorded interviews in any manner. NTSB has determined that this investigation requires that the flight crew interviews be audio recorded and transcribed by a court reporter to ensure the highest degree of accuracy, completeness, and efficiency. As a result of the flight crew’s repeated unwillingness to proceed with a recorded interview, subpoenas for their testimony have been issued.

short but packed with interesting details.

[ATC Watcher]

Indeed . Subpoenas to get a report from a crew after an incident? Never heard of this before,The US legal system is amzing . Can they do like politicians and " play the 5th amendment " and say nothing ? Wating with impatience to read Willow Run 6-3 comments on this .

[BFSGrad]

Indeed . Subpoenas to get a report from a crew after an incident? Never heard of this before,The US legal system is amzing . Can they do like politicians and " play the 5th amendment " and say nothing ? Wating with impatience to read Willow Run 6-3 comments on this .Seems the sticking point is the audio recording and associated transcription even though I suspect that method of interview is standard procedure for the NTSB. Eventually the interview transcript will be made public; e.g., TransAir 810 interviews.

[BFSGrad]

From the linked statement it seems APA would be OK with a transcribed interview but not any type of interview that is recorded electronically.

Allied Pilots Association Statement on National Transportation Safety Board (https://www.alliedpilots.org/News/ID/11069/Allied-Pilots-Association-Statement-on-NTSB)

[WillowRun 6-3]

I'm not convinced that audio recording and transcription by a court reporting service are standard at this stage of an NTSB inquiry (maybe these are routine but the fact has eluled my experience). Also, the prelim report refers to an "accident".... which I don't think is the correct terminology for what occurred.

As to the privilege against self-incrimination, under the Fifth Amendment of the U.S. Constitution:
As a first point, while politicos might have given "taking the Fifth" a poor reputation, the Constitutional privilege against self-incrimination is a very significant right (just for the record).

The privilege generally applies to criminal proceedings. The NTSB process at this time is far from criminal in nature. But..... the privilege against self-incriminatiin may be asserted by a witness in any type of proceeding if there is reasonable possibility testimony by that witness would incriminate the witness in future criminal proceedings. And a subpoena does not disarm (it doesn't over-ride) the privilege.

My moderate-to-strong inclination here is to rely on what I understand is the very high level of effectiveness and professional competencies of the labor organization which represents American's pilots. They may, for example, sense a search for fault-finding and scapegoats is looming. Rather than investigation of the incident giving fair or equal time and consideration to the way ATCOs communicate at JFK, and/or to the set of changed procedures the F/O reportedly was required to review, the APA reps and legal counsel may see Board or other highly visible authorities trying to tag the AA crew as wet laundry needing to be cycled, or hung out as the case may be, to dry.

With FAA reauthorization pending soon in Congress, and with FAA part of a Cabinet agency headed by a political appointee whose aspirations, not to say ambitions, for higher office are well-known, it strikes me as not surprising at all that labor organization reps and legal counsel are playing this match very cautiously. Not surprised at all. (I know aviators on the line wear suit coats as part of their uniforms, not jackets; still, back in the City of Chicago, County of Cook, State of Illinois, we would refer to this cautiousness as, "My clients are not gonna wear the jacket on this.")

EDIT: the above written and posted prior to seeing APA statement in post immediately preceding mine; no further comment, and this post's content not edited.

[BFSGrad]

NTSB Statement on the Use of Recorded Interviews in Aviation Accident Investigations

[ethicalconundrum]

You bet they can and should refuse to offer any testimony. NTSB is playing hardball, and they are facing a min of a job action, and a maximum(depending on what they say) of potential criminal charges down the line. The US govt is no longer the kind older uncle. They are going for blood by issuing a subpoena for testimony. The only thing I would say under oath is 'I refuse to answer, and assert my constitutional rights', even if they aren't citizens.

This kind of stuff is BS, and needs to stop. It most certainly will not benefit safety. The pilots crossing without authorization are going to take a rip. Get it done, and move on. Not just the 5th Amendment at work here, but we also have something called the 'Pilot bill of rights' which offers significant legal protection. I am almost certain both pilots of the offending runway plane have counsel already. If I were their atty(solicitor) I would seek to quash(remove for cause) the subpoena due to missing any criminal intent. The pilots offered to give information, but not enough for the feds. Tell them to pound sand, it's not going to get better with age.

[VHOED191006]

Sorry, but that's ridiculous. Over 300 people nearly died, and they're refusing to give their account of the situation and their insights/thinking at the time? I believe that it's quite frankly an insult to what the industry has been working towards - the highest level of safety possible. Don't most of us believe Just Culture? And so what if the interviews are recorded electronically? Nothing from the final report can be taken as liability nor can it be used to appoint blame. The basic question in this investigation that must be answered is: 'why did a 737 at a high speed almost crash into a 777?'. For the sake of the industry, you must have an openness and willing to be transparent in explaining from your perspective what happened in a particular instance. If you don't, then the result can end up with blood.

[blind pew]

CVR overwritten on both aircraft; the system appears not to have evolved since the Trident crash in 1972. Imho the constraints of the FDR should be applied as in a non erasable full time CVR record if only for learning lessons.

[ethicalconundrum]

Sorry, but that's ridiculous. Over 300 people nearly died, and they're refusing to give their account of the situation and their insights/thinking at the time? I believe that it's quite frankly an insult to what the industry has been working towards - the highest level of safety possible. Don't most of us believe Just Culture? And so what if the interviews are recorded electronically? Nothing from the final report can be taken as liability nor can it be used to appoint blame. The basic question in this investigation that must be answered is: 'why did a 737 at a high speed almost crash into a 777?'. For the sake of the industry, you must have an openness and willing to be transparent in explaining from your perspective what happened in a particular instance. If you don't, then the result can end up with blood.

From the prelim report: "Flight crew statements were received." From the perspective of the FAA, they have completed their mandated reporting of an 'incident'. Let the feds make of it what they will with the info received. It really doesn't take a lot of energy to figure out what went wrong here.

[VHOED191006]

From the prelim report: "Flight crew statements were received." From the perspective of the FAA, they have completed their mandated reporting of an 'incident'. Let the feds make of it what they will with the info received. It really doesn't take a lot of energy to figure out what went wrong here.
But that's only statements, not answers to questions in an interview. From the perspective of the NTSB, they're not fulfilling their requirements. You can't just assume that they've gotten confused with what they heard. You've got to consider other factors.

[AerocatS2A]

But that's only statements, not answers to questions in an interview. From the perspective of the NTSB, they're not fulfilling their requirements. You can't just assume that they've gotten confused with what they heard. You've got to consider other factors.
Right. A statement is essentially an answer to questions you’ve decided to ask yourself.

[WideScreen]

But that's only statements, not answers to questions in an interview. From the perspective of the NTSB, they're not fulfilling their requirements. You can't just assume that they've gotten confused with what they heard. You've got to consider other factors.
A written statement has probably been gone through lawyers, implying "nasty" things are filtered out. Potential additional questions/answers will go the same route.

A recorded audio statement will clearly mark answer hesitations and lawyer intervention, so, yeah, can be highly valuable, especially when pesky lawyers did ruin the value of the earlier write statement. And where witnesses show "hesitation" as an investigator, you know you hit a relevant "sweet" spot to check out on, further on.

[WillowRun 6-3]

Which is why, in criminal matters, the Constitution guarantees the right to trial by jury of one's peers.

If you believe that "hesitation" denotes something to hide, then if we ever meet up in person, remind me to tell you the story of the cross-exam that went off the rails (off, badly) for opposing counsel when she pounced on a "hesitation" in an answer. (The witness had nothing to hide but the meat loaf sandwich for lunch, well . . .). And yeah, we won at trial.

[WideScreen]

Which is why, in criminal matters, the Constitution guarantees the right to trial by jury of one's peers.

If you believe that "hesitation" denotes something to hide, then if we ever meet up in person, remind me to tell you the story of the cross-exam that went off the rails (off, badly) for opposing counsel when she pounced on a "hesitation" in an answer. (The witness had nothing to hide but the meat loaf sandwich for lunch, well . . .). And yeah, we won at trial.
The NTSB is not a trial for guilty and/or punishment, but an ac/incident investigation to prevent things happening again. For this, you need those fine details and clues that can be available in a spoken statement and have been lost when lawyers are involved. Things like a hesitation or just a sigh, which in turn can be a reason for ac/incident investigators to ask for more details, clarification, etc.

Just to give a (very) stupid example. When the FO gives a statement, something like, "I told the captain, blablabla" and then sighs, the investigators can ask further and maybe found out, the captain was (for example) mentally absent at that moment. Something potentially highly relevant, but probably not make it into a written statement, where lawyers scrap out everything that can not be proved (how did you conclude the captain was, etc ?).

[Bergerie1]

There are systems which are reasonably mature which could assist both pilots and ATCOs. Worth a look:-
https://www.skybrary.aero/articles/a...e-systems-asas (https://www.skybrary.aero/articles/airborne-separation-assurance-systems-asas)
https://www.asas-tn.org/workshops/5th-workshop/session-3/5_ATSA_SURF.ppt
https://www.asas-tn.org/workshops/final-seminar-paris-14-15-april-2008/session-3/1_Runway_incursions.ppt (https://www.asas-tn.org/workshops/5th-workshop/session-3/5_ATSA_SURF.ppt)

[WillowRun 6-3]

No, WideScreen, I'm very well aware of the purposes of safety inquiries at the Board and otherwise. I'll assume you are also. So, if attorneys are allowed to be present, what is the predicate for legal counsel to say anything? -- the inquiry isn't a contested proceeding, and it isn't under oath. I would guess that if legal counsel are permitted to be present, it is in a non-speaking role. Union reps, probably treated differently, but I don't know.

Beyond the logic which yields to blaming legal counsel for something they might not even be able to do, I think you're missing the point of my presumptive support of APA and the crew. The FAA or the Secretary or some other political player is looking for someone to shoulder the lion's share of the blame and not just limited to this specific incident alone. You may have seen or heard about the recent House T&I hearings to start the reauthorization process - more than incidental semi-alarmist handwringing over the SWA "meltdown", this incident, the Austin TX incident. Of course the Board investigation will be defended as just a fact inquiry, which it is, but: if one believes that there are not significant political pressures around this particular incident, and protestations about the merely factual nature of Board inquiries aside - at least from the cheap SLF/attorney seats I'm in, the crew and APA have reasonable cause to be concerned with being made into examples -- beyond the specifics of the actual incident.

I'll just agree to disagree that your archetypal investigator drills deeper factually just because some sigh or hesitation occurs. Pressures in courtroom cross-examination are even more intense, which is why I used the example. Maybe you've sat in a ton more interrogations than I have. ​​​​​c'est la vie.

[Denti]

CVR overwritten on both aircraft; the system appears not to have evolved since the Trident crash in 1972. Imho the constraints of the FDR should be applied as in a non erasable full time CVR record if only for learning lessons.

At least in EASAland 25 hour CVRs are mandatory. That requirement might not have made it across the pond, but even on the east side there is no retrofit requirement. No idea about other types, on the A320 CVR and FDR have been integrated into one single unit (CVDR) which holds both datasets on one solid state memory, to give a higher chance of finding all data there are two of those integrated units on board, one in the tail in the traditional section 19 and a second one in the E&E (90VU) compartment, both with identical memory and crash proofing.

[BFSGrad]

Here’s how the NTSB investigators kicked off the first (face-to-face) interviews with the TransAir 810 pilots:

Today we’ll be using a digital recorder. After the interview we will send that recording out for transcription and the party members will have an opportunity to review that transcription to correct it for typographical errors.
Once again, the purpose of the investigation is to determine probable cause and prevent reoccurrence. Our role is not to assign fault, we’re here as simply for you to share your insight with the goal of preventing accidents. Having said that, however, we cannot offer any guarantee of confidentiality or immunity, and the transcription of this interview will eventually be made part of the public docket for this investigation.

As previously discussed, Mr. Katt [attorney], you can direct the witness to -- actually he’s not actually the witness -- you can direct Mr. Ryan [FO] to answer a question, you can request a short break to confer with him, if needed, but if you would please refrain from answering any questions for him or on his behalf that would be appreciated.

The pilots were allowed to have a representative present, which in both cases was the same attorney. In the follow-up telephone interviews, both pilots declined to have their attorney present. The FO was also a family law attorney. TransAir (Rhoades Aviation) is (was?) a non-union air carrier.

[FlightDetent]

This will sound enormously stupid, why not use an analogue recorder? Does it?

[DaveReidUK]

At least in EASAland 25 hour CVRs are mandatory.

Became effective January 2021, I believe. I'd be surprised if it's retrospective.

[ethicalconundrum]

I don't discount the possible or likey probative value of further investigation, including in-person interviews, and even recording of voice, in-person visual clues(eye-roll, scowl, laugh, smile, brow furrow, etc). There is a mandatory reporting requirement for the crew, and by making their report/statement they have now fulfilled the regulation. After that, one party(NTSB) decided to go all 'we will MAKE you testify in person' with the threat of subpoena. For me, at that point - I have now decide that my voluntary cooperation has come to an end, and we have just entered what I consider adversarial territory. At this point, there will be no more discussion, interview, examination, evaluation, or reaction without benefit of counsel(solicitor). If that doesn't satisfy, issue the subpoena, bring me in to the appropriate venue, swear me in, and you can hear me say the famous words; 'I decline to answer on advice of counsel as my statement MAY be used against me in a court of law'. They we all file out, and let the media find out who the bad guy is/was.

They may want my help, and may ask and I may offer. NTSB/FAA/fedguv demand my help - go **** in one hand and demand in the other and see which get filled first. As someone who has been on the wrong side of Johnny law more than once, the feds are not there to help you, or be your buddy, pal, friend, defender.

YMMV, objects in mirror, pro driver on closed course, contents have settled and may cause anal leakage.

[ethicalconundrum]

But that's only statements, not answers to questions in an interview. From the perspective of the NTSB, they're not fulfilling their requirements. You can't just assume that they've gotten confused with what they heard. You've got to consider other factors.

If one is not satisfied with the mandatory reporting method by the FAA as it involves incidents they can seek to have the language changed. If they want to require in person interviews, in person recordings, and repeated investigation techniques then have that added to the FARs by issuing a NPRM, gathering the replies from the notice, and then publish the new rules. I do not know for certain, but pretty sure anything like that would violate the legislature approved Pilot Bill of Rights act(public law 112-254) which would clearly trump any administrative regulation.

[WideScreen]

No, WideScreen, I'm very well aware of the purposes of safety inquiries at the Board and otherwise. I'll assume you are also. So, if attorneys are allowed to be present, what is the predicate for legal counsel to say anything? -- the inquiry isn't a contested proceeding, and it isn't under oath. I would guess that if legal counsel are permitted to be present, it is in a non-speaking role. Union reps, probably treated differently, but I don't know.

Beyond the logic which yields to blaming legal counsel for something they might not even be able to do, I think you're missing the point of my presumptive support of APA and the crew. The FAA or the Secretary or some other political player is looking for someone to shoulder the lion's share of the blame and not just limited to this specific incident alone. You may have seen or heard about the recent House T&I hearings to start the reauthorization process - more than incidental semi-alarmist handwringing over the SWA "meltdown", this incident, the Austin TX incident. Of course the Board investigation will be defended as just a fact inquiry, which it is, but: if one believes that there are not significant political pressures around this particular incident, and protestations about the merely factual nature of Board inquiries aside - at least from the cheap SLF/attorney seats I'm in, the crew and APA have reasonable cause to be concerned with being made into examples -- beyond the specifics of the actual incident.

I'll just agree to disagree that your archetypal investigator drills deeper factually just because some sigh or hesitation occurs. Pressures in courtroom cross-examination are even more intense, which is why I used the example. Maybe you've sat in a ton more interrogations than I have. ​​​​​c'est la vie.
Ouch, sorry stepping on your toes.

I think, you did not get my point. Let me write again.

The NTSB investigation/interviews is/are not a court-room.

The AA crew did hand-in a written statement. A statement, the NTSB was not satisfied with and as such, wanted to hear the AA crew in person. Initially just verbal, later with a certified note taker, later as a last resort with audio tape. All of these in person interviews were turned down by the AA crew.

My issue (and presumably the NTSB too), is that the written statement is probably a lawyer "corrected" version of what the AA crew wrote/created themselves, omitting those items, the NTSB is interested in. As such, my critics around negative lawyer influences, as far as it goes for in/accident investigations. The NTSB just wants to know the real version, not the lawyer downgraded one. Even when a lawyer would be present during the verbal interview, it will be much clearer, when the lawyer influence jumps in. Giving the interviewers the opportunity to add questions, etc.

A very clear example of bad lawyer influences is what we publicly see happening in the Trump insurrection investigation, where all important witnesses suddenly have a loss of memory ("proving they are not suitable for the job they did, a better memory is required for that job"), and persistently declare "I have no recollection of blablabla".

Only after Cassidy Hutchinson kicked out her Trump paid lawyer, revising her statements and started telling the real truth to the investigation committee, it became clear, that pesky lawyers had influences witnesses at a large scale in this investigation. To have had a few more witnesses with regrets and the investigation would have dived in much deeper with significantly more clarity around the Jan 6 insurrection attempt. A pretty shameful happening, in a country where honesty, law and order are considered predominant. Not.

Just check, when another presumed Russian spy/murderer is caught somewhere in the world and brought to trial. The moment, they keep denying their guilt, despite overwhelming evidence of the opposite, you know, they are indeed Russian spies........ Yes, I know, innocent, until proven otherwise, but that is the proof part, not the factual guilty.

Regarding your presumable fear to victimize the wrong person(s), that is more an issue due to the FAA hindering and persistently wanting to ignore the NTSB's recommendations. Not to say, the moment the NTBS is able to investigate a case properly (and they are rigorous, as the recent past has shown), the real parties to blame are pretty well identified. Though to come to that point, there is a need to get the facts first, all the facts.

And, we can see, the NTSB does do an excellent job, even producing "alternative" investigation reports, when the official one is obviously politically or commercially motivated watered down, like in the ET302 737MAX crash report.

[WideScreen]

I don't discount the possible or likey probative value of further investigation, including in-person interviews, and even recording of voice, in-person visual clues(eye-roll, scowl, laugh, smile, brow furrow, etc). There is a mandatory reporting requirement for the crew, and by making their report/statement they have now fulfilled the regulation. After that, one party(NTSB) decided to go all 'we will MAKE you testify in person' with the threat of subpoena. For me, at that point - I have now decide that my voluntary cooperation has come to an end, and we have just entered what I consider adversarial territory. At this point, there will be no more discussion, interview, examination, evaluation, or reaction without benefit of counsel(solicitor). If that doesn't satisfy, issue the subpoena, bring me in to the appropriate venue, swear me in, and you can hear me say the famous words; 'I decline to answer on advice of counsel as my statement MAY be used against me in a court of law'. They we all file out, and let the media find out who the bad guy is/was.

They may want my help, and may ask and I may offer. NTSB/FAA/fedguv demand my help - go **** in one hand and demand in the other and see which get filled first. As someone who has been on the wrong side of Johnny law more than once, the feds are not there to help you, or be your buddy, pal, friend, defender.

YMMV, objects in mirror, pro driver on closed course, contents have settled and may cause anal leakage.
Yep, reporting is mandatory. A report, with effectively "I have nothing to tell", is also a report. lE, legal obligation fulfilled, usefulness zero.

These regulations are from a time in history, when it was considered an honor to contribute substantially. Nowadays, the contribution is generally torpedoed down into something useless. With the consequence, the NTSB starts to use more formal instruments to enforce substantial cooperation.
All due to the tendency to use NTSB reports in court to blame people, ruin their lives, etc.

Legal/lawyers does/do have a substantial influence on this happening/tendency.

[ATC Watcher]

Qoting WideScreen :The NTSB investigation/interviews is/are not a court-room.
That is the key sentence. By allowing lawyers into incident invstigations we will learn nothing valuable, because the onus wiil no longer be on what actually hapenned in order to proppse recommendations to pervent it happenning again , but the focus is on avoiding or deflecting blame and criminal prosecutions. Fortunately only a very few countries are following that , but it is a worrying tendency. If we continue down that path safety will never improve.

I pesronally find it a shame for a large Professional pilots union to attack the NTSB as they did instead of working together with them to get rid of judicilary interferences in their work . The Union has the money and the political lobby power to change things, attacking the NTSB instead wil not.do any good. As for the NTSB to issue subponeas to crews, that will also not advance safety in any way.

[spornrad]

The solution to this classical conflict of interest would be 'just culture'. FAA / prosecutors grant immunity, crew has no reason and no right to refuse full cooperation.
Way to learn from honest mistakes to avoid them in the future.

[DaveReidUK]

The NTSB investigation/interviews is/are not a court-room

Except that, in a way, they are:

however, we cannot offer any guarantee of confidentiality or immunity, and the transcription of this interview will eventually be made part of the public docket for this investigation

We shouldn't underestimate the influence of The Court of Public Opinion, or more specifically that of the experts (and "experts") in the media, who will be all over the testimony once it's in the public domain.

[ATC Watcher]

Except that, in a way, they are:
We shouldn't underestimate the influence of The Court of Public Opinion, or more specifically that of the experts (and "experts") in the media, who will be all over the testimony once it's in the public domain.
Agree to a certain extend, but this was an incident , not an accident . Nobody involved woke up in the morning that day saying : "I am going to create an incident today" so no need for any criminal investigation , lawyers and judiciary to get involved. . Let the invstigators work alone. If during the investigation the NTSB discover someone did had these thoughts in the moning , then they shoud pass it on to the judiciary.

As to the Court of public opinion once it becomes public, what ? an incident a year or two later ? will be old news . But again this seems to be more a US issue .

[blind pew]

In context perhaps one should remember the November Oscar incident in 1989 where Glen Stewart refused to go into the office and was hung out to dry eventually topping himself after being prosecuted in a civil court by the authority.
It was after a long flight, with sick crew members, in crap weather with an autopilot that wasn’t fit for purpose (at that time) and had been given dubious permission to ignore the copilots lack of qualification for said approach, The relevant tech log pages were “lost” although a mate had the aircraft the next week and the coupled approach was again displaced off the ILS and entered in the tech log.

[Lake1952]

Immediately after the "incident ", the crew of AA106 was told to make a phone call to the tower. I don't know if that call was recorded. I haven't heard anything about that call. The flight did not depart 31L for LHR for a half hour. While that time lapse may be perfectly normal at JFK, I would like to know what was said on that call. Were other calls made by the AA crew prior to takeoff, perhaps to their company dispatch? Was any consideration whatsoever given to cancelation or delaying the AA106? Or did the AA crew only learn about the gravity and notoriety of the "incident " after they arrived in London?

[ATC Watcher]

Immediately after the "incident ", the crew of AA106 was told to make a phone call to the tower. I don't know if that call was recorded. I haven't heard anything about that call.
Nowadays all calls to ATC operational ops rooms are to be recorded , I do not know the specific JFK procedures but it would be extremely unlikely if they were not recorded. As to have not heard about it , again talking from my Europan experience , those recordings can only be accessed by the persons involved including the supervisor and the incidents investigators, no way they would be leaked to the outside, The person doing this will be fired .

[BFSGrad]

...those recordings can only be accessed by the persons involved including the supervisor and the incidents investigators, no way they would be leaked to the outside, The person doing this will be fired .
Harrison Ford is still waiting for someone to be fired.

[WillowRun 6-3]

The Chairman of the Transportation and Infrastructure Committee of the House of Representatives kicked off the hearing last week (February 7) to begin the process of FAA reauthorization. (For readers not quite familiar with U.S. legislative processes, authorization creates the legal mandate for federal agencies to conduct their activities; appropriations legislation is the funding. Hence authorization (and periodic reauthorization) holds by far the heavier policy implications and effects. Sorry for the pedantic detour, but after all, this is PPRuNe, where information is cheap, cheap, yet understanding may be dear.)

The Chairman, Congressman Sam Graves, stated as follows:"[R]ecently there have been incidents that reemphasize why getting an FAA reauthorization done on time is critical. On January 13th, a runway incursion occurred at JFK International Airport when two passenger planes nearly collided as one crossed an active runway. And just this past weekend, at Austin International Airport, a cargo plane was attempting to land on the same runway where a passenger plane was beginning to take off.It shows that even following the safest decade in our history, our aviation system is clearly in need of urgent attention. As Mr. Boulter says in his testimony, complacency and stagnation are equal threats to a safety culture. The previous conventional wisdom for regulating safety focused on addressing concerns after aviation accidents. Now, the FAA seeks to mitigate risks before accidents happen.In addition, the National Transportation Safety Board (NTSB) has several open safety recommendations that warrant review. The Committee will be reviewing all such recommendations while reauthorizing the NTSB as part of the FAA bill." [Note: David Boulter is the Associate Administrator for Aviation Safety (Acting) of FAA.]

In reporting on the opening reauthorization hearing, the Wall Street Journal (Feb. 8) noted that current NTSB Chair Jennifer Homendy (in an interview) had cautioned against, with regard to the two runway incursion incidents, saying "this is a trend." The WSJ article also quoted a former investigator for FAA and NTSB as saying, "two incidents so close together is more of a coincidence than an indication of a systemic problem." (This SLF/attorney isn't using that person's name here since he is not afaik a public official or public figure.)

However, note the comments by former NTSB Chair Robert Sumwalt, as reported by the Journal:
"the close calls didn't appear related or part of a broader trend. Both appear to have resulted from human error, [Sumwalt] said, and the outcomes point to the exceptional safety record of the U.S. air-traffic system and aviation industry. 'It shows that given there there were failures, that there was enough resiliency or enough redundancy in the system to have prevented a catastrophic event in both cases.', Mr. Sumwalt said." (emphasis in quotation added)

So, first, with regard to a video representation of the events (and/or non-events) in Austin that is highly inaccurate, count this poster as one who sees no utility or usefulness whatsoever in such a video, for informing, aiding, facilitating, or otherwise being helpful, in the nascent public policy and legislation process which ultimately will lead to FAA reauthorization. Wanting to play with computers and video programs? - fine, no problem. But posting such output here is only about ten (10) years or more behind the times (see, for example, the thread on the accident in August 2013 in Birmingham AL, UPS Fl. 1354, in which thread a good number of posters contributed very informative graphics and computational items.) And, needless to remind the serious pro's who populate this forum, the video showing Air Canada 759 in San Francisco in 2017 was indeed very informative, helpful, and so on.... but it was evidence, not fun-time.

Second, I think there is a fair question - possibly even an important question - about what meaning to take away from both incidents, whether looked at as only quite marginally related or connected to each other - or as connected in some as yet-undefined manner beyond being in close temporal proximity. Is it actually random chance that small steps in each incident - for example, the Southwest flightcrew not "aborting" their takeoff roll (if they even had heard the FedEx call) - contribued to a serious tragic accident not occuring? Or is it something other than random chance, something closer to what most people fluent enough in English to parse definitions would identify as "resiliency" or "redundancy"?

Or, . . did a previous poster make this all less interesting, by saying there are no "reserves" - the term I had tried to apply to what Chairman Sumwalt called resiliency and redundancy - but instead there are "layers and back-ups"?

[ATC Watcher]

Harrison Ford is still waiting for someone to be fired.
My remark was for my ATC Centre in Europe , and I believe valid for most European ANSPs.. I cannot speak on what the FAA does..

[ATC Watcher]

@ WillowRun6-3
Second, I think there is a fair question - possibly even an important question - about what meaning to take away from both incidents, whether looked at as only quite marginally related or connected to each other - or as connected in some as yet-undefined manner beyond being in close temporal proximity.
For me they are not related and are both very different in nature. One JFK was crew related was saved by a back up ( which you call reserve ) system : the ASDE which alerted the controller of the runway incursion.
The secondi ncident in Austin was ATC related and was not saved by any back up. I still maintain good luck saved the day. Not by any action by any of 3 parties involved, and not by an external factor like in JFK , since there is no ASDE in Austin .

The only possible ( emphasis on possible) common feature in both cases would be ATC staffing .Things like how long were the controllers on duty on the day and the days before, were they working alone where there should have been two, etc..there the FAA migh have to look at themselves.
The pace of R/T delivery on both cases , too fast in one , too slow and sounding tired in the other might give a clue. But I am fully speculating here.
​​​​​​​
Another issue coming from the Austin case is standard phraseology , totally disregarded by all 3 participants. If I was the US regulator I would definitively have a serious look at this. As should the correct phraeology had be used , that would have been a safety layer ( what you call reserve) that could have saved the day in a far better way..

[ethicalconundrum]

Except that, in a way, they are:

"however, we cannot offer any guarantee of confidentiality or immunity, and the transcription of this interview will eventually be made part of the public docket for this investigation"


We shouldn't underestimate the influence of The Court of Public Opinion, or more specifically that of the experts (and "experts") in the media, who will be all over the testimony once it's in the public domain.

The bolded is the key takeaway from the NTSB position. Without beating the dead horse too much, the way it goes is the feds want pilots to testify, in person, with recordings and failing to do that, one will be subpoenaed. And, when forced into the venue(court, hearing room, whatever), will be sworn in, and then will give testimony, and that testimony may be the fount of a civil or criminal case, as the NTSB lawyers decide. As professional pilots on here, be careful who you choose to hang out to dry. Someday, it may be your turn on the clothesline.

[Chiefttp]

ATC Watcher,
How can you say “Good Luck saved the Day” in the Austin incident. The FedEx crew had good situational awareness and knew the Southwest crew was cleared on to the runway. That’s why he asked if he was still cleared to land. Then when Tower queried the Southwest crew if they had started their takeoff roll, and Southwest didn’t respond, FedEx initiated his go around while asking SW to abort. So in my mind FedEx had very good situational awareness, not so much for SW and ZERO for the Tower controller. If luck was involved it was how both jets , on the same runway heading didn’t converge vertically as SW took off on climb out, and FedEx was climbing out on the go-around. The other 800lb gorilla in this story is the Tower Controllers employment qualifications and prior record. There are rumblings under the surface.

[ATC Watcher]

ATC Watcher,
How can you say “Good Luck saved the Day” in the Austin incident. The FedEx crew had good situational awareness .
Once again , Fedex initiated the go around by himself was a good move but there after all SA was lost to everyone. . Fedex commanding SW to abort was most probbaly undestood by the controller as coming from the SW hence the clearance to turn right when able, and not giving any avoid instructions to the Fedex who lost sight of the SW below him as he pitched up and the SW could not see where the Fedex was as it was coming from behind. No one ordered a side step by the Fedex so collsion was avoided by the diffrent rates of climb. (All this if the ADS-B and minimum separations given are validated of course ).So yes : no-one was taking care of separation anymore, and luck did save the day ..
The other 800lb gorilla in this story is the Tower Controllers employment qualifications and prior record. There are rumblings under the surface.
You mean those on the .65 web site ? I would not touch that site with a bargepole to do an investigation ;
This is the first post in there accusing the controller :Rumor has it the controller in question faked a hate crime at his old facility which he then leveraged to get into his new facility and possibly the reason he certified as well. This is a rumor I got from the internet and does not necessarily represent fact..
Which of course it becomes and slandering the guy continues in a harder tone afterwards, mentioning his skin color and even his supposed religion . Internet at its best.

[FlightDetent]

Since WR double posted his piece #163 https://www.pprune.org/accidents-close-calls/651189-ntsb-probe-fedex-southwest-close-encounter-austin-11.html#post11384585 (https://www.pprune.org/accidents-close-calls/650801-aa-106-jfk-13-jan-23-a.html#post11384577) identical in the AUS thread, my reply there: This last one turned out as a non-accident due to pure coincidence. And the incident happened without really breaching much if anything at all.

Chiefttp IMO what stopped that accident from unfolding was one crew's cognitive performance exercised well in an area they were not supposed to be visiting as per their job requirement or training. Thus not a layer nor backup.

As an example from a computer chair, my first gut reaction was I may had taken a self-authorized break left 90 deg, staying at 2000', and not attempt to control the other plane to safety. The G/A is a layer and this time it did not resolve the situation satisfactorily.

[KRviator]

When did the rules change about testimony to the NTSB cannot be used in a subsequent civil or criminal action against the crewmember involved? What are the crew scared of?

Hell, Australia is a third-world backwater in terms of civil aviation and even we have that protection down here, to ensure you can speak frankly without fear of being pineappled for your testimony.

[WillowRun 6-3]

Reporting by Wall Street Journal today (Feb. 15, evening, on website) states that the American pilots will comply with the subpoenas issued by NTSB and appear for interviews.

[Compton3fox]

I think the one issue with the controller that appears to be a foul was that it appears (from the recording) that he didn't challenge the lack of runway assignment in the crew readback.


Given the lack a definite departure runway as per previous comments at JFK, ensuring a readback of the Departure runway seems critical. Of course that takes up airtime but never the less, it should always be required.


Isn't the key bit here "Cross runway 31L"? Even at the little puddle jumper airports I fly into there are big red signs telling you what the ahead runway is and the holding point.

Surely you would want to visually confirm the identity of any active runway you are about to enter?
Having read the whole thread, i suspect the PF thought they were departing on 31L so was expecting to be cleared to cross 4L and that's where Conformation or expectation bias stepped in. Hence he expected to see 4L, which he did, so he crossed as per the clearance he thought he had. Ironically, being JFK based & very familiar with JFK may make this *more* likely as once he had the departure runway set in his mind as 31L, he knew how to get there in his sleep! If this was an unfamiliar airport to the crew, they would very likely be paying far more conscious attention. My made up term but u get the idea...

Did someone check the runway was clear before crossing? I don't know but that would be a question to ask as I make a habit of checking every runway before I cross. Have I ever not done this? I don't know but I suspect I have based on what I know about how the brian works...

[aa73]

Unconfirmed report from the Delta crew that the 777 had very minimal exterior lights on during the runway crossing, most likely only the taxi light.

Not sure how it is at your airline - but AA has historically never required all exterior lights to be on when crossing runways. It was never really trained or mandated in our manuals, and until only recently it became a “suggestion” during low visibility and/or night conditions. “Suggestion” - not “requirement.” That has always been shocking to me. Personally as a 737 CA I switch on my lights (landing and runway turnoff) every time when crossing and if the FO does not switch on the strobes/wing illumination I turn those on too. About half of them do it, mainly due to it having been a requirement at their previous airline, not because AA trained them to do it.

Compare this to the Delta flight ops manual which REQUIRES all exterior lights to be on when crossing runways - every time no matter the conditions.

Moral of all this, illuminate your aircraft anytime crossing runways for safety.

[punkalouver]

Given the lack a definite departure runway as per previous comments at JFK, ensuring a readback of the Departure runway seems critical. Of course that takes up airtime but never the less, it should always be required.



Having read the whole thread, i suspect the PF thought they were departing on 31L so was expecting to be cleared to cross 4L and that's where Conformation or expectation bias stepped in. Hence he expected to see 4L, which he did, so he crossed as per the clearance he thought he had. Ironically, being JFK based & very familiar with JFK may make this *more* likely as once he had the departure runway set in his mind as 31L,
..

Shouldn't matter if done in a manner designed to minimize incidents like this. That would be: The controller says "American 106 cross 31 Left". The pilot on the radio reads back "Cross 31 Left American 106". The pilot taxiing says out loud "Cleared to cross 31L" and looks at the hold short sign or painted markings that have the runway number and says "31L cleared to cross, clear left". If they find themselves saying 4 Left so soon after saying 31 Left, it should trigger doubt and the aircraft does not proceed until confirmation that they are cleared to cross 4L. At first, they will think that ATC made the error and just want confirmation to cross 4L. After some back and forth with ATC, it will eventually become apparent to the AA pilots that they had become confused and eventually things get figured out. But everything is figured out safely because the pilots never moved beyond the 4L sign because they never heard the words "cleared to cross 4 Left".

If you freeze this video at 13:52, you will notice that you can see the hold short painting markings and the runway signs even when very close to the hold short line. Think of a clearance to cross the runway as a clearance to move beyond paint markings/signs of the same number.

https://www.youtube.com/watch?v=3YyScI9vSUc

[WillowRun 6-3]

Presumably this incident will receive a good deal of attention next week, on Wednesday 15 March, when FAA convenes a specially arranged safety summit.

The website of a prominent aviation law firm (with which I'm not affiliated at all) posted a link for livestream of the summit.
"The FAA’s March 15 safety summit is being live streamed: https://bit.ly/3muh30Z" (https://bit.ly/3muh30Z)

(I'm posting this item on this particular thread, because of all the recent incidents, it was the one at JFK that produced some difficulty, if not drama, for some of the aviators involved - referring to the short-lived dispute over the format of interviews with NTSB investigators.)

[Compton3fox]

Shouldn't matter if done in a manner designed to minimize incidents like this. That would be: The controller says "American 106 cross 31 Left". The pilot on the radio reads back "Cross 31 Left American 106". The pilot taxiing says out loud "Cleared to cross 31L" and looks at the hold short sign or painted markings that have the runway number and says "31L cleared to cross, clear left". If they find themselves saying 4 Left so soon after saying 31 Left, it should trigger doubt and the aircraft does not proceed until confirmation that they are cleared to cross 4L. At first, they will think that ATC made the error and just want confirmation to cross 4L. After some back and forth with ATC, it will eventually become apparent to the AA pilots that they had become confused and eventually things get figured out. But everything is figured out safely because the pilots never moved beyond the 4L sign because they never heard the words "cleared to cross 4 Left".

If you freeze this video at 13:52, you will notice that you can see the hold short painting markings and the runway signs even when very close to the hold short line. Think of a clearance to cross the runway as a clearance to move beyond paint markings/signs of the same number.

https://www.youtube.com/watch?v=3YyScI9vSUc
I still think knowing your departure runway before taxi and reading it back is not to be underestimated. I don't know for sure in this case but it seems likely the PF *thought* they were departing 31L, hence my answer to the previous poster. This may explain why he thought he was cleared to cross 4L. I am not convinced your process would solve this as it doesn't address the bias that may override what seems like a solid approach. After all, the actual clearance to cross was clear but was not digested fully by either crew. *my assumption based on what happened next* I am theorising why. You may disagree but we for sure need understand and account for its cause in whatever solutions are put forward.

[punkalouver]

I still think knowing your departure runway before taxi and reading it back is not to be underestimated. I don't know for sure in this case but it seems likely the PF *thought* they were departing 31L, hence my answer to the previous poster. This may explain why he thought he was cleared to cross 4L. I am not convinced your process would solve this as it doesn't address the bias that may override what seems like a solid approach. After all, the actual clearance to cross was clear but was not digested fully by either crew. *my assumption based on what happened next* I am theorising why. You may disagree but we for sure need understand and account for its cause in whatever solutions are put forward.

Knowing your departure runway helps a lot. But there can be other situations such as unknowing becoming lost at an airport can lead to an improper runway crossing. Ensuring one properly identifies a runway that they are cleared to cross will prevent most improper crossings.

We visually verify the proper runway that we takeoff from, using markings/signs. If we see that the markings/signs do not match up with the clearance(and the runway we were planning to use), we do not takeoff. The same thing should apply while taxiing. If we see that the runway markings/signs does not match up with what we were cleared to cross, do not cross.

[Compton3fox]

Knowing your departure runway helps a lot. But there can be other situations such as unknowing becoming lost at an airport can lead to an improper runway crossing. Ensuring one properly identifies a runway that they are cleared to cross will prevent most improper crossings.

We visually verify the proper runway that we takeoff from, using markings/signs. If we see that the markings/signs do not match up with the clearance(and the runway we were planning to use), we do not takeoff. The same thing should apply while taxiing. If we see that the runway markings/signs does not match up with what we were cleared to cross, do not cross.

I argee but despite all what has been stated, it didn't work in this case and there are many other examples too. So unless we understand why and use this understanding to take whatever action is needed, it will continue to happen.
I just listened to the audio again and the AA PFs question to JFK TWR after the possible pilot deviation message was: "The last clearance we were given, we were cleared to cross. Is that correct?" He seems to be validating they were cleared to cross. He didn't question which runway they were cleared to cross. I think this is significant as it potentially implies he was still unaware they were in the wrong place. IE he genuinely believed he had crossed the right runway and was seeking Conformation that there were cleared to cross. Based on the way he phrases the question, it implies he was pretty sure they were cleared. IE Question asked in the positive sense...
When the tower replies, without commenting on what the clearance was but did say "We're departing Rwy 4L you were supposed to be departing Rwy 4L... you are currently holding short rwy 31L" Only at this point does it appear to dawn on tha AA PF that something it not right. Or so it would appear from the tone and content of the reply.. "Err.. Roger. AA106 heavy"
The tower controller just stated the facts. Which runway was being used for departure and where AA106 was located. This seems to be a text book response. No judgement was passed.

The other question I and others asked is why didnt they see the Delta Jet already moving on 4L when they looked before crossing? Or did no one on the FD look... or..? This seems to be the last chance of saving the day before the incursion but it didn't. We need to find out why, if at all possible.

[punkalouver]

I argee but despite all what has been stated, it didn't work in this case and there are many other examples too. So unless we understand why and use this understanding to take whatever action is needed, it will continue to happen.
I just listened to the audio again and the AA PFs question to JFK TWR after the possible pilot deviation message was: "The last clearance we were given, we were cleared to cross. Is that correct?" He seems to be validating they were cleared to cross. He didn't question which runway they were cleared to cross. I think this is significant as it potentially implies he was still unaware they were in the wrong place. IE he genuinely believed he had crossed the right runway and was seeking Conformation that there were cleared to cross. Based on the way he phrases the question, it implies he was pretty sure they were cleared. IE Question asked in the positive sense...
When the tower replies, without commenting on what the clearance was but did say "We're departing Rwy 4L you were supposed to be departing Rwy 4L... you are currently holding short rwy 31L" Only at this point does it appear to dawn on tha AA PF that something it not right. Or so it would appear from the tone and content of the reply.. "Err.. Roger. AA106 heavy"
The tower controller just stated the facts. Which runway was being used for departure and where AA106 was located. This seems to be a text book response. No judgement was passed.

The other question I and others asked is why didnt they see the Delta Jet already moving on 4L when they looked before crossing? Or did no one on the FD look... or..? This seems to be the last chance of saving the day before the incursion but it didn't. We need to find out why, if at all possible.

When I see an incident like this(and many others), I understand that I am the kind of pilot who could definitely make the same mistake. Therefore, I try to figure out a procedure that is most likely to prevent it from happening again.

I like to share my solution on what can be done to prevent a situation like this. I guarantee you, they didn’t use my method. Those who do use my method will most likely prevent what happened that night in JFK.

We know why they crossed the wrong runway. Because they didn’t confirm that the runway they were cleared to cross was the runway they were crossing. If one already has a personal procedure to ensure that the the runway they just heard a clearance to cross matches the hold short markings and to always question any discrepancy, it will go a long way toward preventing this sort of error.

The similar procedure that is now widely used to prevent wrong runway/taxiway takeoffs seems to work well as we rarely hear about it anymore.

What the pilots said after the incident is not really relevant except to prove what we already knew…..that they were confused.

In a certain way, it really doesn’t matter why they were confused(in terms of the validity of this procedure). The procedure is based on the reality that pilots become confused sometimes for a variety of reasons and this is the safety net.

[Compton3fox]

When I see an incident like this(and many others), I understand that I am the kind of pilot who could definitely make the same mistake. Therefore, I try to figure out a procedure that is most likely to prevent it from happening again.

I like to share my solution on what can be done to prevent a situation like this. I guarantee you, they didn’t use my method. Those who do use my method will most likely prevent what happened that night in JFK.

We know why they crossed the wrong runway. Because they didn’t confirm that the runway they were cleared to cross was the runway they were crossing. If one already has a personal procedure to ensure that the the runway they just heard a clearance to cross matches the hold short markings and to always question any discrepancy, it will go a long way toward preventing this sort of error.

The similar procedure that is now widely used to prevent wrong runway/taxiway takeoffs seems to work well as we rarely hear about it anymore.

What the pilots said after the incident is not really relevant except to prove what we already knew…..that they were confused.

In a certain way, it really doesn’t matter why they were confused(in terms of the validity of this procedure). The procedure is based on the reality that pilots become confused sometimes for a variety of reasons and this is the safety net.
I think confused is the wrong word here. That implies they weren't sure what was happening and/or where they were going etc. I think the PF was sure about where he was going, he was simply wrong. That's very different from being confused. In these situations, no matter how good a process is, the brain is very good at convincing you, you are doing the right thing and will wash over the best of processes.

I agree a good solid process, like the one you described, helps reduce deviations and provides a safety net and although we don't know what process was being followed that night, I agree it likely wasn't yours but even yours will get circumvented at some point. If say it's too complex, steps will be skipped. If it's not validated, it may hold flaws we don't know about etc..
Good discussion BTW.

[Lake1952]

AA 106 thought they were taking off on 31L despite being cleared to 4L and from that point on, classic confirmation bias took over. Confirmation bias is pervasive in much of human endeavors....politics, investing, medicine and certainly aviation.

One of the most extreme examples of confirmation bias in aviation was the Comair CRJ crew at LEX. They were so certain that they were on the right runway, and they hurled down a 3000 foot unlit runway, passing the correct lit runway on the way. Tenerife was also largely due to this bias as well.

[Compton3fox]

AA 106 thought they were taking off on 31L despite being cleared to 4L and from that point on, classic confirmation bias took over. Confirmation bias is pervasive in much of human endeavors....politics, investing, medicine and certainly aviation.

One of the most extreme examples of confirmation bias in aviation was the Comair CRJ crew at LEX. They were so certain that they were on the right runway, and they hurled down a 3000 foot unlit runway, passing the correct lit runway on the way. Tenerife was also largely due to this bias as well.
Exactly and this is supported if you listen to the exchanges between AA and TWR.

[aa73]

One of the most extreme examples of confirmation bias in aviation was the Comair CRJ crew at LEX. They were so certain that they were on the right runway, and they hurled down a 3000 foot unlit runway, passing the correct lit runway on the way. Tenerife was also largely due to this bias as well.

Actually the Comair crew never crossed their departure runway (22).. they never even reached it: 26 was on the way, and the one they lined up on.

[Lake1952]

Actually the Comair crew never crossed their departure runway (22).. they never even reached it: 26 was on the way, and the one they lined up on.

Sorry, but on their unfortunate takeoff roll down the unlit 26, they crossed the lit runway 22 used by air carriers. I didn't mean they crossed 22 on their taxi to 26, but they did cross 22 on their roll.

I remember a discussion about the accident and how powerful the power of confirmation bias can be. When was the last time any of you air carrrier pilots have had to take off on an unlit runway a half hour or so before sunrise without being NOTAMed to death about the lights being OTS? And after just a few seconds into their roll, a long lit runway goes by...

[aa73]

Sorry, but on their unfortunate takeoff roll down the unlit 26, they crossed the lit runway 22 used by air carriers. I didn't mean they crossed 22 on their taxi to 26, but they did cross 22 on their roll.

I remember a discussion about the accident and how powerful the power of confirmation bias can be. When was the last time any of you air carrrier pilots have had to take off on an unlit runway a half hour or so before sunrise without being NOTAMed to death about the lights being OTS? And after just a few seconds into their roll, a long lit runway goes by...

Ahh, gotcha, sorry I thought you meant in the taxi out

regardless.. due to their strong bias as you mentioned, prob no chance they would have recognized their error that far into their takeoff roll as they crossed the lit (correct) runway.

[Lake1952]

Ahh, gotcha, sorry I thought you meant in the taxi out

regardless.. due to their strong bias as you mentioned, prob no chance they would have recognized their error that far into their takeoff roll as they crossed the lit (correct) runway.

But having rolled onto an unlit 3500 foot runway in the pre-dawn darkness, Comair could have still saved the day with a simple, "Tower, lights please."

[MarkerInbound]

The Comair crew even commented “That’s weird with no lights” at the start of their takeoff. FAR 139 requires runways used for air carrier operations to be lit so I’m fairly certain that was their first dark runway airline takeoff. But they were so locked into thinking they were doing it right that warning signs like a dark runway were overlooked.

[punkalouver]

The Comair crew even commented “That’s weird with no lights” at the start of their takeoff. FAR 139 requires runways used for air carrier operations to be lit so I’m fairly certain that was their first dark runway airline takeoff. But they were so locked into thinking they were doing it right that warning signs like a dark runway were overlooked.
Verify the runway sign(or painted number) matches the runway you are planning to depart from.

[Lake1952]

But in the case of AA 106, they thought they were departing on 31L despite clear instructions to taxi to 4L. And the bizjet crew in the BOS incident believed they were cleared for takeoff. Tackling confirmation bias is a difficult task which must include CRM.

[punkalouver]

But in the case of AA 106, they thought they were departing on 31L despite clear instructions to taxi to 4L.
Already discussed on how to prevent an incursion in such a case, if you go back about ten days in the thread

[Compton3fox]

Already discussed on how to prevent an incursion in such a case, if you go back about ten days in the thread
The power of bias is shown in these examples and I know you think your process will solve the issue but i didn't see everyone agreeing. It has been discussed but not concluded. It has been shown that the brain will, in some circumstances, see, hear or feel what it expects to, not always what is actually there. Or in the comair case, see what is there, question it but still proceed. I don't know the answers but I don't think just adding more cross checks alone will overcome bias.

[Lake1952]

In the late 70s and 80s, the concept of CRM expanded enormously in response to several well publicized accidents. It remains an important theme in the culture of aviation safety to this day. In many ways, the insidious nature of confirmation bias relates to CRM. But aviation needs to now shine an intense spotlight on confirmation bias and figure out what to do about it.

Another piece of the JFK puzzle I have not heard brought up...how many clues and cues did the crew of AA 106 ignore on their taxi before crossing 4L? How many takeoffs on 4L did they have a chance to see pass before them before the DL flight? Assuming that 4R was the main arrival runway, how many landings did they have a chance to see? Was there a "string of pearls " in the sky to their right perpendicular to their taxi route? Or was the crew thoroughly busy with their checklists. Did they notice that while taxiing parallel to 31L for a mile or more, there wasn't a single takeoff on that runway?

Obviously this is not a easy problem to solve... there's simply not enough frequency time for everything to be be verified ad infinitum.

[punkalouver]

The power of bias is shown in these examples and I know you think your process will solve the issue but i didn't see everyone agreeing. It has been discussed but not concluded. It has been shown that the brain will, in some circumstances, see, hear or feel what it expects to, not always what is actually there. Or in the comair case, see what is there, question it but still proceed. I don't know the answers but I don't think just adding more cross checks alone will overcome bias.
Simple cross checks can overcome these biases, especially if one has already thought to themself that they will not cross a runway until two variables align.....runway crossing clearance and runway sign. Think of it as a kind of two-factor authentication.

That is why most airlines now have their pilots independently crosscheck the runway number for departure. I believe that this is a direct result of the Comair accident.

There are all kinds of simple cross checks we can do. Check that the flight plan that we received has the proper flight number(along with things such as the weight and balance form. Check that the logbook on board is the proper one for the aircraft, checking the proper frequency before transmitting, checking that the MEL listed on the paperwork is the same MEL number written in the logbook, check that there is no ground equipment that would create a hazard is at the gate area prior to taxiing in an focusing on marshalling signals etc.

Many of these things are not written down as procedures, they are just good practice.

I certainly think that taking approximately one second to look at a runway sign is a better solution than saying that one doesn't know the answers and simply accepting that it will not be possible to overcome biases.

[Compton3fox]

In the late 70s and 80s, the concept of CRM expanded enormously in response to several well publicized accidents. It remains an important theme in the culture of aviation safety to this day. In many ways, the insidious nature of confirmation bias relates to CRM. But aviation needs to now shine an intense spotlight on confirmation bias and figure out what to do about it.

Another piece of the JFK puzzle I have not heard brought up...how many clues and cues did the crew of AA 106 ignore on their taxi before crossing 4L? How many takeoffs on 4L did they have a chance to see pass before them before the DL flight? Assuming that 4R was the main arrival runway, how many landings did they have a chance to see? Was there a "string of pearls " in the sky to their right perpendicular to their taxi route? Or was the crew thoroughly busy with their checklists. Did they notice that while taxiing parallel to 31L for a mile or more, there wasn't a single takeoff on that runway?

Obviously this is not a easy problem to solve... there's simply not enough frequency time for everything to be be verified ad infinitum.
Very important questions! Also, was 31L being used for departures too? At Kilo Alpha I think? No.matter how many clues, bias is very powerfull and subconscious, so not easy to spot by the conscious mind... Agree it will difficult to solve but only by examination in great detail and ensuring we have the ability to do that examination (availability of CVR for example) will we succeed.

[Lake1952]

Very important questions! Also, was 31L being used for departures too? At Kilo Alpha I think? No.matter how many clues, bias is very powerfull and subconscious, so not easy to spot by the conscious mind... Agree it will difficult to solve but only by examination in great detail and ensuring we have the ability to do that examination (availability of CVR for example) will we succeed.

The CVR on AA106 flew to London that night, rendering it useless.

[WillowRun 6-3]

FAA issued a SAFO (Safety Alert fir Operators) by its Flight Standards Service - addressing the several incidents recently.

The SAFO is linked in this news release by FAA, as of March 22, 2023.
https://www.faa.gov/newsroom/faa-urges-airlines-pilots-and-others-take-key-safety-actions

[WillowRun 6-3]

http://www.faa.gov/other_visit/aviation_industry/airline_operators/airline_safety/safo
A SAFO contains important safety information and may include recommended action. Besides the specific action recommended in a SAFO, an alternative action may be as effective in addressing the safety issue named in the SAFO. The contents of this document do not have the force and effect of law and are not meant to bind the public in any way. This document is intended only to provide clarity to the public regarding existing requirements under the law or agency policies.

Subject: Aviation Safety Call to Action.
Purpose: This SAFO informs the aviation community that several highly visible and notable recent events demonstrate the need for continued vigilance and attention to mitigation of safety risks. This SAFO applies to all aircraft operations conducted under Title 14 of the Code of Federal Regulations (14 CFR) Parts 91, 91 subpart K (Part 91K), 121, 125, 129, and 135.

Background: In recent months, a number of notable and high visibility events have occurred in the National Airspace System (NAS). While the overall numbers do not reflect an increase in incidents and occurrences, the potential severity of these events is concerning. Six serious runway incursions have occurred since January 2023, including an incident at John F. Kennedy International Airport in New York City involving a taxiing aircraft narrowly avoiding a departing aircraft and a landing aircraft coming within 100 feet of a departing aircraft at Austin-Bergstrom International Airport in Texas.
In February 2023, the acting Administrator of the Federal Aviation Administration (FAA) issued a Call to Action comprised of a series of events and initiatives to ensure focus and attention on risks to the aviation system. Senior leaders from the FAA, other government agencies, commercial and general aviation operators, labor partners and others attended a Safety Summit on March 15, 2023. Attendees discussed these recent incidents, as well as ways to enhance aviation safety.

Discussion: Safety management requires adapting to continual change. Effective safety management is designed to detect emerging safety issues, assess the level of risk and address those risks through mitigations. Those mitigations may be a change in processes, procedures or training. Operators should evaluate information collected through their safety management processes, identify hazards, increase and improve safety communications with employees and enact mitigations. Safety management systems,
policies and procedures must be able to account for a high rate of change.

Recommended Action: Recent events have highlighted several areas of focus. Directors of Operations, Chief Pilots, Directors of Training, Check Airmen, Directors of Safety, Program Managers, Pilots and Operators should review the following items and consider taking additional steps to ensure operations are conducted at the highest level of safety, including changes to procedures or training, if appropriate.

• Use all available internal communication processes to specifically highlight recent events and existing issues.
• Reinforce adherence to published processes and procedures, including checklists, Air Traffic
Control instructions, and internal company procedures.
• Ensure pilots and flight attendants have the same understanding of what “sterile flight deck” means and the risks associated with extraneous communication during this time.
• Explore helpful resources available for all pilots at: FAA Safety Team - FAASafety.gov.
• Encourage crews to diligently follow Crew Resource Management (CRM) procedures and principles to control workload and reduce distractions.
• Encourage personnel to identify and report existing and emerging safety issues through voluntary reporting programs and understand the usefulness of the voluntary reporting system.
• Review information about runway safety here: Runway Safety | Federal Aviation Administration
(faa.gov).
• Review the following previously published SAFOs:
o17012, High Collision Risk During Runway Crossing;
o11004, Runway Incursion Prevention Actions;
o08001, Flightcrew techniques and procedures to enhance taxi, pre-takeoff, and after landing
safety to reduce the risk of runway incursions.
• Apply Safety Management System principles to analyze safety data and assess risk associated with emerging hazards. Evaluate existing risk mitigations to determine if they are effectively controlling risk, or if additional action is required.
Contact: Questions or comments regarding this SAFO should be directed to the Air Transportation
Division at [email protected] or the General Aviation & Commercial Division at 9-
[email protected].

[Lake1952]

Can I ask a question... the AA 106 crew refused to be interviewed on multiple occasions if the interview was going to be recorded in any way. What possible excuse is there for their adamant refusal to be recorded? What rules and regs are in place that might require cooperation with an FAA or NTSB inquiry?

I have other questions about why that flight was allowed to complete its flight to LHR? What was emotional state of the flight crew? They were told to make a phone call. Who had the authority to tell that crew to return to their terminal? The tower supervisor? Obviously the event took on a life if its own by the time they arrived in London. What happened to that flight crew after reaching the UK? Did they crew a return flight two days later as they would have done on an ordinary rotation?

[WillowRun 6-3]

Maybe that's more than one question....

Based on several statements and other information publicly available at the time of the interview controversy, and despite holding only SLF/attorney status here, I'll take a swing at part of this. The primary basis articulated by the labor union representing the AA aviators was that recording of interviews imposes an unwarranted and unnecessary atmosphere of formality and even an adversarial posture, whereas the interview is supposed to be more aligned with purely investigatory factual inquiry and reflective of a "just culture" approach., (This is only my description of the APA's position and I'm not quoting anything or anyone.)

My understanding is that there was an airline labor-management issue that was part of the context, or background or backstory - or at the very least, the issue very likely could have been part of the context. IIRC, the APA had objected to some changes in flight deck procedures being implemented - perhaps "imposed" would be the correct word - by management without participation by and/or consultation with APA that the union understood to be required by the collectively bargained labor agreement. The issue concerned how the airline management requires the PM to deal with checklist items in the time period immediately prior to departure, including taxiing to position for takeoff. I apologize for imprecise terminology if I've used any here - although I'm pretty certain APA had stated concern over a change management had imposed about the specific workload and task items to be completed by the PM during that time period.

The change had been imposed just a short time before the incident The linkage is that some reporting at the time indicated that the PM of the incident flight had been focused on getting these new procedures done correctly, to the detriment of what otherwise would have been more complete and/or intense focus on clearances for taxi and takeoff. (I realize the thread has included comments on the tasks to be completed by the PF, and it could be that APA's actual interactions with management had included discussion of how the new procedures would affect not just the F/O (in this case) but the PIC as well.)

If this much is correct, then in addition to a general sense that the incident was ripe for finding a bus, grabbing a couple of pilots, and tossing them under said bus, the union also was concerned with not compromising its issue over the implementation of the new procedures - and also guarding against its position being co-opted by the inevitable attention-payers swarming in - competent press and media; the rest of the press and media; the Congress, et cetera. (My observation of a "general sense" of blame looming over this incident relates back, and it relates all the way back, to the way the Executive Branch has run the Department of Transportation and the FAA, but let's not get political or anything.)

As to how and why the crew continued on to LHR and the rest, I can't say anything useful.

[Chiefttp]

Any updates or reports on this incident? It’s been over a year.

[BFSGrad]

NTSB Docket (https://data.ntsb.gov/Docket?ProjectID=106577)

[Nfield750]

If the report is true - Its a bit naughty for the AA flight crew not to report the runway incursion and deliberately allow the voice recording to overwrite. I thought that civil aviation had got over the blame culture and was willing to admit and learn from mistakes, which is how safety is improved. At least in this chain of events, the controller was able to intervene before the last resort of the 737 crew seeing the 777 on the runway, possibly too late

Investigators detail how an American Airlines jet crossed a runway in front of a Delta plane at JFK (msn.com) (https://www.msn.com/en-gb/travel/news/investigators-detail-how-an-american-airlines-jet-crossed-a-runway-in-front-of-a-delta-plane-at-jfk/ar-BB1hs1zX?ocid=msedgntp&pc=DCTS&cvid=9aff9206d5364583a4b31b7d721a1094&ei=32#comments)

[ATC Watcher]

I must admit I do not have the courage to read the 294 pages of the crew interviews on the Docket, but reading the msn article resume , I find it noteworthy to read that 2 of the 3 crew on that aircraft apparently admitted they did not know where they were on the airport. during taxi. There might be reasons for this distraction or lack of monitoring but this needs serious addressing by the airline I would say.

[Request Orbit]

Having read them, all three crew had a slightly different immediate reaction to being on the runway, but all independently seemed to think the “red glow” from the RELs didn’t illuminate until the cockpit was already over the runway.

Interesting difference in perspective between how close the AAL pilots and the ATC staff perceived the event to be, even with a bit of hindsight.

[BFSGrad]

I must admit I do not have the courage to read the 294 pages of the crew interviews on the Docket, but reading the msn article resume , I find it noteworthy to read that 2 of the 3 crew on that aircraft apparently admitted they did not know where they were on the airport. during taxi. There might be reasons for this distraction or lack of monitoring but this needs serious addressing by the airline I would say.
No courage required, just time. Both pilot and ATC interviews worth the read. Please don’t trust the typically abysmal media reporting to form an opinion.

But regarding 2 of 3 pilots being somewhat out of the loop, correct. CA inexplicably reverted to original departure/taxi brief for 31L even though he acknowledged being aware of and briefing the revised 4L departure. IOW, he was driving the aircraft where he intended and believed he was cleared to go. FO was head down at a critical point (B/J/K intersection) doing admin and believed aircraft was crossing 31L as authorized when she went head up. FB was head down doing admin from mid-B until actually crossing 4L. He was also aware 4L was the intended departure runway.

Troubling inconsistency between pilot interviews and performance study regarding the runway entrance lights (REL). All pilots said that, by the time RELs illuminated, aircraft was already midpoint 4L such that they couldn’t directly see the RELs and could only see the red glow off to the side of the aircraft. Performance study states (p. 17) that RELs illuminated as the aircraft crossed the 4L hold bar. Pilots should have been able to directly see RELs #3-6 and possibly #2. Only REL #1 would have definitely been out of view.

Also wondering about exterior lighting. Appears AA was operating exterior lighting consistent with company procedures, which gives the pilot the option of when to use logo lights. Do 121 pilots turn on logo lights when crossing runways? No mention of this in the interviews. Seems a logo light (if this aircraft was so equipped) would have given the DL pilots the best opportunity to see the crossing AA. Based on interviews, the DL did not see the crossing AA until after the RTO had been initiated. IOW, the RTO was initiated based on ATC instruction instead of observing the crossing AA.

[DIBO]

Because the incident wasn't ATC related, I had a rather quick reading through the most interesting parts of the ATC interviews and from one of the interviews I copied a few text-lines that were rather ... interesting :uhoh:

about ASDE-X:
* So the ASDE and me don’t have a good relationship because it’s as far as I’m concerned, it’s totally hit or miss.
* There’s a flashing thing that’s broken, that’s been broken, which I’ve reported probably 100 times on the ASDE forms. You’ll find them with my initials on them
* I’ve seen - without getting into storytelling, I watched an entire emergency response team line up on the runway with an aircraft on final, not show up on the ASDE and the ASDE not go off. I watched five firetrucks line up on 22 Right with an aircraft while we were using 22 Right, thinking they were on Yankee and the ASDE didn’t go off and the local control had to send the aircraft around on 22 Right.
* Yes, so the MLAT will flash, and it says MLAT not working. Great. So I have reported that hundreds of times to the FAA, officially on paperwork, in writing, and I keep getting told it’s a Verizon issue. Now that means nothing to me. That’s your problem. I’m telling you that the equipment that you’re telling me I need to rely on that people don’t die, especially when we can’t see out the window
* -- doesn’t work properly. And I actually was yelled at by my manager for saying, on a relief briefing, that the ASDE is either out of service or not working properly. That was after I ran a runway sweep with a vehicle that the vehicle never showed up on the runway and the ASDE didn’t show them at all.
* So that’s the culture that goes on about the ASDE is that I’ve become more of a pain in the ass for constantly filling out this form to say that the ASDE doesn’t work where it goes downstairs and probably just goes into the garbage, because it’s a no, quote-unquote, known issue. They take it out of the logs, so it doesn’t cause FAA issues at wherever the FAA has to deal with that issue being on their log every day.

about pilots:
* No, I expect that all pilots will listen to nothing I say at all times, because that’s how they operate.
* They all switch, they do it all the time, it’s a common thing, they -- because they’re hotshots, and I’m just a, you know, a radio operator and they know everything, and they do this all the time. They switch frequencies when they’re not supposed to. They act like a bunch of amateurs all the time. Every day, all day dealing with having to say things. And let me tell you something. After COVID, it has been 10 times worse, and I’m sure that this is not a JFK issue, that this is a NAS problem. Ten times worse after COVID where you’re having to repeat simple, basic, taxiing instructions to every pilot from top to bottom, from seniors to, you know -- from Saturday pilots to international pilots and everyone in between. You’re having to say things four and five times, basic instructions to pilots.
* Because if you give them more than three things, their brain turns to mush and they forget everything, and then they say, say again, and -- or they say something wrong and you can’t work ground control, because you will be constantly talking to the same pilot three and four times. So if you don’t break it down into turn left Bravo, hold short of Kilo. Because if I gave a Bravo, Kilo, cross 31 Left, forget about it. She’d be -- you know, some of them can, obviously, some of them can't, but you can’t work that way because you’ll -- you’re just working uphill.

[blind pew]

It’s all well having a go at the flight crew but it can be a bloody difficult job especially setting priorities with the bull**** of paperwork, checklists and when they can be completed, fatigue, strange airfields and languages (I include American regional accents in that sans le Français ) without the ridiculous long duty days and jet lag.
The airlines want more and more blood out of the stone. In my day 650 flying hours left me knackered and as soon as I could afford it I took unpaid leave as a 40 year old..ok electronics have changed things and you don’t need the flying skills of 30 years ago but it is still a hands on job and airfields aren’t the easiest thing to navigate even in daylight. I’ve called for a follow me at home base for the local captain who got lost whilst I was doing the after landing checklist.
BA managed to take the wrong taxiway Twice in SA hitting a building the second time - that takes some doing.

[WillowRun 6-3]

It’s all well having a go at the flight crew but it can be a bloody difficult job especially setting priorities with the bull**** of paperwork, checklists and when they can be completed, fatigue, strange airfields and languages (I include American regional accents in that sans le Français ) without the ridiculous long duty days and jet lag.
The airlines want more and more blood out of the stone. In my day 650 flying hours left me knackered and as soon as I could afford it I took unpaid leave as a 40 year old..ok electronics have changed things and you don’t need the flying skills of 30 years ago but it is still a hands on job and airfields aren’t the easiest thing to navigate even in daylight. I’ve called for a follow me at home base for the local captain who got lost whilst I was doing the after landing checklist.
BA managed to take the wrong taxiway Twice in SA hitting a building the second time - that takes some doing.

The sesnse of "having a go at the flight crew" pervaded the initial phases of the aftermath of this incident. As is widely known, the union for American pilots - assessing only from the public statements at the time - had significant concern about the workload factor. And IIRC how the company had imposed changes on procedures without having engaged properly or sufficiently with the union about these changes first. The changes, in turn, were reported as having been a factor in why one or more of the pilots did not realize their taxiing mistake (the F/O was concentrating on getting through the procedures, as recently changed).

These concerns were, at least in part, the basis for the union challenging the process NTSB intended to follow for gathering information from the pilots (recording of interviews).

Not having worked through very much of the Docket, I don't know whether any statements from the union, about the dispute with the company about procedures being changed, or about the process for interviews, were provided to NTSB and are somewhere on the Docket. But even so, as an SLF/attorney (whose practice for many years focused on employment and labor law matters) I have to wonder what views are held inside the union and obviously not being publicized - or not publicized yet.