JAL incident at Haneda Airport

[waito]

Another Swiss Cheese Hole!

I did my best to replay that A350 flight on FR24
Position at first Tower Contact, that's 11.3NM out and already lined up on Final course!


Position when receiving landing clearance, 6NM out:




Assumed Position when the Dash-8 crossed the Hold Short Line on C5 (not really confirmed!). 2.7NM out.


That means the A350 flew a normal long Final and should have been visible all the time for the Dash-8 crew. Especially when crossing the Holding Point C5. If the Dash-8 crew believed to enter an active runway, they should have checked the approach path. And it must have been super easy to detect that large landing aircraft. Further out another A/C (JAL166) approaching.

[LOWI]

So it seems the Coastguard aircraft made the runway incursion.

The question still remains, why didn't the JAL see the aircraft on the runway?

In the sim we always practice low level go arounds due to runway incursions. An experienced crew on finals would not just be looking for an aircraft on the threshold but also any potential aircraft crossings further down the runway.

Plus listening out on ATC really shows good SA especially if aircraft are told to taxi and hold short. Sometimes my finger gets a bit too happy over the TOGA when an aircraft is told to hold short and I see they may potentially cross the hold short line.

[mike current]

Reports of Runway stop bars unserviceable since 27th December.
How does an airport like Haneda allow that to happen for nearly a week? In winter! Surely on site engineers at any airports would fix that as a matter of priority.

Does anyone know the runway safety systems available to RJTT ATC? I'd expect them to have a full A-SMGCS system with runway incursion monitoring and warning.

[markbingo]

Originally Posted by skiguy84

I can't see if anyone has discussed this yet, but do the JAL A350s have a tail camera and/or nosewheel camera? And if so, does anyone know if those feeds are recorded in the black box?

I flew a Finair A350 recently and it had those cameras installed, but I doubt they would keep the whole flight recorded, let alone in the black box. That would be a lot of data to store, especially on long haul (which I know this flight wasn't).

It does feel like it could be something for manufacturers to consider though (if they haven't already). Basically a "Dashcam" for aircraft.

[Yancey Slide]

Originally Posted by EDLB

position and hold (stop on the runway but no take off clearance),

I thought "position and hold" was replaced with "line up and wait" donkeys years ago now?

[OPENDOOR]

I was just wondering if anybody has actually heard what the tower said and if it was clearly intelligible.

[BFSGrad]

Originally Posted by hitchens97

In Europe, "Clear to Land" by ATC means the runway is clear and will remain clear until after you land, whereas in the US, it's still "buyer beware", and there may well be approved aircraft movements on and across the runway,, before you land.

In the U.S., a landing clearance is just like every other clearance; i.e., it is predictive. For example, if you are cleared to climb from 10,000 ft to FL250, it doesn’t mean that the point in space where you eventually will be is clear at the time the clearance is issued. Instead it means that point in space will be clear when you are predicted to arrive at that point.

While there has been much discussion about why JAL didn’t see CG or why CG didn’t see JAL, there’s little mention of why ATC didn’t ensure that (1) CG followed ATC direction, and (2) runway 34R remained clear for an aircraft granted clearance to land. From what has been reported so far, it’s not as if CG darted onto the runway just as JAL was touching down.

A clearance to land on a runway, regardless of when that clearance is issued, has little value if ATC doesn’t continue to actively manage that clearance as the aircraft approaches the clearance point.

Haneda tower is listed as the 6th tallest in the world at 380 ft and is sited about 1.25 statute miles from the threshold of runway 34R. Even at night, the tower controllers should have had an unobstructed, commanding view of the accident location.

[DIBO]

Originally Posted by OPENDOOR

I was just wondering if anybody has actually heard what the tower said and if it was clearly intelligible.

Given the correct readback of JA772A found in the transcripts, TWR transmissions must have been sufficiently intelligible.
The LiveATC recording extract can be found here but quality is very poor (certainly for my old ears )

Originally Posted by waito

17:45:11
(JA722A): TOWER JA722A C.
(Tokyo Tower): JA722A Tokyo TOWER Good evening, No.1, taxi to holding point C5
17:45:19
(JA722A): Taxi to holding point C5 JA722A No.1, Thank you.

[mike current]

Originally Posted by BFSGrad

In the U.S., a landing clearance is just like every other clearance; i.e., it is predictive. For example, if you are cleared to climb from 10,000 ft to FL250, it doesn’t mean that the point in space where you eventually will be is clear at the time the clearance is issued.

It should be. That's why we use headings. Any controller that ignores that principle will have an incident at some point.

15 years as an ATC instructor and examiner.
"I thought they'd pass wider than that"
"I thought they'd climb faster"

[MissChief]

"We have departure"

Originally Posted by waito

I OCR converted into text, rearranged and marked bold the passages of the two accident aircraft. Please somebody crosschecked for OCR errors
The last line is MY comment, that's the impact time. It coincides with the CCTV timestamp, I hope that's accurate

17:43:02

(JAL516): Tokyo TOWER JAL516 spot18.

(Tokyo TOWER): JAL516 Tokyo TOWER good evening RUNWAY 34R continue approach wind 320/7, we have departure



17:43:12

(JAL516): JAL516 continue approach 34R.



17:43:26

(DAL276): Tokyo TOWER DAL276 with you on C, proceeding to holding point 34R

(Tokyo TOWER): DAL276 Tokyo TOWER good evening. taxi to holding point C1.

(DAL276): Holding point C1, DAL276



17:44:56

(Tokyo Tower): JAL516 RUNWAY 34R cleared to land wind 310/8



17:45:01

(JAL516): Cleared to land RUNWAY 34R JAL516.



17:45:11

(JA722A): TOWER JA722A C.

(Tokyo Tower): JA722A Tokyo TOWER Good evening, No.1, taxi to holding point C5



17:45:19

(JA722A): Taxi to holding point C5 JA722A No.1, Thank you.



17:45:40

(JAL179): Tokyo TOWER JAL179 taxi to holding point C1.

(Tokyo Tower):JAL179 Tokyo TOWER good evening, No.3, taxi to holding point C1

(JAL179): Taxi to holding point C1, we are ready JAL179.



17:45:56

(JAL166): Tokyo TOWER JAL166 spot 21.

(Tokyo Tower):JAL166 Tokyo TOWER good evening, No.2, RUNWAY 34R continue approach wind 320/8, we have departure, reduce speed to 160 knots.



17:46:06

(JAL166): Reduce 160 knots RUNWAY 34R continue approach, JAL166 good evening



17:47:23

(Tokyo Tower):JAL166, reduce minimum approach speed

(JAL166): JAL166



17:47:27

(comment by waito: Impact Landing JAL516 A350 into waiting JA722A Dash-8)

Despite JAL 516 being informed "we have departure" and subsequently JAL 166 being told "we have departure", no departing aircraft was cleared to line up on 34R. And then JAL 516 was cleared to land, which in Japan means the runway is clear for the aircraft (JAL 516) to land.

I have never heard "we have departure", incidentally. "Departing traffic", or "Land after" are standard, I think.

It is conceivable that the Coastguard Dash-8 assumed (wrongly) that it was the departing aircraft, despite having been cleared to holding point C5, and having acknowledged this.

My thought is that this was a tragic error of the Coastguard Dash-8 crew, compounded by the fac that RHS on the Dash-8 had not visually checked that the approach was clear, which should also have been evident on TCAS, unless it had been switched to STBY in error when entering the runway. The JAL 516 crew may well not have spotted the incursion well down the runway, what with light glare and the likelihood that incursions are usually at the take-off threshold.

Full credit to the JAL516 crew, flight and cabin, for keeping the aircraft on the runway, and executing a successful evacuation in difficult circumstances.

But I do wonder about "we have departure". This is non-specific and can lead to confusion or wrong assumptions. The single "Go Around" instruction to JAL 166 also lacked urgency. "Go Around, I say again Go Around" would have been clearer, although I imagine the crew on JAL166 would have seen the ensuing fireballs quite easily.

Linguistic/communication issues at play here, I feel.

[DaveReidUK]

Originally Posted by MartinM

The Dash-8 for sure had a transponder, but FR24 cannot read all of them.

No, FR24 decoded all the transmissions from the Dash 8.

In fact you can download the data from their website, but there's little point as all it contains are 15 minutes' worth of repeated transmissions of ICAO24 (867BEE), Mode A squawk (7320) and Mode C altitude (0).

[Iron Duck]

Originally Posted by MissChief

Despite JAL 516 being informed "we have departure" and subsequently JAL 166 being told "we have departure", no departing aircraft was cleared to line up on 34R. And then JAL 516 was cleared to land, which in Japan means the runway is clear for the aircraft (JAL 516) to land.

Also a lack of curiosity on the part of JL516: being informed of a departure prior to their landing, but then no aircraft being instructed to depart before their landing clearance being issued.

[scifi]

Just looking at the relevant Notam. It is a page full of U/Ss, including the holding point C5 U/S, and the TW lighting U/S.

[OPENDOOR]

Originally Posted by DIBO

Given the correct readback of JA772A found in the transcripts, TWR transmissions must have been sufficiently intelligible.
The LiveATC recording extract can be found here but quality is very poor (certainly for my old ears )

Interesting, the link suggests the transmission was; "TWR: CostGuard772A Tokyo Tower, good evening number one taxi to hold abeam .... Charlie 5"

I haven't been able to open it but that is quite a bit different meaning to the transcript.

[physicus]

Based on the transcript, my mental picture as JAL516 is that JA722A who was cleared to taxi to C5 and hold would be taking off behind us, and before JAL166.

There is absolutely nothing in that transcript that would hint at any missed opportunity by JAL516 to pick up a hint of something going wrong.

The Swiss cheese tragedy is the late switch of JA722A to the TWR frequency *after* the landing clearance was given to JAL516.

Perhaps there's an opportunity to reconsider how far out a landing clearance should be given to lower the window duration during which someone is approaching with clearance to land expecting a clear runway.

[DIBO]

Originally Posted by OPENDOOR

Interesting, the link suggests the transmission was; "TWR: CostGuard772A Tokyo Tower, good evening number one taxi to hold abeam .... Charlie 5"

I haven't been able to open it but that is quite a bit different meaning to the transcript.

You've asked for the recording, that's what you've been directed to.
One may assume by now that the official ATC transcript, is the only reliable thing we have to go by...

And aahm.... there was a reason why I've include "but quality is very poor (certainly for my old ears)"

[xcris]

JA722A Tokyo TOWER Good evening, No.1, taxi to holding point C5
JA722A: Taxi to holding point C5 JA722A No.1, Thank you.

Sound&Clear. Nothing "ambiguous" here.
Or are we questioning the official transcript of the ATC recordings & maybe CVR with "evidence" from an enthusiast-run website?

[geopat]

Maybe a silly question, but how did the crew know to release the 3 safe chutes out of a possible 8 available.

It seems an unbelievable split second decision that saved many lifes.

[xcris]

Could be the "number one" [for departure] the "killer", triggering in pilots' minds the idea that they were entitled to go first, thus line-up once they arrived at the runway? Especially if they previously used to take-off in the same manner? I would be curious...
How many special-ops flights did this crew in that day and the previous? How did they develop?

[jumpseater]

Originally Posted by BFSGrad

I

A clearance to land on a runway, regardless of when that clearance is issued, has little value if ATC doesn’t continue to actively manage that clearance as the aircraft approaches the clearance point.

Haneda tower is listed as the 6th tallest in the world at 380 ft and is sited about 1.25 statute miles from the threshold of runway 34R. Even at night, the tower controllers should have had an unobstructed, commanding view of the accident location.

A runway visual scan is much like a cockpit instrument scan, things can be missed. At night the potential to miss stationary anticol lights on a lit runway the best part of 2km away on the far side of a well lit apron is to me understandable. There’s also other tasking for the ATCO, eyes inside as they phone co-ordinate with radar the 722 departure, watching other traffic under their control for example. So if you’ve spent time in a busy tower cab environment you can understand how this might slip through another hole.