Having read all of the posts so far ... sounds like the holes on a very large Swiss cheese almost all lined up together.
1. Parallel runway lights switched off, giving impression that taxiway is the parallel runway.
2. Visual approach for noise abatement with no auto tuning of the ILS
3. Pilots at low point of circadian rhythm and fatigued

Far from being a long way away from disaster, this look to be very close ....

I disagree with that, the last modern plane I used to fly (ejet) was tuning automatically frequency and inbound course with what inserted in FMS. It was happening around 30nm from destination.
But everybody in my compagny were tuning manualy inbound and ils/vor/loc, ... frequency in the standby frequency box during the briefing.

I don't know the airbus or boeing, but I like to set my approach aids myself and not just relied on FMS.
Just my 2 cents.

Finally when I do visual approach, I have always backup (ils, vor, fms ...) To check what I m doing.
I read too many article of plane landing wrong airport (Ryanair in Milan) or wrong runway ... I don't do "just" visual.

Of course they wouldn't know there was an airport there until after they moved in :rolleyes:

Long Haul:---thanks for the lesson mate.:) My A330 and 777 before that do indeed automatically tune the ILS but if they don't I do know how to myself and indeed do on occasion. We certainly check the whole NAV RAD during the approach briefing and setup before TOD and adjust as required.

As to HIAL being too bright and therefore distracting on a clear night......ok well ask ATC to turn them down a stage..........certainly it should have been on and it is a fantastic aid to find the runway in all weather down to CAT 2 ....Much better than a black hole to land on like I had in Jakarta 07L last month.....:ooh:

RTZL on or off? Way better if they are on ( at an appropriate level )

It's not hard for the PM ( and the PF higher up ) to scan the PFD on approach at 1,000' and 500' etc to monitor the LLZ/GS as well as the IAS/VS and then make appropriate trend callouts. ( as they already are required to do on all approaches anyway in my Airline )

Indeed it should be taught that if the info is displayed for your runway then check it.......for this very reason
Especially in places like SFO with close parallel runways and taxiways.

Flocks:---good practice that will serve you well.:ok:

CVR secured!?
 

I assume the CVR has been removed, if not we lack a critical tool to learn from this mess.
I challenge you FR24 experts to check if the aircraft departed as planed with in the morning.
If it did there is two cases:
1 The CVR pulled and new inserted and the inbound CVR on its way to NTSB.
2 The aircraft departed and the CVR was over written and we will never know what happened in the flight deck.

If the aircraft is grounded due to lack of CVR we are good.
Anyone?

Finally a balanced viewpoint instead of sensational nonsense...

CTV News Channel: 'Situation was unusual' | CTV News

Basic FR24 skills suffice:

No Visual Night!
 

Norwegian Chief Pilot just commented on this incident and stated that a visual night approach is prohibited in his company.
This is the case in most EU companys.
I am a big fan of visual approaches , but seldom do it as we have to do a separate brief.
What I do is call visual and intercept ILS or LNAV-VNAV path no later than platform alt with HDG/ VS +, VORLOC/GS armed .
All low risk and wasting 31 seconds...
Done plenty of yank and bank in the old days, seen to much and read to much to be the slightest interested to satisfy the local ATC agenda.
Love FAA and the safe execution milions of flights, but they are overdue for a big one.
Conveniently this one is on the Canadiens , but take notes.
Visual night approach to a big international airport, give me a f.... break.
Seriously.

To be fair to Mackey, he was clearly being interviewed on the day following the incident before the full facts had emerged. Nevertheless he appears to have completely misread what actually happened, presumably not yet having heard the ATC recording.

So maybe not such as "balanced viewpoint" after all..

Is this picture and video similar to what CA would have seen?

28L unlit (except for red "X"?), 28R illuminated, Taxiway C green centerlights.

https://storage04.dropshots.com/phot...713/073903.jpg

I'm with you on that - especially when fatigue is a factor! Another airport where the 'greenies' rule:yuk:

These photos are an interesting addition to the conversation; and we are still waiting confirmation about the radiating status of the 28R LS. I see PAPI's on LHS of rwy threshold; I see standard green threshold lights; I see standard approach light & touchdown zone lights. That looks like a runway. I can't believe taxiway C had anything like this, so how can you line up on a strip of concrete, at night, that you can't see too well anyway, that doesn't have appropriate lighting. That's all you can see at 4nm, not the concrete/tarmac. I'm also surprised they could see a/c lights hidden in all the airport lighting. They would have been 2 candle power compared to what was surrounding them. But they said they could see something odd.
What approach did they brief at TOD? It can't have been only some kind of visual; they must have briefed an IFR approach of some kind and set the nav boxes for it. If 28R ILS was in the air it should have been tuned and identified, surely? How far out were they when cleared for this odd, last minute offset visual? To an unfamiliar crew that would be a huge cheese hole, right there, if it's last minute and unbriefed. (A bit like ATC asking you to take the next exit just as you touch down. It's a bit late now, mate)
If 28ILS was in the air, and if it was tuned and displayed, what was PM looking at on finals? It would have shown almost full scale deviation at 4nm. What calls does AC have on their approaches? It does seem like 2 heads outside ands no-one managing in the office inside. Were they in autopilot CMD or manual? If A/P, was CMD what was it following; FMS, VOR or ILS? If manual were the FD's on and what were they following? It would seem that they should follow 275 until passing the bridge and intercepting 284, by what ever method laterally.
If they were lined up with the taxiway, which seems invisible in the photos, what were they using for glide path guidance?
One reason that dismissing this as a minor event is not a good idea, is that no-one will investigate and find out how the holes lined up. Damn sure it will happen again. Time to plug the holes.

The mind boggles........

Speaking as a "greenie":

Would you claim that this Air Transport Pilot was so fatigued that he didn't recognize the runway - but alert enough that had he been provided ILS he could have landed safely?

I am willing to believe that some (perhaps most) ATPs would prefer ILS over visual, but I am not ready to believe that so few of them find visual such a handicap that it could be characterized as "not a good idea".

'An unusual set of circumstances that created something that's probably going to turn out to be a non-event'.

Yep, I'm sure the FAA and the NTSB will agree. ;)

Captain Mackey does do a good job of explaining the runway setup and construction at SFO. As DaveReidUK said, this was an early assessment and more information on the incident has emerged.

From the 0700Z approach control reel at liveatc.net, on the second approach AC 759 was vectored for a visual approach behind an AA B-777. Other planes were still using the FMS Bridge Visual though.

I don't believe any of the posted plots of AC 759 overflying the taxiway, including mine, came from FlightRadar24. The data in close to the runway on FR24 didn't look too good to me but you can definitely see who the players are as the incident evolves.

That's what I've been told by a friend who flew A320's years ago. Can anyone with recent experience on legacy A320's confirm this?

Did the AC A320 have GPS for the Nav Display? On some older planes as we've discussed on other threads, ADS-B, TCAS and EGWPS may have GPS data but the ND might not display it depending on the installation. I've flown older '75's that were like this for a while before a subsequent update.

Nothing to do with 'greenies', though I'd buy 'nimbies'. I live about 25 miles south of KSFO, under the BDEGA (from the north, with a teardrop turn) and SERFR (from the south) approaches. The amount of fuss in neighbourhood forums, town meetings and such, especially about the latter, is just unbelievable - people claiming they can't leave their house during the day because of the noise and so on. There are quite a lot of people who file a complaint with the airport EVERY time a plane flies overhead.

Now, do you think these people wear vegan sandals and walk everywhere they go? Or do you think they take the cheapest flight they can to travel several times per year? Yeah, right. But it doesn't stop them complaining.

Could any white lights on four evenly-spaced airplanes on the taxiway create some kind of a runway-like appearance?

Need help with that shoehorn, it looks mighty heavy?

Its called compromise, it's how society generally functions. Believe it or not, and contrary to the belief of many here, aviation is not the single most important priority in the world, and does not have to be satisfied first at all times before any other decisions can be made, which must be made in deference to the wishes of a few pilots on the internet.

Noise regulations generally strike a good balance between airports operating efficiently and not causing unnecessary disturbance. What can be wrong with that unless you just get a kick out of annoying people? In which case you probably haven't reached the metal age to be flying a plane.

This forum confidently told me after Asiana that only Asian pilots did things like this, perhaps some reflection is in order?

Well it's more likely once you've had that Notam that the approach lighting is U/S.


Reading the flight times above, they were an hour late, due before midnight, getting there after midnight. Does the noise abatement approach only come into force at a certain time.

I have done many FMS Bridge visuals in the A320 family. You do not have the ILS raw data because it is an FMS approach. You have VASI and the electronic glidepath. You can hard tune the lLS through the RMP but no one ever does.

As others have said, you have to work to get it down to 1800 feet at SAMUL. You are given the clearance with no time to waste getting it in the descent out of 11000. Speedbrakes are usually required with speed changes given as well. Late night arrivals have less restrictions on speed so the tendency is to keep the speed up.

It's a massive dark hole coming in over the water. With 28L out and the lights off the tendency would be to focus on the lights you see - 28R and the taxiway. Preconceived expectations of what you should see can be hard to overcome. And the difference in lining up on the taxiway and the runway would not be apparatus to the tower until close in. If they were doing the PRM approaches then yeah, radar controller would have caught it a few miles sooner but Late night operations in good weather have minimum staffing. It works fine that way all the time.

Given that the procedure in question provides positive course guidance provided to the runway, the same screwup could be just as evident on an ILS.

The approach is advertised on the ATIS. The ATIS is digital so they could have known for hundreds of miles or even at the departure gate in Canada if so inclined and concerned enough to look. It is tied into the STAR after the approach is line selected (closing the discon, perhaps different on your jet). The instructions from SFO TRACON on initial contact are at ARCHI, join the FMS Bridge routing and only after sighting certain bridges and the airport are they cleared for the approach. From first hand experience, this isn't a particularly challenging procedure to fly having done so numerous times and the routing drops you exactly on CL with VNAV all the way down. AC aren't infrequent visitors to SFO either.

Well, then, let's all sign off our call signs and just go home. And if we're taking a civil air transport flown by an ATP to get there, let's not worry whether noise abatement restrictions have a statistical, or unquantifiable but nonetheless real, detrimental impact on safety. After all, the whole Swiss cheese model, why, that's just the concoction of a "few pilots on the internet." :ugh:

Correct me if I'm wrong, but the angled approach to KSFO 28R has nothing to do with noise abatement and everything to do with the fact that you can't fly simultaneous IFR parallel approaches to two runways that are only separated by 750'.

This incident has nothing whatsoever to do with noise abatement. As a society we constantly deal in levels of risk and decide what is acceptable. Given there hasn't been an incident as far as I'm aware in the history of aviation where noise abatement was considered a contributing factor, perhaps they've got it right?

About angled approach to KSFO 28R - I'm making no claim about current approach designs at all (and also, not claiming that D-R-UK is wrong). The noise abatement factor had been raised by other poster(s) in context of visual approach at night under certain conditions. If that association does not exist in fact, then nothing about differing points of view regarding different approach designs or their components should have elicited any comments about noise abatement (including mine).
As to whether noise abatement as such is relevant, I disagree, neila83. I can make no claim about this incident's causes, but I'll take your word for it (unless and until some future analysis of the airspace architecture reveals that concern for noise abatement compliance yielded a suboptimal approach design). But the lack of an incident where abatement was a contributing factor is just a nice historical fact. As airspace gets more and more heavily utilized, and as complications of CRM, automation of the flight deck, fatigue, and other factors threaten to drill into the Swiss cheese with greater and greater impact, I don't agree that past compromises necessarily are a template for contemporary and near-future decisions. After all, isn't the reasonableness of the compromises being saluted undermined to some extent (not completely) by the simple fact that very nearly everyone who complains about airport noise knew the airport was there before they moved close to it?

With all the technology we have at our disposal today, ('Go-Pro/High Res' Cameras, cockpit display screens, electronic flight bags, etc), surely it should be possible to provide crews with visual data, annotated as necessary, in addition to the charts which you obviously have to have.
It surely wouldn't be difficult to photograph each R/W approach from, say, 2 or 3 points on the centre-line, downwind, base-leg or whatever, in day/night and VMC/IMC, under the heading of...'This is what it will look like when you're lined up, or bang out of cloud.
O.K., you might not need it for every R/W at every airfield, just the ones where confusion may occur or has done in the past.
It's often said that 'a picture is worth a thousand words', and the acquisition of a library of visual images would be easy and relatively cheap to assemble. Just put a few cameras in cockpits for a month, and it's done.

There's one that I'm aware of - a Saratoga vs a Skyhawk at Oakland 30 years ago, although the noise abatement procedures at the time were put in place without FAA approval. Not particularly relevant to this case but it was a contributing factor. More background at '87 Oakland Midair

This is already done by Jepp for a number of different airports if one cares to go that deep into the charts. As you mentioned, it is not done for all runways, but certainly is done for those runways I operate to where a little more "heads up" on the local area is advised.

However, I wonder whether such a chart would have done any good, as the pictures I've seen in the Jepps are always taken during daylight and could not cover all the variables of different airport, city, and sky lighting conditions. A picture does cover a thousand words, but only if those are the thousand words you're looking for.

"Tower, just want to confirm. This is Air Canada 759. We see lights on the runway there. Across the runway. Can you confirm are we cleared to land?"

I can only imagine that he didn't know that 28L was out so deduced the two things he was seeing were 28L and 28R when they were actually 28R and Charlie. Is there any other possibility that I am not considering?

If so, this doesn't have anything to do with noise abatement.

Helps that a couple of the taxi a/c were 87s. That beacon stands out like nothing else

There seems to be some confusion regarding the serviceability or otherwise, of the approach lighting systems for 28R and 28L at the time this event occurred.
In The U.K., any failure, or irregular functioning of approach-aids, constitutes 'Essential Aerodrome Information', and 'shall' be passed to aircraft, by Aerodrome Control' The use of 'shall', means a mandatory instruction.
Notwithstanding the possible promulgation by NOTAM, the information is usually re-inforced by transmission to each individual a/c, or by inclusion on the arrival ATIS. Was this done, I wonder?

O.K, I know that USA and U.K. ATC systems operate differently, but, someone is bound to say that 'lessons will be learned'.

Earlier on in the discussion, someone mentioned the responsibility of the TWR controller, regarding 'looking out of the window'. When I started my ADC training, it was drummed into us that it was essential. It's where the action is.
Having said that, I've never been to KSFO, but looking at the charts, the angle of view and distance, to the 28 thresholds, might make detection of an errant approach to either R/W, or it's parallel taxiway, difficult to detect without the use of in-VCR technology.

Maybe, maybe not. The 'FMS Bridge Visual' approach is the FMS version of the 'QUIET Bridge Visual' approach (they have the same ground track). QUIET in the name strongly suggests noise abatement. Of course the approach may also provide the ability to fly simultaneous approaches as you suggest. But in this case 28L was closed, so the only approaches were to a single runway (28R). And yet they were still cleared via the equivalent of the 'QUIET Bridge Visual' approach and not the ILS. Excuse the pun, but that 'sounds' like noise abatement was a factor.

Bieve,
Or could the 'QUIET' designation mean that the airport is relatively quiet, due to a lack of traffic, and therefore nothing is happening on 28L?

Also might have something to do with arrivals and departures into SJC and OAK. I know when we are in the transition from the DYMND 3 arrival to the FMS Bridge Visual, traffic landing in OAK are below us and traffic departing SJC make an immediate right turn and are just west of us. The step down altitudes keep us above OAK arrivals, the offset keeps us clear of aircraft departing SJC.

How can one mistake 28 left is closed or not...there's a big lighted "X" on the threshold.

Cactusbusdrvr:--- fair enough I understand the dive bomb high speed approach into a dark hole over water approach to SFO is quite challenging but maybe now you'll take the bold airmanship step of manually tuning the 28R ILS and selecting the LS push button on. It's the simple things that will save the day and add another layer of safety for just these scenarios.

It might save your ass too from an embarrassing stuff up to something a whole lot worse.

Live and learn.

Oh and "it's works fine that way all the time". Ahhhhhhh, it would seem that is not the case.

AC have dodged a bullet and in doing so have highlighted a problem that needs addressing.

Here is the TSB Canada incident report with the estimates quoted in the media that AC 759 overflew taxiway C for a quarter of a mile before the tower called the go around and they missed the first two aircraft by 100 feet :eek:.

It's odd that it took 4 pages to mention this. Maybe it's done otherwise in the eu, but unless it was just closed seconds prior, the huge, lighted, white X would have been there.
At the threshold.
Facing east.

OK, point taken. In fact, a bit of digging shows that the airport does actually specify:

"To reduce the impacts of aircraft noise in surrounding communities, particularly between the hours of 2300 and 0700, the Airport encourages the use of the following procedures.

(cut)

(4) Use the Quiet Bridge Approach to Runway 28L/R."

Rules and Regulations: San Francisco International Airport

"Finally a balanced viewpoint instead of sensational nonsense.."

Canadian prelim report:
"The closest lateral proximity between ACA759 and one of the four aircraft on taxiway C was 29 feet."

What is 29 feet between friends? :-)