AAL 331 Kingston final report

[PENKO]

You don't need complicated systems to prevent long landings. As Mozella stated, you know exactly what you are doing when you land long, there is no mistaking it. But pilots tend to reason the risk away: it's a long runway, don't wanna go around so close to the ground, any time now..

What we need is a change in attitude regarding long landings, passed down from training management. Send every pilot a warning email when he lands long (it IS recorded in flight data monitoring) and convey the message that landing long is NOT acceptable in any circumstance. That a balked landing is mandated EVERY time you see those last blocks of touch down zone disappearing below you windscreen.

Hammer it into the pilot workforce that just as not being stable at 500 feet and continuing is a 'crime', so are long landings.

[BOAC]

Originally Posted by PENKO

You don't need complicated systems to prevent long landings. As Mozella stated, you know exactly what you are doing when you land long, there is no mistaking it.

- so true. But what will our amateur psychologists and those marketing expensive software do then?

[alf5071h]

PENCO, et al, you didn’t need complicated systems to avoid CFIT (so was the thinking several years ago); now with a simple, robust computer, based on accurate data, the industry has a very reliable safety device … if only everyone would use it.

Yes the industry does need to change its attitude to long landings, or the major contributions such as tailwind and flare training with high GS. If we continue to think that we always know what we are doing, or why we do it, then we are very mistaken.
Why does the industry land in tailwinds … a balance of economics and safety; sometimes the balance is misjudged, but by whom, pilots or management.

[BOAC]

Alf - you just cannot put EGPWS and some 'automated' long-landing system in the same frame. It is more than possible to be IMC/at night and lost and then warning of terrain is vital, but if you are on a visual approach and are 'lost'........................ Do you need a warning that you are still airborne and the touchdown zone has gone behind your feet?

Why do we land in tailwinds? Why not? If the performance is there, what is the problem? The 'performance' assumes competent pilots,of course, so should we mandate a minimum of 10kts headwind perhaps?

[aterpster]

BOAC:

Why do we land in tailwinds? Why not? If the performance is there, what is the problem?

No problem with a clean and dry runway and touchdown on speed within the SOP landing zone.

[DOVES]

BOAC:

Why do we land in tailwinds? Why not? If the performance is there, what is the problem?

And why not in the opposite direction (conditions permitting)?

[BOAC]

I cannot believe those last two posts!!

Aterp - thus you are choosing to ignore wet runway performance as the work of the devil I take it?

Doves - what if you cannot? Have you never been to such airfields? I assume you would not land at such - just a little limiting for your employer, I feel - not to mention any passengers/freight.

[DOVES]

In the case we are dealing with, even if I do not want to go into details, there was a magnificent RNAV procedure for the reciprocal runway.

Why to pull the rubber band when there is a risk that it breaks?

We have always been taught to choose the safest side!

[Aluminium shuffler]

Doves, reread the thread. The case has already been put and basically agreed that there are times it is safer, not just easier, to take a tailwind landing than a headwind. Want some examples? I'd take a tailwind on the ILS at RAK and ACE any time. Suggesting that one must always land into wind for safety shows a very basic level of airmanship - there are many more elements to an approach than wind direction.

Now in this case, the wind was so strong and the other factors so contributory that landing with the tail wind was clearly wrong. It it has been 5kts with all the same other variables, I'd say 50/50 and with nil wind the ILS would be better than the VNAV. But don't criticise others' airmainship if you don't understand the factors involved.

[PENKO]

alf, anyone can make a navigational error and end up approaching high terrain, no matter how hard we train and how vigilant we are, so EGPWS is a good thing. But landing long is usually not a mistake, it's an attitude, it's a deliberate act. You know you land long. If you flare and you do not touch down within five seconds, you know you are leaving the best part of the TDZ behind you. If you cross the threshold and you do not hear 'fifty', you know you are in a bad position. We have all been there, haven't we. Do we really need another box to shout at us? Or should we change our attitude?

Actually I'm not against warning systems with exotic acronyms at all. I'm just saying that landing long is mostly a matter of attitude. Be it corprorate attitude, flight deck attitude, individual attitude, you name it. It's attitude change we need. We need to regard the long landing in the same light as bing unstable at 500'.


Anyway, to switch focus if I may: why do some of you prefer an ILS over RNAV? I sense a certain fear, a certainnreserve and I don't think it has anything to do with the higher minimum..
An Airbus will happily fly any RNAV approach, so what is the hesitation about? Unfamiliarity?

[aterpster]

BOAC:

Aterp - thus you are choosing to ignore wet runway performance as the work of the devil I take it?

I didn't say that. What I implied and what I meant is that more factors must be factored into landing with a tailwind on a wet runway, particularly when a significant rainfall is occurring.

The first order of business would be to land into the wind if possible, which was the case with this accident.

But at LAX, for example, with 12,000 feet of runway, it's no big deal for any civil transport.

[DOVES]

…there are times it is safer, not just easier, to take a tailwind landing than a headwind…

I know! I remember when I was a young pilot in the right seat of a SE-210 in the first seventies performing night approaches to LIMJ. We preferred to land on Rwy 29, also with 5 Kts tail wind rather than to perform the particular circling for Rwy 11. But we used to brief “We will deploy ‘Le parachute’ after landing!”
But in this case:

Accident: American B738 at Kingston on Dec 22nd 2009, overran runway on landing 03:22Z Dec 23rd
MKJP 230325Z 32011KT (bell) 22000 (2200?)+SHRA (bell) BKN014 FEW016CB (bell) SCT030 BKN100 21/19 Q1014 RETSRA (bell)
MKJP 230300Z 32008KT 3000 +SHRA BKN014 FEW016CB SCT030 BKN100 21/20 Q1014 RETSRA

While the aircraft descended to 4000 feet the approach controller advised the winds were now from 320 degrees at 15 knots (bell) and queried whether the crew was still to land on runway 12 which the crew replied to in the affirmative

The aircraft was almost at Max Landing Weight.

7) The flight crew did not use the RNAV (GPS) Rwy 30 approach, and land into wind on runway 30.

10) The flight crew elected to land with flap 30, rather than the flap 40 recommended for short field and tailwind wind Landing

14) The aircraft crossed the runway threshold 20 feet above the ideal height, and landed long.

19) The aircraft touched down at 4,100 feet (at 140 kts = almost 20 sec [begin to count 1001, 1002, 1003,….!!!!!]) from the runway threshold

22) There was a 14 knot tailwind component when the aircraft landed.

24) The flight crew did not conduct a go-around when the long landing made this necessary.

26) There was reduced friction on the runway, as evidenced by the longitudinal deceleration rate recorded on the FDR. The presence of melted rubber balls on some of the aircraft’s main landing gear tires could also be an indication of reduced friction.

3. The captain’s visibility may have been impeded by the heavy rain on the windshield (bell: flooded rwy?), and the rapid movement of the windshield wipers.

And as you play the ball, not the player: we must learn from this!

http://www.pprune.org/8476954-post71.html
http://www.pprune.org/8477548-post79.html

Don' t conduct T.O. and Landings (operations) in the proximity of TS/Cbs!!!

[SLFinAZ]

I'm sorry but if your so "task saturated" that your unaware you've used up 50% of the runway on the flare then you don't belong in the pointy end of the plane. While I've done my best to read and understand all of the subtleties in the end this is simply command responsibility.


The PIC (regardless of who's PF/PM) has final responsibility for all souls on board. It's a heavy burden and requires a specific psychological make up. While a pilot may be technically sound and have tremendous experience as measured in hours that does not mean he's actually capable of command.


Not every FO is qualified for command but from my understanding of the current system it's almost entirely seniority based unless someone is less then competent. This is in fact the real problem with Airbus, that in fact command authority can be engineered to the lowest common denominator. To the point that you even refer to a large, complicated and entirely unforgiving marvel of technology as a "bus"...and start to treat those responsible for it's operation accordingly.


As for this the fact that the captain still has 4 stripes is a major indictment of the system. Fundamental command failure is just that.

[slf4life]

Recognizing the float

If I may, given the focus/tasks required would there be a difference between a small field like MKJP and vastly larger ones in terms of visual references and noticing you are long?

Even in the conditions that prevailed, the emergency equipment and terminal area are comparatively close to the single runway, and I'd imagine provide visual reference for the two pilots, especially if they had previous experience with the field?

I ask because one of the interviewed pax was an old acquaintance. Even he was aware through his small window that something was unusual when the fire equipment location (about halfway down the field I think) went speeding by.

I've no doubt this experience will stay with the crew. I'm sure they are people of good conscience, and will learn from hard lessons and get better. I also hope that those who were injured behind 'the pointy end' have no lasting ill effects, physical or otherwise.

[Aluminium shuffler]

SLFinAZ, well said - shame we don't have a "like" button! There are continuing excuses for these guys not going around. Even if you can't see any designators at night, just the fact that you haven't touched down within a few seconds of the flare should tell you to go around. Really, there are no legitimate excuses for what happened here, or in so many other over-runs.

[alf5071h]

PENCO, as before I agree that attitude is important (see http://www.pprune.org/safety-crm-qa-...ml#post3835921 also #9, #11, before that). However, I disagree with the view that the decision is always deliberate or something which without-fail can be controlled; the reasoning for this is described by others in HF guidance materials, and in particular ‘A Life in Error’ by James Reason – his latest (final) book which provides an excellent overview.

Also based on the views of Reason, blame and punishment (as in sending an e-mail to errant parties) to aid behaviour change may not be as successful as imagined. Better results – a change of attitude / behaviour, might come from explaining the risks in the operation (education / training / experience) and having clear operational boundaries and guidance, aiding decisions, deliberate or otherwise to be based on critical issues, which hopefully would provide a better balance between safety and human desire, i.e. knowing what parameters are important and why, what are the limitations or assumptions in data, or how far information might be extrapolated; the devil is in this detail, where improvements in professional attitudes might help.

[alf5071h]

BOAC, the devil is indeed in wet landing performance; as above, the basis of and the assumptions in this data are often overlooked. (open sucking eggs icon)
Certificated ‘wet’ landing performance is based dry runway test data (some manufacturers’ spot-test wet landing distances, but only in particular conditions).
The safety distance factors are to account for differences between test conditions and variability in normal operation, i.e. greater air distance than certification, and ‘average’ piloting margins in speed, TCH, flare time, etc. However, the safety factor may not be sufficient for gross deviations, nor for a combination of accepted max deviations – excessively fast and long (see AFM/FCOM).

‘Wet’ distances have an additional factor to account for the runway conditions – the braking action on a wet surface. The wet landing distance safety factor in certification (1.67 – 1.92) assumes a particular value of mu, lower than the dry mu to extrapolate the dry landing distances. Thus the wet distances are only as good as the assumed wet mu, which may be considerably different from the actual (un-measurable mu) in operations – depth of water, basic runway friction, runway material and texture, grooving or not, tyre tread condition, … ; many of these items are unknown (unknowable) to the crew. These differences again use up the safety factor. The baseline wet safety factor may not provide the same level of safety as for a dry runway, and as conditions deteriorate, wet landings involve increasing risk.

The wet certification limit is at 3mm water depth (flooded = contaminated) where different methods of distance calculation (and assumed operation) are used, again with further risk – this is explained in AMC CS 25.1591 (FAR may differ!!!). However, the gap between the assumed wet performance (AFM) and the actual conditions involves many other variables, including flight accuracy, which in near limiting conditions may leave no margin for error, or even be insufficient due to inaccurate reporting – water depth, wind. (close sucking eggs icon)

Hence my question back at #64, 13 May. ‘Wet/good’ relates to the certificated wet data, but ‘medium/fair’ probably does not, and has a different basis for calculation and operation (Boeing ‘actual’ distances), which are more aligned with the risks of operating on a contaminated runway and thus requiring a change in crew’s attitude in the use of this data. Thus the lack of reference to water / wet in the ‘medium/fair’ category might be misleading and not cue an appropriate attitude to risks involved.
It is possible, perhaps more probable with hindsight, that even if the crew in this accident had calculated the distance with factors for the braking action, and landed in the ideal position, with max braking, then the aircraft might not have stopped in the distance available.

Ref Managing Threats and Errors during Approach and Landing note slides 25 / 26

[Aluminium shuffler]

alf, you're stretching logic to pardon the crew and blame the system by suggesting that if they had landed in the right place that the possibility of a run-off still existed. Perhaps it did, but that was not the cause of the accident - their excessive speed from poor runway choice and, more critically, their failure to land in the right place are the causes. Both errors could have gone without harm if they had done the right thing and firewalled the thrust levers to go around. They decided to press a very bad landing and therein lies the prime cause.

[Sky Slug]

When I underwent training on a RJ, it was emphasized that most Canadian airports don't have grooved runways (for whatever reason, big winter airports like MSP/CHI/BOS/NYC seem to handle ice okay). It was said that a long landings could end up in the snow banks as has happened in YOW multiple times. That was on a regional jet. A 737-800 with a tailwind and a long landing, good luck.

I guess my ALPA dues will save my ass if I do something ridiculous at my major. ALPA representation was rarely seen in my regional. You screwed up your career was essentially gone.

[hifly787]

Love these first world cockup justifications. Blame ATC . BA at Joburg putting a 744 wing into a building . ATC fault . Land on the wrong runway with huge aircraft . Innocent mistake. Not one word of sympathy for the Lion Air P1 not being informed of heavy showers on approach or wet runway.
Keep it decent guys.