In Explanation (for those who may have come in late)
Quest
During my IR training I can recall being told about all the instrument errors that occur during various situations but I can't remember anything about this. I've been trawling through my memory ever since the GF accident and the first discussions of the somatogravic effects on Pprune and just can't recall this problem yet it seems such an important physiological phenomena to be ignorant of.
Did these guys get themselves into trouble ONLY because they did NOT stay on the instruments and flew visually without any visual references - or am I missing something?
http://members.ozemail.com.au/~aupa/PDF_files/DNT.pdf
http://www.pprune.org/ubb/NonCGI/For...009550-15.html
Barry
Answer
Barry
They were turning 360 degrees to try and recapture the centre-line on short finals - always a challenge that's best left unattempted - when they shot through the centre-line and so had to declare a missed approach and reposition for another approach. Instead of a wings-level missed approach to a safe height, levelling off, stabilising and only then turning to a good repositioning heading, they were given a clearance by ATC that had them doing it all at once (overshooting, cleaning up, accelerating, climbing, turning, communicating and levelling). All these significant accelerations and distractions in a light and frisky aircraft posed a physiological mish-mash of sensations that then had suddenly superimposed upon them the complete loss of visual cues. This critical loss of "visual" was caused by their ATC-instructed immediate left turn away from the bright lights of Manama and the island of Bahrein into the dark-pool of the inky blackness of the overwater Gulf. You could NOT have set up a better scenario for that pitch-up somatogravic illusion. And of course the real killer with that one is that both the PF and PNF get it simultaneously. The powerful and instinctive reflex response to the sudden onset of the pitch-up illusion is to forcefully lower the nose (which causes greater acceleration and stimulates an even stronger instinctive reflex). It has been the cause of more night and IMC accidents than it has ever been given credit for. Low experience levels in the flight-deck don't help. Neither does an airline culture (within GulfAir and elsewhere) that calls for the captain to file a report on why a missed approach became necessary. The pressure upon pilots to then abbreviate the procedure (and ATC to comply) is a given and would have been a cause factor in this accident. None of the built-in alpha-floor protection features of the FBW Airbus were going to stop them from falling for this trap.
Since I raised the issue of somatogravic illusion, I have sensed in local Gulf News reports and the interim Official factual report, that they are trying to decry or dispute this, and find some other aspect that downplays the human error angle and seeks to either lay blame elsewhere (misleading instrumentation) - or leave the cause safely "unknown". The Final Report's due out now, so I guess we shall see if those suspicions are correct.
I'd have to agree with you that there is scant training carried out for airline pilots in avmed and physiological pitfalls. This will become even more significant as the number of military-trained airline pilots decreases in future. It's not something that can be done effectively or induced in a simulator. Significanty I had an email from an airline Chief Pilot (and a Pprune-post) thanking me for highlighting the phenomenon, simply because he'd never heard of it. It makes you wonder - particularly as there's no shortage of academic writings upon the subject.
To answer your second para question, the dichotomy for "contact" non-IMC night-flight in VFR conditions has always been that it must be an instrument/visual "mix" (i.e. instrument flight but necessarily utilising visual contact cues). That was brought home to me very effectively when another Vampire first night solo just ahead of me forgot about the runway change for that second wave of jets and turned fatally left off 18 into the Darling Ranges at Pearce (instead of right). Fatal oversight, like hindsight, is easily come by.
**************************
Ques
It's starting to come back to me. I remember reading about a crash in Canada
where a light a/c , a potent turbocharged c210 from memory, took off from a dark country airport and just flew into the ground a mile or two upwind not far off the extended centre line of the runway. The investigation did mention something
about an illusion which I didn't understand at the time.
I seem to have lost the link to the Canadian civil aviation site, do you have any URL's as I'd like to try and find it and re-read it in the light of my new found knowledge. Obviously someone is well aware of the problem but it's not public enough. Barry
Answ:
Barry
It will be one of those accessible via
http://www.tsb.gc.ca/ENG/
(actually
http://www.tsb.gc.ca/ENG/reports/air/RptAvi_Indx.html )
So could you be a mate and, if you look through them, pass me the URLs of:
a. Any other somatogravic-related accidents
Belcheek
SomatoGravic Illusion http://www.tsb.gc.ca/ENG/reports/air...ea93c0169.html somatogravic illusion suspected
"Somatogravic illusion may have adversely affected the pilot's performance during the acceleration stages of the take-off and initial climb. "
2.6 Behavioral Factors - Illusions and Disorientation
The forward acceleration of the Cessna 310 aircraft is sufficient to produce a powerful illusion of increasing pitch. Under extremely dark night conditions, with restricted outside visual references, a somatogravic illusion could cause the pilot to erroneously conclude that the aircraft was rotating to an increasingly high pitch angle. This illusion would be intensified if the pilot were denied accurate visual information because of a weak instrument scan, degraded eyesight, or poor ambient external lighting.
also
http://www.tsb.gc.ca/ENG/reports/air...ea96c0002.html
I think this somato...... illusion is so insidious it should be part of the pre takeoff briefing before every flight into night/ IMC in multi crew ops just as you turn onto the runway. Wouldn't have helped the poor chap in the URL above.
Barry
Barry
Research has shown that pilots are easily overtaxed by (and don't interpret) audio alarms - in particular when things are already coming unglued. It doesn't help to have a real tyro first officer who was overawed by his captain (and likely suffering the same sensory overload and selective radial scan breakdown). My experience of Arabs especially was that they were sometimes prone to throwing in the towel and leaving the outcome to the Supreme Being (with suitable incantations). That's why Arab students are never allowed to spin solo. In the finite time that you must take to permit spin recovery actions to take effect, an Arab cadet was always likely to suffer an attack of doubt, take hands and feet off and decide that it was Kismet that he should die. Nothing racist about this, it's purely an innate cultural and religious belief that they lean upon heavily when stressed. I saw it often.
When a very powerful sensory impulse is calling for an instinctive reflex reaction (which will unfortunately compound the problem) you can expect that the accompanying audio alarm is going to be beyond the stressed pilot's recognition and response threshold. You cannot "freeze-frame" a dynamic situation like they do regularly (for discussion) in a simulator. Which means of course that you have to look at this particular accident in terms of the time-scale of occurrence, recognition and the time available for reaction/recovery at the profile height, speed, altitude and attitude that they were at. It's easy to say that "surely they'd have been taught to wholly trust their instruments". That's a great concept of course, but it requires absolute self-discipline and some prior exposure to the sensations of entrapment. They would have had neither of these pre-conditioning absolutes. I call them "absolutes" because in reality, without this sensory training and a solid avmed understanding of the process...... well it is all so much lip-service. There is evidence in the DFDR readout of an attempted recovery, however it ended up being quite half-hearted and to me simply reflects the totally abject confusion that reigned once they had lost visual cues and were suddenly struck dumb and essentially powerless by that overwhelming physiological illusion. That's simply, factually and historically how it happens. CRM is no help when it's highly likely that both pilots were riding that same phenomenon.
Thanks for the research offer Barry. I always value your insightful inputs. I think that perhaps you are starting to realise and recognise one of the really basic truths about developing trends in airline aviation and crew quals. Aircraft are designed to be increasingly forgiving and reliable but once things start coming off the rails, it better be a very straightforward malfunction or you will soon be beyond your training, experience and limited systems knowledge - and in the province of confusion, despair and improvisation. That's what begets the typical, nowadays, pilot-error induced accident outcome, moreso than any overconfidence, negligence or risk-taking. And I guess I'd have to add that, like the GF072 crew, you don't really need a malfunction to qualify you for a visit to Never-Never Land (as in "never done this before - so what's it doing now?"). You only have to stick your neck out a little bit - to suddenly find that you shouldn't have.
If you read the speech (URL below) by Benoit Bouchard (Head of Canada's Transportation Safety Board), you will see that that is what he is saying in a nutshell, whilst still trying to put a positive spin on that admission.
http://www.ntsb.gov/events/symp%5Fre...t_bouchard.htm