Atlas Air crash should spark an overdue debate about piloting

Recent releases from the US National Transportation Safety Board’s investigation of the Atlas Air Boeing 767-300F fatal crash in February 2019 contain a vital message to the industry about loss of control in flight (LOC-I).

Unfortunately, the message could be overlooked, or not taken seriously, as it has been many times before.

The Atlas Air crash, however, finally negates a common reason for unconsciously dismissing the seriousness of a LOC-I accident.  This unconscious dismissal, among “Western” observers at least, is caused by the mindset that says: ‘It happened to a non-Western carrier’; the implication being ‘What would you expect?’.

Such pilot reactions to the Lion Air and Ethiopian 737 Max accidents flooded the web, particularly in the USA, and are still out there. The latter two accidents, however, involved an aggravated version of LOC-I, precipitated by a confusing technical distraction.

Right now, in the Atlas Air investigation, the NTSB is testing evidence that suggests pilot disorientation by somatogravic illusion might be pivotal in what happened. During descent toward its destination airport the aircraft finally dived steeply and at high speed into the surface .

A synopsis of the basic accident details can be found on the Aviation Safety Network.

A common example of somatogravic illusion – which is an acceleration-induced illusion – is the feeling that airline passengers get when their aircraft begins to accelerate along the runway; they perceive the cabin to be tilted upward, but a glance out the side window shows the aircraft is level, the nosewheel still on the ground.

Visual input, if available, is the dominant human sensory input, and it will correct the illusions caused by the reaction of the body’s balance organs to a linear acceleration.

The Atlas Air 767 freighter was inbound to Houston Intercontinental airport from Miami, and the flight phase in which things began to go wrong was a routine descent, the crew receiving vectors to avoid weather. As the aircraft was descending, in cloud, through about 10,000ft, cleared to 3,000ft, the crew were flying a vector heading of 270deg, and were told to expect a turn north on a base leg to final approach for runway 26L. All pretty normal.

There was a pilot call for “flaps 1”, the aircraft leveled briefly at 6,200ft, climbed very slightly, and its airspeed stabilised at 230kt. But shortly after that the engine power increased to maximum, and the aircraft pitched about 4deg nose up.

It is at this point that somatogravic illusion appears to have kicked in powerfully with the pilots. They had no external visual horizon because the aircraft was in cloud.

According to the NTSB, almost immediately the aircraft began a dramatic pitch down to -49deg, driven by elevator deflection. The airspeed ultimately increased to 430kt, and although the pitch-down angle was eventually reduced to -20deg, impact was inevitable.

The factor the NTSB is examining now is what triggered the sudden – apparently unwarranted – massive power increase. The cockpit voice recorder has captured a sound that may indicate the activation of the go-around button on the power levers. But neither of the pilots mentioned a need for go-around power.

About ten seconds after the power increase, caution alarms began to sound. The inquiry says the control column remained forward for ten seconds. According to “The aircraft transitioned from a shallow climb to a steep descent. Five seconds after the alarm commenced, one of the pilots exclaimed, ‘Whoa’, and shortly afterwards, in an elevated voice: ‘Where’s my speed, my speed’. Three seconds later, a voice loudly declares: ‘We’re stalling.’”

The flight data recorder gives the lie to the pilot’s stalling perception, because the angle of attack at that moment was safely below the stalling level.

During these remarks the thrust levers were brought to idle for about 2s, then were advanced again to their high power setting. During the transition from nose slightly up to nose steeply down, there were negative g-forces for nearly 11s.

Puzzling unknowns still lurk: like why a pilot exclaimed “where’s my speed?” when the indicated airspeed was rapidly increasing. Was it a fault of instrumentation, or of pilot instrument scan or perception at a moment of confusion?

The simple fact is that, every time a big engine-power increase takes place in flight, forward acceleration combined with a pitch-up moment caused by the underslung engines, is inevitable.

Just as inevitable – if this happens at night or in cloud – is somatogravic illusion in the pilots. “For this reason,” says the NTSB, “it is important that pilots develop an effective instrument scan.”

Develop? It’s a bit too late to develop a scan!

Recognising that acceleration brings with it the risk of disorientation, pilot conditioning should be to ignore all other sensory inputs except the visual, and with no external horizon that means concentrating totally on instruments, believing them, and controlling aircraft attitude and power accordingly.

Recurrent training must keep pilots alive to this risk, and to its remedy, but it clearly does not do this at present. Not for Asian, African nor for American pilots.

LOC-I has, since the late 1990s, been the biggest killer accident category for airlines. LOC-I linked to somatogravic illusion has frequently occurred, two of the most dramatic recent examples being the March 2016 FlyDubai Boeing 737-800 crash at Rostov-on-Don, and the August 2000 Gulf Air Airbus A320 crash at Bahrain International airport. Both occurred at night; both involved a go-around.

The FlyDubai pilot reaction to somatogravic illusion was a dramatic push-forward into a steep dive, like Atlas Air, and the aircraft smashed steeply into the runway.

The Gulf Air manoeuvre was an abandoned night visual approach from which the captain elected to climb and turn into a 3,000ft downwind leg to make a second approach. In the latter case, the changes in attitude and power were less dramatic, but as the captain advanced the power and began the climbing turn to the left over the night sea, he would have lost the airfield and town lights and had to transition fully to instruments. The aircraft described a shallow spiral into the dark water.

Somatogravic illusion makes instrument flying essential, but more difficult because of the need to reject the balance organs’ misleading input. A clear natural horizon in daylight completely overcomes those misleading feelings, and although the instrument panel – especially in modern flight-decks – provides an intuitive visual display, it is artificial and still not as compelling as the real thing.

But there is a long list of LOC-I accidents in the last two decades that involved more subtle sensory inputs resulting in pilot disorientation, and everybody died just the same.

Think of the old expressions associated with instrument flying skills.

First, there is its antithesis: “Flying by the seat of your pants.” Anybody who believes that is possible in IMC or on a moonless night is fated to die.

Then there is the original name for the skill: “Blind flying”; that was in the days before the artificial horizon was invented, when the airspeed indicator, altimeter and turn-and-slip indicator sufficed for accurate flying, possibly assisted by a vertical speed indicator.

Further clues as to the fascination – even mystery – surrounding early blind flying skills are the descriptions of what it feels like when things are going wrong: “The Leans” described the situation in which your perception of what the aircraft is doing is not what the instruments tell you. Finally there is the extreme example of “The Leans”: Americans used to call it “vertigo”, Europeans “disorientation”. That is when your senses are screaming at you that the situation is not what your instruments say – you don’t even know which way up you are.

Nothing has changed just because aircraft now have LCD displays.

It is time to go back to basics, to re-discover pride in precision manual instrument flying, and regain that skill which no pilot truly believes s/he has lost, but which automated flying has stolen away silently, like a thief in the night.

PS: Good blind flying is not a stick-and-rudder skill, it’s a cognitive skill.






Ordinary airline fatal accidents are back

After an apparently near-impeccable year for airline safety in 2017, the traditional accidents are returning. In the last week a Russian carrier has fatally lost a twinjet, and an Iranian airline a twin turboprop.

It would be closer to the truth to say these accidents – or at least the risk they represent – never went away, it’s just that a year is too short a time in which to measure the true safety performance of an industry.

The previous story in this blog sequence investigated the part that luck plays in airline safety, and it concluded that it still plays an unacceptably big part in an industry that confidently tells its passengers it has high standards.

The Russian loss involved a Saratov Airlines Antonov An-148 regional twinjet. It had taken off from Moscow Domodedovo airport in snow, bound eastward for Orsk, but after about 6min its climb became a rapid descent and it hit fields at high speed.

Early data from the investigation suggests the trigger for the fatal sequence of events was a disparity in airspeed indicator readings, probably caused by ice build-up in the external sensor because its heater had failed. The crew saw the disparity developing between the airspeed indicators and tripped out the autopilot, but failed in their attempts to fly the aircraft successfully relying on instruments that included misleading  airspeed data.

Less detail is known about the Iran Aseman Airlines ATR72, but it was on a flight from Tehran to Yasouj among Iran’s south-western mountains. The destination is cradled in a valley, and the aircraft hit mountains about 30km north of the city in its early descent. The mountainous terrain was under complete cloud cover and snow.

The Saratov case provides more evidence of pilots’ unpreparedness for “limited panel” instrument flying. Air France 447 was the most famous example of pilot inability to cope with instrument flying when the airspeed sensors were temporarily compromised by ice, but the final report revealed that there had been six other recorded occasions in the same aircraft type (A330) where pilots had coped successfully with unreliable airspeed readings.

Airline recurrent trainers need to go back to basics with instrument flying, because it is increasingly clear many pilots all over the world are losing this crucial skill.

Loss of reliable airspeed information is unsettling, and it usually causes the autopilot to trip out, so airlines should ensure their pilots are able to cope with this situation.

In stable flight, whether level, climbing or descending, airspeed is a product of engine power and pitch attitude. All pilots with time on a particular type should know approximately what power will produce the performance they want.

So if they become aware that airspeed indications are compromised, it makes sense to adopt straight and level flight (if at a safe altitude) while sorting out a Plan B. That way the attitude is stable, and if the pilot selects the power setting that will produce a safe airspeed, all is well. However unsettling it is to see an airspeed reading that is clearly wrong, and which would be dangerous if true, it is a plain fact that the correct pitch attitude for S & L flight plus the correct power setting will produce the correct airspeed.

In the case of the Iran Aseman ATR72 the cause of the crash isn’t known yet, but there was no emergency call and that range of mountains contains many aircraft wrecks. If it turns out to be a classic case of CFIT (controlled flight into terrain), the issue will be one of three-dimensional navigation on instruments. Yes, such approaches are demanding, but these procedures should be the lifeblood of crews working for Iranian domestic carriers, for whom approaches into airports surrounded by mountains is their daily work.

These airlines – and others – have to ask themselves what is missing in their pilots’ skills, and why these skills are missing at all. Finally, they have to ask what they need to do to replace skills that have lapsed.

If the investigators’ final verdict is that pilot error was a factor in these accidents, the fault lies squarely with the carriers for failing to ensure, in their recurrent training regime, that their pilots have the living skills their passengers have a right to expect.

And again, if that verdict were to be delivered by the investigators, the airlines should worry about whether their existing crews could fail in the same way tomorrow.