When a pilot dies mid-flight

Are passengers at risk? Plus: Plenty of flotsam and jetsam, but no real answers in Air France crash

Published June 18, 2009 10:18AM (EDT)

Some quick notes, If you'll allow some cribbing from past columns, on the in-flight death of the Continental Airlines captain earlier Thursday ...

This was a highly unfortunate event, obviously, for the captain and his family. From a safety standpoint, however, it was really a non-issue. Remember that all commercial flights carry at least two fully qualified pilots, captain and first officer, who are able to operate the aircraft in all regimes of flight, in good weather or bad. A transoceanic flight, such as Brussels, Belgium, to Newark, N.J., on which crew members take scheduled rest breaks, would have been carrying a minimum of three pilots -- a captain and two first officers, one of whom would be designated a "relief" first officer.

The first officer is known colloquially as the copilot, but he or she is not an apprentice or a helping hand. First officers perform just as many takeoffs and landings as captains do.

Captains and first officers will typically take turns at the controls. On a two-leg day, for instance, the captain will fly the first leg, and the first officer will take the second. Flying Newark to Brussels, the captain would have been the "flying pilot," as it's called, with the first officers sharing the various other duties. On this afternoon's return trip, one of the first officers would be at the controls, with the captain assuming the other duties. This is industry standard. Both on-duty pilots are plenty busy, but only one is physically at the controls. The captain, of course, always has command authority -- and a somewhat bigger paycheck. (Moving from first officer to captain is strictly a function of seniority.) 

Long and short, when the Boeing 777 touched down at Newark, there were two fully rated pilots at the controls -- exactly as there would have been normally.

Obviously, crew incapacitation would be a serious issue during a critical phase of flight -- takeoff and landing, namely -- but pilots are trained to recognize and react to it. Apparently the Continental captain suffered cardiac arrest during the low-workload cruise portion of the flight.

Protocols can be somewhat subjective, but airlines have their own rules and procedures governing what to do if a passenger or crew member passes away. The media has reported that the passengers were not made aware that the captain had died. Presumably the first officers, together with staff on the ground, decided this was the best course of action. Operationally, nothing about the flight would have been handled any differently.

When somebody on board falls ill, the crew will communicate with company personnel and medical specialists on the ground, while also soliciting help from doctors, nurses or any other health professionals who happen to be on board. Commercial planes carry a cardiac defibrillator and EEMK (enhanced emergency medical kit). Flight attendants receive some emergency medical training, but are not paramedic certified.

An interesting angle to this story: The mandatory retirement age for pilots in the United States was recently increased from age 60 to 65. The Continental captain was 60. He could just as easily have been 59 -- or 40 -- but you will probably hear some I-told-you-so rumblings from those who were opposed to this rule change from the start.

Pilots over 60 need to meet stringent, twice-yearly medical exams. Stipulations also prevent two over-60 pilots from being paired together on the same flight.

One more small point, but something that keeps jumping out at me when I peruse the various media stories: They refer to the doctor who reportedly "went to the cockpit" to examine the captain. This is impossible. Under no circumstances would a passenger be allowed into the cockpit while aloft, even to assist a gravely ill crew member. Most likely the captain had been removed from this station, and the doctor is referring to the cockpit area.

- - - - - - - - - - - -

From the remarkable photo archives at Airliners.net, here is a shot of the doomed Air France Airbus A330 (registration F-GZCP), taken at Charles de Gaulle airport on the morning of May 31. It would later depart for Rio de Janeiro, with a scheduled return to Paris in the evening, as Flight 447. 

Over two weeks have passed now since the airplane disappeared in stormy weather off northeast Brazil. Investigators and salvage teams have been scouring the ocean, hauling in ever more bodies, luggage and airplane parts. There is plenty of flotsam and jetsam, but alas no real answers. Recovery of the elusive black boxes appears more and more unlikely. 

All we have to go on, pretty much, is a confusing series of automatic system status messages, relayed from the flight just prior to its disappearance. They seem to indicate a cascading series of faults and failures, including faults in the jet's primary and backup flight instruments. It is improbable that these instruments failed completely, as an aircraft's internal systems are engineered with such redundancy as to render a scenario like that all but impossible. But something seems to have happened that was affecting the plane’s instruments and controls.

We continue to hear about the possible role of malfunctioning "pitot tubes." I downplayed this theory in my last column, but it's certainly worth looking at. The issue of how and why the crew found itself in the midst of a presumably violent storm in the first place is the obvious and most critical question, but yes, once they got there, a malfunctioning pitot system may have contributed to an eventual loss of control and, most likely, an in-flight breakup. 

Pitot tubes are among the many sundry probes, sensors and detectors (there can be two dozen or more) dotting the exterior of a jetliner. Along with other sources (GPS, inertial reference systems, etc.) they help determine speed. This airspeed data is then fed to the plane's flight control computers. Apparently there has been a known problem with the pitot tubes on some Airbus models, including the A330. All such probes are heated and monitored for proper operation, but this particular design has a drainage issue that can, under rare conditions, result in the misinterpretation of airspeed. 

That alone would seem unlikely to have brought down the aircraft, but industry sources familiar with the A330 contend that unusually severe weather can affect the plane's air data processing systems, providing erroneous data to critical instruments, including backup instruments. We've been hearing a lot about icing, but this was likely just one of multiple problems brought on by violent weather. Lightning, severe turbulence, icing, hail -- each, perhaps, had a part to play.

Some have theorized that haywire pitot probes and/or a bizarre electric anomaly could have resulted in a deadly overspeed. One A330 pilot I spoke with raised the possibility that Flight 447's computers suddenly believed the aircraft to be at a speed much lower than it actually was. In theory, this could have commanded a rapid, automated power-up of the A330's engines. The crew, meanwhile, battling several simultaneous problems, would not have been expecting this, and may not have reacted in time. The plane quickly exceeded its maximum speed, the thinking goes, went out of control and crashed. 

Flying at higher altitudes, airspeed is very important. Flying too slowly can be dangerous, as can flying too fast (minimum and maximum speeds vary, depending on weight and altitude). In addition to overstressing the airframe, flying too fast can result in aerodynamic buffeting, loss of roll authority, a hazardous pitch disruption known as "Mach tuck," or even, in extreme cases, a stall. 

A plane stalls when, roughly put, the wing runs out of lift. Not only are there low-speed stalls, as most people are familiar with, but there are high-speed stalls as well. It's not only a question of how efficiently the wing is moving through the air, but how fast the air itself is flowing around the wing. As this airflow nears the speed of sound -- aka Mach 1, which itself varies with temperature -- a shock wave builds, separating the airflow from the surface and destroying lift. 

Thus, at upper altitudes, where the air is very thin, we find an aerodynamic paradox: The higher a plane flies, the faster it needs to go to maintain lift; but the faster it goes, the closer it gets to that shock wave. You're stuck between going too fast and too slow at (almost) the same time. These high- and low-speed boundaries will eventually meet at a realm called "coffin corner" -- a scary buzzword that has been making the media rounds of late. 

But jets are engineered to fly at high altitudes, and have been doing so safety for the last 50 years. This proverbial razor's edge is only relevant at a point well beyond the average commercial jetliner's performance envelope. Coffin corner is not where commercial airliners hang around, even at their maximum weights and at maximum certified altitudes. It's possible that Flight 447 was victimized by high-speed complications, but according to one expert, it's doubtful that a coffin-corner stall was one of them, even in severe turbulence with rapid airspeed fluctuations.

"Commercial airliners do not operate in this region, plain and simply," says Chris McCann, a retired U.S. Air Force test pilot. "It is true that an aircraft approaching Mach 1 can experience localized supersonic flow on various locations on the fuselage, wings and tail, but this is seldom an issue for a swept-wing commercial jetliner. Neither is it something that can happen 'in seconds' by accelerating the aircraft. It is very improbable that a stall could develop sufficiently on the lifting surfaces -- not before something else causes bigger problems."

McCann says that an overspeed presents two more likely, and still potentially dangerous, complications. First is a major reduction in aileron (bank) authority due to shock waves building along the wings. This could lead to an uncontrolled roll. Second, if a shock wave causes laminar separation, this can impart a significant upset -- buffeting, or even structural failure -- of the horizontal tail (that is, the pair of smaller, aft-mounted wings that control a plane's nose-up, nose-down pitch) as it is struck by the energized and turbulent airflow. The chances of either of these things happening is substantially greater during heavy turbulence, or if the crew is exerting heavy forces on the controls.

"Regarding the pitot tube icing issue," says McCann, "it's entirely possible that they were a complicating factor, but I doubt they were causal. Once an aircraft penetrates a no-kidding thunderstorm, all bets are off. In extreme turbulence, at night, with large quantities of super-cooled moisture pelting the aircraft -- and probably a fair amount of dazzling lightning -- my guess is that the aircraft got into an unusual attitude from which it could not recover, particularly if the primary flight instruments and controls were degraded."

And something you’re liable to hear more about in the weeks ahead: the use of composites in aircraft construction. Carbon-fiber composites are strong and lightweight, but stress cracks, gaps and other defects can be very difficult to detect during routine inspections. Ever since the crash of American Airlines Flight 587 outside Kennedy airport in 2001, Airbus has been dogged by accusations that its reliance on composites renders its planes prone to structural failure under certain, albeit highly extreme conditions. In that accident, pilot overreaction to a wake turbulence encounter, together with an overly sensitive rudder system, resulted in the fracture and separation of the airplane’s composite vertical tail. Eight years later, mine weren’t the only eyebrows raised when television footage showed the tail of Air France 447, virtually intact but very much separated, floating in the Atlantic near Brazil.

Speculation, for now, is all we've really got. And it's entirely possible that we'll never know for certain what happened. Plane crashes, tragic as they are, are almost always instructive, leaving us with one or more valuable lessons. This one, maybe, not so much -- aside from the clear directive to give violent storms as wide a berth as possible.

- - - - - - - - - - - -

Do you have questions for Salon's aviation expert? Contact Patrick Smith through his Web site and look for answers in a future column.


By Patrick Smith

Patrick Smith is an airline pilot.

MORE FROM Patrick Smith


Related Topics ------------------------------------------

Air Travel Ask The Pilot Business