The island was already behind them.
The whole trip is a short hop, the kind of leg a crew flies several times in a morning without thinking about it.
Twenty one thousand feet over open water, the autopilot holding the heading it had been given, the airplane doing exactly what it was told.
On the ground, controllers were calling.
Other crews on the same frequency tried too, one after another, and for eighteen minutes the calls went out and nothing came back.
Both seats were occupied.
Neither man in them was awake, and the one in the left seat had been sleeping badly for years without knowing it or being told so.
What sleeps you without letting you rest
Obstructive sleep apnea is not insomnia. The sufferer sleeps.
The airway collapses repeatedly through the night, oxygen drops, the brain surfaces just far enough to reopen it, and the whole cycle repeats without ever reaching the surface of memory.
Morning arrives with eight hours on the clock.
What the body actually got was eight hours chopped into fragments, which produces daytime somnolence and, in severe cases, measurable cognitive impairment.
The cruel part is the self report.
A person with this condition will tell you honestly that he slept fine, because as far as he can remember he did, and there is no moment he can point to and call a bad night.
Why a medical exam does not catch it
A routine aviation medical checks the obvious things.
Vision, hearing, blood pressure, heart, a conversation about medications and a form the applicant fills in himself.
None of that finds a collapsing airway.
Diagnosis needs a sleep study, an overnight recording of breathing and oxygen, and nothing in the routine examination triggers one unless the pilot volunteers a symptom he does not believe he has.
So the certificate is issued, correctly.
The examiner asked the questions on the form, got truthful answers, and cleared a man whose sleep was quietly broken in a way neither of them had any way to see.
What the investigation established
The flight left Honolulu for Hilo on the morning of 13 February 2008, a hop between islands.
On board were 40 passengers, one flight attendant and the two pilots, in a regional jet at flight level 210.
Radio contact stopped at 9:40.
Controllers tried the frequency, then tried again, then asked other aircraft in the area to try, which is the standard sequence and it took a while.
It came back at 9:58, by which point the airplane was 26 nautical miles southeast of the airport it had been cleared to land at, still cruising.
The schedule behind it was relentless rather than dramatic.
Both pilots had reported for duty at 5:40 in the morning on three consecutive days, and the first officer was carrying a documented sleep debt of one to nearly three hours across the preceding 72.
They had been on duty under four hours when it happened.
What the finding actually said
Investigators named the cause without softening it.
The captain and first officer inadvertently fell asleep during cruise, and contributing to it were the captain’s undiagnosed obstructive sleep apnea and the crew’s recent work schedules.
He admitted something else as well.
He had often napped in the cockpit before, and he described that morning in terms anyone would recognize, the warm sun coming through the windscreen and his eyes simply closing.
The same shape keeps recurring.
Two pilots on an A320 slept for 28 minutes at 36,000 feet with 153 people behind them.
A watch officer who dozed off put a 443 foot ship onto a Norwegian shore, stopping 23 feet from a sleeping man’s bedroom wall.
What changed and what did not
The consequences landed on the two men.
Both were fired, the captain’s certificate was suspended for 60 days and the first officer’s for 45, which is the part of this story that is easy to arrange.
The recommendations went somewhere else.
What the safety board asked the regulator for was a proper fatigue risk management system, an approach that treats scheduling as a hazard to be measured rather than a roster to be filled.
The medical gap took much longer to close.
Screening for sleep apnea in pilots became a live argument for years afterward, and it was resisted precisely because it would catch people who feel fine and pass every other test.
Which is the whole problem in one sentence.
Nobody on that flight deck was doing anything wrong that morning, and the airplane still flew past its destination with nobody flying it.
The recommendation letter has the schedules and the ask. The incident report has what the captain said.
Read the whole thing?
Get the week's signal, not the noise
Our sharpest reporting on energy, climate and nature — free, once a week.
