The Crash Detectives

by Christine Negroni

  By the time the plane neared Antarctica, Flight 901 had been in the air for about four hours and forty-five minutes. This wasn’t normal air travel. Air New Zealand wanted to delight and awe its customers, so it staffed the flight with an Antarctic expert, who provided a running commentary. Sir Edmund Hillary, who along with Tenzing Norgay was the first to summit Mount Everest, was on some flights. On this one, though, was Hillary’s friend Peter Mulgrew, retired from the New Zealand Navy. Mulgrew had accompanied Hillary on an expedition to the South Pole in the late 1950s. Passengers moved around the cabin with glasses of champagne, looking out the windows and listening to Mulgrew pointing out the sights. They could mosey up to the cockpit, where an open door invited them to see the view head-on.

  Cameras clicked as the plane passed the ragged mountains of Victoria Land to the west. In total, more than nine hundred images taken on the flight were developed and studied, photos that would play an important role in exposing flaws in the official conclusion of what happened to Flight 901.

  Two hundred eighty miles north of McMurdo Station, Collins was invited by air traffic control at McMurdo Center (Mac Center) to fly down to fifteen hundred feet via radar, through a layer of clouds below him at eighteen thousand feet. This was good news: it meant that the passengers could get their first close-up look at the scene below. Then Collins saw the clouds breaking up, so he descended in clear air instead. On the way down, he made a modified figure eight. First he flew a three-hundred-sixty-degree turn to the west and then to the east, ending up once again aiming south toward McMurdo Station.

  After completing the second orbit, Captain Collins reengaged the autonavigation system to be sure he was back on the course the airline had provided and continued his descent. Had the DC-10 been on the track shown at the pilots’ briefing, this latest group of airborne explorers would have had the same experience as the travelers on earlier flights. But once Collins reactivated the automatic navigation, Flight 901 was locked onto a path that would take it straight into the side of Mount Erebus.

  The cockpit and flight data recorders recovered at the crash site reveal that the plane was flying at around fifteen hundred feet when the first alert sounded with Terrain pull up! Seconds later, Captain Collins instructed the first officer, “Go-around power, please.” The recording ends six seconds after the initial warning.

  There were two big questions for those assigned to investigate the crash: why was the plane flying so low approaching high terrain, and why had the pilots not seen the mountain in front of them? For Air New Zealand, the answer to the first question came within hours.

  As soon as they learned that the plane was missing, flight dispatchers Alan Dorday and David Greenwood looked at the navigation information given to Collins and compared it with the previous Antarctic flight. They saw that the two paths were different. Captain Collins was not routed over flat sea ice, as the previous flight had been, but directly over land, and not just land, but the highest terrain for miles around. The pilots were flying low over a mountain because they thought they were over the sea. It was that straightforward.

  Anyone who has ever screwed up probably understands wanting to escape the consequences, and it must be exponentially greater when lives have been lost as a result. People at the airline gave in to that impulse. It began with a decision to hide the route shift from investigators so that when the country’s chief air accident investigator, Ron Chippindale, left New Zealand for Antarctica the following day, he was unaware of the critical error that led the pilots to make their fatal descent.

  Accompanying Chippindale for the ten days of on-site investigation was the chief pilot of Air New Zealand, Ian Gemmell. Unlike Chippindale, however, Captain Gemmell knew about the change in flight path, because flight dispatcher David Greenwood had told him.

  Gemmell was by Chippindale’s side throughout the investigation, as a technical adviser. How much he influenced Chippindale’s early view of the accident is a matter of opinion. What is not disputed, however, is that materials discovered at the crash site were tampered with, and some things just went missing. The most intriguing item was Captain Collins’s small ring binder—and it was the absence of this item that led people to suspect that Gemmell was behind the disappearance of evidence.

  Forty-five-year-old former air force pilot Jim Collins was a notorious list maker and note taker. He took his ring binder with him in his flight bag whenever he flew.

  As they worked at the crash scene recovering bodies, two New Zealand police identification officers, Stuart Leighton and Greg Gilpin, found the ring binder not far from where Captain Collins’s body lay. The binder had about thirty pages, mostly blank, except for a few in the front, which contained numbers. Sergeant Gilpin characterized it as flight-related information. When the binder arrived in New Zealand, however, those pages were gone. No one could satisfactorily explain what happened until 2012.

  On his deathbed, Captain Gemmell told documentarian Charlotte Purdy that the airline had removed the pages. Purdy, whose uncle was the flight engineer on Flight 901, was producing the film Operation Overdue, an account of the retrieval of the victims from Antarctica. That someone whom Gemmell did not name removed the pages made sense to Purdy, but by then, all of New Zealand already knew about how the airline sought to cover up its mistake by keeping the route change secret and blaming the pilots for flying too low.

  Yet there on the ice in 1979, any indication that Captain Collins did not know about the route over terrain would have derailed that plan. This is why the question of Gemmell’s access at the crash site and his influence over Ron Chippindale was so important.

  Gemmell always insisted he did not know about the alteration of the flight coordinates until after he returned from Antarctica on December 8, but his coworkers told a different story. “I certainly told Ian Gemmell,” said David Greenwood, the dispatcher, when testifying about sharing his discovery with the chief pilot the morning after the crash.

  These conflicting recollections did not happen in a vacuum, and it wasn’t just Collins’s binder pages that went missing. His atlas and flight papers were never found, either. And in the days following the accident, items were taken from the home of First Officer Cassin. Anne Cassin told me her husband Greg’s body had not even been recovered from Antarctica when she returned home from an errand to find that her in-laws had handed over boxes of documents to an Air New Zealand pilot who had come to the house. “Insurance details, Greg’s time sheets, bills, receipts, personal letters, bank statements, check stubs, flying books—everything had gone,” she said, the memory still vivid more than thirty years later. “The person chosen to liaise with the dead air crews’ families and Air New Zealand stole every single item that I’ll call paperwork from my home.”

  When she realized what had happened, Anne Cassin went to the coffee table where her husband had left the folder containing notes on his Antarctica briefing. Most of the contents were gone. Anne Cassin was a thirty-one-year-old mother of three and a private pilot,7 but this early in the probe, she couldn’t imagine why details of her husband’s last briefing mattered to the airline. In February, it became clear when Air New Zealand’s mistake in changing the route to McMurdo without notifying the crew was leaked to a newspaper.

  Cassin and the pilots at Air New Zealand understood for the first time: The airline was saying the pilots knew they were flying over Mount Erebus. But the briefing notes indicated otherwise. That’s why they were being snapped up. Collins’s notes were gone and so were Cassin’s. And all this time, Air New Zealand’s chief executive Morrie Davis was ordering the shredding of crash-related documents.

  That odd request came within days of the accident, when the airline boss said that all original relevant material should be collected and everything else destroyed. Davis later explained to the lawyer for the airline pilots union that he was motivated by a desire to prevent leaks to the media.

  In June 1980, Chippindale, the chief air accident investigator, released his repor
t. The probable cause of the disaster was “the decision of the captain to continue the flight at low level toward an area of poor surface and horizon definition when the crew was not certain of their position.” The airline was cited for the inaccuracies in the briefing to the crew. The country’s Civil Aviation Division, or CAD, was told it should have paid more attention to how Air New Zealand was operating the flights. But it was the pilots who dominated Chippindale’s report.

  Regardless of whether the pilots thought they were flying over high or low terrain, everyone would have returned home safely if the crew hadn’t descended below sixteen thousand feet, Chippindale said. There was a certain logic to this, reinforced by what the airline told Chippindale: pilots were absolutely prohibited from flying below sixteen thousand feet on the Antarctic Experience flights until directly over McMurdo.

  It had taken Chippindale six months to come to his conclusions. One month later, a special Royal Commission Inquiry would begin examining all his findings and the information from the airline on which it was based. The national disaster was on course to become a national scandal.

  Peter Mahon, a Christchurch lawyer and longtime judge in Auckland, was appointed to conduct the evaluation. After ten months, the commissioner found fault with practically every aspect of the probable cause report. It was full of inaccuracies and supported by sloppy fact-finding techniques, Mahon found. Curious interpretations of the truth were exposed in Chippindale’s conclusions.

  In one example, Chippindale wrote that officials with McDonnell Douglas and Air New Zealand pilots told him that when approaching Erebus, the pilots would have seen the volcano depicted on the flight deck radar monitor, so they must have ignored the approaching mountain. Yet Bendix, maker of the radar, said the air in Antarctica was too dry for the mountain to be reflected on radar. When Chippindale was asked to provide names or notes to back up his contrary information, he could not.

  After a committee assigned to transcribe the cockpit voice recorder spent a week at the office of the NTSB in Washington, DC, debating every utterance and sound on the tape, Chippindale and Gemmell replayed the tape in Chippindale’s home and revised the transcript. Chippindale then went to the United Kingdom and had another go at the transcript alone, revising it yet again.

  “That’s just not done. You cannot do that,” aviation safety specialist John Cox told me. In his years as an accident investigator for ALPA, Cox has listened to and participated in the transcription of half a dozen cockpit voice recorders and calls them the most subjective aspect of the investigation. If Chip-pindale thought he’d heard something new or different in the cockpit voice recording, Cox says the process would have been to recall the entire CVR committee. “You sit down and say, ‘We’ve listened again. We think the following corrections need to be made.’”

  That did not happen. When the words were published in the final report, fifty-five changes had been made to what the CVR committee had earlier agreed to. Words and phrases no other contributor heard had been added, and an entire exchange the committee had agreed to had been excised.

  The all-important question was whether the pilots had truly been ordered to remain at sixteen thousand feet on approach to the continent. Controllers in Antarctica seemed to have been unaware of the restriction, because they cleared Flight 901 to fifteen hundred feet. Mahon was convinced the prohibition was never heeded because flights often operated between fifteen hundred and three thousand feet. Newspaper reporters who made the trip wrote about the low-level flights, as did writers for the airline’s marketing material.

  The official probable cause report and the report of the Royal Commission disagree on just about every key fact except one: Air New Zealand and the Civil Aviation Division, which was responsible for regulating the airline, did not take seriously the risks associated with flying in Antarctica.

  When the flights began, there were a number of requirements designed to mitigate the dangers. The crews were to be provided with flight charts on a topographical map. They were not. Two captains were to be on each flight, but that requirement was swapped for one captain and two first officers. All captains were to fly one flight with supervision. That rule was abandoned because the briefings were supposed to be good enough to eliminate the need for actual flight experience.

  Above all, the special circumstance ignored by CAD and Air New Zealand is the one that sent Flight 901 into the mountain: the failure to recognize and train the pilots for the singular nature of flying in Antarctica.

  People who knew Collins and Cassin to be thoughtful and conscientious aviators were outraged at the lopsided way the accident was attributed to the pilots. Still, everyone was baffled by one question: why steer a course into the side of a volcano? The day was clear: the photos captured in the cameras of the passengers revealed expansive landscapes, and sunshine to the east and west. Presumably, the view was the same out the cockpit windows, or Collins would not have continued to note that he was flying in visual conditions.

  In spite of that, the official accident report concluded that the pilots were flying in cloud and that descent was foolhardy. The CAD’s chief, E. T. Kippenberger, had a theory that “Captain Collins must have been suddenly afflicted by some medical or psychological malady, which made him oblivious to danger looming in front of him.”

  Ultimately, a more sensible answer, one in keeping with the evidence, came from Gordon Vette, an Air New Zealand pilot. Vette had been a captain on an Antarctic Experience flight. He’d flown with every crew member who died at Mount Erebus. He knew the pilots had been fooled into believing they were over McMurdo Sound. Reading the cockpit voice recorder transcript, he learned how the ground references on approach to Mount Erebus bore an unlucky and uncanny resemblance to what the crew would have seen on the McMurdo Sound approach. So he wondered if another trick had been able to hide from the crew a twelve-thousand-foot obstruction. Remarkable as it seems, the answer was yes.

  When Collins took the plane below the layer of overcast, the summer sun was behind the plane. Ahead were forty miles of unbroken white: the white of the sea ice stretching to the horizon, where it blended indistinguishably with a sky bleached white by the sun’s diffused rays above the unbroken clouds. The powerful effect of light meeting white eliminated the visual borders created by texture, shadow, and depth. Under these conditions, there would be no perception of the lines that separate one item of the landscape from another.

  Dr. Arthur Ginsberg, director of the Aviation Vision Laboratory at Wright-Patterson Air Force Base, confirmed Vette’s hunch that there was a reason for the pilots’ improbable and inexplicable flight into the side of a volcano. The flat white carpet in front of the plane rising at an inclination of thirteen degrees and then nineteen degrees would not have been perceived by the crew. A pilot not familiar with the illusion would have flown straight into it.

  Whiteout was not unknown at the time of Air New Zealand’s Antarctic Experience flights. In his report, Chippindale devotes a page and a half to it. But whiteout (and more specifically, sector whiteout, where visibility is affected in just one direction) was not part of the training of the Antarctic crews.

  Several movies and even more books have been produced dissecting the Erebus scandal. In Verdict on Erebus, his book about the case, Peter Mahon says that when he accepted the assignment, he thought Chippindale’s report to the Office of Air Accidents Investigations was solid. He went in thinking he’d rubber-stamp a well-researched, well-reported analysis, but he discovered that nothing was as it seemed. What Mahon, an aviation outsider, saw, like a mountain dead ahead, was a government-owned airline that had engaged in what he ultimately called an “orchestrated litany of lies.”

  Mahon did not have the final word on Air New Zealand Flight 901. Chief Inspector Ron Chippindale wrote a rebuttal, claiming Mahon’s finding “abounds in errors,” and to this day, two conflicting reports account for one disaster.

  How different is that really from the Arrow Air crash in Canada, where neither probab
le cause was considered credible, or the Hammarskjöld accident, investigated four times and still considered incomplete? These troublesome investigations show that the search for truth does not always result in certainty, and that ambiguity may be the best cover-up of all.

  When the Metrojet flight from Sharm El Sheikh in Egypt to St. Petersburg, Russia, crashed on Halloween 2015, confusion reigned once again. Metrojet executives immediately announced that nothing was wrong with the plane or the pilots, and British prime minister David Cameron offered that it was likely that a bomb brought down the airliner. Under international agreement, it fell to the Egyptians to lead the probe into what had happened, and for months they said it was too soon to say.

  That didn’t matter, because a deluge of stories concluded that terrorism was involved; even a front-page story in The New York Times said the Egyptians could not be trusted to carry out an impartial investigation.

  A few days after the crash, I spoke to National Public Radio’s Robert Siegel on the program All Things Considered. I told Siegel this should be no mystery: all the tin was there; the flight data and cockpit voice recorders had been recovered. What I failed to consider were the external factors, how each government would posture.

  The Russians were responsible for supervising the safe operation of the airline. Had they failed to do so? European airlines were frequent users of the Sharm El Sheikh airport. Had they paid enough attention to the security there? Egypt had the most at stake; its economy is largely reliant on tourism. Had it been diligent enough protecting the safety of the tourists at its airports? Would it bamboozle, as the Times incautiously suggested, or were its investigators acting responsibly by holding off on drawing conclusions?