There are only a few ways this can be explained: there was a power failure on the airplane, the software failed, or something interfered with the connection between the antenna and the satellite, such as the plane flying upside down so that the fuselage was between the antenna and the satellite. All three possibilities are extremely remote. Some clues, however, have not been pursued.
One week into 2008, a Qantas Boeing 747 was on approach to Bangkok from London with 365 people on board. It was a clear and sunny afternoon—which was fortunate because as Qantas Flight 2 passed through ten thousand feet, it lost electrical power. The autothrottle, autopilot, weather radar, and many other systems, including the automatic control for the pressurization system, simply stopped. Only the captain’s flight display worked, albeit in “degraded mode.” The plane landed safely, but once it was on the ground, its doors could not be opened because the outflow valves failed to automatically release the cabin pressure.
On the Boeing 747 and other Boeing jetliners, including the 777 and the 767, there is a galley located above the electronics and equipment room, called the E&E bay. On Qantas Flight 2, a flood from the galley above caused water to flow into this area. This was not a one-time event. During its investigation, the Australian Transport Safety Bureau discovered that electronics equipment in the bay had been “repeatedly subjected” to liquid beyond what it was designed to handle. When the ATSB set out to find similar events, it turned up five on large jetliners, four on Boeings, and one on an Airbus A300—and those were just the ones serious enough to have caused a safety event in flight.
I’ve learned that on airplanes with galleys located above electrical equipment, mechanics often see leaking.
“The 777 has an avionics bay below the first-class galley. When a crew reports water, it is required by the manual to inspect the avionics bay for leaks from water penetration,” I was told by a mechanic for a major American airline—I’ll call this mechanic Fred because he does not want me to use his real name. A few days after Fred told me this, he sent a video in which I could clearly see water dripping onto the floor of a cramped and noisy equipment room.
“Where did you get this?” I asked, thinking Fred had found the footage on a YouTube-like service for aircraft mechanics. But no, he’d shot the video himself, on a Boeing 767 that came under the care of his wrench shortly after our initial conversation about Qantas Flight 2.
I started to think that maybe some water-induced intermittent electrical problem could have produced the various failures on Malaysia 370, including the puzzling power down and subsequent restoration of communication at 2:25 a.m. that no one has yet been able to explain.
So I asked the ATSB, when Flight 2 lost most of its power on January 7, 2008, did it cause a termination in the link to the satellite? Could this issue with water damaging the electronics affect satellite communication? The ATSB did not know.
“I am unable to provide you with a definitive answer as we would need to establish a detailed understanding of the load-sharing arrangements on the aircraft, interaction with AC BUS 4, not to mention the electrical system that supports the SATCOM,” Julian Walsh, the acting chief commissioner, replied in an e-mail. “This was not an aspect of the original investigation,” he told me.
I do not know if it is part of the Malaysian investigation into what happened to the plane, which had the registration ID 9M-MRO, because the team doing that investigation does not answer questions.
Once Malaysia 370’s last radar echo faded—the one showing it somewhere at the northern tip of Sumatra—Fariq made a final turn. No data suggest when, but the plane turned south and flew on for five hours more until it ran out of fuel. This final turn is the point where I believe Fariq’s deprived brain reached its limit. Like Number 14 fixed on the four of spades, Fariq was locked onto some thought. I asked airline captain Pete Frey to try to explain Fariq’s last action. I wasn’t intending to insult Pete, a longtime friend, by suggesting he might know what it is like to be without cogent thought, and thankfully, Pete didn’t take it that way.
“Who knows what he’s doing? He doesn’t know what he’s doing,” Frey said after considering my question. “He’s lost sense of time, so now he thinks he’s back there. Maybe he’s thinking, ‘I’ve got to head north, and where am I? I’ll go this way.’ By the time he realizes he’s lost, he says, ‘Now I’ll turn around and go back, but I don’t know where I’m going back to, so I’ll just head south. I’m too far north.’”
When you consider how muddled Fariq’s mind must have been, you can see many ways in which MH-370’s bizarre flight path can be explained.
“All you really have to say is at this point,” Frey told me, “he’s struggling with intermittent abilities, and it’s not enough.”
Center of Confusion
Commercial aviation is both more and less advanced than people think. The pilot’s preflight programming can enable the machine to take off, fly, and land, though any pilot will tell you there’s a lot more to the job than just getting the airplane into the sky and back down again (as you will see in part 5 of this book).
“If we have it set up for where the plane is going to fly the flight plan, we could go to first class and have a meal, and it would do those things,” said James Blaszczak, now retired, who flew the Boeing 777 for eight years for United before going on to fly the 787 Dreamliner. Yes, that’s pretty impressive. Yet at any hour around the world, hundreds of planes are flying isolated in the sky, communicating only sporadically with the ground. In this respect, MH-370 is more like the Hawaii Clipper.
The 1930s-era, four-engine Martin 130 flying boat was robust and comfortable enough that up to thirty-two passengers, seated in bamboo and rattan chairs in the lounge, could be served hot meals prepared in the galley by uniformed and gloved stewards. After dining, they could retire to berths made up with blankets and pillows.
A nine-man flight-deck crew was responsible for operating the flight that would get travelers from California across the Pacific to Manila, a five-day journey with overnight stays in island hotels built and operated by the airline.
The crew consisted of a captain, four copilots, a flight engineer, a radio operator, a navigator, and a pilot studying basic navigation from him. Navigation was by celestial observation combined with a calculation of elapsed speed and time, the dead reckoning Pete Frey explained earlier. The navigator could get an assist by using a direction-finding system, antennas that locked onto broadcast radio stations.
“We would tune in certain stations we knew we could use. In those days, one of our favorite things were the high-powered commercial stations along the California coast,” said Ed Dover, a radio operator with Pan Am from 1942 until 1948, who later spent thirty-three years as an air traffic controller. “KGO broadcast such a strong signal we could hear them out to sea.”
Using an antenna shaped like a figure eight, radio operators like Dover could note the areas of strongest and weakest signal and use that information to determine the direction of the station. “You could draw a line on the map. We knew where the station was; on land, the transmitter was on the map, so we could match the direction in terms of compass direction and say, ‘Okay, down that way, that’s where we are in relation to where the station was.’” You have to love the term used when navigators and radio operators shared their information to ratify their calculation of position: they called it the center of confusion.
If that kind of direction finding seems like something from the Stone Age, communication technology wasn’t much better. The weekly Pan Am transpacific flight from San Francisco to Manila left California on a Wednesday in coordination with two ships leaving on Monday, one from San Francisco, the other from Honolulu. The ships would be midway through their cruises when the Pan Am Clipper passed overhead, giving the airliner a navigational fix and, as significantly, a degree of comfort that someone—even someone eight thousand feet below, in a dark ocean—knew where the airliner was.
Still, darned if those big a
irships didn’t land right where they should have, without a single computer assist.
It is easy to be dismissive of the early rudimentary systems when air travel began. But that a twenty-six-ton flying boat could take off from San Francisco and arrive in Honolulu eighteen hours later was nothing short of a miracle. It was state of the art, as they say.
So people were shocked to learn, four generations later, that with all their fancy navigation and communications equipment, many airlines were still in the dark when it came to knowing exactly where their airliners were when crossing the oceans. As with the Pan Am Clippers, the challenge for overwater flights remains that airplanes are outside the range of land-based radar. If they are going to communicate their position, it will have to be via satellite, which is expensive. As Daniel Baker, the founder and CEO of FlightAware.com explained, the “center of confusion” has given way to the “cone of ambiguity.”
An airline sending a position report via satellite every fifteen minutes can cover one hundred fifty miles before sending the next position report, Baker said. Should the location be more precise?
“Satellites don’t know where the airplane is. The airplane sends the signal, and that requires the airplane be pointed in the right direction, that is, belly down. If there is an upset on board—a loss of control, the plane is headed straight down or upside down—it can’t get a satellite signal, because it is pointed in the wrong direction,” he said, adding, “That’s a big challenge.” And just in case you are thinking, as I did, that he was talking about really unlikely possibilities, he ticked off a few disasters to illustrate his point: EgyptAir Flight 990 and Air France 447. “We are at the limits of technology,” he said.
Acausal Connections
Seventy-six years separate the Hawaii Clipper and Malaysia Flight 370, and yet we see striking similarities. Both airliners were modern and spacious, and the pilots in command were highly trained and experienced. Upon closer look, however, we see that maybe all wasn’t as it seemed. Was Leo Terletsky, forty-three, “one of the most distinguished flight commanders,”3 as Pan Am claimed—or was he afraid of flying and so volatile that most pilots were unwilling to work with him, as Horace Brock wrote?
Pilots who flew with Zaharie Ahmad Shah, captain of Malaysia 370, said he was a passionate aviator and a mentor to younger pilots. “A gem of a guy, a real professional, enjoying the best time of his life,” one of them told me. Yet a few journalists—quoting unidentified sources, mind you—painted a picture of a political fanatic.
Pan Am’s Martin 130 flying boat was a marvel of aviation; custom-built to help the airline span the oceans, but the company’s chief pilot in the thirties complained that it was “unstable on every axis and a pig to fly.”4 While the Boeing 777 airliner is widely used and considered by pilots to be pleasurable and reliable, the list of things that could have contributed to the 2014 disaster exposes unappreciated hazards on the airplane.
If the Pan Am Clipper experienced something catastrophic, the area where it could have gone down is relatively small. If it was hijacked, though, the search zone becomes enormous, because the amount of fuel on board allowed the plane another eleven hours of flight. With Malaysia 370, the plane’s satellite indicated the Boeing 777 flew for five hours and fifty-four minutes after mysteriously powering up at 2:25, so predictions about where it ultimately came down could be no more precise than an area from sixty thousand to six hundred thousand square miles.
On a stormy night above the Atlantic Ocean on June 1, 2009, an Air France Airbus A330 with 228 people on board disappeared on a flight from Rio de Janeiro to Paris. Though the plane was five hundred miles outside radar coverage, investigators had a last-known position based on an ACARS message transmitted via satellite. The plane must have entered the water within forty miles of that. Even so, it took five days for the first bodies and floating debris to be discovered, and two years to locate the plane.
Some of the credit for finally finding the submerged airliner goes to Metron Scientific Solutions, a company staffed with pencil-wielding mathematicians who used probability, logic, and numbers to conclude that the likely resting place of the plane was a narrow slice of ocean that had already been checked.
“A lack of success tells you about where it is not, and that contributes to knowledge,” said Larry Stone, chief scientist at Metron. Talk about having a positive point of view. The Metron method is based on Bayesian probability, the theory of eighteenth-century statistician and philosopher Thomas Bayes, whose first published work, Divine Benevolence, was equally optimistic because it attempted to prove that God wants us to be happy.
Using Bayesian logic to look for missing airplanes, as interpreted by Metron, involves taking all kinds of input about the missing thing (even conflicting input) and assigning levels of certainty or uncertainty to each. Everything gets a weight, and everything gets revised as things change. New information, as Stone so cheerily described it to me, is often negative.
In the search for Air France 447, scientists, mathematicians, and underwater technologists were involved in a very difficult bit of detective work. They covered a surface area of eighteen thousand square miles and a debris field nearly three miles deep. Two search seasons after they started, the wreckage of the Airbus was found on the edge of a plain not too far from the beginning of a steep and rugged underwater mountain range.
Many smart people contributed to the search but Olivier Ferrante, then the investigator in charge of the Air France 447 probe for France’s Bureau d’Enquêtes et d’Analyses, said they benefited from an additional, highly uncertain element: luck. “The fact that the airplane was on flat terrain” was important to seeing the debris on the sonar pictures, Ferrante told me. “We saw man-made debris, and we identified it in the picture. That was luck. If it had been a couple of kilometers to the east or north, or close to the cliff, we wouldn’t have seen it.”
I bring up the Air France search in this context because it is the event most like Malaysia 370 in terms of how the newest technological developments are being pushed to do even more within the cone of ambiguity, defined earlier by FlightAware’s Daniel Baker as the number of miles a plane can travel between satellite position reports.
While the ACARS messages helped to narrow the area where the Air France 447 airliner might be located, the difficulty in getting a more precise location prompted the satellite company Inmarsat to beef up some of its network by adding new data to the communication transmission. Two additional tidbits of information enable calculations of a plane’s location based on how long it takes a message transmitted from the plane to arrive at its destination.
“Inmarsat did modify its systems to add the so-called timing and frequency information to the handshake messages,” Ruy Pinto, an engineer and the chief operating officer at Inmarsat, told me at the company’s futuristic high-rise headquarters in London. This newly added information would become useful the weekend Malaysia 370 went missing. First, it showed that the plane had not crashed right away, but had flown for hours. Later, the timing and frequency data allowed the company to determine that the plane flew south into the Indian Ocean.
“If MH-370 had occurred at the time of the Air France disaster it would not have been possible to make the analysis that we ended up doing,” Pinto told me.
What’s missing with Malaysia 370 is even the basic information used by the French, because in the case of MH-370, no ACARS messages were sent after 1:07 a.m. This meant the search area would be massive.
It would be nearly a year and a half before the first debris from Malaysia 370 was found on a beach on Réunion Island, off the coast of East Africa. By then, the wing section had traveled too far and arrived too late to provide any clue as to where the airplane landed in the water. At its smallest, the search area is three times larger than the one in which Air France 447 was found.
Still, the discovery of the wing flap was useful in one way: it shut down the theory that once it disappeared from radar, the plane had turned north,
toward Asia and the Caucasus Mountains. One of the more popular proponents of that line of thinking was Jeff Wise, a CNN talking head who wrote a book, The Plane That Wasn’t There, describing an elaborate plot that required dismissing some of the data from Inmarsat.
“All the inexplicable coincidences and mismatched data went away,” Wise wrote about his alternative scenario, getting a big spread in New York magazine. “The answer became wonderfully simple.”
He was not the only person who thought the plane was hiding in a remote part of the world. Thomas McInerney, a retired lieutenant general and military analyst for Fox News, told the network’s morning news program in 2015 that the plane could be in the “the Stans,” referring to the countries whose names include the suffix -stan. “That airplane can fly nonstop from the Stans to the United States, New York, or Washington, DC. It could be a future trigger for events against the country.”
I’ll leave that kind of worrying for Fox News watchers. I’m more concerned with a disturbing discovery made while working on this book: for all the apparent effort to try to solve the mystery of MH-370, authorities may not be as committed to finding out what went wrong. That also would not be unprecedented.
Cover-Up
Air crashes have the potential to reveal government secrets and failures, company malfeasance, or all the above. This is even more pronounced if the airline is owned by the government, as is the case in Malaysia and many other countries.
In the case of Malaysia 370, the airline seems to have had a most embarrassing secret to keep: that before the plane flew off into oblivion, the company already knew it was unable to track its airplanes as frequently as required. After the plane went missing at the end of March 2014, twenty-six countries were donating personnel, aircraft, and ships to look for the jet. What would they say if they knew that a year earlier, Malaysia Airline executives had been warned about just this kind of problem? In fact, they had.
The Crash Detectives Page 5
