The MH370 Miracle (updated)

If after nearly five years the disappearance of MH370 is still regarded as an unsolved (and perhaps unsolvable) mystery, that’s because something that happened in the course of MH370’s vanishing is generally talked about as if it were unremarkable when in fact it is ridiculously unprecedented to the point of being virtually impossible. And if that fact could be more generally understood, the case would seem a lot less mysterious.

Call it the MH370 miracle.

OK, back up. Here’s the story of MH370 in a nutshell: a plane takes off and vanishes from air-traffic control radar. Weeks later, it turns out that the plane had reversed course, flown through an area of primary radar coverage, and then vanished from that. It’s gone. It’s dark. Off the grid. There is absolutely no way that anyone is ever going to know where this plane went.

Then a miracle happened. Something that has never happened before in the history of air travel and in all likelihood will never happen again. It’s this: three minutes after disappearing from primary radar, the plane began sending out a signal. A signal with unique and wonderful properties.

A Miracle Signal.

The general public has never heard about the remarkableness of this occurrence. It has been glossed over entirely. The ATSB and the mainstream press talk about the signal as something generated as a matter of course, like the cell phone data carelessly left behind by a fugitive criminal. But the Inmarsat data set is not like that at all. Not only was it not normal, it was unprecedented and produced in a way that cannot be explained.

Those of you have have been following the case know what event I’m talking about. At 18:25 UTC, MH370 starts sending signals to one of the satellites in the Inmarsat fleet: Inmarsat-3 F1, aka IOR, hovering in geostationary orbit over the equator at 64.5 degrees east.

The standard story goes like this: “Then scientists studying the signal realized that it contained clues about where the plane went.”

What gets omitted is how completely bonkers this is. In fact, there are two insane things going on here.

The first is that the signal exists at all. As far as I know, never in the history of commercial aviation has a plane been flying around with all of its communications equipment turned off, except for this one piece of gear. It. Just. Never. Happens.

Not ever.

Now at this point you might say, “But these were extraordinary circumstances, the plane had just been hijacked.” Okay, sure. But the point is that in order for this signal to exist, the hijacker/s must have done something extraordinary to the electrical system, something that pilots never normally do, that isn’t called for in any checklist, and for which there is no rational explanation that anyone has been able to come up with. And yet—voila! There it is, right at the exact moment it’s needed.

The second insane thing about the Miracle Signal is that it just so happens that this fortuitously appearing signal has embedded within it revelatory information—the BFO. Now, under any but miraculous circumstances, you would not expect this signal to tell you anything about the planes position or velocity. It’s a communications signal. It’s designed specifically to not have any navigational information. But lo and behold, it turns out that because of a quirky convergence of happenstance, the system is working in an unusual way that does indeed provide a hint—but only a hint!—of where the plane is going in a way that cannot be cross-checked with any other source.

This is the astonishing convergence:

1. The plane is equipped with a piece of equipment called an SDU that was manufactured by Thales. If this box had come from the other leading manufacturer, Rockwell Collins, there would have been no navigational information in the BFO data.

2. The plane is flying under the footprint of a satellite that is past its design lifespan and has run low on the fuel it requires for stationkeeping and so has started to wander in its orbit. If it had been functioning as designed then there would be no navigational information in the BFO data.

3. The subsequent path of the plane is along a north-south axis. It turns out that the information can’t be used to tell you where the plane traveled with any precision where the plane went, but it can tell you whether it went north or south.

4. The path lies entirely under that one satellite’s area of coverage. If the path had crossed over into another coverage zone, the connection would have been transferred over and the nav information would have been lost (but the direction of flight would have been unambiguously confirmed). For instance, if MH370 had flown east at IGARI instead of west, it would have flown into the coverage zone of POR, aka Inmarsat-3 F3, and re-logged on with that satellite.

5. The path lies entirely over water. If the plane had turned south earlier, or had headed east instead of west, it would have passed over land and either potentially been spotted or detected on radar.

6. MH370 used a satcom service called Classic Aero. If it used a newer, higher grade of service called SwiftBroadband, the transmissions between the plane and the satellite would have included position information.

So when we talk about MH370, and how the plane went into the southern Indian Ocean, what we’re talking about are six hours of data whose existence is nearly as miraculous as a little baby Jesus lying in a manger.

Imagine if the authorities had announced back in March, 2014: “We’ve just realized that after it disappeared from military radar, the plane waited three minutes and then started broadcasting a Miracle Signal. This Miracle Signal has never been seen before, and will never be seen again, and it happens by a crazy quirk of circumstance to provide us a super intriguing clue, but instead of questioning how it came to be, we’re going to just accept the data it’s giving us and treat it as unimpeachable fact.”

Oh, and that’s not the end.

The really gobsmacking thing about the miracle signal is that it took place in the context of a bunch of other equally unlikely things. Namely:

— If the data are accepted as face value, the only explanation for the plane’s behavior is that the captain hijacked his own plane. That means a man with no manifestations of stress or mental illness spontaneously decides to commit mass murder/suicide.

— On top of that, he decides to do it in a way that no suicidal pilot ever has: by waiting impassively hour after hour until his fuel tanks to run dry.

— Once his fuel tanks run dry, he dives the plane toward the ocean, thinks better of it and glides it in what just happens to be the right direction, and then dives it into the ocean once more. This sequence of events is psychologically implausible but is the only way to explain how the plane could have ended up outside a seabed search area the size of Great Britain.

— The debris then floats in a way that is not reconcilable with any known drift model. Despite being adrift for over a year, none of it picks up any biofouling organisms more than a few months old, and the flaperon picks up goose barnacles that somehow manage to grow in the open air.

I’ve always said that the 18:25 reboot is the crucial clue that lies at the heart of the MH370 mystery. What I’m saying now is that it should be understood in even stronger terms: the extreme improbability of the Miracle Signal means that it can’t be construed as an unintentional byproduct of a “normal” suicide flight. It just could not have occured that way by happenstance. It must have been engineered.

One of the most common things you hear from normal people (by that I mean non-obsessives like present company) about MH370 is, “I just can’t believe that in this day and age a modern airliner could just vanish.” Of course, they’re absolutely right. Things don’t just vanish, except in one context: magic. Magicians make rabbits disappear out of hats, then make coins disappear behind kids’ ears, they make themselves disappear behind clouds of smoke.

If you don’t like the idea that something inexplicably miraculous is the handiwork of a magician, then your other option is to suppose that events have been arranged by sheer luck. Indeed, every “innocent” explanation that anyone has proposed to explain the vanishing of MH370—like a pilot suicide scenario, or a lithium battery fire, or accidental depressurization—assumes that the fact that the plane was never found is due to an incredible chain of coincidences.

And sure, bad luck happens in life, but once the odds get astronomical—when you start having to start calculating the odds that a rabbit could spontaneously teleport out of a top hat—then it’s time to start thinking about possible sleights-of-hand.

Here’s a historical analogy. In May of 1942, a Japanese fleet invading New Guinea was attacked by US aircraft carriers in the Coral Sea, suffering heavy damage. How, just half a year after Pearl Harbor, had America’s thinly stretched naval forces managed to intercept the Japanese task force amid the vastness of the Pacific? There were two possibilities. Either the Americans had just gotten lucky, or they had managed to break Japan’s naval cipher, JN-25. The former was a stretch, but the admiralty was certain that the Americans couldn’t have broken their code. A mentality later branded as “Victory Disease” convinced them that they were vastly superior to their enemy. They themselves couldn’t imagine how to break their most sophisticated code, so there was no way the Americans could have done it. The Japanese Navy had nothing to fear.

Then, bad luck hits again. As the Japanese carriers are moving against Midway Island, lo and behold, the beleagured American fleet not only shows up but gets a jump on them, sinking all their aircraft carriers and turning the tide of the war. How lucky could those gaijin get? Apparently really lucky, because the Japanese leadership didn’t understand their codes had been broken until they’d signed their surrender on the deck of the U.S.S. Missouri.

The search officials tasked with finding MH370 were in the same camp as the wartime Japanese. As far as they were concerned, it was inconceivable that they were dealing with an adversary capable of outwitting them. When I asked Mark Dickinson, vice president of satellite operations, how Inmarsat could be certain that the MH370 data hadn’t been tampered with to mislead investigators, he dismissed the idea out hand, saying: “whoever did that would have to have six month’s worth of knowledge of what would happen, in essence have to know how the data would be used.”

To be fair, some among the Japanese leadership were suspicious of the Americans’ good luck all along. And in the case of MH370, some of us have long smelled a rat. Earlier this year David Gallo, the man who found AF447, wrote, “I never accepted the satellite data from day one,” adding: “I never thought I’d say this….I think there is a good chance that MH370 never came south at all. Let’s put it this way, I don’t accept the evidence that the plane came south.” And this fall we learned that investigators conducting the last extant investigation into the disappearance of MH370 are looking into the possibility that the Inmarsat data could have been hacked.

So far, these skeptics are still in the minority, but I think that their numbers will continue to grow. A more people become aware of the circumstances of the MH370 miracle, the penny will continue to drop.

UPDATE 12/23/18: It seems to me that mysteries can be divided into two categories. 

The first I’ll call mysteries of indeterminacy. When Amelia Earhart and Fred Noonan departed Lae Island on July 2, 1937, they were flying a primitive aircraft, by modern standards, and relied on the most rudimentary form of navigation. It’s no wonder that they never made it to their intended destination. What we don’t know, and may never know, is where exactly in the western Pacific they crashed.

The second type I’ll call mysteries of inexplicability. When a magician puts a ball into a closed fist, then reopens it to reveal that nothing is there, you’re astonished at how he could have done it. 

In science, the question of whether an unknown planet X lurks at the edge of the solar system is a mystery of indeterminacy. The struggle to reconcile quantum mechanics and relativity is a mystery of inexplicability.

MH370 started out looking like a mystery of indeterminacy. The authorities had a good data set in hand, and developed an analytic method to generate a search area. They were extremely confident that, while they didn’t know exactly where the plane had landed, a few hundred million dollars worth of brute-force seabed scanning would give them an answer.

They turned out to be wrong. The plane wasn’t there. So now we understand that what we’re really grappling with is a mystery of inexplicability. There simply are no simple, widely-accepted explanations for how it could be that the plane wasn’t found. This kind of problem needs to be tackled in a fundamentally different way.

230 thoughts on “The MH370 Miracle (updated)

  1. That whole press conference seems iffy to me. Why would a fisherman come forward nearly 5 years later to reinforce a southern arc crash hypothesis? Why are they so desperate to ‘push’ his sighting by rubber-stamping it with an oath on the Koran? How much… (cough)… did he get paid for all this acting…? It just seems all very staged imo

    And where is the full press conference from start to finish? (We only have a 2min clip edited for the papers)

    In contrast, Raja Dalelah was immediately rubbished but this random dude fished in all the way from Indonesia 5 years later…

    Seems a lot like smoke and mirrors to reinforce the idea MH370 crashed in the sea…

  2. @Gysbreght

    I agree with everything you said in your last message to me and even your statement that ” when the fuel tanks are empty there is no fuel to shift around.” You assume that all of the tanks are empty because the 7 hours duration is reasonable for the initial fuel load. But you really can’t know whether there was 15 minutes of available fuel after 00:19. I agree this is not likely.

  3. @Hank McGlynn: “But you really can’t know whether there was 15 minutes of available fuel after 00:19. I agree this is not likely.”

    Thank you. Glad we agree.

  4. The reason magic works is that nobody thinks you will go to so much trouble.

    A rough quote from the TV series Jonathon Creek but an interesting thought for the MH370 miracle.

  5. @PSOher, Indeed! I’ve thought a lot about this quote from the magician Teller:

    2. Make the secret a lot more trouble than the trick seems worth. You will be fooled by a trick if it involves more time, money and practice than you (or any other sane onlooker) would be willing to invest. My partner, Penn, and I once produced 500 live cockroaches from a top hat on the desk of talk-show host David Letterman. To prepare this took weeks. We hired an entomologist who provided slow-moving, camera-friendly cockroaches (the kind from under your stove don’t hang around for close-ups) and taught us to pick the bugs up without screaming like preadolescent girls. Then we built a secret compartment out of foam-core (one of the few materials cockroaches can’t cling to) and worked out a devious routine for sneaking the compartment into the hat. More trouble than the trick was worth? To you, probably. But not to magicians.


  6. @PSOther @JeffW
    “The reason magic works is that nobody thinks you will go to so much trouble.” Well it does indeed seem like someone put a lot of thought and planning into the MH370 disappearing act. On that point I can agree with Penn and Teller. Let’s ask Penn and Teller who they think did it – probably does not help but let’s see what they think.

  7. @TBill, I reached out to Teller, got through to his handlers, requested an interview, got denied. I’ve been thinking about reaching out to another magician who could talk about this general point, i.e. the disappearance of MH370 as a feat of stage magic, but wasn’t sure it would move the ball much.

  8. For those people who are of the firm opinion that a triply-redundant 777 could not possibly suffer a complete loss of main bus power, the incident of the LATAM 777-300 last month (Dec 2018) will make interesting reading.

    The right backup generator failed and tripped the rest of the system, including the APU output. Only the standby busses powering one VHF radio, some displays and the emergency lighting remained. It landed on battery/RAT power.

    Allegedly, it was not a failure of power generation but of distribution – the breakers tripped. Once the engines were shut down after landing, power returned.

    “On Dec 22nd 2018 The Aviation Herald learned that the electrical problems started with the failure of the right hand backup generator, as result both transfer and both converter circuit breakers popped leaving all electrical busses unpowered although left and right main generators as well as their generator controls, the APU and the left backup generator remained operative (see ELEC page below, the lines to the L MAIN and R MAIN should be green not white and these symbols should be green, too). As a result almost all systems except a few connected to the standby busses lost power, only systems remaining powered are: stand by busses (powering one VHF radio) and emergency lighting, the left inboard and outboard and upper center displays as well as the left CDU remain also powered (according to FCOM via the batteries and/or RAT – with the RAT operating also the right hand – first officer’s – displays are being powered).

    On Dec 23rd 2018 the airline reported that it was not a failure of power generation but a failure of power distribution. The causes of that failure are now being investigated by CENIPA. After landing, when the crew shut the engines down, power returned, as result the cameras came back online and the crew could see that there was no fire on the landing gear or excessive heat on the brakes, hence the crew queried the need for an emergency evacuation as suggested by emergency services.”

    Not saying this is what happened to MH370, but it shows there are paths to circumvent the triple redundancy that Boeing haven’t imagined possible.

  9. @Scuttle, Really interesting, thank you. I’ve heard speculation that, instead of isolating the left AC bus, the perps might have deliberately turned off all electrical power, though I think it would be hard to come up with a rationale for this.

  10. @Jeff Wise: Try Andrew Mayne. He’s a magician and co-host of a podcast called “Weird Things”. Sounds like this might be right up his alley.

  11. Hi Jeff, What if the both variables of input for Inmarsat to base their calcultions on were wrong. The military admitted In a press confrence that the military radar was an unidentified blob that was seen when looking back at recordings and that they wern’t saying it was MH370. If this unidentified blob wasn’t MH370 their wouldn’t be a starting point for Inmarsat’s calculations. The second variable – speed, could be greatly reduced if the planes right engine had failed. Nearly all the debris that has been found is from the right side of the plane, which makes me believe the debris could be a result of engine failure.

  12. @Matt Betts, I don’t think there’s any plausible reason to doubt that the primary radar track corresponded to MH370, as it matches the secondary radar at one end and the Inmarsat data at the other. The high speed of the track is not consistent with a plane with one engine out.

  13. New analysis of Hydroacoustic signals from CTBTO hydrophones indicate possible MH370 crash location MUCH closer to DIEGO GARCIA
    Scientist based in UK, who previously analysed the underwater sounds picked up by hydrophones located near Perth, Australia & Diego Garcia, BIOT, has now reevaluated the signals and suggests the location of the crash could have been much closer to DIEGO GARCIA.
    Some interesting points:
    – completely ignored the BTO arcs and only the timeline considered
    – the data matches locations much closer to Diego Garcia
    – the locations correspond to areas where military action may have been taking place (Diego Garcia is a US/UK military base)
    – mysteriously, 25 minutes of the hydrophone data is missing

  14. @CliffG

    “completely ignored the BTO arcs and only the timeline considered” + “mysteriously, 25 minutes of the hydrophone data is missing” sounds like the data set is being massaged to suit a hypothesis rather than the other way around.

    Of course part of the problem is the veracity of the data set and whether some of it is there to make us believe a sequence of events at variance with the truth. Only delving into the minds of the perpetrators would answer this question.

  15. As much of a stretch as that BFO/BTO ping data is, the notion that it was deliberately spoofed is even more of a stretch. Only people within Inmarsat would have known how to properly spoof the signal to cause a convincing deception. And those people would have had a much better way of producing the deception: just plant the data.

    There is an article worth reading for anyone who wants to understand Inmarsat’s involvement in organizational terms:
    The decision to include the BFO/BTO data in the recordings was made by Inmarsat. Here’s a “satellite today” article that describes Inmarsat’s involvement:

    To be clear, I do not think that the signal was spoofed – nor do I think that the data was planted. But planted is within the realm of possibility, spoofing is not.

    Also, if it was planted, the deception would be only that the plane stayed airborne until it ran dry of fuel and was heading generally south. The analysis that went beyond those conclusions involves presumptions about how the plane was flown (or not flown), deep knowledge of the 777, and using data that is, at best, at the limits of its precision.

  16. @Scott, Thanks for the link. I continue to be baffled by people who blandly assert that a hypothesis is not possible simply because in their gut it strikes them as unlikely. If you have some reason why spoofing is not “within the realm of possibility,” please share it.

  17. @ ScottO
    The following excerpt from the article you linked is quite enlightening:

    For certain spectra, maintaining a geostationery satellite in a precise position relative to the Earth is essential. Fortunately for Inmarsat’s L-band services, this degree of precision is not essential in terms of the speed or reliability of the voice and data connection provided to a user at sea, on land or in the air. Due this “pointing tolerance,” Inmarsat is able to extend the life of its L-band satellites by using less fuel for station keeping.

    This older Inmarsat satellite which received signals from MH370 on the L-band has a ‘global’ beam, instead of a ‘spot’ beam required for other spectra, hence the socalled “pointing tolerance”. Plus, these satellites are in geostationary orbit which is the highest orbit (around $36000 km).
    So Inmarsat has received data from a satcom terminal with very relaxed pointing requirements, and which is located nearly $36000 km away.

    In many of the papers written by Inmarsat staff, they have validated the assumptions for their analysis of BFO with data from other aircraft whose positions were known, and arrived at a ‘nominal’ distance of the aircraft from the satellite.

    But there appears to be no mention of validating the BTO data.
    What if the BTO data is misleading?
    Or worse, what if the BTO data was planted by someone within the company, working on the instructions of the UKAAIB?

    Remember, it was the UKAAIB which first got involved with the Malaysian Govt. in the search for MH370, and it was the source of information that Malaysian PM used as justification for declaring the plane lost in the Indian Ocean.

    It behooves all MH370 researchers to get as much information as possible from others sources, and rely less on the satellite data ‘bubble’.

  18. @CliffG: The DSTG validated the path reconstruction method using both BTO and BFO. The path cannot be reconstructed from BFO alone. BFO contains only the North-South component of airplane ground speed, and BTO is needed for the East-West component. The BTO could be calibrated for the accident flight while the airplane was stationary at a known location before leaving the gate. The relation between BTO and the distance to the satellite is very simple compared to the many factors that affect BFO.

  19. @All

    Article in the Australian about the MAS engineer that rang Capt Zaharie for about 45 minutes before the fateful 9M-MRO flight.

    It turns out they were related and the nature of the phone call seemed quite innocent. The article seems to exonerate the Captain. Attests to his flying knowledge and experience but didn’t go so far as to understanding SDU reboot etc. Behind a paywall:

  20. @SteveBarratt, Thanks, that’s extremely interesting. The existence of this call has been known for a while and was fuel for conspiracy theories. Clearning this up leaves little remaining room for evidence of anything suspicious on Zaharie’s part. Could you cut and paste the text?

  21. A curious coïncidence:

    From the article in The Australian:
    “Mr Zulhaimi last called Zaharie on February 2, 2014”

    February 2, 2014 was also the date of the simulation file fragments in a Shadow Volume on a diskdrive found near Zaharie’s home computer but not connected to that computer.

Comments are closed.