Sydney Morning Herald: Everyone will sound like a conspiracy theorist unless we find MH370

When news broke in 2014 that a Malaysia Airlines 777 had gone missing, no one imagined that, nine years later, we still wouldn’t know what became of flight MH370.

It once looked like closure was imminent. Soon after the plane vanished from radar screens, scientists at the UK-based satellite communications company Inmarsat announced they had found recorded signals automatically transmitted from the plane. By using some complicated mathematics, they were able to work out where the plane must have gone into the remote southern Indian Ocean.

They turned over their findings to the Australian Transport Safety Bureau (ATSB), which was entrusted with the search because the flight’s presumed end point was within Australia’s marine jurisdiction.

All that remained was for ships to scan the seabed and collect the wreckage. Yet when the seabed was scanned in the area the scientists had calculated, the plane wasn’t there. Still optimistic, officials expanded the search area. But it wasn’t there either. Finally, they threw in the towel.

Then, to everyone’s surprise, a previously unknown private company came along and continued the search on their own dime. Still no plane. In the end, an area the size of Great Britain was scanned but the plane was nowhere to be found.

In the years that followed, the world mostly forgot about MH370. But not everyone. For the family members of the disappeared, the nightmare has never ended. They remain stuck in a shadowland, unable to grieve or to hope, as several of them compellingly describe in the recent Netflix documentary MH370: The Plane That Disappeared, which I was a part of.

But it’s not just the family members for whom we need to solve this jumbo-sized mystery. The flying public need to know they can get on a plane and not vanish. We can’t close the books on MH370. We must begin again, from square one, and persevere until we find the answer. If science can find a Higgs boson, it can find a 70m-long airplane.

The question is where to start, and the answer comes down to the issue of why the search has failed so far. Did the official investigation just get unlucky? Or did they make a big mistake?

Many people, including search officials themselves, favour the first explanation. They point to the fact the ocean is a vast place, so it’s not unsurprising the search came up empty. In their view, the plane simply must be there somewhere, either just beyond the defined search area or perhaps in a seabed crevice that the aircraft unluckily fell into.

Those eventualities are certainly possible, but, I’d argue, quite unlikely. Yes, the ocean is big, but the Inmarsat signals that the searchers were working with had well-defined margins of error. Bigger than a GPS error, but equally quantifiable. If you do the math, it turns out to be very hard to find a spot outside the search area where the plane could have gone.

That leaves the possibility that the search officials made a mistake. That when Inmarsat carried out their initial analysis they made a fundamental error that predestined all their subsequent efforts to failure.

Such a thing has happened before. When Air France flight 447 went missing over the Atlantic in 2009, investigators laid out an underwater search area using probability calculations much like those later used for MH370. When their seabed scan came up empty, they revisited their initial assumptions and realised they had made a shaky assumption about the black boxes’ acoustic pingers. They redrew their probability maps and found the wreckage within days.

In the case of MH370, various initial assumptions could have been at fault. The Netflix documentary describes several theories, none of them proven and most of them criticised as “conspiracy theories” by people who would prefer to think that the official analysis was correct. Yet where has that official analysis got us?

The only reasonably satisfying narrative that was ever on the table – that a suicidal pilot flew the plane straight and fast over the southern Indian Ocean until he ran out of fuel, before nose-diving into the sea – is at this point itself seriously problematic.

As they say in addiction recovery, the first step to getting better is admitting you have a problem. Now it’s clear the official analysis has failed, it’s time to start the search again from scratch. Australia should set up an all-new independent commission that can take a hard, fresh look at every piece of evidence from the case and revisit every assumption one by one.

The hour is late, but it’s not too late. Nine years on, let’s roll up our sleeves and get to work. It’s time to find that plane.

This article originally appeared on April 7, 2023 in the Sydney Morning Herald and WAToday.

94 thoughts on “Sydney Morning Herald: Everyone will sound like a conspiracy theorist unless we find MH370”

  1. @Peter, Thanks for this. I continue to be baffled as to why people are so angry/offended by the spoof scenario. Victor seems to lump me with the wacky fringe conspiracists in one breath, and then in the next say that the only reason he gave up his own thinking along these lines is that he realized that Blaine Gibson was too nice of a guy to have carried out something nefarious!
    I’m reminded once again of William Langeweische’s claim that the reasons why one can’t control a 777 from the electronics bay are “too numerous to mention” and so doesn’t mention any. If people have rock-solid evidence that the plane couldn’t possibly have gone north then I wish they would share it.

  2. @Ventus45, I wouldn’t think there’s great significance in those tracks, they probably just plugged a couple of likely speeds into the ping rings to see what tracks they would generate. Anything from March 2014 I would take with a grain of salt, they were just starting to figure out what was going on.

  3. Jeff Wise. The references for both the Data 3 connection and the Satcom link in the cabin for Cabin crew use are in the Safety Investigation Report.

    I was a 747 captain for QANTAS and the QANTAS 747 had the same Classic Aero Satcom system with both those capabilities.

  4. @Jeff Wise
    Two words explains all of that: Active pilot

    Channeling Hammerstein, we have been carefully taught it had to be a ghost flight and/or if the pilot was alive, he did nothing except maybe glide at the end. This is denial logic that will not find the crash site in my opinion.

    By the way I want to make sure you are up-to-date on newer learnings from ATSB about the sim cases.

  5. @BillTracy
    What new sim data from the ATSB Bill ?
    I haven’t seen anything new from the ATSB.

  6. @Mike, Thanks for this. To spare me the search, would you mind copying and pasting the relevant section? If it’s an image you can email it to me at jeff [at] jeffwise.net. Appreciate it.

  7. @Jeff Wise:
    Could you please remind me:
    How does your theory of a northern path with spoofed BTO values (and landing in Kazakhstan) account for the nosedive-like BFO values ?

    (PS: I will get back to you on your other comment above.)

  8. @Jeff Wise:
    correction to my comment above:
    How does your theory of a northern path with spoofed BTO BFO values (and landing in Kazakhstan) account for the nosedive-like BFO values ?

  9. @Peter, If the BFO values were spoofed, then they were deliberately altered so as to mislead, so the nosedive-like values could either be interpreted as a) a deliberate deception, or b) an artefact of the disconnection process. Note that when the SDU logged on at 18.25 it prodcued values that Inmarsat scientists were unable to interpret.

  10. @Jeff Wise:
    ok, got you. But I thought the idea was “that you only need to change 1 parameter inside the SDU to create a track that looks like it’s going south, when it’s really going north” (source). The spoofing you suggest in your post above is way more complicated than that! Isn’t that moving the goalposts a quite a bit ?

    “when the SDU logged on at 18.25 it prodcued values that Inmarsat scientists were unable to interpret.”

    Wasn’t that explained here ?

  11. @Jeff
    Re: Flight Sim Data
    Let’s start with background where we left off on your blog in late 2017-

    BACKGROUND
    The sim data had became a hot topic on your blog again in Oct_2017 when an ATSB report disclosed flight times (which had been undisclosed). ATSB stated the flight times were consistent with flight MH150 to Jeddah which the pilot flew in Feb_2014. Surprisingly ATSB also stated that ATSB was in possession of *complete* sim data files, which are still secret. Recall until then we had been trying to interpret “fragments” of files that were leaked to you and others. The file fragments were apparently redacted from complete files prior to the leaking…others disagree with my “redacted” inference, in any case, the complete sim files exist, we are simply not entitled to see them.

    Finally by way of background, just as we started hot debate on your blog and others, almost at the exact same time, enter Ocean Infinity in late 2017 with a proposal to start searching again…so the revived sim data discussion quickly fell to back burner status.

    NEW SIM DATA LEARNINGS
    From my personal perspective, several years later 2020/2021, I decided to make a final attempt to study the sim data. I reached out to several independent investigators and learned that ATSB had been selectively leaking some new hints about the still-secret portions of the sim data.

    I saw the need to advise the MH370 community, but I did not want to take credit as others (Mick Gilbert) had got the data from ATSB. My write-up was covered in the early days of Richard Godfrey’s then new blog:

    See link to my pdf paper: New Interpretations of the Pilot’s Home Flight Sim Studies
    https://www.mh370search.com/2021/02/04/guest-paper-by-bill-tracy/

    One important new finding is Line 1 of the data which gives us the file type:
    [Main]
    Title= Program generated temporary flight
    So these files are MicroSoft FS system temporary files, and not files saved by the user.

    Another new finding was that there was no LNAV flight path showing on the FS9 map, so the old idea of the pilot manually moving the aircraft along a presumed path to NZPG was not confirmed. Pilot was still manually moving the aircraft, but we do not know exactly how he decided where to move the aircraft icon.

    >>Overall the sim data conclusions remain the same: the data looks very suspicious like a plan to divert (possibly MH150) to the SIO. I must disclose others claim the sim data is innocent play/practice, but for me that is a denial position.

    We can discuss other findings.

  12. @Jeff Wise: I might be missing something, about for years I thought that your northern path is a “mirror path” of the IG’s southern path, created by changing 1 parameter inside the SDU. You said as much:
    “you only need to change 1 parameter inside the SDU to create a track that looks like it’s going south, when it’s really going north” (quoted from above). Victor Iannello apparently also pursued this possibility back at the time.

    To that you replied:

    « If the BFO values were spoofed, then they were deliberately altered so as to mislead, so the nosedive-like values could either be interpreted as a) a deliberate deception, or b) an artefact of the disconnection process. »

    But the kind of spoofing you suggest in (a) is way more complicated than the simple change of 1 parameter inside the SDU, which cannot account for the nosedive-like BFO values.

    So either MH370 nosedived in Kazakhstan (why?) or how do you otherwise suggest the spoofing worked if not by changing a single parameter inside the SDU? The more complex the spoofing has to be, the less likely.

    I hope I was able to get my point across ?

  13. @Jeff Wise:
    bottom line: If the BFO values can be altered arbitrarily (as you seem to suggest in (a)), then MH370 could have flown pretty much anywhere …

    I thought your northern path to Kazakhstan is a “mirror path” of the IG’s southern path, created by changing 1 parameter inside the SDU ?
    Am I misconstruing your theory ?

  14. @Peter, The two metadata sets, BFO and BTO, tell us two different things. The BTO values reveal the path that the plane took — but these are mirrored, north and south. The BFO value tells us which of these paths is correct. I have proposed that the BFO values can be spoofed, but the BTO values cannot. Thus, being able to create arbitrary BFO values would not allow you to fly anywhere.
    The interpretation of the BFO values was based on the fact that the algorithm that gives rise to them assumes that the plane is flying at a level altitude. They only are able to break the north-south symmetry because the system was not working as planned. If the plane is climbing or descending, its BFO value does not tell you anything about whether it went north or south. That’s why the final BFO values have no bearing on the plane’s presumed path.

    Hope that helps.

  15. https://www.reddit.com/r/UFOs/comments/15p14tp/megathread_mh370_relevant_posts_regarding_mh370/?utm_source=share&utm_medium=ios_app&utm_name=ioscss&utm_content=1&utm_term=1

    Video evidence of the disappearance of MH370. Thoroughly discussed in this subreddit. I implore you to watch the videos and try to debunk the evidence. UAPs are real as acknowledged by the US gov’t. The USA knows what happened to MH370. It would make sense that any gov’t that lost track of a commercial airliner would send eyes into the sky to make visual contact on the airliner. Please read through the thread and view the countless efforts of those on this subreddit to uncover the truths.

  16. @AnonK, I have deleted the initial response I posted here, it was immature and insenstive on my part. I apologize.

    Suffice to say, I do not find recent claims made about UFOs and MH370 to be plausible.

  17. I know that MH370 (boeing777 200er) is designed to fly up to 13900 km. What’s the evidence that it had only 7 hours left. What was the fuel status at the departure?

    If there’s an evidence that it can fly more than 7 hours, then I’m sure, after looking at the simulator track on the map, there’s only one scenario that what the pilot is thinking. Yet, it is hard to believe.

  18. @SM, Unless a flight is intending to go the maximum distance a plane is designed for, it will take off with tanks that are less than full. This was the case with MH370. It only had enough fuel to fly for about seven hours. To take off with more than is necessary would mean expending more energy, and hence burning more fuel, which is wasteful

  19. Apologies if you have addressed this elsewhere but what is the significance of omitting the data point shown to the nok at the Lido hotel?

  20. @Truman, You mean the second half of the primary radar data? No one knows why the Malaysians won’t release it, but it could be quite helpful in refining the probablity distributions of the plane’s end point.

  21. Yes that’s what I meant. Logically, isn’t it most likely the case that whatever it shows is helping support their conclusion as to the final resting place?

  22. @Truman, We don’t know what their logic is. The Australians did state (if memory serves) that they excluded the 18.22 radar return from their Bayesian modeling. My personal suspicion is that if they included it, it would have shown a much higher probability distribution of endpoints in Kazakhstan, because combined with the 18.25 ping ring it means that the plane was turning north when last spotted. In other words, in a scenario it which MH370 went north, there was no “final major turn” — it just kept going straight.

  23. Jeff- (this came up again recently because I was trying to study the WSPR route) the Bayesian analysis went very close to 1822 radar point but did accept it as the exact location. The Bayesian analysis was apparently done before we had a complete understanding of Arc1. Today Arc1 is seen as the point where the N571 OFFSET occurred, because there is an apparent delay in the forward BTO progress towards the satellite. This in mind, it would probably be possible to accept 1822 point exactly and still be consistent with Arc1 if we were to redo the Bayesian today.

  24. @TBill, Given the distance between the 18:22 radar contact and the location of the 7th arc, we can infer that the plane turned to the right, i.e. to the north. One plausible speculation I’ve heard is that the reason for the reappearance of MH370 on primary radar is that as it turned the plane presented a larger apparent cross section and so the target was able to be reacquired.
    This turn would, interestingly, align the plane with the northern path to Kazakhstan, and it’s long been my contention that if CSIRO had included the 18:22 data point it would have resulted in a probability distribution that heavily favored Kazakhstan over the Southern Indian Ocean (before the BFO data was taken into account of course.)
    Obviously this turn north is hard to incorporate into a southern scenario, so it was suggested that the reason for the turn was that the pilot was performing an “offset” to remove himself from a busy airway to avoid other traffic that might be on it. This idea has never made much sense to me because the odds of hitting another plane over the ocean in the middle of the night are vanishingly low. However the “offset” is an article of faith in certain circles.

  25. First sentence makes no sense, distance 1822 to Arc7 does not infer a turn right. Second point, at 1822 we have no explanation of the secret military radar data to be able to say much of anything. Clearly MH370 appears to most of us to be heading forward near N571. Nobody is saying there is right turn before or at 1822. The offset at 1825-1828 does not really have a definitive direction to it, IG assumes it is right offset, my guess it is really a left offset or jog south over to airway B466. All we really have is an apparent delay 1825-1828 which is assumed by most to be some action to get out of incoming air traffic coming down N571.

  26. @TBill, Sorry for the slow response. The distance from 1822 to the 7th arc is too short; unles the plane dramatically decelerated, the only way to explain would be by a turn — either a gentle turn to the right, or a hard turn to the left that would have sent the plane on a due-south heading right into Indonesian airspce and over the coast of Sumatra. Based on its final report, the Australian authorities also believe that the plane turned north at 18.22. You can read about it here: http://www.jeffwise.net/2016/09/19/how-mh370-got-away/
    PS there was no traffic coming the other way and even if there were the chances of hitting it dead on in all that airspace is vanishingly slim.

  27. Jeff-
    Per yours above, left FMT turn does not mandate flight thru Indonesian airspace. There is plenty of time to fly around Indo FIR space, still rendezvous with Arc2, and head 180 south, which is my guess what was actually done. Around ISBIX, Indo FIR may have been violated but that is so far out-to-sea, who cares?

    There was indeed air traffic coming inbound (to KLIA) coming down N571. which seems to explain some evasive effort by MH370…offset of N571 of some type (I say maybe left).

    I listened to the second podcast so far, I did not see big objection but I thought could have been a little more accurate about Vietnam’s response…Vietnam ATC responded fairly quickly but made mistake spending to much time making emergency radio calls, and delayed informing KLIA until they realized MH370 was way lost.

  28. @TBill, Thanks for your feedback on the podcast, always good to get your input. Do you think that if Ho Chi Minh had followed procedure more closely the outcome might have been different?

    About that 18:22 turn: this would be separate from the “final major turn” to the south, which presumably took place closer to 18:40. If the turn took place here the plane would have crossed over Aceh east of Banda Aceh.

    There would be no need to change course to avoid oncoming traffic; airspace is three-dimensional so you’d just need to be at a different altitude.

  29. @Jeff

    The last part of that transcript reads as follows:

    interviewer:
    So, let me ask you this, before the debris was found, and all we had indicating that the plane went south was the BFO data, how did you guys know the BFO hadn’t been tampered with? #00:50:33-3#

    Neil Gordon:
    All I’ve done is process the data as given to me to produce this distribution. What have we done to understand the accuracy of BFOs? As I’ve said, we’ve looked at all these other flights, where we know where everything is so we can get a feel for what the accuracy is and what it tells you about the flight paths. And we’ve done that on previous flights of the accident aircraft, other 777s in the air at the time of the flight.
    #00:51:09-6#

    interviewer:
    I actually asked this question of Inmarsat, and they said, “the equipment manufacturer is who you’d have to ask.” So, that’s who we have to ask.

    There is no timestamp after your last statement, and there is no response from Neil Gordon.

    Did the interview actually END there ?
    OR – has the transcript been “truncated” ?
    Is there more ?

  30. Dear Jeff,

    Keep up with the good work, much appreciated!

    What we might be able to find is what (physically) happened – but “finding the plane” seems improbable to me. Sadly, the best chance probably is that someone of those “in the know” will eventually leak something that will lead us to physical evidence about plane’s fate.

    I think years ago it became clear that the biggest challenge of the MH370 investigation is ‘sorting the evidence” – excluding reliable from unreliable information. I’m writing here to point out what I think is a) evident but b)overlooked (for anyone who cares to remember TWA800) : it seems very improbable that – if the plane somehow “crashed”/disintegrated – only such a tiny fraction of the plane’s material was (ever) recovered or accounted for. That’s even if we accept the (few) parts that were recovered as legit. From here I draw a “preponderance of evidence” conclusion: whatever happened to the plane, it ended up (somewhere) being (more or less) intact. In my humble opinion it seems very, very improbable to effectively conceal (scoop and cover up) every single vestige from a 777 disintegration – weather (no, more so) in the heavily-monitored South China Sea immediately after the accident, or in the SIO, or anywhere inland.

    What do you think about this? As an aviation expert, how unusual is to – ever – find so little material? Ergo, how probable is any scenario where the plane stays intact in all the possible “catastrophe” scenarios (considering the excessive vertical speed at the last handshake, considering fuel exhaustion, over open water etc.)

    Thank you and good luck!

  31. @Kerguelon, Thanks for your comment. I agree that a major challenge with MH370 is sorting useful evidence from not-so-useful. But I think it can be done, and it’s what I’m trying to do with my podcast, where I look at the evidence in detail and explain how the authorities used it to figure out where to search, and where they might have gone wrong. I hope you check it out, the show page is deepdivemh370.com.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.