MH370 Second Interim Statement Released

Today, March 8, 2016, Malaysia’s Ministry of Transport released its second annual interim report into the disappearance of Malaysia Airlines Flight 370, as required by international treaty. This document was highly anticipated within the community of independent MH370 investigators, as many of us hoped that it would close some of the huge gaps in our understanding. Unfortunately, it was only three pages long and contained no new information about the disappearance itself. Instead it merely restated the most basic facts of the case and indicated that a more complete final report would be issued “in the event wreckage of the aircraft is located or the search for the wreckage is terminated, whichever is the earlier.”

Given that Martin Dolan, chief commissioner of the Australian Transport Safety Bureau, has said that the search will end if nothing is found by July, and that the high-probability zones of the search area have already been scoured, it seems likely that we will have to wait at least four more months before a substantial reckoning is made.

The new report is not entirely devoid of information, however. For one thing, it acknowledges that while the main wreckage of the plane has not been found, “a flaperon was recovered in the French island of Réunion on 29 July 2015 and was determined to have been a part of the MH370 aircraft.” No mention is made of two pieces recently discovered in Mozambique and Réunion which may or may not have come from MH370. A cryptic undated press release recently issued by the Ministry of Transport both suggests and denies that the Mozambique piece came from a 777. So the possible relevance of these new finds remains ambiguous.

The final page contains a list of eight subjects on which “the Malaysian ICAO Annex 13 Safety Investigation Team for MH370” (aka “the team”) will be working on in preparation for the final report. One of these items is “Wreckage and Impact Information (following the recovery and verification of a flaperon from the aircraft).” Again, no mention is made of the new pieces, and the emphasis on “impact information” suggests that the focus will be on what the deformation of the flaperon indicates about the nature of the crash, rather than what the marine life found on the flaperon tells us about how the flaperon floated and where it drifted. On the bright side, the inclusion of this item suggests that the French are sharing information about the flaperon with Malaysian authorities, and so the public will learn at least something about this important clue in the foreseeable future.

Another item is “Flight Crew Profile.” As it becomes increasingly likely that the plane did not fly south on autopilot alone, the possibility that the plane was flown on a suicide mission by one of its two pilots comes increasingly to the fore. The “Factual Information” report issued a year ago indicated that the captain’s psychological condition had been evaluated, and it was determined that he showed no signs of suicidality. It seems to me that the inclusion of this item suggests that investigators are looking at this topic more closely.

Other items include “Air Traffic Services Operations” and “Organisation and Management Information of the Department of Civil Aviation (DCA), Malaysia and MAS.” These, hopefully, will examine some of the curious goings-on that occurred after MH370 disappeared, such as the fact that Malaysia Airlines told air traffic controllers who were looking for the plane that it was flying over Cambodia. Likewise, the item “Aircraft Cargo Consignment” should explain why the plane’s manifest showed that it was carrying a large quanitity of mangosteens, when this fruit was not in season at the time.

While these latter investigations may shed light on chronic irregularities at the airline and at the civil aviation department, I don’t think that they will tell us much about MH370’s disappearance. Others, no doubt, will disagree.

Finally, one of the biggest takeaways for me in the report is the notable absence of radar as a topic of discussion. There is a great deal of uncertainty and ambiguity surrounding the primary radar trace that MH370 made after it turned around at IGARI and headed west toward the Andaman Sea. We hoped that some of these riddles would be cleared up in today’s report; instead, it seems likely that they will never be resolved.


128 thoughts on “MH370 Second Interim Statement Released”

  1. Dennis,

    “I am more concerned about the Springer”

    Springer pays no royalties to authors… but the authors and journal will definitely enjoy impact points and citation count.

  2. @Oleksandr

    I was unaware of that. Thank you. I will keep you on the short list of coffee table people.

    I will definitely not use Springer for my revised CI theory. I have held off publishing it until the flaperon forensics are made public.

  3. @DennisW

    So interesting that you mention Kalman, to me at least. Tibor Kalman : “I am interested in imperfections, quirkiness, insanity, unpredictability. That’s what we really pay attention to anyway. We don’t talk about planes flying; we talk about them crashing.”

    That’s not the Kalman you’re talking about, though they would be friends.

  4. Oz,

    Do you have any document describing functionality and use of B777 ADIRU in detail?

    Almost two years ago I was looking at “uncorrected gyro” scenario (Duncan’s), and I was very surprised that uncorrected heading, which was found as a result of BFO/BTO residual minimization, pointed almost exactly to N at KLIA. I recall the difference was smaller than 0.5 deg. I have realized that it was not coincidence, but likely it was the result of ADIRU alignment at KLIA.

    Is it correct to say that B777 ADIRU cannot be aligned if it detects motion?

    At a glance some forums say it is possible to align ADIRU of B737 in the air in case it fails: entering heading from an external instrument would be required in such a case. In addition some FMS functions will be lost, particularly navigation by waypoints. I don’t know if this is true and if this is applicable to B777.

  5. @Oz,

    I agree thet the fibers can be easily distinguished by their colour.

    I’m on my phone and havent got the images handy. From memory, the panel’s material had a brownish appearance, neither black nor white.

    I’ll have a closer look when I get access to the images to see if I can make out some fibers and guess at their colour.

  6. @Oleksandr
    Although the question was not directed at me, I comment.

    ADIRU can not be aligned in flight. The heading has to be set into the backup system SAARU (Secondary Airdata And Refeference Unit).

    In case of ADIRU failure
    -Backup attitude is available from the SAARU.
    -The SAARU cannot navigate.
    -After 3 min the heading must be entered manually at the “set hdg” prompt on the POS INIT page.
    -PFD attitude is supplied by the SAARU.

    The following report sheds some light on its operation.

  7. @Brock

    Just on your search area issues:

    1. As is clear from the DSTG book, the Bayesian analysis does not extend beyond 00:11 and the book uses a nominal across track probability function to match the ATSB +/-40nm selected area. The Bayesian analysis is validated on the other flights and none of those crashed, so the principle cannot be carried forward. The DSTG work deprecates the BFO data so it not surprising that the two BFO data points after 00:11 do not determine the cross-track area.

    2. The simulator work discussed in the report indicated that turns after 2nd engine flame-out were possible or even probable. Hardly orders of magnitude lower. The reasons for the symmetric search are in the ATSB report. As has been widely discussed, selecting the cross-track search area is the most problematic as there is almost no data. The fact that the search has yet to move to its last, inside the arc, quarter indicates that the search organisers rate this area low on their priority list.

    3. The ATSB theory gives the published search area based on a clear theory. Until that is searched out to an appropriate level of probability no significant conclusions can be drawn – half-cut searches prove nothing. The fact that the original theory continues to be pursued does not suggest incompetence, as you suggest.

    Agreed, when the current search is done all theories could be reviewed against a null result. However, it is clear that only one (large area) search is possible for MH370, the resource and determination will not allow another.

  8. @Oleksandr,

    No further to add from RetiredF4.


    Imagine a jar of honey that someone has put in the fridge by mistake. That’s the sort of color you are looking for (you might find some white fibres but it’s easy to recognise the honey).


  9. For those of you that believed that the ATSB would take possession of the part from Mozambique before Malaysia would get it, this article claims the part arrived in Malaysia today.

    There were previous reports that Malaysian authorities would accompany the part to Canberra. Instead, it was taken to Malaysia. The ATSB lost in the battle for the initial possession of the part.

  10. @moonkoon – “The aeroflot link on Boeing fasteners posted by RetiredF4 may be unreliable.” The bolts on page 139 were certainly intended to be funny. I’m not sure if the remaining pages are accurate.

    @DennisW – If either or both of the new pieces are determined to have come from a B777, it would increase the probability that 9M-MRO crashed into the Indian Ocean. I don’t have a lot of faith in the drift studies so the search area shouldn’t move due to these finds, but they do reduce the probability of the spoof or the switch plane at IGARI theories.

    As for motive I think it is possible that there was an initial motive for the diversion and search for landing sites, then another motive for the FMT.

  11. Here is what Joao de Abreu, the director of Mozambique’s National Civil Aviation Institute, had to say about the condition of the part recovered in Mozambique after his organization studied it:

    Abreu was also quoted Friday by state news agency AIM, saying that any claim that the debris belonged to the missing Flight MH370 was “premature” and “speculative,” according to All Africa. He also expressed doubts that the debris may not be from the missing Boeing 777 as the object was too clean to have been in the ocean for the past two years.

    However, he reportedly said that “no aircraft which has overflown Mozambican airspace has reported losing a panel of this nature,” First Post reported, citing AIM.

  12. I’ve only just (I know, and you may laugh) realised that one of the other accidents alluded to in the area was of course Egyptian 961 in 1996. That was a 767, and I haven’t seen any references to it in as far as the parts recovered being potentially left over from back then.
    Has anyone any thoughts on this? I’m wondering what the significant differences would be, perhaps the structure would be different.
    In any case I hope it’s possible to tell.

  13. @Susie, from the look of it those pieces can’t be from 1996.
    Joao de Abreu’s assessment is very important IMO because it’s the first time that we learn the informed opinion from someone whose colleagues have actually seen and examined the debris personally. Apparently they support our notion that the condition of the piece simply isn’t compatible with a 2-year drift.
    Unfortunately this early assessment has gone fairly unnoticed in the media coverage.

  14. Re latest finds.
    Now we need some serious forensics:
    can these have been in the ocean for two years?
    If not, where did they come from, where were they?

  15. @Richard: my “orders of magnitude” comment referred strictly to turns executed by an active human pilot, post-flameout. If you restrict your thinking to that subset for a moment, you will see the absurdity of assuming a pilot runs out of fuel, continues on course for three minutes, then pulls a 135 degree left turn, straightens back out, and gets as far from the 7th Arc as possible. If you compare this scenario to, say, continuing to fly straight, you will see what I mean.

    In pilotless scenarios, I of course would allow for random spirals – of a direction and radius experts tell me to expect – in the unpiloted scenario. Those experts – informed by flight simulator results – tell us that, with respect to search width, the unpiloted scenario is searched out, no matter how deprecated or deprecatable the BFOs may be.

    Accordingly, I had inferred that any search width extension would reflect additional weight being given to PILOTED scenarios. Otherwise, the DSTG has recommended expanding the search width in defiance of all published evidence.

    @Dennis: you’d asked about my sense of BFO error. I have tried always to follow the lead of our resident experts – but sadly, it seems these days that means we’re to ascribe to the BFOs whatever precision makes our theory look good – or rationalizes our search decisions.

    Sorry to hear your first experience with a boycott Didn’t go well. I assure you that this next one will go much better (the airline industry has thin margins which are sensitive to utilization stability, and would not long suffer a concerted consumer disruption). And I see the voluntary decision of an individual to “put our money where our mouth is” as one of the purest expressions of free market capitalism available to us. The only alternative is to lobby some nanny government to come save us from our ignorance. I’d rather we roll up our sleeves, and fix this MH370 public accountability gap ourselves.

  16. @Brock

    you said:

    “@Dennis: you’d asked about my sense of BFO error. I have tried always to follow the lead of our resident experts – but sadly, it seems these days that means we’re to ascribe to the BFOs whatever precision makes our theory look good – or rationalizes our search decisions.”

    Actually, the Bayesian Methods Book provides a very clear example of what BFO accuracy should be expected. Figure 5.4 shows the BFO drift for a typical flight. This data is the ONLY window I know of into the measured accuracy of the oscillator chain. The drift shown in Figure 5.4 agrees very well with estimates I made a very long time ago. I said a BFO accuracy of +/- 5Hz was a reasonable estimate, and I stand by that claim.

    In my own work I have used +/- 2Hz as a BFO constraint largely to deflect undue criticism. Neither the physics nor the data set we do have supports even +/- 2Hz accuracy.

  17. @DennisW:

    Baysian Methods Figure 5.4 shows a steady drift of about -1.3 Hz per hour, plus some periodic error at approximatelu the Schuler frequency. I’m not sure about the sign of the error, but the text suggests error = measured – calculated.

    Are positive and negative drifts equally probable?

Comments are closed.