Update on MH370 Drift Modeling Enigma

Last month I wrote, in a post entitled “Nowhere Left to Look for MH370,” that recently refined drift models produced by Australia’s CSIRO contradicts both their own premise (that the plane crashed on the 7th arc between 34S and 36S) and an alternative idea presented by intependent researchers (that the plane crashed near 30S).

I’ve only just become aware that CSIRO director David Griffin weighed in on the matter a few weeks ago in a letter to Victor Iannello, which Victor published on his blog. He essentially confirmed the points I raised.

He wrote, for instance, that:

As you correctly pointed out, a 30S crash site would, according to our model, have resulted in debris washing up on Madagascan and Tanzanian shores a full year earlier than was observed. That is a discrepancy that is hard to set aside.

He also wrote that:

The other factor against 30S that we find very hard to discount is that 30S is right in the middle of the zone targeted most heavily by the surface search in 2014. This is the “other evidence” that Richard overlooked. Please see Section 4 of our Dec report, and Fig 4.2 of the April report.

Griffin also defends, rather weakly in my estimation, the idea that the early arrival of the “Roy” piece in South Africa does not contradict his preferred 34s-36S crash point. However, I don’t think this really matters, since there is so much other compelling evidence against it.

The conundrum, therefore, becomes even more impenetrable than before: the evidence indicates that the plane did not go into either of these “hot spots.” And it indicates even more strongly that it did not go anywhere else in the southern Indian Ocean.

The way to solve conundrums is to open up your thinking and to check for implicit assumptions that my be incorrect. In this case, the obvious follow-up question is: is it possible, given the data in hand, that the plane could have gone somewhere else?

Australian officials remain puzzlingly unwillingly to acknowledge the issue.

77 thoughts on “Update on MH370 Drift Modeling Enigma”

  1. @StevenG
    Did not mean to imply he could not leave through the cockpit door, only an effective exit to avoid crew and passengers.

  2. @DennisW @Freddie
    Talk is cheap. You gotta show BTO + BFO matching that flight path story. Perhaps you are partially correct about the verbal story, but if so, it would appear to me that there was another option which is 180S due South to oblivion, if the negotiations failed.

    Here is how I currently interpret Z’s FS9 Simulator data to McMurdo waypoint:

    Z may have done that work to see how his plan would work on the new MAS night time flight to Jeddah. If you notice, the flight path 1090E NZPG on SkyVector almost perfectly circumnavigates the Indonesian airspace boundary, which would have been much more critical 2 or 3 hours earlier in the evening when the radar ops were potentially still active before shutting down for the weekend.

    For whatever reasons, Z had to instead use the Beijing flight, but he still circumnavigated Indonesia (similar to the Sim studies) via @Nederland’s “go around” idea (IGOGU APASI TOPIN ISBIX) or via Victor’s simpler LAGOG switchback to BEDAX then 180S CMH path.

  3. @DennisW
    correction above…use 1090E NOBEY for simulator path approximation because neither SkyVector nor PSS777 model have NZPG (need to manually enter)

  4. @TBill

    The BFO and BTO matches are easy, and I published that a long time ago. The objections are that the flight path was not at a constant track (violating the IG principle of the way pilots like to fly airplanes), and it did not use waypoints (JW objection).

    I postulate the reason that Shah chose the Beijing flight is that his objective was to demonstrate the diversion as early as possible, not to sneak off into the SIO. The lack of motive for the popular SIO flight paths has always been troubling to me since I harbor the belief that people do things for a reason.

    No matter. The searching is over, and I see no possibility of it being resumed.

  5. @DennisW
    Show me your path if avail…I know your Cocos path on your blog.

    I do not agree on that point. I see deep spot in SIO as a target either generally deep region or specific spot.

  6. BEDAX is very close to Sabang and Lhokseumawe radar stations. I never understood the reasoning for why MH370 would have avoided Indonesian airspace for so long (instead opting to fly into Indian airspace), only in order to then blatantly return into Indonesian airspace, passing close to the main military bases.

    Either the PIC assumed all Indonesian radar were turned off on that night or he wasn’t sure about that. In the first instance, he would have simply flown over Sumatra, in the second he would not have returned to waypoints such as BEDAX. The distance between BEDAX and Sabang is around 80 or 90 NM, well within radar range, regardless of altitude.

  7. @TBill
    DennisW Freddie “Talk is cheap …… “

    We are covering old ground – the track was put up as NILAM – LAGOG – NOPEK – BEDAX – ISBIX – PCCNG – YPXM – 7th arc.
    About 465kts slowing to 370kts after BEDAX.

  8. @Freddie

    Interesting. Do you have BTO/BFOs for that route for the ping intersections?

  9. @TBill

    if that was his goal he’d easily find out about JORN and its coverage(actually being interested in planes he most probably knew already), he wouldn’t fly towards but away from it

    it simply doesn’t make sense for myriad of reasons

  10. @TBill

    I have a couple of CI paths on my blog (both created before the simulator data was revealed), take your pick. If I were to create a path now, I would lean towards Freddie’s posted above.

    However, that would be pointless. One can create a path to any terminus on the 7th arc South of Sumatra by tweaking track, speed, and altitude changes, particularly since the DSTG Bayesian Methods document has created a much broader interpretation of the BFO constraints. The ISAT data is simply not useful for defining a terminus. It is useful as a very generic qualifier. You seem not to realize that. I am not going to waste any more time with it.

  11. @Nederland
    I think I am closer to 139 degrees T on your NIXUL to UPROB (not ? (re: Arc2 BFO)

    OK I am not currently in support of that path because the very slow speed conflicts with Brain Anderson’s 495 speed calc at approx. 19:50

    On the assumption BTO/BFO are weak and cannot be used without massaging the data, we’ve failed to look at the one spot that that data says the aircraft actually went.

  12. @TBill

    OK I am not currently in support of that path because the very slow speed conflicts with Brain Anderson’s 495 speed calc at approx. 19:50

    On the assumption BTO/BFO are weak and cannot be used without massaging the data, we’ve failed to look at the one spot that that data says the aircraft actually went.”

    The minimum speed at 19:40 is about 410 knots at 180 degrees. The maximum speed is the maximum speed of the aircraft.

    I would not characterize the BFO/BTO as weak. They are what they are. A terminal location cannot be derived from that data without making additional assumptions even if that data were perfect.

  13. @DennisW “The minimum speed at 19:40 is about 410 knots …. “

    That speed of 410kts sits well with our early on calculations regarding the speed crossing 19:41 as the aircraft continues slowing to 370kts before ISBIX and holds a steady 370kts in a loiter through to after 22:41 on its way to CI.

  14. @Nederland
    I see now NIXUL is secondary to TOPIN

    Your path is somewhat similar to @Nederland but I think he is going faster thru ISBIX to BEBIM where he could also easily also divert to CI and probably Jakarta if he wanted to.

  15. Freddie’s route sounds logical, but I’m not convinved it matches the BTO/BFOs in the latter stages of that flight. Anyone has a spreadsheet for that?

    I was also working on the assumption that there was some intention to land the plane on Cocos or Christmas Islands, Learmounth or Indonesia. That’s why BEBIM makes perfect sense as it is located at the intersection of all possible destinations. Something could have happened at some point.

    I could not, however, work out a route that does not lead into a generally southern direction.

  16. @ Nederland
    “….I’m not convinved it matches the BTO/BFOs in the latter stages of that flight ….”

    What altitude are you using to calculate the BTO’s?
    What type of turns have you allowed for?
    How long are you allowing for time around Christmas Island before proceeding to the 7th arc?
    I am using the final version of Duncan Steel’s ISAT circles adjusted for different altitudes.

    “….That’s why BEBIM makes perfect sense ….”

    The motive was to loiter while coming round below Sumatra before landing safely.
    To go out to BEBIM is counterintuitive to this objective, I would have said that a route from ISBIX direct to CI would have been more likely but the BTO’s don’t fit.

  17. @Freddie,

    In my paper posted above I have used a medium altitude of 320,000 ft (9.75 km). The BTO are very slightly different, say at FL290 or FL350. At the end of the day it wouldn’t matter.

    As to turns, I basically have the one slight turn (and change of speed) at BEBIM, that’s it (after MH370 had circled around Sabang radar). I don’t allow any turns close to the pings. I really think this is decisive, like it is unlikely MH370 made a ‘freak’ turn while passing a ring (in that case you have arbitrary BFOs).

    I have no doubt that you have adjusted your route to match the BTOs/ping rings. But how about BFOs? The BFO tell us that MH370 flew in a generally southern direction. Do you have calculated BFOs as well for your route and, if so, do you have BFO values for your route?

    No criticism, just asking…

  18. A small break from the SIO search. Going back to the clues, ATSB website mentions is the FAQ
    “Why did the search take so long?
    From the point where the aircraft flew beyond the range of radar, there was very little information in relation to its location nor trajectory. It has been difficult for some people to accept that, in a world with Google Earth, smartphones and GPS, not everything is being tracked all the time. But MH370’s transponder – the unit that broadcasted its location – had ceased to function. There were no confirmed sightings of the aircraft. The only clues about its route were periodic satellite data unit communication signals. These signals had never been intended to provide information about the aircraft’s location; they were ”handshakes” between the aircraft and the ground earth station indicating only that the aircraft satellite communication system was still functioning. They continued for hours as the aircraft flew south into the Indian Ocean, a vast and largely empty expanse of water with little routine surveillance.

    I find surprising that the ATSB considered that “the ONLY CLUES about its route were periodic satellite data unit communication signals.”

    I also find surprising that such satelite data subject to significant assumptions and interpretation (i actually admire all the work on it) have been given more weight than the other clues such as radar data (although note released, were actual raw radar data considered in the investigation?), witness account (why the official investigation did not consider them?) and debris (not sure what forensics other than reverse drift and barnacles study have been considered, maybe more forensics such as crack analysis/fracture testing, C14 dating of the parts, etc. could give us more information).

    The same ATSB website mentions that no one have sighted the plane despite a number of sighting report claims. I am wondering how those people pretending to having seen something feel about this statement. Those are clues with equivalent weight to study all relevant scenarios especially if as @Jeff mentioned the possibility of data tampered has been raised.

    Another piece of clue which i find interesting is the transcript of the air ground communication about the aircaft position at 18:35 UTC saying that MAS Operation cannot use the fax because the computer is downloading the aircraft position. This merits also forensics as the timing of the data download is more like an ACARS message download than a FlightRadar style download not sure what system MAS used such as it is so slow. The latter also would not prevent the use of fax. Do more ACARS messages than reported actually exist? Maybe i am wrong but i want to point out that all clues should be examined systematically in details and clearly we do not have an official convincing story on all of the elements.

    I also could not see a complete reconstruction of the timeline showing ALL the elements evolving STEP by STEP including plane instrument status, ACARS messages, telephone calls, radio comm with ground, radio com with MAS, ATC comm, sat data, other plane locations, radar data, witness data, etc. (if one reconstruction exists please send me the link).

  19. @HB

    i might add to your list, that i didnt hear of any forensic analysis, how long the debris spent in salt walter, exposed to the SIO. It killsme, that it seems imposible to detemine thiese exposition. Salt water is a very aggresive liquid. It should leave traces on a flaperon and from these traces, one should find a forensic method to determine how long a piece was exposed to salt water. Is our science really so dumb, as not be able to do that?

  20. @CosmicAcademy

    The French are in the possession of the flaperon. It is being held part of a terrorist investigation. Quite possibly they have information on salt water attack on the aluminium composite but the nature of the legal process precludes this be published in a conventional sense.

  21. @CosmicAcademy

    i am not sure about salt water whether salt growth for instance can give us additional information, a forensic expert could tell us. However, I cannot agree more than a scientic and systematic method of investigation could help us find what has happened faster as opposed to an inductive method currently proposed.

    This paper sums up what should have been done and what should still be done.


    Also from another industry but still revelant here: http://frigg.ivt.ntnu.no/ross/reports/accident.pdf

    Note the requirement for “Independence” in the investigation. This is probably the most important requirement.

    Also note the requirement to interview potential witnesses.

    When looking at this, sorry to say but the investigation has barely started it seems.
    Specifically for aviation industry, i cannot comment whether the investigation complies with international practices but for other industies it certainly won’t.

  22. @Freddie
    I always liked your path, and feel it might make a good paper to show 2 flight paths coming off the same root path: one path diverting to Jakarta and a different path diverting to SIO from the same root path, both meeting BFO/BTO. That could be illustrative.

    >> Hey, I can’t find those Duncan Steel ISAT circles, are they still on-line?

    So MAS/MY knew from the very beginning, from the unanswered sat phone calls, that MH370 was in the air for 7 hours, and yet they dragged us thru a 10-day search off-shore Malaysia assuming MH370 crashed shortly after take off. Also Inmarsat was telling MY from Day2 that it was apparently a very long flight.

  23. @Nederland
    I get a good path as follows:

    FMT to a 180S CTH heading happens almost exactly at 19:41 right on L896 exactly due North of ISBIX. With winds, BTO/BFO are closely matched and MH370 basically ends up due South of ISBIX around 93.67E.

    I am at M.84 and FL350 the whole time. The flight gets to FMT exactly at 19:41 which is nice, but I do not think that leaves much time for a descent at 1840. So I feel the the actual flight path may have been simpler, towards LAGOG and then cut the corner to NISOK, maybe with a descent at IGOGU (or SAMAK) to duck below the UAE coming up from behind.

    >>More generally stated, the Inmarsat arcs are telling us (quite obviously) MH370 may well have gone straight South of ISBIX (93.67E) or BEDAX (93.76E) or BULVA (94E). That’s an obvious flight path, and teh end point is unsearched inside of Arc7.

Comments are closed.