Thoughts on Australia’s New MH370 Report — UPDATED

end-of-flight

Earlier today, the Australian Transport Safety Board released a document entitled “MH370 — Search and debris examination update.” Perhaps occasioned by the recent completion of the towfish scan of the Indian Ocean seabed search area, the document updates earlier ATSB reports and offers some intriguing insights into what may have happened to the plane. Some thoughts:

— The first section of the report expands upon an assertion that the ATSB made in an earlier report: that the BFO values recorded at 0:19 indicate that the plane was in an increasingly steep dive. Indeed, the newly published calculations indicate that the plane was in an even steeper dive than previously reckoned: between 3,800 and 14,600 feet per minute at 00:19:29, and between 14,200 and 25,000 feet per minute at 00:19:37. On the lower end, this represents an acceleration along the vertical axis from 37.5 knots to 144 knots in eight seconds, or 0.7g. On the higher end, this represents an acceleration along the vertical axis from 140 knots to 247 knots, likewise about 0.7g. If the plane were freefalling in a vacuum, its acceleration would be 1.0g; given that the airframe would be experiencing considerable aerodynamic drag, a downward acceleration of 0.7 would have to represent a near-vertical plunge, which a plane would experience near the end of a highly developed spiral dive.

— The second section describes end-of-flight simulations carried out in a Boeing flight simulator in April of this year. These tests were more detailed than others carried out previously. Evidently, modeled aircraft were allowed to run out of fuel under various configurations of speed, altitude, and so forth, and their subsequent behavior observed. Thus, the exercise modeled what might have happened in a “ghost ship” scenario. Notably, it was found to be possible for the plane to spontaneously enter the kind of extremely steep dive described in the previous section. This being the case, the report states, the plane “generally impacted the water within 15 NM of the arc.” This is not surprising, considering that the plane had already lost altitude and was plummeting straight downward. This offers a tight constraint on where the plane could plausibly be if the 0:19 BFO analysis is correct.

— The third section describes the results of debris drift modeling that has been informed by tests involving replica flaperons “constructed with dimensions and buoyancy approximately equal to that of the recovered flaperon.” An important point not addressed by the report is the fact that the French investigators who tested the buoyancy of the flaperon were unable to reconcile its observed behavior with the observed distribution of the Lepas anatifera barnacles found growing on it. So when the French ran their own drift models, they had to run them twice, one for each buoyancy condition. Apparently the Australians overcame this paradox by discarding one of the states.

— The third section notes that, according to modeling carried out by the CSIRO, debris which entered the ocean in the southern half of the current search area would not likely reach Réunion by the time the flaperon was recovered. Meanwhile, debris that entered the water significantly north of the current search area would reach the shores of Africa much earlier than the time frame in which pieces were actually discovered there. Using this logic, the report concludes that the northern part of the current search area is probably correct. However, this seems dubious reasoning to me: one would expect a gap between the time debris arrives in Africa, and the moment when it is discovered. Also, debris can move quickly across the ocean, only to be trapped in a local gyre and move around randomly before beaching. Therefore I think the argument that the pieces couldn’t have originated further north is flawed.

— The fourth section, describing the damage analysis of the flap and flaperon, is the most interesting and newsworthy of all. In short, it makes a persuasive case that the flaperon and the inboard section of the right-hand outboard flap (which, rather remarkably, turn out to have been directly adjacent) were in the neutral, non-deployed state at the moment of impact. Assuming this is correct, this eliminates the IG’s flutter theory, as well as the widely discussed theory that the flap was deployed and therefore indicative of a pilot attempting to gently ditch the plane. Proponents of these theories will continue to argue on their behalf but in my opinion they were dubious to begin with (given the shredded condition of much of the recovered debris) and are now dead men walking.

— No mention was made of Patrick De Deckker’s exciting work with Lepas shells.

— Overall, the thrust of this report is that the plane went down very close to the seventh arc in a manner consistent with a “ghost ship” flight to fuel exhaustion, exactly as the ATSB has assumed all along. There is, however, one very large elephant in the room: the fact that Australia has spent two years and $180 million demonstrating that the plane’s wreckage does not lie where it would if this scenario were correct. Therefore it is not correct. The ATSB’s response to this conundrum is rather schizophrenic. On the one hand, it has recently floated the idea of raising another $30 million to search further—presumably the small remaining area where a plane just might conceivably have come to rest in a ghost-ship scenario, as I described in an earlier post. On the other, it has today convened a “First Principles Review” consisting of experts and advisors from Australia and around to world to scrap their previous assumptions and start with a clean sheet of paper. This implies an understanding that they have proven themselves wrong. I wonder how many assumptions they will scrap. Perhaps, as Neil Gordon mused in his interview with me, that the plane wasn’t really traveling south at 18:40? Or perhaps they’ll dare to go even deeper, and contemplate the provenance of the BFO data… ?

— A postscript: Richard Cole recently posted an update of the seabed search (below). I’m intrigued by the fact that the Fugro Equator has deployed its AUV near the northern end of the search zone. When I interviewed him for my last blog post, Fugro’s Rob Luijnenburg told me that the northern end of the search zone was flat enough that it could be scanned by the towfish alone; there was no need for an AUV scan to infill the craggy bits. So why is the AUV looking there now? Especially given that it’s very close to an area just reinspected by Dong Hai Jiu 101’s ROV. Another MH370 mystery.

UPDATE 11-2-16: I emailed Rob Luijnenburg and he immediately responded: “The AUV is scanning in a section in the north part of the priority search area in the very rugged terrain south of Broken Ridge (the east -west mountain range at approximately the 33rd parallel)… Generally the AUV is deployed in spots of extremely rugged seabed to complete the 120,000 sq km priority area survey.” Worth noting is that if the search gets expanded northeastward, it’s going to be into very rough terrain indeed.

richard-cole-11-2-16
courtesy of Richard Cole

495 thoughts on “Thoughts on Australia’s New MH370 Report — UPDATED”

  1. @Johan, @Boris, This has been covered here before — apparently the 7th arc lies outside JORN coverage, and what’s more it wasn’t being used at that time (24/7 coverage being possible but not always in effect.)

  2. @Jeff:
    Thanks. I was hoping for someone’s intervention. Maybe still indicative that he kept aloof (note wikipedia’s piece on mh370 and JORN, though).

    @DennisW:
    Thanks. Maybe it works as a scare… so that the barbie won’t have to be cancelled.

  3. @JeffW @DennisW
    Some MH370 bloggers had suggested a pilot trying to avoid JORN would have flown an extreme south-westerly flight path. Just trying to understand if a hijacker or Z would have agreed with those blogger’s fears of JORN detection. In hindsight of course we now know it was not seen by JORN, and flight path now suggested by many is quite north-easterly, towards JORN.

  4. @DennisW:, @TBill:
    I actually read the article Dennis suggested now. That was enlightening.

    Here is a quote:
    “The expensive hi-tech system’s capabilities came under fire three years ago for its failure to detect asylum seeker boats entering Australian waters.

    In the previous two years more than 200 boats landed without detection.

    The explanation at the time was that the system was looking only at vessels the size of the 57m Armidale class patrol boat.”

    It is a good business idea though: the operational qualities are classified.

  5. @DennisW
    As USAF had operational X-37B shuttle, which is considered as orbital testing vehicle, probably pumped by reconnaissance devices too, flying in low earth orbit at 7.8km/s it can maneuvre anywhere and be above any earth point during 90 minutes or so, I can imagine that after MH370 vanished as dangerous “tomahawk” but as at least ping activity was detected (I expect every hand and foot was looking for the missile…) then there was plenty of time to catch him somewhere by diff-comparing images (not manually, no) to find something what isnt expected to be there. But YOU are THE sat guy here – do you agree?

  6. @strangelove

    “But YOU are THE sat guy here – do you agree?”

    Not me. ALSM is the most qualified SAT guy on this blog.

  7. @Gysbreght

    –“When I wrote “hardly any loss of lift” I was referring to the fact that the curve of lift coefficient versus angle of attack showed no marked drop at the stall AoA of about 10 degrees. The stall warning started at 02:10:51.3 (see figure 66 of the Final Report) and the loss of lateral control, deterrent buffet, and a slight drop in normal loadfactor indicated that the airplane was stalled about 6 seconds later.”

    You used up an entire paragraph to say absolutely nothing in support of “hardly any loss of lift.” Some readers will think you used some fancy terms and maybe know what you’re talking about. Me, I don’t.

    Where are you seeing the lift curve in the AF447 report? I’m not seeing it. I see a generic example of one in Fig 88 but in Fig 66 I see a squiggly blue line. That’s the G-meter. It’s bouncing all over the place because the plane was in turbulence. It has nothing to do with the lift coefficient or the curve or anything other than the Gs from one second to the next.

    I don’t have exact data on the Airbus airfoil and neither do you but I have plenty of experience with supercritical airfoils at Reynolds numbers over 5 million and I can tell you that when it comes to C/l they pretty much have a bell shape at alphamax. Increase alpha and C/l drops. Period.

    Assume a deck-angle alpha of 45 degrees and add a rough average of the design incidence of the wing itself and you have an average foil alpha somewhere around 47-48 degrees. When I had to reverse-engineer the B777 airfoil for X-Plane there was no good data so I ended up loosely basing it on the 0714 and I think the mid-foil had an alphamax of something like 15deg at zero mach and Reynolds 75-100million and maybe 10 degrees at M0.89 (I really don’t know because X-Plane computed compressibility for us). Assuming the Airbus is similar, 48 alpha means you’ve got a wing (of continually varying incidence) running (an average of) 33-36 degrees past alphamax.

    I’ve never seen an airfoil plotted past about 18 degrees because there’s no point. In that part of the curve you have a massive loss of lift. AF447 was in that part of the curve from 37,000’ on down.

    There’s about half a dozen other variables between paper plots and real-world stall characteristics of an airplane (fuselage lift, for example) but your little blue G-meter line in Fig.66 of the AF447 report ain’t one of them.

    –“When an airplane descends at approximately constant speed on a flight path that is 45 degrees down, that means that the balance of thrust and drag is about equal in magnitude to the lift force, and that the resultant force of thrust, lift and drag is about equal in magnitude and opposite in direction to the weight of the airplane.”

    I get a migraine even trying to decipher that paragraph, much less see how it supports “hardly any loss of lift.” I just can’t penetrate the language you’re using. Are you trying to tell me that lift equals weight in a descent with engines at full power pitched up 16 degrees?

    Forces in flight are theoretically always equal until there’s an acceleration (which is pretty much always, which is why it’s theoretical). In a theoretical unaccelerated descent (and I really don’t think AF447 qualifies as unaccelerated), there are VERTICAL and HORIZONTAL components of three of the four forces. VERTICAL of lift plus VERTICAL of thrust equals weight PLUS the VERTICAL component of drag. Ditto HORIZONTAL of thrust equals the HORIZONTAL of lift PLUS the horizontal component of drag.

    And how does that help you defend “hardly any loss of lift?”

    You don’t speak in the terms I’m accustomed to from anybody who knows this stuff so let’s agree to just stop. Everyone else can decide what our respective levels of credibility are.

    What’s clear is that MM is on the record as saying AF447 should be its own thing and cease to stink up Jeff’s otherwise worthwhile comments thread.

  8. Tbill,Johan others..

    Here’s some technical info about Jorn…

    “The Jindalee network is a system of over-the-horizon-radar which bounces high frequency radio waves off the ionosphere. When these waves hit a metal object of sufficient size some of them are reflected back and computer systems process these signals and detect objects far from the Australian coast. However, this system is different from ordinary radar in that the high frequency waves do not sweep over an area, but rather focus on one area at a time. In the diagram below (Figure 1) the area examined is called the radar footprint. There are three systems in this network, one located at Longreach Qld., one in Alice Springs NT., and one in Laverton, WA. “Radar data from these sensors is conveyed to the JORN Coordination Centre within the Air Force’s No 1 Radar Surveillance Unit (1RSU) at RAAF Base Edinburgh in South Australia. 1RSU is tasked by higher headquarters to operate the JORN capability on a daily basis.”

    And

    First there are a number of factors which are relevant to its performance, such as the state of the ionosphere, weather conditions and solar disturbances. However the most important reason why it might not have been detected is that it was not working at the time or it was not looking in the right place. Remember that it does not “sweep” like ordinary radar, and, quoting from the JORN Fact Sheet:

    “JORN does not operate on a 24 hour basis except during military contingencies. Defence’s
    peacetime use of JORN focuses on those objects that the system has been designed to detect,
    thus ensuring efficient use of resources.”

    Further, John Blaxland of ANU’s strategic and defence studies centre said “the JORN system would probably have to have been programmed to look for MH370 in advance.”

    https://www.airforce.gov.au/docs/JORN_Fact_Sheet.pdf

  9. @Matt Moriarty:

    Why don’t you ” reverse-engineer ” the DFDR data in the accident report and get a lift curve beyond the stall for yourself?

  10. I enjoy reading the comments on this blog and try to visit each day.

    I am a bit baffled by the focus on AF447’s final minutes. Unless someone thinks the person flying MH370 at the end was pulling back on the yoke (which I don’t believe anyone is asserting), the comparison between the final minutes of the two flights seems irrelevant.

    The firm that helped find AF447’s recorders could be helpful in the MH370 search having found a needle in a haystack previously…regardless of the AoA!!

  11. @NYBanker:

    In my post of Nov. 6 at 2:59 AM I wrote: “If the flight did end similar to AF447, that would imply that there was someone at the controls who had not the intention to die.”

    Examination of the flaperon and the outboard flap segment indicates that as a possible end-of-flight scenario. There is no evidence of an unresponsive crew.

  12. Sidenote on weather. JORN could be used to foresee weather couldn’t it? I wonder what Sweden uses in that regard. Sweden remilitarised Gotland in September through making stationary the troops that was visitng the island then. Now Gotland (Visby, with its oldtown inside the 13th century City Wall to be exact) has received 31 cm snow in two days and 5 more is expected this morning — which is a bit early in the season. So the military will come in handy. Was it in reality the Russian winter that was feared the most?

  13. @Gysbreght, If the person in control of 9M-MRO did not wish to die, why did they fly into the southern Indian Ocean to the point of fuel exhaustion?

  14. @Jeff Wise:
    There are several possibilities that I could think of, but that would be idle speculation. My starting point is the condition of the recovered debris, in particular flaperon and outboard flap, and also the ‘no step’ piece.

  15. @Gysbreght, Respectfully, that’s not a very good place to start, because despite some very strong claims from some commenters that they can tell exactly how the plane crashed based on the size and condition of the fragments retrieved so far, the state of the crash is still so far very ambiguous. High-speed crashes can leave some fairly sizable pieces; low speed crashes can produce small ones.

    I think we can be fairly confident, however, that the moment before impact, the plane was not configured in the manner of AF447, as there was no engine power available. Likewise, there is no reason to imagine that the plane would be held in some kind of unfortunate semi-stall (“mush”) by a pilot who had become disoriented in meteorological conditions, as none prevailed. Could the pilot have been deliberately holding the plane in a mushy stall? Certainly. He could have been standing on his head and singing “Hello, Dolly.” As DennisW has many times pointed out, turning the plane to the south and flying to fuel exhaustion is an irrational thing to do, so trying to come up with reasons to justify it is a fruitless exercise. The best we can do is to define the range of what is physically possible and statistically plausible as narrowly as we can.

  16. @Jeff Wise: You write as if you are sold on the “ZS did it” fiction. There are other possibilities.

  17. @Jeff Wise, you wrote: “The best we can do is to define the range of what is physically possible and statistically plausible as narrowly as we can.”

    A person at the controls at the end of the flight who was not a pilot and did not want to die is physically possible. Can it be rejected on the basis that it is not “statistically plausible” ?

  18. @matt (non-Moriarty):

    You disappeared on me. I promised you an answer.
    I still believe the (political) martyr is reading too much into Z. I haven’t revisited the videos but I know others here didn’t see much suspiscious in them. There is for me not enough credibility in that. As it stands. Not as a sole explanation. He must have been completely whacko in that case, and that ought to have shown. As I have said before: you can’t get the idea to promote democracy through murder and terrorism (if you are not living in a dictatorship, and then you don’t kill the innocent ones.) There is no lesson to teach anyone sane from a deed like this one. Certainly not his children. You may do a lot of silly things but not this. No way. It can be done for revenge, “impotence”, devastation, dishonor, disappointment, insurance money, lack of will to live — perhaps with a shroud of self-assuring good will — but not as a good deed in itself.

    You spoke of the fallacy of self-exemption, but this is the fallacy of the fallacy of self-exemption. It doesn’t lead to anything but self-exemption with an exponential figure. You don’t need to be a professor of philosophy with a hut in the Schwarzwalds to figure that out (but that one didn’t). There is no way he could have believed in it.

  19. The U.S. astronauts have voted reports the news. I thought only earthlings could vote. But Trump will of course need the extra help.

    Let the elections begin.

  20. @Gysbreght
    Your graph of flight altitude after IGARI was referenced/reposted over on Reddit. I was noticing the time scale on your plot suggests only about 4 minutes above 38000 feet. Am I reading that correctly? Was there an earlier discussion I can go back to get the background?

    @Johan, @Aaron
    Thank you for the JORN discussion. I had read the fact sheet, but that seemed to be written after MH370. But sounds like they would need to know MH370 was heading down there.

    Maybe that explains the apparent “loiter” in the Andamans to head north for a while and then double back south to confuse any possible visual or other tracking. Maybe he knew the Singapore AWACS could be over there.

  21. @Gyabreght@Jeff

    “There are other possibilities.” (Than ZS did it.)

    None that can withstand even a cursory examination. You guys are talking nonsense again.

  22. @TBill:
    “your plot suggests only about 4 minutes above 38000 feet. Am I reading that correctly? Was there an earlier discussion I can go back to get the background?”

    I don’t recall much discussion at the time. Translating the speed variations into altitude variations at constant total energy is quite simple. At 500 kt TAS you get 44.3 ft altitude for 1 kt TAS. At lower speeds proportionally less. So the altitude trace is a sort of mirror image of the true airspeed trace.

  23. @strangelove

    >X-37B capabilities

    The X-37 has been flown to date in relatively low inclination orbit (45degrees) so can only access latitudes in the band 45S to 45N, not whole Earth. All reconnaissance satellites (if the X-37B was carrying such equipment in 2014 – it is rather small for telescopic cameras of the size carried by the US Keyhole series) are limited in capability to adjust their orbit. The revisit time (time between passages over the same spot on the ground) can be adjusted by change of orbit altitude if a particular area becomes important, such as at time of war. However, it is expensive in fuel so is not performed for minor (i.e. non-defence) issues.

  24. @Gysbreght:
    🙂 . Where do I turn to pick up my service ribbon?

    The single survivor non-pilot coming to at fuel exhaustion is a very rare flower, though.

  25. –“A person at the controls at the end of the flight who was not a pilot and did not want to die is physically possible. ”

    Hallelujah!

    At long last, the shroud of coyness is lifted and an actual offering is made as to the mechanics of a belly-flop-finish to MH370.

    I imagine it feels that a great weight has been lifted. Congrats, man.

  26. @Jeff

    So take her wrap, fellas
    (aft yoke)
    Find her an empty lap, fellas
    (full aft yoke)
    Dolly’ll never go away again!

  27. @Matt Moriarty:

    Your mindset must be extremely limited if you are unable to discuss “the mechanics of a belly-flop-finish” on a technical level without guessing who might have been responsible for it.

  28. @Gysbreght

    My mindset is limited by whatever I perceive to be the surest path to solving the mystery of MH370. If I’d gotten my wish and the search had been turned over to the experts, maybe we’d all be exercising our typing skills a little less.

  29. @Matt M

    I feel the terminus swayin’…while that ole SSWG keeps on playin’
    Their old favorite scenario from way back when…

    In the meantime, my “crime fighting” Macbook screen has not had a new “sticky” since the simulator revelations.

    http://tmex1.blogspot.com

  30. @DennisW

    Ok, everybody now…

    Hello, Dolly! Well, hello, Dolly!
    It’s so nice to have you back where you belong…

  31. @Johan: “The single survivor non-pilot coming to at fuel exhaustion is a very rare flower, though.” If you narrow the range of possibilities that much, you may be right again.

    My window is about five hours wide, and is not limited to a single survivor. Why are you talking about survivors anyway?

  32. @Matt Moriarty & Dennis W – Great News:

    Hello Dolly is coming back to Broadway! It will star David Hyde Pierce & Bette Midler. Previews begin 3/15/2017 at the Shubert Theatre.

  33. @LaurenH

    Bette Midler! See you in line at the TKTS! (Do they still have that in Times Square?)

    (Btw, if a comment thread was ever going to go off the rails, Nov 8, 2016 is the perfect day, no?)

  34. @Aaron:
    “…JORN does not operate on a 24 hour basis except during military contingencies. Defence’s peacetime use of JORN focuses on those objects that the system has been designed to detect, thus ensuring efficient use of resources.”

    Wow, that’s two statements from the AGDD which directly contradict themselves (see previous post).

    Perhaps this is the real reason why JORN didn’t see anything (from the AGDD site again)…

    “…Given (MH370’s) range from individual OTHRs, the ionospheric conditions and a lack of information on MH370’s possible flight path towards Australia, it is unlikely that MH370 would have been detected if the system had been operational.”

    Looks like it would be a piece of cake for a hostile country to knock out key Australian defence targets, providing they went in at night and used a very very large plane. Looks like JORN is a complete waste of $2billion.

  35. @All
    Between all the thrusts and parry’s.

    There is a sharp disagreement among competent men as to what can be proved and what cannot be proved, as well as an irreconcilable divergence of opinion as to what is sense and what is nonsense.
    —ERIC TEMPLE BELL. Debunking Science

    Cheers Tom L

  36. @Lauren H, @Moriarty:

    Betty on Broadway, not a day too early. Bad thing the SEK is going down in flames.

    Matt, pulling you ferociously back on track again, there is an obvious connection between Midler and MH370 in that guy Blaine Gibson. So youre not that far out.

Comments are closed.