Yesterday, officials responsible for locating missing Malaysia Airlines Flight 370 announced that their two-year, $150 million search has come to an end. Having searched an area the size of Pennsylvania and three miles deep, they’ve found no trace of the plane.
The effort’s dismal conclusion stands in marked contrast to the optimism that officials displayed throughout earlier phases of the search. In August, 2015, Australia’s deputy prime minister Warren Truss declared, “The experts are telling us that there is a 97% possibility that it is in [the designated search] area.”
So why did the search come up empty? Did investigators get unlucky, and the plane happened to wind up in the unsearched 3 percent? Or did something more nefarious occur?
To sort it all out, we need to go back to why officials thought they knew where the plane went.
Early on the morning of March 8, 2014, MH370 took off from Kuala Lumpur en route to Beijing. Forty minutes passed the last navigational waypoint in Malaysian airspace. Six seconds after that it went electronically dark. In the brief gap between air-control zones, when no one was officially keeping an eye on it, the plane pulled a U-turn, crossed back through Malaysian airspace, and then vanished from military radar screens.
At that point the plane was completely invisible. Its hijackers could have flown it anywhere in the world without fear of discovery. But lo and behold, three minutes later a piece of equipment called the Satellite Data Unit, or SDU, rebooted and initiated a log-on with an Inmarsat communications satellite orbiting high overhead. An SDU reboot is not something that can happen accidentally, or that airline captains generally know how to do, or that indeed there would be any logical reason for anyone to carry out. Yet somehow it happened. Over the course of the next six hours, the SDU sent seven automated signals before going silent for good. Later, Inmarsat scientists poring over the data made a remarkable discovery: due to an unusual combination of peculiarities, a signal could be teased from this data that indicated where the plane went.
With much hard work, search officials were able to wring from the data quite a detailed picture of what must have happened. Soon after the SDU reboot, the plane turned south, flew fast and straight until in ran out of fuel, then dived into the sea. Using this information, officials were able to generate a probabilistic “heat map” of where the plane most likely ended up. The subsequent seabed search began under unprecedented circumstances. Never before had a plane been declared lost, and its location subsequently deduced, on the basis of mathematics alone.
Now, obviously, we know that that effort was doomed. The plane is not where the models said it would most likely be. Indeed, I would go further than that. Based on the signal data, aircraft performance parameters, and the available autopilot modes, there is a finite range of places where the plane could plausibly have fetched up. Search vessels have now scanned all of them. If the data is good, and the analysis is good, the plane should have been found.
I am convinced that the analysis is good. And the data? It seems to me that the scientists who defined the search area overlooked a step that even the greenest rookie of a criminal investigator would not have missed. They failed to ascertain whether the data could have been tampered with.
I’ve asked both Inmarsat scientists and the Australian mathematicians who defined the search area how they knew that the satellite communications system hadn’t been tampered with. Both teams told me that they worked with the data they were given. Neither viewed it as their job to question the soundness of their evidence.
This strikes me as a major oversight, since the very same peculiar set of coincidences that made it possible to tease a signal from the Inmarsat data also make it possible that a sophisticated hijacker could have entered the plane’s electronics bay (which lies beneath an unsecured hatch at the front of the business class cabin) and altered the data fed to the Satellite Data Unit.
A vulnerability existed.
The only question is: Was it exploited? If it was, then the plane did not fly south over the ocean, but north toward land. For search officials, this possibility was erased when a piece of aircraft debris washed ashore on Réunion Island in July of 2015. Subsequently, more pieces turned up elsewhere in the western Indian Ocean.
However, as with the satellite data, officials have failed to explore the provenance of the debris. If they did, they would have noticed some striking inconsistencies. Most notably, the Réunion debris was coated completely in goose barnacles, a species that grows only immersed in the water. When officials tested the debris in a flotation tank, they noted that it floated half out of the water. There’s no way barnacles could grow on the exposed areas—a conundrum officials have been unable to reconcile. The only conclusion I can reach is that the piece did not arrive on Réunion by natural means, a suspicion reinforced by a chemical analysis of one of the barnacles by Australian scientist Patrick DeDeckker, who found that the barnacle grew in water temperatures that no naturally drifting piece of debris would have encountered.
If the plane didn’t go south, then where did it go? Not all the Inmarsat data, it turns out, was susceptible to spoofing. From the portion that wasn’t, it’s able to generate a narrow band of possible flight paths; they all terminate in Kazakhstan, a close ally of Russia. Intriguingly, three ethnic Russians were aboard MH370, including one who was sitting mere feet from the electronics bay hatch. Four and a half months later, a mobile launcher from a Russian anti-aircraft unit shot down another Malaysia Airlines 777-200ER, MH17. A year after that, the majority of pieces of debris wind up being discovered by a man who had spent the last three decades intimately involved with Russia.
Whether or not the Russians are responsible for MH370, the failure of the seabed search and the inconsistencies in the aircraft debris should undermine complacency about the official narrative. When MH370 disappeared, it possessed an obscure vulnerability that left its Inmarsat data open to tampering. Having spent $150 million and two years on a fruitless investigation, search officials have an obligation to investigate whether or not that vulnerability was exploited.
@VictorI:
That’s another way of putting it. (And on top of that China had a “one child per household” benefit policy for 35 years, changed to “two/HH” only last autumn) and tumbling nativity, meaning that many on the plane might belong to families with few members due to that (although frequently flying Chinese perhaps isn’t exemplary when it comes to birth control)).
Sadly I don’t have dropbox since a while, but I am prepared to believe you. The Chinese could be more forthcoming in this. The current most likely scenario will not get them on their feet much though. But suspiscions they have anything to hide would.
@Ge Rijn, You wrote, “To prove this evidence is wrong you need other evidence at least equaly strong to undermine it. Not well designed speculations but evidence. In Jeff’s scenario there isn’t any evidence still that can stand this test.”
Respectfully, I disagree. I feel that there is considerable evidence that fits a spoof scenario better than a suicide scenario. To wit:
— The reboot of the SDU, and the unexplained 18:25 BFO value
— The absence of wreckage in the seabed search area
— The inconsistencies of the barnacle distribution and Mg/Ca values
— The statistical impossibilities of Blaine Alan Gibson’s finds, coupled with his deep Russia ties
— The blatant shoot-down of a sister plane 4.5 months later
It’s plausible that any one these could be the result of coincidence, but I don’t think it’s plausible to explain them all away.
I completely agree with you.
In which manner the evidence is interpreted and used (e.g. pro or contra a certain scenario) is subjective.
However, where you put an idea along this continuum is entirely subjective.
You are applying your own subjective assessment in deciding how to interpret the data and so-called „evidence“. And subjective assessment is by definition error-prone.
There is no scientific formula or any objective, scientific way to do so.
Especially if the evidence, the science should be based upon, is itself in question, as Jeff points out in the article above.
(Your y axis IS the continuum you just described above. But anyway.)
Ok, then go ahead and put the theories here on x and y axes.
You will see, that there is no scientific formula for doing this.
It’s an entirely subjective choice.
The scientists among you are aware of that, or at least should be.
@Jeff:
I believe DennisW would add the unlikeliness of a depressurization scenario. Due to the neat offset values.
@VictorI,
After further inquiries, I believe that the FMS works like the FSX version for route offsets. It’s back to the drawing board for me. My #4 in the straw-man scenario for the FMT appears to be incorrect. I am told that the lateral offset is removed when a “Direct-To” fix is inserted in the ACT RTE LEGS page. There does not appear to be any documentation in the Honeywell manuals for this, but that seems to be what it does. So we have another case of incomplete or unclear documentation. I am beginning to believe that FSX is a very reliable indicator of the real FMS behavior, even in the case of the EOR heading mode. I have pressed this issue with ATSB several times, but they simply repeat the same statement from Honeywell about the switch determining the heading mode. The only way I know how to prove what it really does is an extended run after EOR in a full-up simulator, just like you did with FSX.
Your experiments with FSX are valuable in understanding these “abnormal” situations where the documentation is not definitive. Thank you for that.
@Johan
Yes, the likelihood of a depressurization before 18:25 is quite small, IMO.
@DennisW
Does the SDU bus shut down impact the furnace as well? If so, why would fairly slow depressurization event necessarily have any additional impact?
So with the search ended we are left with the final statements from officials before the loss of MH370 and its cause is finally buried behind closed doors.
Malaysia is offering a large reward for any privateer who find the plane but, apart from that crumb of help, it looks like the Malaysian government are happy for the planes fate to ever remain a mystery.
I have much sympathy for the NOK, but fear they will never discover what actually happened to their loved ones.
@TBill
My understanding is that an SDU shut down removes power from both oscillator and the oscillator oven.
“Fairly slow” is difficult to quantify. If the change in pressure with the attendant change in temperature is slow enough, the oven can easily chase the temperature change and largely, but not completely, compensate for it. If the change in temperature is more rapid, the oven cannot respond with sufficient agility. Oscillator ovens do not have unbounded compliance, meaning they are typically designed with sufficient power to compensate for diurnal temperature changes in an outdoor environment. Not a temperature change associated with a rapid (whatever that means) decompression.
@DrBobbyUlich: I am happy to run these simple test cases in the PMDG 777 model. The FCOM for the B777 is included in the documentation package, with the implication that the behavior of the model closely matches what is claimed in the FCOM. Caveat emptor. (I will say that the fidelity of the PMDG model greatly exceeds the much older PSS model.)
I have been doing other studies of the dual flameout at the 7th arc. I am less convinced that the log-on of the AES at 00:19 was due to electrical power to the left AC bus from the APU. The power to the left AC bus may have been supplied by one of the main generators, which were driven by the torque supplied by the windmill effect. (This effect can be used to restart an engine in flight in the event of dual engine flameout and no APU.)
I also am more inclined to believe that after fuel exhaustion, the autopilot remained engaged until the speed reduced to the point where the backup generators, also driven by the windmilling spools, could not hold up the right transfer bus. There are some implications on timing, speed, bank, rate of descent, and distance of the terminus from the 7th arc that I am still trying to understand. Unfortunately, the behavior is a strong function of unknowns such a rudder out-of-trim and the stabilizer trim at the point the autopilot disengages.
@DennisW:
“Rapid” may not be what you intended — to those who have visited the Wikipedia page, where a stipulated terminology is used. I quote in excerpt:
“Uncontrolled decompression is an unplanned drop in the pressure of a sealed system, such as an aircraft cabin, and typically results from human error, material fatigue, engineering failure, or impact, causing a pressure vessel to vent into its lower-pressure surroundings or fail to pressurize at all.
Such decompression may be classed as Explosive, Rapid, or Slow:
Explosive decompression (ED) is violent, the decompression being too fast for air to safely escape from the lungs.
Rapid decompression, while still fast, is slow enough to allow the lungs to vent.
Slow or gradual decompression occurs so slowly that it may not be sensed before hypoxia sets in.
[…]
Rapid decompression:
Rapid decompression typically takes more than 0.1 to 0.5 seconds, allowing the lungs to decompress more quickly than the cabin.[1][5] The risk of lung damage is still present, but significantly reduced compared with explosive decompression.”
@Victorl
Thank you for your excellent work.
https://www.dropbox.com/s/qdg1jcnc7nfxahp/2016-05-18%20China%20and%20MH370.pdf?dl=0
Are those events considered as unfortunate coincidences?
@all
The linked article below appeared yesterday. It is good, IMO. Numerous IG inputs that I happen to agree with.
https://www.theguardian.com/world/2017/jan/19/why-hasnt-mh370-been-found
VictorI-
I’d be curious to see what PMDG thinks engine flame out looks like visually for low fuel. I’d also be curious to know what it really looks like too. I see nothing online for that case.
I am trying to dump fuel in PSS to see what that looks like, but for some reason not getting fuel to jettison yet. Also wondering if cabin depressure make a visual plume.
@Johan
My contention is that the BFO data suggests that a decompression event did not take place, but if it did that it had to be slow.
I would even seriously question whether power was removed from the SDU given the accuracy of the 18:25 BFO value. As Victor pointed out, it is possible that power was removed and restored, and that the oscillator recovered to its nominal frequency. It is even possible that some serendipitous confluence of events took place such as an aircraft maneuver that masked an oscillator error. Who knows?
@TBill: There is nothing visually exciting at flameout. If you look at the front of the engine, the fan slows. At some point, the RAT drops down and you can hear it whirring away.
@DennisW:
I do of course realise that you are speaking partly of something else, and that you know what you are talking about — but since you hesitated, and, since I have theimpression that people (from how they ask questios) here have read the wiki page. In short: it appears probable — if no other technical or other limitations will set the threshold — that the oven you are referring to is not meant to handle decompression events whether slow or not so slow, without a noticeable offset slip, when change is quick or substantial and protracted.
It would of course be interesting to learn more about what it would actually have had to endure during a depressurisation event.
@DennisW:
Possibly “prolonged” instead of “protracted”.
@DennisW:
I didn’t see your last response to me before I wrote again.
@Johan
Yes, that is my opinion.
@DennisW:
Thanks for the Guardian article. It swayed for a moment at 3/5, as if the tone was lost for a moment, but was very well written and conceived. Ha ha — yes it got many things covered. I hope the author is reading our pages.
By the way, just to get it off my table: the famed “descent” or “turn south” value couldn’t be the result of depressurization could it? Someone would have brought that up. Not the proper table column…?
@Johan
An interesting sidebar (perhaps only interesting to me) is the notion of oscillator training. Training is the common vernacular although it is not exactly appropriate, IMO. The idea is that if you have a temperature sensor (consistency being important, not accuracy) and a primary reference such as GPS or a cesium, you can write code that extracts the two main causes of oscillator error – that would be temperature and aging. After you learn the oscillator behavior with a suitable training period (typically several days of temperature cycling), you can use the derived temperature and aging coefficients to control the oscillator when GPS or the cesium is removed. It works very well particularly when the primary reference is available periodically to update the training data. This is what is done for CDMA base station clocks. Even though GPS is theoretically available at all times the reality is that in the US GPS antennas, especially in rural areas, are the favorite target of redneck marksmen. We ended up selling a lot of replacement antennas until the service provider designed industrial strength shielding for them.
As a second sidebar antennas is the correct plural form in this case. Antennae being the preferred form when insects are being talked about.
@all
Some enterprising person should check the ISAT data against Benford’s Law.
https://en.wikipedia.org/wiki/Benford's_law
That’s been done, @dw
The official searching is over. Perhaps for ever.
Sometimes it’s best to step back to first principles, to look at the original information, as it happened. Redacted and filtered as it is, the PPRUNE MH370 thread is a good wayback machine for as-it-happened news, hunches and reactions. Apart from a few disputed structural and electromagnetic relics, nothing much has changed since then.
I’m sure that if so much as a stray photon was emitted from that flight after the turnback, apart from the SDU pings, every eye in the area would have detected it and acted on it; I don’t think a two-way negotiation is a viable situation at all. What was Captain Shah doing via WeChat whilst waiting to taxi – if these reddit rumours are true, this could be as conclusive to one theory as forensic evidence proving that supposed 9M-MRO debris actually came from 9M-MRD would be to another.
Someone wrote of dreams earlier; I haven’t had any MH370 sleeping nightmares, but I can still feel the visceral certainty that I had seen:
a) A mystery that may never be solved to anyone’s satisfaction, and,
b) The perfect crime committed.
When the full horrors of what could have happened between diversion and last attempted handshake became feasibly imaginable, I think investigators chose to concentrate on the least disturbing option of search – the ghost ship without the live man at the controls. Was that a chandelle to FL450 just after IGARI and a merciful decompression, or did two-hundred and thirty-nine people… We will probably never know.
My first instinct, as the bemused magnitude of this disappearance slowly dawned on us, was that some sort of sacrifice or blood tribute – I have mentioned this here before – had taken place, the knife plunging out of the equonoctal sky into wherever the PIC believed Osama bin Laden had been dumped overboard several Springs before, perhaps even as a cover for a different motive or ideology. Looking at subsequent depth of evidence, the nihilism of the Germanwings incident and my own psychological suspicion, I’d go for Broken Ridge as the final destination, but something about the floatation and other debris evidence makes me think back to what I first felt. This whole event was weird from the moment of breaking news. And I can’t see an airframe, that supposedly crippled, flying and transmitting sattelite data for that length of time.
This must be Hell for the NOK.
@PaulS
Do you know the result?
@VictorI,
You said: “I have been doing other studies of the dual flameout at the 7th arc. I am less convinced that the log-on of the AES at 00:19 was due to electrical power to the left AC bus from the APU. The power to the left AC bus may have been supplied by one of the main generators, which were driven by the torque supplied by the windmill effect.”
In order for the AES to log on at 00:19, it must have lost power for more than a few seconds which was then subsequently restored. If the power was first supplied by one engine-driven generator (IDG or backup) and then by a different generator, I think the switchover would be rapid and the AES would continue operating normally so there would be no re-boot. Are you suggesting an engine-driven generator lost drive power after flame-out and then regained it during the descent (because of increased windmill speed)?
@DrBobbyUlich
That is what he is suggesting. Also, presumably, he considers that after
flameout, the RAT is contributing electrical power.
@DrBobbyUlich: First, the SATCOM is fed by the left AC bus. If the APU didn’t supply power for the power up of the SATCOM, it would be from an IDG, not a backup generator. The backup generators supply power to the transfer busses, and the transfer bus can’t feed the AC bus (although an AC bus can and does feed a transfer bus).
Assuming no APU start up, for the autopilot to disengage, power must have been lost to the transfer busses, which was fed by the backup generators. At this point, power was already lost to the AC bus because the IDGs cut out at a higher spool speed (and therefore IAS) than the backup generators. In order for the SATCOM to restart, the IAS must have increased to the point where the IDGs were again supplying power to the AC busses.
So yes, during the descent, the IAS and the rotational speed of the relevant spool might have increased to the point where the IDG was again producing power and maintaining the voltage on the left AC bus.
Malaysia is withholding evidence that could help the investigation.
http://www.news.com.au/travel/travel-updates/incidents/malaysia-military-agencies-deliberately-withholding-radar-data-that-could-help-find-mh370/news-story/a2197eee178ba5807988c3454ff7561e
@VictorI
Malaysia sucks, no doubt about that. People questioning a politically motivated diversion need to rethink their positions carefully. There is little doubt in my mind that the Malays not only knew of the diversion after IGARI, but were complicit in the events following that diversion.
@ Dennis Do you believe that Z accomplished his goal with this diversion? What do you suppose was that goal? It seems to me the only victims were those on board the plane and the NOK. Perhaps a gross miscalculation by Z or something went awry with his plan?
@Shadynuk
I believe Z acted out the last details of a failed negotiation with the Malay government. What that negotiation might have been about is subject to speculation.
I doubt the windmill effect supplying power. without startup procedures, there would be no exciter field power so no magnetic field, so no induced current
@DennisW, I could not agree more as it relates to MY government. Collectively we have often mentioned their withholding of critical evidence and sabotaging the investigation. It begs the question: at what point was the IG aware of this? And why continue a search when you know critical pieces of the puzzle are missing? This makes no sense. If only NOK could obtain this data using legal avenues or by other means. Wishfull thinking on my part, no doubt.
@All, If MY placed more than 2 calls, and conveniently left this information out, would that alter the ISAT data as we know it today?
@ Dave Opperman
“I doubt the windmill effect supplying power. Without startup procedures, there would be no exciter field power so no magnetic field, so no induced current”
If that is your interpretation of events, then it implies that the ‘start-up’ was attempted.!!
@DW it was discussed briefly on reddit some time back https://www.reddit.com/r/MH370/comments/4uaifv/a_btobfo_compliant_route_that_heads_for_simulator/d5pbi7r/
@Gysbreght, DrBobbyUlich. Final nose dive.
Here are some indicative figurings for the nosedowns after fuel exhaustion which would lead to the last two BFOs.
I disregard any acceleration due to gravity in the 8secs between them.
Using the:
http://www.atsb.gov.au/media/5771773/ae-2014-054_debris-update_2nov2016.pdf
page 11 Cases A and B BFO outcome extremes, at say 30,000 ft, IAS 175 knots (280 TAS), Case A (minimum ROD) results in an 8 deg nose down at 0:19:29 rising to 31 deg at 0:19:37.
The Case B (maximum ROD) results in a rise from 30 to 62 deg.
For comparison the Ulich 4,600 fpm rising to 9,100 results in 9 to 19 deg.
At a higher speed, 250 knots IAS (349 TAS); Case A results in 6 to 24deg, Case B, 24 to 45, Ulich 7 to 19.
This gives an idea of initial and final nosedowns and the differences.
Gysbreght has mentioned a piloted stall after using speed to maintain height and a zoom earlier a zoom I remember; and there could be a deliberate nose down for other reasons.
Whatever the cause energy height can be saved as potential energy, or used in staying airborne or, in diving, converted to kinetic energy. Much of that can be converted back to p.e. again after a pull-out or again used to stay airborne.
Now, what would be the height lost, even bearing in mind that much of that would be recovered or used after the dive? At the average descent rate, in Case A the loss would be 1,230 ft plus recovery height, Case B 2,613 ft plus recovery and Ulich 910 ft, also plus (of course this does not include height lost in getting to the 0:19:27 nose down in the first place).
I think the figures and energy saving do support a view I have been belabouring recently that neither the BFOs nor the Boeing simulations, nor for that matter the flap positions and IFE non-connection, rule out a glide entirely.
@David
You’re absolutely right. The amount, nature and and condition of the wreckage do not support an uncontrolled impact, either.
@Jeff Wise. Earlier I posted,”I cannot reconcile two of your statements:
• After the loss…“ Inmarsat scientists poring over the data made a remarkable discovery: …… a signal could be teased from this data that indicated where the plane went.”
• “…..the scientists who defined the search area overlooked a step that even the greenest rookie of a criminal investigator would not have missed. They failed to ascertain whether the data could have been tampered with.”
Why would they if at the time of the flight the SDU role in establishing route was unknown, so there was no motive to tamper”?
I think you had some issues at the time and may have missed this but I would welcome a response.
Also you said, “.. DeDeckker,…found that the barnacle grew in water temperatures that no naturally drifting piece of debris would have encountered”.
Can you point to that finding please?
@David: I agree that a glide scenario cannot be excluded because a dive could have been the result of controlled inputs, and thus still able for the BFO anomalies. The question is important because before we go contemplating truly outlandish hypotheses, the more pedestrian explanations (e.g., simple straight & fast flight paths) ought to be excluded first. Given the early flight behavior (no apparent attempted landings) and the existence of flight simulator coordinates to the SIO, the possibility of controlled inputs right up to the bitter end cannot or at least ought not be excluded IMHO.
https://www.dropbox.com/s/qdg1jcnc7nfxahp/2016-05-18%20China%20and%20MH370.pdf?dl=0
The markers on the chart are about 120 nm south of the 7th arc, and show that there are vast swaths of sea area that were reachable by a glide even after fuel exhaustion that were never searched AFAIK. (Thanks to Richard Cole for the overlays.)
I was just taken to task by @airlandseaman on twitter for pointing this stuff out: His response: “No evidence of glide. ZERO. Meanwhile, sim trials, 2X BFO ob’s, lack of 00:21 IFE = steep descent.”
Not sure what the significance of no signals at 00:21, but at least some simulations conducted by Byron Bailey had controlled dives with a 12,000 ft/min (120 knot vertical component) that would account for any observed BFO values.
@ROB: Mr. Exner is still holding to his flutter theory. Can we exclude that for sure? And what about the ATSB’s theory that the flaps were retracted? Trailing edges remind me of the Hudson River ditching attempt.
Sorry: wrong link to chart….
Here is the correct one:
http://i.imgur.com/qQppOL0.png
@David: I agree with your recent observations. It is incomprehensible that the ATSB in its analysis completely ignores the opinion of Inmarsat’s engineers quoted below. They rely entirely on tests conducted by the SDU manufacturer on a stationary unit in the laboratory. The SDU may behave differently in flight, as observed by DrBobbyUlich recently:
FROM: THE JOURNAL OF NAVIGATION, Ashton et al, (Inmarsat), “The Search for MH370”:
@David, Sorry for not responding to your earlier comment, yes, the backend got a bit messed up and I lost track of things (it seems to be sorted out now.)
To address your questions:
“Why would they if at the time of the flight the SDU role in establishing route was unknown, so there was no motive to tamper?” I’m suggesting that the abduction of MH370 was a zero-day hack–that is to say, the attackers exploited a vulnerability that the owners of the system themselves weren’t aware of at the time. This necessarily implies very sophisticated hijackers. The ATSB and the IG alike find it inconceivable that the perps were smarter than they are, and this is a major reason why they refuse to even entertain the idea. Personally, having seen how the Russians have rolled up the West on MH17 and the US election, I do not find it hard to believe at all. Our collective arrogance regarding Russia is one of the main reasons they have been able to wreak havoc undetected.
“.. DeDeckker,…found that the barnacle grew in water temperatures that no naturally drifting piece of debris would have encountered” See http://jeffwise.net/2016/12/16/the-flaperons-path-to-reunion/
@KarenK, You wrote ” If MY placed more than 2 calls, and conveniently left this information out, would that alter the ISAT data as we know it today?” Yes, if more calls had been placed, they would have shown up in the raw Inmarsat data.
RE SDU in flight: See also ATSB/DSTG “Bayesian Methods” fdigures 5.3 and 5.4.
These errors have nothing to do with OCXO warm-up drift.
@Jeff Wise.
Jeff, I still have difficulty seeing how the Russians would have not only anticipated the discovery of the relevance of satellite ping data to aircraft route but been so confident that it might be imminent that they went to the trouble and risks of falsifying the data.
About De Dekker, at your site’s introduction you said, “The closest we see to this pattern would, I suppose, be drifter 71030 or 41337. But neither of these experienced water temperature lower than 21 degrees. (Of the four drifters that didn’t fit on the graph, the coldest temperature experienced was 23 degrees.)
To my mind, this suggests that the flaperon may not have arrived at Réunion through a natural process of drift.
To be sure, there are other possible explanations for this apparent anomaly”.
I thought that might be overtaken by our 2nd January exchange after Richard Godfrey’s work was published which rather superseded the earlier drifter data.
You did say you had e-mailed De Dekker about his recent visit to France. I assume without response?
@David, You wrote, “I still have difficulty seeing how the Russians would have not only anticipated the discovery of the relevance of satellite ping data to aircraft route but been so confident that it might be imminent that they went to the trouble and risks of falsifying the data.” One simply must imagine that the Russians were clever. If you find this hard to do, go back and look at press coverage of the MH17 shoot-down. For over a year their misinformation campaign fooled literally everyone.
You also wrote, “I thought that might be overtaken by our 2nd January exchange after Richard Godfrey’s work.” Godfrey’s work overlooked both seasonality, which obviously plays a crucial role in water temperatures, and Lepas growth rates, which tightly constrains the time-frame in which the observed temperature change could occur.
@David, PS, I have not yet followed up with De Deckker.
@Warren
Mr Exner is holding to his flutter notion That is a shame, really, because there is very little evidence for it. In theory, it’s not impossible for the flaperon to have been wrenched off due to flutter, if only one actuator was powered and the aircraft reached a very high speed. However, there is no way the RH outboard flap could have suffered the same fate. Very little came off the LH wing. If flutter was involved in any way, we would also have more bits from the LHS. Boeing themselves won’t have any truck with flutter. The IG got together to disprove the pilot suicide theory (they couldn’t deal with the notion a professional pilot could do such a thing) The ATSB liked what the IG were saying, because it simplified things no end: they could discount a final glide, and make the search area more manageable. The ATSB have bent over backwards to keep alive the unpiloted dive at flameout scenario (MY had no qualms over that, for pretty obvious reasons). The flaps may have indeed been retracted at impact, simply because Z would have been aiming for a high speed entry after his glide, to ensure the hull was breached and the plane sunk quickly. You would only want to extend flaps if you were aiming to ditch in an intact state – clearly not what Z was intending Imo.