If you were leading a high-profile international aircraft investigation, in command of the world’s most qualified technical experts and in possession of all the relevant data, would you bother listening to a rag-tag band of internet commenters, few of whom actually work in the space or aviation industry, and none of whom have access to all the data?
Most likely, you’d say: certainly not! But as time goes by, and the puzzle remains curiously impenetrable, you might find it worthwhile to pay a listen to what the amateurs were saying. You might even abandon some of your own conclusions and adopt theirs instead.
This appears to be the case in the search for Malaysia Airlines Flight 370, which disappeared en route from Kuala Lumpur to Beijing back in March. From the beginning, the authorities running the investigation — first, Malaysia’s Ministry of Transport, and later the Australian Transportation Safety Board (ATSB) — held their cards close to the chest, releasing very little information about the missing plane and maintaining a posture of absolute conviction. The investigators’ self-confidence reached its apex in April, when their methodology led them to an area of ocean where underwater accoustic signals seemed to be coming from pingers attached to the plane’s black boxes. Officials assured the press that the plane would be found in “days, if not hours.” But then it wasn’t. A scan of the seabed found nothing; the pingers were a red herring (perhaps literally!). Back to square one.
Meanwhile, on the internet, a group of amateur enthusiasts had come together from all around the world to trade ideas and information about the missing flight. The group, which came to call itself the Independent Group (IG), emerged from various online comment threads and eventually grew to about a dozen individuals. This was a truly spontaneous, self-assembling crowd: there was no vetting of credentials, no heirarchy of any kind. (Full disclosure: I count myself among this group.) Basically, if you seemed to know what you were talking about and could comport yourself in a collegial fashion, you were accepted into the crowd.
While the mainstream press was reporting the ATSB’s pronouncements as received wisdom, the IG was raising red flags. IG members were among the most vocal critics of the ATSB’s contention that the accoustic pings probably came from black-box pingers. And later, after a public outcry led Inmarsat to release a trove of data received from the aircraft, and the ATSB issued a report explaining how it had come to identify its current search ear, the IG dove into the new information with abandon, quickly identifying holes in the data and weaknesses in the official approach. In a pair of papers, the group recommended its own search area, hundreds of miles to the southwest of the ATSB’s officially designated zone.
Today, the ATSB has released an update to its earlier report, explaining why it has decided to reassess its conclusions and move its search zone to a new area — one that overlaps, as it turns out, with the IG’s recommended area. (In the graphic above, the white bracket shows the ATSB area; I’ve added a yellow dot to show the IG area.) Needless to say, this has caused elation within the ranks of the IG, who see the move as vindication of their methods, and indeed validation of their combined efforts over the last few months.
A few observations on the new report:
— One of the reasons the ATSB gives for the shifting of the search area is the recognition that Inmarsat data related to an unsuccessful ground-to-air telephone call attempted at 18:40 indicated that the plane had already turned south at that time. The IG had been basing its analyses on this data point for months.
— Since the June report, the ATSB has improved its BFO model by taking into account various factors — such as temperature shifts caused by the Inmarsat satellite passing through the Earth’s shadow and the mis-location of the Perth ground station in an important Inmarsat algorithm — that IG member Mike Exner has been working through in detail for months.
— The ATSB has fundamentally changed its approach in how it is assessing the plane’s likely path. In its June report, the focus was on what I call the “agnostic” approach: it generated a large number of flight paths based on as few initial assumptions as possible, then graded them based on how well they fit the timing and frequency data received by Inmarsat. This resulted in a population of potential flight paths that fit the data well, but did not make any sense in terms of how a plane might be flown. Some of the routes, for instance, involved multiple changes in heading and airspeed. Today’s report explicitly excludes such flight paths. The ATSB and the IG alike now assume that the last several hours of the flight were conducted without any human input — the crew were presumed to be incapacitated by hypoxia or other causes, so the plane flew on autopilot until it ran out of fuel and crashed. This has been the IG’s starting point for ages, and the fact that the ATSB has now adopted it is a major reason for why the two group’s search areas have now converged.
— You can see in the graphic above how an emphasis on matching the Inmarsat data will tend to lead you in one area (“Data error optimisation”) while an emphasis on routes that comport with real-world autopilot functioning will lead you to another (“Constrained autopilot dynamics”). To be sure, they overlap, but the peak area of one is far from the peak area of another. I think it’s important to realize this, because it helps us to understand why it has been so hard to get a handle on where MH370 went, why the official search area keeps moving, and why knowledgeable people have been furiously debating possible flight paths for months: the BFO and BTO data just do not match up that well. In order to arrive at its recommended area, the IG has been willing to accept much wider deviations from Inmarsat data points than the ATSB has been comfortable with.
— Finally, it’s worth nothing that the ATSB approach is superior to the IG’s in one important regard: it is at heart statistical, looking at families of potential routes rather than proposing and assessing one at a time. There is a tendency, as an individual–and I have fallen into this myself–to cook up a solution, run an analysis, and to be so impressed with the result that one wants to shout about it from the rooftops. (Ask me about RUNUT some time.) The IG has come up with a search area essentially by pooling together a bunch of individual solutions, each of which is generated by a different set of procedures and different set of assumptions. It’s a herd of cats. To really move the ball forward a more rigorous approach is needed, one that takes each procedure and sees how it would play out if the assumptions are methodically modified.
The upshot is that, since the early days of the investigation, the attitude of search officials has changed radically. Once dismissive of amateurs’ efforts to understand the incident, they have clearly begun to listen to the IG and to turn to it for insight and ideas. Indeed, you could say that since the release of Inmarsat data and the issuance of the ATSB report in June, the search for MH370 has become effectively crowdsourced: a de facto collaboration between the professionals and a spontaneous assemblage of knowledgeable experts.
UPDATE:
The overlap between the ATSB’s analysis and the IG’s is more evident in the image below, courtesy of Don Thompson. It shows the fan of values calculated by ATSB to match likely autopilot settings.
It is obvious from the hourly arcs that MH370 made additional turns. The closer spacing between arcs indicate a more tangential course relative to the satellite. And the further spacing between arcs indicates a less tangential course. If MH370 had been flying straight, the arcs would either uniformly increase and/or decrease. Instead we see that in some cases an hour of flight time, resulted in a significant change in distance from the satellite, and in other cases an hour of flight time resulted in much less change.
@Gysbreght
The temperature analogy is not similar at all. The AES Doppler is reduced from more than 1000 Hz to a few 10s of Hz, a factor of >10 typically. Thus, the S/N is reduced by >X10. This is what I meant by a “weaker signal”.
@airlandseaman,
Just to be pedantic (apologies) is the S/N not similarly reduced when one changes from Kelvin to Celsius?
More seriously, if you consider that it really the AES frequency compensation that tells us the speed and direction in which the airplane is moving, because the AES calculates it from those parameters, is your S/N really applicable?
P.S.
Sure, the BFO signal is weak, because the movement of the satellite is small relative to the distance between satellite and aircraft. But does that explain the shape of the probability distribution?
While eating red cactus pears this morning:
“On board a ship with fools at another gutless, witless bit of PR theatre over the search for MH370”
“The disquiet over MH370 won’t end with the finding of its wreckage. There is a respectable view that it’s a race between recovering important clues from the bottom of the Indian Ocean, and various intelligence bodies deciding to disclose more as to what they know.”
Ben Sandilands (@planetalking)
http://t.co/YUAKQ6piwS
So the Indons do have radar?
Indonesian jet fighters force down Aussie plane
THE AUSTRALIAN OCTOBER 23, 2014 12:00AM
Print
Save for later
Peter Alford
Correspondent
Jakarta
TWO pilots, understood to be Australian, were being questioned by Indonesian civil aviation officers last night after their plane was forced down by two air force jet fighters.
The men were named as pilot Graeme Paul Jackline and co-pilot Richard Wayne MacLean, who had flown from Darwin yesterday morning in a Beechcraft plane, call sign VH-RLS.
Indonesian Air Force spokesman Air Marshal Hadi Tjahanto, said two Russian-built Sukhois were dispatched from the fighter jet squadron base in Makasar, southern Sulawesi, after the Beechcraft was detected on radar crossing into Indonesian air space over Yamdema Island, eastern Indonesia.
Marshal Hadi said once it was confirmed the aircraft had not obtained permission to enter Indonesian space, “we forced it down”.
The Beechcraft was instructed to land at Sam Ratulangi Airport at Manado, northern Sulawesi. The pilots told aviation officials at the airport they were flying to Cebu, in the central Philippines.
Marshal Hadi said the interception was co-ordinated between the air force, Transport Ministry and the Foreign Affairs Ministry.
The plane was forced to land about 10.30am Indonesian time and the pilots were still being questioned late last night.
Five Australians were detained by Indonesia for nine months in 2008-09 after their aircraft landed at Merauke, Papua, without them having entry visas. The five Queenslanders were each sentenced to several years jail but their convictions were overturned and they returned to Australia in June 2009.
You are quick on the draw Matty-Perth 😉
And BTW, I wonder if people remember this.
On March 19, it was reported that Malaysia CONFIRMED that it had RADAR DATA showing that MH370 flew OVER (not around) another country. But they wouldn’t say which.
Can you guess?
https://twitter.com/nihonmama/status/452723093908123648
No mystery, Nihonmama. The canonical primary radar track shows MH370 passing over the southeasternmost corner of Thailand shortly after making its 180 at IGARI.
Good point.
@Dr. Ulich
I have been taking a closer read of your monograph, “The Location of MH370.” I’ll focus here on your analysis of the 18:25-18:28 BFO data. On re-reading, it seems that the turn that straddles the line-of-sight to the satellite is not directly inferred from that particular message sequence, but rather from the change in BFO between 18:25, shortly after the plane was tracked headed NW by radar, and the BFOs from the abortive call to the cockpit at 18:39. (The phone call data didn’t make it into the Journal of Navigation paper, incidentally.) I don’t have a problem with the inference of a left/southward turn by 18:39. However, you state in your discussion that a left turn would actually contribute only a negative trend to the BFO, and yet the spike in the 18:25-18:28 sequence is highly positive.
Parenthetically, I did mention in an earlier post that there might be a scenario involving a lag within the air terminal compensation which might allow a turn to manifest as a BFO transient. You do not invoke such an assumption and I have no idea if a system latency of seconds is even plausible, so I’ll evaluate your analysis on its own logic.
You tweak your model to deal with the problem by incorporating a high-impulse climb into the same maneuver, and then declare everything’s okay and that the data reveal a turn in the 18:25-18:28 timeframe — which happens to support another part of your model for the total flight path. The logic seems quite tortured. A physical interpretation of the BFOs supports only a rapid climb, and actually argues against a southward turn at that juncture by your own analysis. It would be more logical to infer a climb and assume that the turn must have occurred later, sometime before the subsequent call attempt and therefore unrepresented by BFO data — only, that would’t be favorable to your preferred model.
I will also note that your discussion of routine “step altitude” changes frankly seems to be a rather desperate attempt to minimize the implications of the unusual climb implied by the BFO spike.
I’m afraid I therefore have to revise my preliminary assessment — I have serious difficulties with both the logic and the presentation in your analysis of the 18:25-18:28 dataset.
?!
Local militants linked to ISIS have threatened to annihilate non-Muslims in Malaysia. Police Inspector General: it’s mostly just “talk”
The Malaysian Insider
http://t.co/6Qbx2E65E2
kthxbye
“What happened in the first four hours when MH370 disappeared?
I am watching with some amazement, the amount of money being expended in the search of the southern Indian Ocean for MH370. I am not convinced by the official version of the final moments of MH370. Nor am I convinced that it is anywhere near the southern Indian Ocean and I am quite familiar with Doppler effect, satellite handshakes and all the other high tech stuff that is being promulgated!
Everyone knows that Malaysia has a military radar system which monitors ALL flights in its area of responsibility. The ex-Deputy PM, Anwar Ibrahim, who the current authorities keep trying to silence, recently stated on BBC TV that he had authorised a state of the art military surveillance system to be installed whilst he was Deputy PM of Malaysia.
Neither Malaysia nor Australia seems to wish to make this information public and could be accused of covering up vital information which would help the families and independent investigators to work out what happened.”
Des Ross
Aviation Business (Asia Pacific)
http://t.co/b9UTLIDCYp
@Nihonmama: The article from Des Ross is excellent. He poses many of the questions that serious investigators are asking. Although I disagree with his assertion that the plane is not likely in the SIO, I do agree with his assertions that Malaysia is keeping evidence in its possession from public view.
Malaysia deliberately misdirected the SAR operation into the South China Sea when it had clear radar evidence to the contrary. This action alone demonstrates that Malaysia has been less than honest and leads me to question anything that cannot be verified by another source. I also question whether the military on the day of the disappearance was explicitly told to “not see the plane”, which would explain why there was no military intercept, why the SAR operation was initially in the South China Sea, and why it was changed only after the US disclosed to the world that the plane was likely in the SIO. Unfortunately, Malaysia is leading the investigation, and in this role has an opportunity to hide what is embarrassing for it.
I have seen no evidence that the UK, the US, or Australia have misled the search of the aircraft in the SIO, although I do believe intelligence bodies have much more data as to the “who, what, why, and how,” especially in the early hours of the flight, than has been publicly disclosed. I do think there is an honest effort on the part of those countries to find the plane.
@Victori
Roger that.
And, who told Malaysia Ops to disinform ATC that the plane was cruising safely through Cambodian airspace and that “signals had been exchanged”? My oh my, who could it be?
Well, it would appear that a bit of a general consensus is begiinng to build in the ranks of the peanut gallery…
We began with wondering whether the behavior of the Malaysian authorities was grounded in incompetence on the one hand or obfuscation on the other. If we now conclude that Malaysia does indeed have a rather robust air defense system; and that obfuscation has been the central driver from the start; then there are people in Malaysia whom are better informed regarding the circumstances surrounding the ‘disappearance’ of the aircraft then they are allowing. From here, we can then begin exploring in a more focused fashion the possible motive for such behavior. Motive will in turn indicate a higher perp and/or a clearer meta-frame for the loss of MH370. The Malaysian authorities appear to want to frame the loss of the aircraft as a ‘mystery’, or at the very least they are simply rolling along with the prevailing public bafflement as to what happened. Why?
As Des Ross has so indicated by way of his ‘coulda-shoulda-woulda’ analysis of the actions of miltary and clvilian aircraft tracking systems operators, the Malaysian authorities should not be so baffled. Even if the actual loss of the aircraft is grounded in the gross incompetence of the operators of Malaysia’s aircraft tracking systems, one would expect the Malaysian authorities to pursue and expose the lapses in process and command-and-control and bring those responsible to the bar. Indeed, eight months on, one would expect at least some surfacing of the results of any inquiry into such clear and evident lapses.
Des Ross, clearly in the know as a veteran aviation and air traffic management specialist, is flabbergasted while framing the response to MH370’s deviation from its intended flight path as ‘criminal negligence’. Perhaps the head that woulda roll is that of the Minister of Defense and acting Minister of Transportation, Hishammuddin. At the very least, I find it ironic that the man ultimately accountable from a 360 perspective for Malaysia’s response to a bugaboo airliner has escaped unscathed. Perhaps the motive for the deliberate obfuscation is to be found in avoiding accountability for the demonstrations of outrageous incompetence on the part of Malaysian civil aviation and military authorties. The Prime Minister of South Korea resigned over the ferry disaster, and individuals get fired for viewing soft port at work, every day. How, then, has Hishammuddin been allowed to get away with what amounts to being stone-cold drunk or asleep on the job? Perhaps his blurting out references to ‘shooting down the aircraft’ in television interviews or his shameless disposition as a victim of these events indicates his very real fear of possibly being held accountable for the loss of an aircraft from his very own flagship airline, which then flew unchallenged through his very own airspace, the very airspace that he was in turn charged with protecting…yes, this is pretty rotten performance on the part of an executive.
The cover up, then, could be wholly attibuted to Hishammuddin advancing favors revolving around power and position to cronies new and old alike in an effort to maintain his role in Malaysian ‘politics,’ if one can even call it that. He is on track to become Prime Minister in the future; the stakes are quite high. Meanwhile, as Des Ross has indicated, a proper forensic investiation involving the Voice Switch recordings found in the various nodes of the air traffic control and military response systems would go a long way towards at least clarifying what went wrong where – but this, too, could implicate the Future PM. Perhaps there is something larger and nastier, such as Luigi’s speculated late night call between the aircraft and the The Boss. Regardless, this would only further be integrated into a damnable account of systemic failure in the systems and org structures under Hishammuddin’s direct control. Certainly, it could be worse – somebody could indeed have been told to ‘not see’ the aircraft, or perhaps the radio transcripts were truncated, but perhaps we can start here, with a cover-up of outrageous incompetence.
Even then, proving a cover-up might not bring down a cronies’ croney, as the leaders of Malaysia are not accountable to anyone, while they marginalize all who oppose them (if you examine Malaysian instituitions from the universtites to the sanitation department you will find them virtually exclusively lead by leading party loyalists).
As for the Americans or the Australians and their agenda, their curious silence informs me that these events are indeed intrinsic to Malaysia, as the flag would have been raised by now on any larger terrorist plot. They would expected to be focused on finding the aircraft to recover the flight recorder and analyzing its deviation from its intended flight path in an effort to learn how to ensure that this does not happen again. In american airspace, this aircraft would have been shot down within 20 mintues; this protocol is already established and has been relatively perfected. What has apparently not been established is a means of ensuring to a reasonably high degree that an aircraft is not hijacked in the first place, as it seems that likely that this is what happened here.
As for the americans not getting up into the faces of the Malaysians or otherwise grabbing their gear in outrage, I believe that we can chalk this up to the rather bland and even uncaring attitude on the part of governments when their interests are not threatened. Why cause trouble when you can smile and shake hands in public while securing wide latitude for the US to engage in a large, clandestine anti-terrorist activities on Malaysian soil? There are people in postions of power in Malaysia clamoring the imposition of Shiriah Law in Malaysia. Were the full extent of US involvement in anti-terrorist activities in Malaysian to become widely known, it coulda lead to a big head rolling down the street – perhaps the same big head? Meanwhile, the US would lose an ‘in’ in the very country where the 911 attacks were largely planned.
Whatever happened in the first few hours of the flight after IGARI quite clearly does not have any form of a constituency, other than the NOK of the victims and the nutjobs of the peanut gallery that can barely get through a day without at least reading the posts on this site.
Don (you lurker): I’ll call my peeps tomorrow.
Nihonmana: apologies for the lack of a proper acknowledgement of your gifts of music and video; I return to Tokyo on Saturday and will indulge myself over the weekend.
Time for bed; apologies, no time for an edit.
Victor, all:
Because I forgot to include it (and proper attribution does matter), the Des Ross article came to me via @PAIN_NET1 – Professional Aviators Investigative Network (AUS).
PAIN is also posting some very interesting things on Twitter about the ATSB, including the scandal involving Pel-Air.
Luigi: your request is in the queue.
Rand: No worries. Hope you enjoy, and if you do (and still buy CDs), Tokyo is the perfect place to pick up their music. Some of SOS’ most amazing work is on the two or three albums released only in Japan. The best import price on Amazon US or UK will be about double what you pay in Japan.
@Luigi,
Inmarsat says the BFO data from 18:25:34 to 18:28 may be unreliable. If this is true, fitting climbs or turns is a waste of time.
If it is not true, and the data are reliable, one can study them as I have done and reach several conclusions. First, it is likely that at 18:25:27 no turn or climb was occurring. Second, no turn alone can explain the 18:25:34 BFO of 273 Hz. A high rate of positive climb is required. Third, the 18:27 and 18:28 BFOs can be explained by three scenarios: (1) a turn to the north by 18:27 followed immediately by a turn to the west by 18:28, (2) a climb at 18:27 ending before 18:28 with no turn at all, or (3) both a climb and a turn occurring simultaneously at 18:27 and continuing through 18:28. I have not found any other scenario that is consistent with the BFO data besides these three. There are insufficient BFO data to allow one to down-select among these three choices based on the satellite data. It appears that a turn was completed before 18:40, but without additional information, the satellite data appear incapable of determining the actual turn time without also invoking a climb. This may or may not be a correct interpretation.
If you read my paper for a third time, perhaps you will find the statement “No turning of the aircraft can make the BFO go as high as 273 Hz at this point in the flight” on page 30. This is in reference to the 18:25:34 BFO value. I concluded on page 31 that an initial (and very brief) ~6,000 feet per minute climb is needed to be fully consistent with the 273 Hz BFO.” Your statements indicate that you still have not understood what I wrote with respect to turn scenarios.
Finally, I choose to not make any “it would be more logical to infer . . .” assumptions with regard to climbing and turning simultaneously. One B777 pilot has indicated to me that if he wanted to both climb and turn, he would do them simultaneously rather than one at a time, but a hypoxic pilot might not do the “logical” thing. All I am saying is that, if you believe all the BFO data: (1) a climb is necessary, (2)the data cannot define the exact sequence of climb and turn events between 18:25 and 18:40, and (3) there is a scenario that is consistent with all the BFO data and my proposed route. It is a matter of opinion, not fact, as to how “logical” that sequence might be.
@Dr Ulich
I don’t doubt that it is possible to torture the model to interpret the spike as indicative of a climb and a turn rather than just a climb. But, that is arbitrary and tendentious. Yet, by sleight of hand, your presentation leads an uncritical reader toward the conclusion that the data “confirms” the early turn indicated by the rest of your model. It absolutely doesn’t. Meanwhile, the red herring discussion of normal “step climb” maneauvers serves to promote and perpetuate the mythology that nothing much untoward was going on in the cockpit — the pilots were just tapping in the odd waypoint here and there, leading to the plane careening around like a billiard ball while peacefully cruising at 35,000 feet. That’s just horse puckey.
The red and green distributions in the figure 5 of the ATSB update report look to me like the addition of a series of Gaussian curves; the North end of the green curve is very ‘Gaussian’ as is the centre of the red curve. By the number of bumps on the curves, I would guess around five Gaussians have been added in the red and at least twice that in the green. My guess is that they have done the following.
The text refers to scoring of the trial aircraft courses against the BTO and BFO data. So:
– The red curve – the constrained autopilot models – as Mike E has pointed out there is not much flexibility in these models so there are only a few speed/course combinations that come close to fitting the BTO data, away from those the BTO errors will rise sharply. Each combination is scored with some combination of the BTO and BFO errors. This score can be translated (not very rigorously) into a probability using the equation for a Gaussian distribution. There seems to be an additional step of blending the points together with a series of Gaussian curves. Whether the shape of this Gaussian is derived from some statistical process or is just for presentation purposes is not clear to me. The BFO fits to these combinations will be what they are (actually not very good, IMO).
– The green curve – the data error optimisation model seems to have evolved since the initial report. In that report (analysis C) the tracks varied significantly at the arcs as would be expected if the fit is chasing the data; in the new version the tracks are quite straight, looking more like the original analysis B than C. In any case the tracks are compared to the measured BTO and BFO and the score is used to generate a probability, presumably by the same formula as for the red curve followed by the blending process to give a final lumpy curve. As Mike E said, the BFO error is a weaker function of bearing that the BTO so the green curve is wider. That doesn’t mean the red curve represents a better guide, however; both BTO and BFO data have to be satisfied at a level appropriate to its errors, with a model of minimum required complexity.
The curves are only a visualisation, if a proper statistical test has been performed it will have been done at the candidate track level. However, as I have commented before the report doesn’t try to do that and defines a search area that covers all the options.
@Richard Cole (richardc10)
Thank you for that well argumented reply to my questions. The graph is the ‘flagship’ of the update, it appears on the title page and at the top of this page in Jeffs blog, so we might as well try to understand what it shows.
Your point about the addition of several Gaussian curves is well taken. The constrained paths are obtained for five autopilot modes, and each mode will have its own probability distribution for the intersections of the 6th arc. For the optimized paths, however, I have some difficulty imagining divisions between distinct groups of paths.
I have no problem with the (red) distribution, but do not understand even the basics of the (green) distribution for the unconstrained paths. When there are no constraints on changes of speed and heading along the path, then a path can be found that matches the BTO and BFO at each crossing – zero errors. So what is then the root-mean-square (RMS) error of the BFO values at each crossing, that is used for weighting and selection of “the best 100 paths”? What are the variables that cause the wide spread of the “distribution in the RMS values” along the 6th arc?
@Luigi,
Apparently you don’t read other people’s work before you criticize them. You continually demonstrate your lack of understanding of both the data and other people’s work, even after three attempts. I never said a turn was needed to explain the 18:25:34 BFO, just a climb. Please stop misinterpreting my work. Why don’t you do something new on your own and stop wasting other people’s time demonstrating your ignorance?
@Dr. Ulich
Anybody can read your paper and reach their own judgment on how you spin that data. But, thank you for confirming my analysis: if that BFO data is to be given a physical interpretation, it reveals a fighter pilot-style high-speed climb maneuver initiated concomitant with the reboot of the satcom, just after the plane pulled out of Malaysian military radar coverage.
A few days ago HH stated that he was 99.9% sure the plane would be found, IF they were looking in the right place. Just in case anyone thought they could find it by looking in the WRONG place.
Today, a Mr. Foley claims the search will take months.
Why, exactly, will it take months, if they are looking in the right place?
I’m curious, because it only took a week or so to locate AF447. I realize that the search area is gigantic, but now I’m coming back to an earlier topic – a favorite of JF. A few days ago I’m pretty sure somebody stated that it would be found “in a few days.”
Why is the search area so big? What is the difference in the probability between the starting point of the search and the edge of the search?
It seems that if the “probability gradient,” if you will, is very steep, the search should last a few hours. By steep, I mean that the starting point is orders of magnitude higher probability, in the minds of the searchers.
The press statements suggest otherwise. Those statements siggest that the probability gradient is very flat – that is, the starting point is only slightly more likely to be the plane’s final location than any other location in the search.
But even if the gradient is flat, then statistically the plane should be found after approximately half of the search area is covered.
These statements lead me to think that they have searched the highest probability areas, and that the remaining areas are all relatively consistent, but low probability.
In other words, they have lost confidence.
Jeff, all:
In response to my post
(October 22, 2014 at 7:13 PM)
You said:
(October 22, 2014 at 9:47 PM)
“No mystery, Nihonmama. The canonical primary radar track shows MH370 passing over the southeasternmost corner of Thailand shortly after making its 180 at IGARI.”
Where does that canonical radar track come from? Because per AP’s article on 03.18, Thailand’s air force spokesman said “the plane never entered Thai airspace”
https://twitter.com/nihonmama/status/485519322362564608
Have the Thais walked back from that statement or did subsequent reviews of the radar data show that the Thai’s were wrong (or lying)?
IF Thais are correct and Malaysia’s statement that MH370 CROSSED over another country is correct, the country MH370 crossed must have been Indonesia.
So why didn’t Malaysia just say that?
Recall (from Daily Mail Australia 09.14.14, but widely reported)
“The head of the Indonesian Police Force claimed he knows ‘what had actually happened with MH370.’ General Sutarman made the comment in front of airline officials and senior police in Jakarta. Malaysian authorities expressed their shock and plan to speak with him”
That same day, Malaysia’s Inspector General of Police, Khalid Abu Bakar, tweeted that Indonesia’s police chief was “wrongly quoted by Indonesian Tempo”:
https://twitter.com/KBAB51/status/511152093017563136
Lucy Barnes response (to me):
“What I have read, and what people in Malaysia have told me, Tempo not only news org at Press conf. All reached same conclusion.”
@Luigi: we’ve been told by experts that the SDU, from a cold power-up, takes a few minutes to warm up before it can deliver accurate BFO values. I have seen no dissenting opinion on this basic property of an SDU.
The lone BFO value you claim as “evidence of a climb” occurs precisely 7 seconds after a cold power-up. This reading is thus about as unreliable as it gets.
Basing a flight path best estimate on a cold SDU BFO seems to me a little like basing a health diagnosis on the thermometer read-out after half a second in your mouth.
Accordingly, the throwing out of that one data point (and the taking of the next two, a couple minutes later, with a grain of salt) is among the LEAST controversial decisions taken by the serious scientists investigating this event.
Including Dr. Ulich, whose work (I feel) you are trying to twist to fit your theory.
The ATSB recently added a paragraph at the bottom of its MH370 page which suggests that debris may soon wash up on the coast of Indonesia.
The first website I checked (link below) suggests, that, in March/Aril/May (only ones I checked) – and starting at the IG’s best-estimate crash site (s38), both currents AND prevailing winds would (over long periods of time) be expected to drift debris due EAST.
How does it get up to Indonesia?
Not pushing a theory; just genuinely perplexed.
Link (scroll to bottom for Indian Ocean): http://www.offshoreblue.com/navigation/pilot-charts.php
@Brock – I share your confusion about the debris. I almost think it is foreshadowing a large-scale revamping of the search area, but for now it’s just odd.
Why now? It’s been 7 months. Do drifting objects converge, rather than diverge? Was Indonesia not watching for debris before? Or did somebody gum up a drift calculation?
Of all the updates, this “check your shores” directive has to be one of the most ridiculous I’ve heard.
Brock, JS:
Re currents (and debris):
Way back in March, this account (in Ipoh, Malaysia) tweeted a current analysis from NOAA, covering 03/5-25:
https://twitter.com/IpohHappenings/status/446926885973262336
In addition to the currents, take note of the convo, which in intercultural terms, can best be described as ‘high context’.
@JS,
Not rediculous according to publicly available drift models. See for example (only one link to speed up posting :o)):
http://adrift.org.au/
There is another NASA site performing a similar task.
These were relayed by @LGHamiltonUSA, @oceankoto on twitter recently.
The drift sims are very sensitive to initial conditions and starting date. But the big picture shows that debris could wash up anywhere on the IO shores round about now, or accumulate en bulk in the IO gyre, depending on said coords.
Cheers
Will
@Will –
The SIO was identified as the probable endpoint in March. It took 7 months to model the drift?
Why now? Why not state in March – “hey, keep an eye out in a few months?”
I stand by my statement. The timing is ridiculous, because either there is new information which has not been fully disclosed, OR, this is old information that they’ve been sitting on for 7 months.
@JS,
The modelling takes seconds, see link I posted. It is the drifting that takes 7 months.
Therefore, I disagree. If it is round about NOW when debris can be expected to turn up on shores, then NOW is the right time to re-iterate that fact to the public and raise interest to go out there and look for it.
This is merely playing to human nature. If nothing happens for a long time, interest and attention wanes.
This human nature thing might also explain the lack of radar detection for MH370, followed by the scrambling of two jets to intercept the small Aussie plane recently. In-attention for one, kick in the butt, over-attention to the next.
Cheers
Will
@Brock,
Just to be clear, the high BFO at 18:25:34 is not seven seconds after a cold power up; it is 2 minutes 47 seconds afterwards. The first transmission, which occurred at 18:25:27 is approximately 2 minutes and 40 seconds after power is applied to the SDU (per the ATSB report), which occurred at 18:22:47. Inmarsat says the BFO value for 18:25:27 is reliable. It’s the ones that occur 2:47 to 5:28 after power up (from 18:25:34 to 18:28:15) that Inmarsat says may be unreliable. It could well be that the first log-on request BFO value after a cold start is slightly wrong because the oscillator has not reached thermal equilibrium, but it seems that other, larger errors contaminate the subsequent transmissions for several minutes (based on examination by Inmarsat of the BFO data for other flights and perhaps including the unreleased, initial power-up data for MH370).
@JS,
A complete search of the two new ATSB “Underwater Search Areas” will take at least 6 months. It’s simple math. The distance along the arc is 1100 km on the ATSB’s map. That is approximately 600 NM. The full width of the (Fugro Discovery) search area is 60 km or 32 NM. The ATSB has not given any reason why the Go Phoenix search area is narrower on their map. Assuming the widths will end up the same, they are looking at side-scan sonar imaging of 600 X 32 = 19,200 square NM. I believe they have been running tracks with a spacing of about 1 NM with a speed of about 4 knots. Thus it will require 19,200 linear seach NM at 4 NM / hour = 4800 hours = 200 days. In fact you have, more or less, 2 1/2 ships available, because the Go Phoenix appears to have a reduced search rate compared to the two Fugro ships. So you can do the search 2 1/2 times faster than with a single ship, but the fraction of time any ship can conduct the search is reduced significantly by weather and equipment problems, plus it takes up to 13 days to make a single round-trip back to Perth for resupply. I would be surprised if the ships actually average better than 50% of the total time collecting side-scan sonar data. So, more or less, the time-on-station inefficiency is offset by having multiple ships, and you are left with approximately 200 days for one complete coverage, barring major problems.
@Brock
>> Including Dr. Ulich, whose work (I feel)
>> you are trying to twist to fit your theory.
You couldn’t be more wrong. I suggest you read his monograph, which takes the BFO values at face value and does not include the caveats I very specifically put forward (drawn from the JoN paper). BTW, the JoN paper does not say that the BFO values are inaccurate or untrustworthy — see my posts above, where I point you to exactly what it does say on the issue.
JS – I agree there could be some sweaty palms at ATSB. When the air search was called off JACC said that any remaining debris would have by now become waterlogged and sunk. At the time don’t forget that sat companies were zooming in on rubbish fields everywhere down there, some hundreds of kilometres away from the search area – indeed the search shifted in any direction that a rubbish field was spotted, and investigated. Now we are told to look out for debris?
Just one tiny lousy bit of 777 showing up about now is going to take a lot of heat off some people. Unfair you could say, when you look at the scale of the issue but they came out semi confident instead of totally cautious. What happens in ATSB-Inmarsat is similar to what happens right here – people become hell bent on finding it and instead of cold dispassionate processing into action you have ego’s and interests colliding everywhere. At the top of their to do list every day it should have said take a cold shower.
Dolan might be good man but whoever runs ATSB knows this will be good for the CV if successful. At some point though you start to drift into but what if??
Thanks, MuOne! Very helpful.
I’ve tweeted a pic showing the results of starting at IG crash coord, and drifting for 8 months, using the model to which you linked me. Everything is slammed up against Australian west coast. Nothing anywhere near Indonesia.
But I must admit to being notoriously slow at getting people’s drifts…
@Matty – exactly. Debris that they didn’t need to look for back then because it would have sank already. Somehow, it unsank and we should keep an eye out for it. I thought people were nuts when they suggested that MH17 was really MH370. I still think they’re nuts. But when somebody pops out of the blue and says, “hey, look over there!” I start wondering.
@Will – I did the drift site. I could not find a point that drifted to Indonesia. It was always South Africa or the southern coast of Australia. But even if I missed place the duckie, I’m still not buying this. You are suggesting, I think, that someone waited to remind us where to look. The modeling isn’t that precise, though. If they were doing it as a reminder, I’d expect one every few months. It’s easy enough to say, “Hey, we’re waiting for X until debris drifts up.”
It’s a lot like missing persons cases in snow covered, frozen over areas. They don’t say, “Ok, start looking… NOW.” They say, “We’re looking, but we don’t expect to find much until the thaw.”
@Dr Bobby Ullich – you are sort of making my point. I understand the calculation of the length of time to search the entire area. But statistically, what is the likelihood that they would find the plane at the very end of that time?
If the entire search area is considered equal probability, the odds are that they’d search exactly half of it before finding a plane. The odds that they would need all six months are slim, just as the odds that they’d find it the first day were slim.
But that is for a search area in which the probability is uniform. Here, we have a much higher probability area defined, which was the starting point of the search. The odds tip far in favor of an early find.
We are now searching areas of ever decreasing probability. Or, in the alternative, we’re searching low priority areas before high priority areas, which is needlessly extending the search.
(Actually, WE are sitting in armchairs, but you get the idea…)
One might reasonably wonder if Immarsat’s hand-waving and studied agnosticism on the 18:25-18:28 BFO values is because they draw attention to the loaded question of what was actually going on in the cockpit of the plane. Not really a “safe” question, is it? Message to geeks, anoraks and trainspotters: “Here be dragons.” So, piss off if you know what’s good for you.
These last few posts, Really good…
And since it’s Friday here (and Sat am across the date line), I’ll part with this question:
How does Angus Houston, who’s sitting in Australia, *know* that “investigators will never be able to connect MH17 shootdown to “a particular group”?
Tweeted by Christine Negroni, who was just (and perhaps still is) in Australia
https://twitter.com/nihonmama/status/525375225660444672
Let me put it another way. Per Immarsat, we just don’t know whether the BFO spike reflects actual motion of the plane or non-specific electronic gremlins. Even if that is an honest assessment, in an unpoliticized, value-free analysis we’d at least be presented with the implications of the former assumption. Only, the response to that would inevitably be — WTF?! Is anyone surprised that they didn’t go there?
JS – The satellite scanning for debris took in an enormous stretch of the arc. A bunch of companies all setting up to “find the plane” and they threw a fair bit at it. Chinese French everyone really. The search jerked to wherever the last rubbish patch was detected. There was lots of excitement, even announcements in parliament, news crews standing by with bated breath – then the uncomfortable back down. No debris. Politics hasn’t helped but to now look to Indonesian beaches smacks of desperation.
Bobby and Mike have detailed the practicalities of the search pretty thoroughly but the surveying can throw up leads also, and indeed there was minor excitement regarding some “hard objects” a while back. Can we infer that there has been very little else to enthuse about? In other words the surveying which admittedly is not an exhaustive search isn’t too encouraging. Could you conclude that it’s very likely not in one piece as some have predicted?
Some say it should be over pretty soon and some want a full 200 days.
Luigi – no I’m not surprised at all they didn’t go there. I have felt that they have stepped around the vagaries and issues from the start and projected a certainty that was not appropriate. It’s been a race for the prestige of finding MH370 pure and simple. Not sharing everything day one was plain wrong. Inmarsat’s game was always predicated on the conviction that it would show up, then it didn’t, then they shifted the game, now it has to show up.
@Luigi: I wouldn’t read too much into a failure to mention Inmarsat’s “disregard 18:25-18:28” in a paper which predates it.
Surely you read the “if Inmarsat says disregard, this is all a waste of time” caveat with which Dr. Ulich opened his first reply to you on this topic. If I were truly interested in his current views on the matter, I’d ask him directly. Or just read Section 5, p. 12 of his latest (October 16) updated analysis, which suggests (to my eyes) that he has responded to Inmarsat’s paper by dropping outright the altitude changes to which you seem determined to cling.
@Matty: I like your cool, common sense, but wonder if you are connecting ATSB’s salvation too strongly to the finding of the plane. If it’s NOT where the data says, I predict the questions will shift to the data’s providers.
If it IS where the data says, I predict questions will eventually circle back to the searchers, who chose to look elsewhere for half a year (most experts concluded “s38” almost immediately – and no new data has emerged since).
@Brock
I was not critiquing Dr. Ulich for omitting mention of Immarsat’s equivocation about the data. I was critiquing you for ignoring the fact that I did address it, while also ignoring that Ulich did take the data at face value in his magnum opus. I’ll also point out to you that Ulich directed me to his analysis of that very data in that very paper in response to my analysis (see exchanges above). Now that I’ve shown how contorted and misleading his analysis is, he is of course keen to play up the gremlins theory. No surprises there!
Brock – I think we are in agreement actually. I think it all will point back it Inmarsat if it comes up empty. It has turned out to be a good arrangement for them where ATSB are feeling some heat while they sit back in apparatus waiting to triumphantly leap out if it shows up. ATSB are left holding the baby while Inmarsat are poised to take the credit.
@Luigi: I see a scientist dutifully fitting wacky paths to curious data (because it’s all he had to work with), then happily removing the wackiness in the paths as soon as the data was downgraded from curious to crud.
You’ve been trying to force altitude changes back into the flight path ever since the IG formally rejected them weeks ago, so it does not surprise me to learn that you see something very different.
@Brock
Yeah, and the IG has such a great track record on this thing!
Just to be clear, a physical interpretation of the BFO data for 18:25-18:28 implies an extraordinary maneuver by the pilot and is not remotely compatible with a routine “step climb” maneuver designed to improve fuel economy — surely the last thing on the pilot’s mind in this instance, in any case!
Comparison of the BFOs at 18:25:27 and 18:25:34 implies an average vertical acceleration of approximately ~14 ft/sec/sec over the interval, using Ulich’s estimate of a 6000 fpm Rate of Climb based on the BFO delta (“The Location of MH370,” p. 31). The general validity of that estimate can be confirmed by inspection of rows 2 an 3 in Table 9 of the JoN paper. That’s an average (not peak!) acceleration of roughly 0.45 g for that 7-second interval.
Judging by the discussion at PPRuNe linked below, a climb rate 6000 fpm is fairly close to the limit of the 777’s capabilities — someone mentions an FAA test flight reaching 7700 fpm.
http://www.pprune.org/questions/104163-777-climb-profile.html
Nihonmama: thanks for the referral to SOS, their grooves are killer. It’s perfect writing music actually, no vocals, just pure, carrying energy that captures a sense of elegant movement – like someone moving through a busy airport with grace. Yeah, I think I see it, too! 😉
@Luigi: a physical interpretation of those BFOs is like a diagnostic interpretation of a single heartbeat: it is meaningless, because it is derived from insufficiently credible data.
I am thoroughly uninterested in debating whether Dr. Ulich did a credible enough job of RATIONALIZING the flaky data (back when many still felt duty-bound to try). No outcome of that debate makes the data any less flaky, or an altitude change theory any more plausible.
The IG and I have gone 12 rounds on performance limits, but I have nothing but applause for their signal data acumen. If the Inmarsat data is authentic, then I’ll take the IG’s sense of it over anyone else’s in the world. To what specific signal data fiasco of the IG’s do you refer, Luigi, when you impugn their track record?