Ocean Infinity Further Expands MH370 Search Area


Fig. 1: The seabed search as depicted in the most recent Malaysian report

When Malaysia announced on January 10 of this year that it had contracted with Ocean Infinity, a US-registered company, to relaunch the seabed search for missing Malaysian airliner MH370, Transport Minister Liow Tiong Lai stated that there was an 85 percent chance that the plane’s wreckage would be found within a 25,000 square kilometer search zone previously demarcated by the Australia National Transport Board. As I’ve noted in earlier posts, Australia’s stated position at the time was that if the plane was not found in this area, which stretched from 36 degrees to 32.5 degrees south latitude, it could offer no rationale for looking anywhere else.

On January 30, the Government of Malaysia released its first weekly “MH370 Operational Search Update” showing the progress of Ocean Infinity’s search vessel, Seabed Constructor. In addition to the ATSB’s 25,000 square kilometer search area, the new report designated two “extension” areas, stretching up to 29 degrees south latitude. (See Figure 2, below.) “The advice to proceed north towards 30S latitude came from Independent Group members,” News.com.au noted, referring to a theory put forward by Victor Iannello that the plane’s captain, Zaharie Ahmad Shah, had hijacked his own plane and set its navigation system for Antarctica.

Fig. 2: The seabed search area as depicted in the first Malaysian update.

That bizarre hypothesis is about to be put to the test: at time of writing–April 18, 2018, 1200Z–Seabed Constructor is working an area around 30.5S. Working at its current rate, it will have soon have finished scouring both the extension areas and laid Iannello’s idea to rest.

What then? With the ATSB’s and the IG’s ideas all exhausted, one might argue that it would be time to pack up and go home. But this is not what will happen. Yesterday, in its 12th weekly update, the Malaysian government unveiled a new supplementary search area, to stretch all the way up to 26 degrees south latitude. (See Figure 1, top.) As far as I know, no one has yet hypothesized a scenario that matches the data and would result in the plane ending up this far north, but hope springs eternal. Perhaps Ocean Infinity, for whatever reason, just wants the process to drag on for as long as possible.

By the way, little attention has been paid to the fact that Seabed Constructor has blasted through the Broken Ridge area of steep, craggy terrain while scarcely breaking stride. This is a testament to the capability of its AUV technology. It also rules out an idea that has been promoted by certain MH370 theorists, to the effect that the captain abducted the plane and headed for Broken Ridge in the hope that the wreckage would never be found there. That idea can now be scratched off the fast-dwindling list of possibilities.

301 thoughts on “Ocean Infinity Further Expands MH370 Search Area”

  1. @Jeff Wise

    On your comment:
    “The Inmarsat data is not compatible with the plane’s absence from the seabed. I find it surprising that that point isn’t universally regarded as obvious.”

    I think it’s not suprising if you consider OI took the search width so far based only on the final BFO’s and the assumption the plane could not have recovered from the indicated steep descent. The possibility of gliding outside the +/-25Nm zone (later even less) has not been seriously considered yet.
    And this option is not impossible at all and therefore still open.

    Even more this option becomes more realistic while still nothing has been found in the designated areas yet.

    The drift-analysis and debris finds are quite compatable with the Inmarsat data at large.
    If they would seriously conflict eachother you would have a point. But they don’t.
    At least you (or anyone else) cannot prove they do rule eachother out in any way.

    The explanation of no succes yet could well be the search width has been taken too narrow at certain areas.
    Even the ATSB took +/-45Nm. OI has shown a lot more conservative in this respect till now.

  2. Seabed Constructor has put to sea.

    03:33:34 utc Entered Outer Port Entrance Channel

    (02:04:30 hours to negotiate the channels to reach AMC 4)

    05:38:04 utc Arrived AMC 4

    05:44:18 utc Moored at AMC 4 – Secured the lines Port Side to Wharf (heading 269 true)

    (17:15:05 hours “turnaround time alongside at AMC 4”)

    22:59:23 utc Moored – Let Go the lines at AMC 4

    23:16:06 utc Underway – Departed AMC 4

    (02:02:50 hours to negotiate the channels to reach the open sea)

    01:18:56 utc Departed Outer Port Entrance Channel

    (22:14:22 hours between entering and leaving the Port Area.)


  3. @Havelock @JeffW @TBill

    TBill said: “Yes there are more “warm” spots. Don’t forget McMurdo path orig. end point was 26.9S which updates to about 28S if you say LRC flight speed, instead of constant speed. 29S-30S is indicated by some of the drift studies. Also we have the underwater sonic data which, if it has merit, points to 26-27S.

    If the aircraft is not found by 26.5S, yes quite a few pin-location hypotheses will have been checked-out unsuccessfully.”
    With fuel, end of flight, flight paths, autopilot modes, plus drifts, as far as I know the IG has some warm spots up to about S26.5. North of that, I do not know of any basis IG has put forward for a crash site. OI will return now to just south of Draak, which is still south of S29–the warm spot of Godfrey’s widely lauded drift study. By the time the search gets to S26.5, Godfrey’s study has the likelihood of a crash site as next to nothing.

    There is still the Geomar/NOAA studies that did have very high probability drift-alone warm spots way north. But many/all of those areas are inconsistent with other IG/ATSB analytics, even if you accept the drift probabilities.

    But there’s not much left to grasp at. Even Mr. Exner has stated on Twitter, “If not found in the current planned search area [up to S26.5 to be searched next two weeks], we [the IG and Ocean Infinity] will regroup and consider where to go next season. Revisit data holidays, further NE, SW, NW, SE? Revisit all assumptions. 7th arc about the only thing reasonably certain.”

    Please note “reasonably certain.” And “Revisit all assumptions.” That is something we can agree on.

    But, for now, let’s hope it’s found fast.

  4. @ErikN

    As you mention the GEOMAR drift study again I like to explain some misconceptions made by some about this study.
    And then I’m talking about the second revised GEOMAR study of may 2016.

    This reverse-drift study was also based only on the flaperon but included more refined data than their first study.
    The result was their higher probability areas moved to the south considerably compared to their first study. And not only between ~23s/~26S but also same higher probability areas around ~32S.

    So the first misconception made by some is this GEOMAR study only indicates higher probability areas north of ~26S. This is just not true.

    The second misconception is this study is based on the best known data about the flaperon. This not true either.
    GEOMAR explains in their study quite clear they don’t know how the flaperon floated.
    So they assumed it floated flat without any windage. They name it as an uncertainty in their study/results.

    Now we know the flaperon did not float flat but flipped every few minutes leaving part of it’s trailing edge sticking above the surface.It had high windage floating faster than previously assumed.
    This fact puts their probability areas even more south.
    All in all this makes the GEOMAR studies outdated for they did not include the latest findings and data, for they were not known at the time:


  5. @Havelock said “If it’s totally independent, strictly speaking you imply that either your, or the drift studies, or both, have zero correlation to the actual underlying debris. ”
    This is the whole point: my study only focusses on the discovery rate pattern, the drift study work performed by Henrik Rydberg focussed on drift patterns and indicates a fraction of debris washed ashore to be between 0.02 and 0.12 (after 24 month), ie 90%+ of debris still out there that could eventually wash ashore with time. It is clearly not the case as informed by the time that no further debris have been discovered by now. This brings to question whether the debris drifted at all from the 7th arc. The reason for questioning this is that the location specific concentration of debris is relativelly high and quite focused on one region and the indicative total number of debris is low (ie all have gone to the same region and none are expected in other regions). This indicates that the source of the debris should be much closer to destination, much much closer. That latter part is my own interpretation. It would be interesting to see a debris drift study based on few 20+ debris, i would be surprised to see a matched origin on the 7th arc.

  6. @Ge Rijn, You wrote, “Now we know the flaperon did not float flat but flipped every few minutes leaving part of it’s trailing edge sticking above the surface.” The ATSB’s investigation of the flaperon was in a sense a microcosm of the investigation as a whole, in that it revealed glaring inconsistencies that, since they could not be explained away in the context of the official understanding of the case, were simple ignored. Specifically, the way the flaperon floated did not match the distribution of the Lepas barnacles. The French noticed this in the course of their investigation but never commented publicly.

    This is a fundamental feature of public discussion of the MH370 mystery: the assumption that all the data must align in a coherent way, because they were generated by an event that ended with a terminus in the southern Indian Ocean. However, as I’ve spent the last four years pointing out, none of the data sets (drift analysis, BFO, BTO, marine biology, etc) matches up with one another.

  7. @HB

    On your comment:

    “This brings to question whether the debris drifted at all from the 7th arc. The reason for questioning this is that the location specific concentration of debris is relativelly high and quite focused on one region and the indicative total number of debris is low (ie all have gone to the same region and none are expected in other regions). This indicates that the source of the debris should be much closer to destination, much much closer.”

    I think your interpretation is not representing reality quite well.
    If you say ‘concentration of location specific debris is quite focussed on one region’ I wonder what you have in mind.
    The region in which debris was found was from Pemba Island/Tanzania till Mosselbay/South Africa.
    This is about ~6000km wide.
    Then the piece found closest to the 7th arc on Rodrigues Island is about ~3000km from mainland Africa.
    This is a huge total region you hardly can call ‘focussed’.

    Then you have to consider the drifting timeframes. The flaperon arrived at Reunion close to ~15 months. We know this cause it was covered with barnacles when found.
    The Roy-piece arrived close to ~22 months after 8-3-2014. We know this cause it was also covered with barnacles when first found.

    In short; such timeframes and huge regions in which the debris was found and the apparant fanning out of the debris along known currents after Rodrigues and Reunion, don’t allow an impact area much closer to the African mainland or islands. At certain areas the 7th arc fits quite well in fact (imo between 30S/35S).

    Your statistics point also to a small initial debris field (of debris that would remain floating for long times comparable with drifter buoys). I think this is accurate.
    Remember based on the 177 historical buoys that drifted through the search area between 39S and 32S only ~45 beached within ~22 months. And only 30 on African shores and Ilands.
    So only a initial debris field of about 150 drifter-like pieces is needed to produce 30 pieces of debris to beach after ~22 months. Till now 29 likely or confirmed pieces have been found. Since nothing has been found anymore for nearly two years.
    So I think the initial debris field was even a lot smaller than 150 drifter-like pieces also according your statistics.

    But your interpretation that this would mean the starting point of the debris should be much closer to the African mainland or Islands is not correct imo. The 7th arc and a (very) small initial debris field would explain all fine.

    If I did not understood you well or you have other explaining views I’m glad to hear.

  8. @Jeff Wise

    I agree the bio-fouling on the flaperon is still a complicated and still not explained problem. There are theories as you know but none of them explain the problem well enough to serve as proof why those barnacles were at places they should not have been considering the attitude of the copy-flaperon while drifting.
    Probably it will remain an unexplainable feature forever. Fact is those barnacles were there and they were between several months and a ~year old. So that flaperon must have been in the ocean for at least a year.

    Then all data sets are impossible to match perfectly but more important imo is the most important data don’t conflict eachother or rule eachother out.

    I mean; if they found confirmed debris within in a year on Java, Sumatra or India they would have known the Inmarsat-data (and other data) could not be right.
    The lack of debris finds anywhere outside the areas where debris was found is well compatable with certain part of the 7th arc in fact and therefore also with the Inmarsat data (and other data, fuel data i.e.).

    This data might not match perfectly but they sure not conflict eachother or rule eachother out IMO.

  9. @Jeff Wise

    To add: Still the big problem with your approach is the same. In a hypothetic world you could be right the SDU was spoofed on the plane before the 18:25 re-boot and the found debris was planted in some way. And the Russians were behind it all planning this with their master-skills diverting the plane to Kazachstan unseen by any radar or anyone else. And no-one years after to give any clue. The perfect crime with still no logical goal or objective known to anyone in the real world.
    A perfect script for your next novel or a next James Bond-film maybe, but not helpfull to find the plane and real answers.

    For if you refute the Inmarsat-data, the drift-data, debris-data and other data as being real and what they are, you have nothing left to find the plane or to find answers anyway.

    It’s about showing facts and evidence.
    If your assumptions hold any merrit you have to offer evidence for them better than the evidence presented. I still haven’t seen any evidence affirming your assumptions in those four years.

    Still you’re ticking on it and agreed still you have not been proven wrong also.
    But what’s needed is evidence.

  10. @Ge Rijn, From my perspective, I have been accumulating evidence over the years, only to run into the problem that you and others refuse to accept what I’ve found as evidence. For instance, I point out that the flaperon float tests and barnacle distribution contradict each other, indicating non-natural provenance; you shrug and say, “Well, there must be some explanation, perhaps it flipped so often that the barnacles were able to stay alive all over.” When I point out that all of the pieces of debris have marine life less than two months old, you either ignore it or say (as you do in your last comment) “they were between several months and a ~year old.” (Check the final Australian report “The Operational Search for MH370,” Appendix F: “It could be assumed the specimens analysed here were quite young, perhaps less than one month.”) When I point out that there is no rational explation of the SDU reboot under any accidental or suicidal scenario, you say that there must be one but we just don’t know what it is. Likewise the impossible timing of the arrival of ROY in South Africa. Likewise the fact that there is no point in the ocean that under any drift model is consistent with the timing of the discovery and the non-discovery (i.e. no debris found in Australia) consistent with the ATSB’s drift modeling. Likewise with the fact that the airplane’s wreckage was not found where you guys so confidently said it would be.

    And there is more.

    All of this evidence accumulated after I presented my hypothesis. All of it validates what I’ve been saying.

    You write, “more important imo is the most important data don’t conflict eachother or rule each other out.” This is the reverse of the actual state of affairs.

    You remind me of the hosts in Westworld who, when shown a photograph of the real world or some other evidence that undercuts the reality of their manufactured world, are programmed to say, “It doesn’t look like anything to me.”

    Great show, btw.

  11. @Ge Rijn

    The issues with the GEOMAR studies are well known by those who care to review the information. I thank you for reiterating them. My comments, which were in fact related to their second drift study, are only to highlight there are still drift-only end points suggested further north. However, as Mr. Wise and you point out, they are not consistent with other evidence or analytics. So, they are what they are, drift-only, based on assumptions, but there is at least something (a study, maybe a weak one) pointing to those impact areas. In regard to comments refuting Mr. Wise’s premises, I have told him I am not convinced, but I also told him I cannot rule it out and it should be considered. He has some serious points that haven’t been or can’t be explained–not the least of which is why we’d trust data from an inexplicably rebooting SDU. I do support thinking outside the box when the box has failed too many times to be trusted. I think you do too. And you know how your ideas are greeted when you are not sufficiently abiding by the prescribed box rules (blue panel, pilot glide, to name a few).

    Anyway, it is my opinion, through all analytics we’ve had access to so far, that the weight of the mathematics and physics (from some combination of ATBS+IG at least) when coupled with debris in SIO (if not put there) supports warm spots up to S26.5. That’s about to be searched. After that, in my view, the powers that be and their influencers have no choice but to see what was missed, consider a glide (“how far”, “east or west”), refute some of Mr. Wise’s points or prove consistency with their analysis, and otherwise reconsider everything…including the 7th Ping Ring.

    Either way, I hope for the best for this search, for all involved. There is an awful lot of work to be done even if the aircraft/debris field is located.

  12. @Ge Rijn; RE “a initial debris field of about 150 drifter-like pieces is needed to produce 30 pieces of debris to beach after ~22 months”

    Yes and about ~20 more debris should have beached somewhere by now.

    Your argument was perhaps still valid a year ago, but now there has been a very substancial amount of time where no debris were discovered and I have tried to analyse that implication. This is the entire point of my argument. The focus is not on what happened to the ~20ish debris found (we know that) but more on what happened to the ~100-130+ ie fraction not washed ashore (in case of seven arc assumption) and more what is that fraction (here it could imply a different location than 7th arc). The drift models outputs a fraction not washed ashore after a time period. In your example 30/150 (but that varies along the 7th arc), the big question is where would be the remaining ~120 pieces after 4 years? Some should have beached somewhere by now and they did not. If you keep the same ratio for the remaining 120 floating debris, ~20 more debris should have beached somewhere by now, and they did not. Even if you account that some may have sunk or not found even if beached, this still indicates that the fraction of debris whashed ashore is much higher than 30/150 and the source location appears incompatible with the seven arc. One possible explanation is that the source of the debris field is much closer to the destination to obtain such an apparent high fraction of debris washed ashore. The seven arc implicitely implies a low fraction of debris washed ashore (~10% or less). And, in any case, the initial debris field could not be in the ~1000+ order.

  13. Or there was no SIO impact/crash as now there is even more doubt about the IGARI turn back and fly over Malaysia datasets.

  14. @SteveBarrat – sloppy data entry for both flyback radar sets?
    what a coincidence ! Even the turn back flight estimates don’t shows to much randomness…

  15. @MH

    “the initial debris field could not be in the ~1000+ order”. I am assuming you are referring to debris washed ashore not spotted in the SIO. 9M-MRO interestingly is the converse of AF-447 were the was a large debris field in the Atlantic Ocean but nothing washed ashore (to the best of my knowledge).

    Regarding the fly over Malaysia datasets didn’t @Gysbreght infer it was sloppy radar data entry on the part of the Malaysian data loggers rather than a completely flawed dataset?

  16. @Steve: “9M-MRO interestingly is the converse of AF-447 were the was a large debris field in the Atlantic Ocean but nothing washed ashore (to the best of my knowledge).”

    In the case of AF-447 all floating debris was recovered from the sea within days after the crash.

  17. Were the 2 Ukrainian passport holders wearing Mechanical Counter Pressure (MCP) clothing to assist with positive pressure breathing during depressurization event?
    French journalist Florence de Changy in her book LE VOL MH370 N’A PAS DISPARU writes the following description of the 2 Ukrainian passport holders:
    ‘Ces deux Ukrainiens arrivent ensemble, dans les derniers minutes de l’embarquement, nettement plus energiques que leurs camarades de voyage. Avec leur physique de commando marine, moules dans des tee-shirts noirs, ils portent chacun un gros sac de cabine, qu’ils font valser sur le tapis roulant du scanner, d’un geste entraine. Entre tous les passagers embarques sur ce vol, s’il fallait tenter de designer deux pirates de l’air, les Ukrainiens seraient les seuls a en presenter les attributs cliches: age, condition physique, apparence, attitude…’


    She says the 2 Ukrainians had the sculpted physique of Navy Commandos and they wore skin tight black t-shirts.
    It seems quite odd that the 2 Ukrainians with identical physiques would wear identical pieces of clothing, unless their purpose was functional.
    If their intention was to intimidate the cabin crew, they may very well have chosen to wear black t-shirts that emphasized their physique.
    But if one recalls the hijacking of the Pegasus flight from Kharkiv, Ukr. to Istanbul, Tur., the Ukrainian hijacker wore a bright baggy ice hockey jersey to probably hide what he was wearing underneath and distract attention.

    QUESTION: Why would Florence de Changy use the specific word ‘sculpted’ to describe their physique? This suggests that perhaps the t-shirts were not merely tight fitting, but possibly ultra tight fitting. Were the 2 Ukrainians merely wearing black t-shirts, or were they wearing specialized tight-fitting Mechanical Counter Pressure (MCP) clothing designed to assist with positive pressure breathing during a depressurization event?

    The SAS is not inflated like a conventional spacesuit: it uses mechanical pressure, rather than air pressure, to compress the human body in low-pressure environments.


  18. SteveBarrat said: “Regarding the fly over Malaysia datasets didn’t @Gysbreght infer it was sloppy radar data entry on the part of the Malaysian data loggers rather than a completely flawed dataset?”

    airlandseaman said on VI on April 15, 2018 at 10:30 am:
    “You made reference to 2 anomalous (sic) records (outliers) being deleted from the data set; what were they please? Answer: There were two records as I recall. One was the AZ value at 17:41:01. That value is obviously wrong, and the error is obviously a missing character (9). The correct value would have been 249.5. In the other case, a single time stamp was off by exactly 1 hour.”

    Then there is a row of data missing between timestamps 31:26 and 31:32. The correct value of 31:32 would have been 31:34.

    Finally several timestamps between 41:09 and 42:08 are off by 1 second, and 2 seconds off between 42:08 and 42:14.

    These errors are typical typographical errors rather than software anomalies. If a value is typed in an EXCEL cell and the ENTER key is pressed, the cursor jumps to the next lower cell. Therefore it is convenient to enter data column by column, rather than by rows.

  19. @CliffG
    I have located a number of articles by Ms de Changy in various publications. One cannot dismiss her as a crank. It is strange that there is no English version of her book-surely there would be enough readers.

  20. @Gysbreght
    Have you already provided us a link to a data table showing reported data vs. what you would surmise the actual data says (with proposed corrections)?

    Can you give me a lat/long/time for the first point to make a user defined waypoint?

  21. @ErikN

    To come back on your comment.
    Like @Jeff Wise I’m also thinking ‘out-of-the-box’ but more literally ‘out-of-the-search-box’.
    Big difference in approach is though that I take the Inmarsat-data for real and Jeff Wise doesn’t. He thinks they must be the result of a spoof before 18:25.
    To me the debris finds and all (better) drift-studies resulting from the found debris are conclusive proof the plane ended in the SIO so the Inmarsat-data must be right in general. Before the flaperon find and other finds the Inmarsat-data could be totally wrong and spoofed but after those finds this is out of the question.
    Unless you come with the impossible explanation all those found debris had to be planted. Maybe the flaperon could have been at that time but all those other 29 pieces is just impossible.
    It’s like keep defending the sun revolves around the earth like some did in the middle-ages.

    My scenario is ‘out-of-the-box’ not for those kind of reasons. I see the data as real. Not perfect but real in essence.
    I took a very close look at the debris over the years as well. The kind of debris found and the kind of damage they show. This all is completly conflicting a high speed nose down impact.

    If the plane entered a high speed descent during the indicated final BFO’s it must have recovered and glided outside the +/-25Nm zone. And ditched, breaking off its tail section or/and another breach of the hull like Asiane 214.

    All I’ve spent researching the past four years can only lead to this conclusion:

    The flight was deliberataly planned in detail from beginning till and by a very skilled pilot. With a specific end-point in mind that would serve the goal of total dissappearance best. And he succeeded till now.

    In my view he deluded us all with that high speed descent after 00:19 letting everyone to believe he must have impacted close to that final transmissions (7th arc).
    He recovered shortly after and glided the plane not that far outside the +/-25Nm area between 32.2S/33S and ditched it. This is the most reasonable area IMO.

    This would explain all. The small number of debris finds. The kind of debris finds. The lack of crush damage, only tension related damage. The intact leading edges of flaperon and outboard flap. The nose-gear door. The left and right wing flaperon closing panels with intact trailing edges. Motivation (to save his family from harm and to put the Malaysian corrupt government at the attention of the world for years to come).

    This is my view. OI is going to search even farther north. I’m almost sure they’ll find nothing still.
    I think they missed MH370 by only ~25Nm when passing the 32S/33S area along the trenches of Broken Ridge.
    We’ll see. Hope I’m wrong.

  22. @TBill: Is this the data table you are interested in:
    https://www.dropbox.com/s/mmkb04p1i0kkwqo/Datatable.pdf?dl=0 ?

    The first column is a counter of radar antenna revolutions. The second column is the corrected time, calculated from the number of revolutions and the rotational speed stated in the header, adjusted for azimuth. The third column is the difference between the corrected time and the corresponding time stamp in the original table. It is shown by the blue triangles in this chart:


    Sorry, I don’t understand your second question.

  23. @TBill: RE your second question: I don’t calculate lat/long coördinates. My focus is on the speeds, and the issue is timing, not the precise location of the track.

  24. @Gysbreght

    Thanks for your comments on AF447.

    Also hitting the tab key means you move horizontally after filling in a spreadsheet cell. It means you enter the data for one subject at a time. This works if independent variables are rows and dependent variables columns. I’m talking about biological research where it’s safer to enter all the data for one patient/research subject at a time eg age/blood pressure/magnesium/cholesterol etc. Entering all the cholesterol levels together is risky for the reasons you state as you can get out of sync.

    Essentially this is a long winded agreement for the point you make.

  25. @SteveBarratt: Thank you for your input. Also the person who did the copying may not have had your EXCEL skill level. I have used EXCEL quite a lot and didn’t know that you can use the TAB key to enter a value.

  26. Ref. the “enter key” behaviour in Excel, this can be changed.

    On a Mac (office 2011) , go to preferences, tab edit and third paragraph down; “after pressing RETURN, move selection followed by a pulldown menu down, right up or left.

    That would be the obvious setting if one fills rows instead of columns.

    Office on MAc is more restrickted in options then in Windows (Bill Gates did not want to give all options to Steve), so am sure it is an option there.
    Sorry for being completely off the subject but feel some of us could benefit for future use……

  27. @ABN397 how utterly bizarre this story, so heavily reported months ago, is being published today, with no current news hook or other reason. The randomness of it makes me wonder.

  28. On the other blog DennisW says on May 4, 2018 at 8:43 pm:
    “The closest approach math allows computing (without any doubts except the estimate of closest approach) the path length. That is important.”

    And at 9:53 pm: “What could be simpler than x = sqrt(R^2 – L^2) ?? “

    Indeed, Dennis’ equation is admirably simple. The only problem is the estimate of the closest approach. The following chart may serve as a complement to Mick Gilbert’s nice explanation of the problem. The blue line shows the closest approach in terms of slant range (Offset) as a function of the distance along track according to Dennis’ equation for slant ranges 11.2 and 15.8. The pink line shows the height of the track against the distance along track for these ranges and azimuths 21.7 and 265.8 degrees. The dashed lines illustrate two examples:

    Red line: For 9.5 NM closest approach the distance along track is 18.5 NM, and the track height is about 47500 ft.
    Green line: For a track height of 32000 ft the distance along track is 21.3 NM and the closest approach is 8.1 NM.


  29. @Gysbreght

    A light read. I have just spent a couple of hours reading this and have come to the conclusion that there is not a snowflake’s chance in hell of us getting anything definitive out of the radar data we have. Just read sections 5 and 6 to see why.

  30. @ventus45: Thank you for your encouraging response. “Just read sections 5 and 6 …” of what?

  31. @DennisW: What would be your estimate of the closest approach distance if you add 19 seconds to all timestamps after the cone of silence?

  32. @Ventus45
    “snowflake’s chance in hell of us getting anything definitive out of the radar data”

    Please give brief analysis of what you conclude from that link.

  33. ventus45: Re: “Just read sections 5 and 6 to see why.”, please be more specific about what you are referring to in sections 5 and 6.

  34. @airlandseaman: You said on the other blog May 6, 2018 at 12:23 pm:

    “The KB PSR radar data is not that far off. It looks correct when comparing the path to the military radar path, and the speeds also generally agree with the scant military radar data reported in the FI. It only seems to be off when the plane is close to the radar head, which suggests that the error is not a simple scaling error, certainly not lbs for kgs, or meters for feet. It is something that only has an obvious significance close to the radar head.”

    Victor Iannello said on May 1, 2018 at 9:12 pm:

    “@DennisW: For now, I’ll share one figure that gives a flavor of what I am investigating. I define a new parameter which is the “time difference from constant speed path”.”

    In the attached plot of D versus time for the PSR data you can ignore the data for 45,937 ft because the airplane cannot maintain that altitude at the recorded speed. You can also ignore the data for 32,497 ft, Range bias = -1.260 NM because you know the range accuracy of the radar is much better than that. What you have left is then the blue dots at 36,000 ft.

    The way I interpret that curve is that the part of the KB outbound track is not in sync with the KB inbound track and with the BW inbound track. I guess that the offset in the Time Difference would be even greater at 32,000 ft, the height indicated by the military radar according to Factual Information.

  35. @DennisW: I’m slightly puzzled by your response at the end of your radar notes document. The length of the path in the COS of ~19.6nm corresponds to closest approach 9.0 nm. Based on the curve in the lower chart on the first page of your document I had expected a greater reduction. The plot you refer to in the text is missing. Would it be possible to provide a plot of range versus time?

  36. @Gysbreght

    I just pulled the 19.6nm off the graph. You are right. I was being lazy. Actually doing the math the COS path value is 19nm and the speed is 452 knots.

    The range vs time plot is showing up as my first figure. I made the corrections in my narrative.

  37. Re KB radar.

    VERY EARLY ON, at multiple televised news conferences within the first few days of May 2014, (all of which I watched in real time, and the multiple repeats in the news and the late news, – I am in Sydney Australia) both Air Marshal Angus Houston, and the then ATSB Chief Martin Dolan, specificall stated, and made a point of stating, on multiple separate occasions, that there was very little confidence in the available radar data, “because” the radar(s) (Unstated which ones) had not been calibrated recently, and specifically, that there was “nothing”, repeat “nothing” (Dolan emphasised that) with regard to altitude. (The JACC was formed and ATSB took over control of the search from AMSA on 1st May 2014).

    That is the baseline position for the data, the condition of the system(s) on the night.

    Recent comments to the effect that the KB radar is good today are irrelevant. After all, they have had 4 years to calibrate it multiple times in that period, as indeed it should be, regularly, perhaps monthly, I don’t know what the acceptable interval is there. What is relevant, is that back then, it obviously had not been calibrated for some time, and was obviously well outside of whatever the acceptable calibration interval was at that time, such that they had such little confidence in the available data, that it was effectively dismissed as useless.

    Four years later, we are examining that same data, and coming up with analysis which is effectively saying that it was flying so high and so fast that is, at best, at, or more likely, well beyond the B777 envelope, both for height and speed.

    Obviously, something is manifestly wrong with the data set. The question is “what” ?

    Various ideas have been floated, timestamp issues, missing scans, even the ways in which possible excel data entry mistakes cause errors in transcription thus produce misaligned rows of data. None of the many and valiant attempts to devine and compensate for these possible errors has produced anything that looks like a reasonably plausible flight path – yet – and although I applaude the efforts expended by many on this, I don’t think it is resolvable. Others obviously think otherwise, and are continuing the effort.

    In any case, calibration issues to one side, I have one more thought to throw in the mix.

    KB is a terminal approach radar, and the magnetic variation is (was ?) about 0.12°W (WMM2015 magnetic declination) 0.00° annual change. (Reference: http://www.gcmap.com/airport/KBR).

    Being a TAR for ATCO use, wouldN’T it’s “north head” be aligned with magnetic north, and not true north ?

    If that were the case, the azimuth data we have is really magnetic, and not true, SO you could correct that by simply adding 0.12° to all the azimuths.

    Wth the azimuths so rotated clockwise, the COS would become more symetrical, and the inbound track would be much closer to a direct overfly of KB. The effect of both would be to reduce the CALCULATED altitudes and speeds to something MORE sensible from a plausible B777 flight envelope point of view.

Comments are closed.