Why we still have not found the black boxes that we found

Some of you are probably wondering why ships in the south Indian Ocean still need to run grids when they already heard pings from those black boxes. Why not just draw a straight line from the ship to the box?

People who read The Hunt For Red October know the answer already, but as a former oceanographer I’ll summarize. In most places the ocean has a sharp boundary between warm-ish surface water and the cold deeper zone where you find giant squids, godzilla and cthulu. We call that sudden temperature change the thermocline. Sound moves faster through warm water, so if it hits the thermocline at an angle it can refract in a weird direction or even bounce off and disappear back into the deep sea. Submarines use this all the time – if you have ears above the thermocline, even when you can hear a sub running underneath it the sound will come and go like light rippling off disturbed water. As a result you have a hell of a time pinpointing where it is. Some whales like to find places where a note can hit the thermocline at just the right angle and reach friends half the world away by bouncing back and forth forever between the warm(ish) and cold zones.

A couple of ships have picked up on again / off again signals that almost certainly come from the flight’s black boxes. Not much else makes a similar ‘ping’ at exact one second intervals. That narrowed the search area way down, but you still cannot pinpoint the box for the same reason that you cannot just shoot at a submarine on the wrong side of the thermocline. They really need some towed arrays or sub-hunting autonomous vehicles in deeper water more or less today, because in another day or two the warrantied thirty day life of those batteries will go kaput.


From reading a little more I see that the boat did use a towed array below the thermocline, but the sea floor is still much, much, much farther down than that. So the thermocline is not specifically responsible here, although the two vertical miles of water the boxes sit in still have plenty of layers with different temperatures and salinities that will make sound go screwy. If you want to get a sense of just how far down the boxes are, check out this helpful graphic. Then pour a drink.

55 replies
  1. 1
    Sly says:

    Some of you are probably wondering why ships in the south Indian Ocean still need to run grids when they already heard pings from those black boxes.

    Wait… did a plane crash in the Indian Ocean?

  2. 2

    All I have to say is there sure are some smart and well educated people who write at this blog.

    Thanks for this.

  3. 3
    eric says:

    ha. all rubbish. the plane is on the tarmac at Cleveland’s Hopkins Int’l Airport.

  4. 4
    lectriclady says:

    Agreed!@The Other Bob:

  5. 5
    lectriclady says:

    Agreed!@The Other Bob:

  6. 6
    gnomedad says:


    ha. all rubbish. the plane is on the tarmac at Cleveland’s Hopkins Int’l Airport.

    … holding Benzazi witnesses incommunicado. The thermocline is a liberal hoax.

  7. 7
    eric says:

    @gnomedad: the cold water proves there is no global warming.

  8. 8
    raven says:

    We also have world class smart-asses!

  9. 9
    dmsilev says:

    @eric: Nonsense. It’s waiting for its gate to become available at O’Hare.

  10. 10
    Botsplainer says:


    Nuh-uh, it’s been reflagged as a Delta jetliner, fitted with lasers and missiles, and headed over here right now with GPS coordinates of Southern megachurches, Legion posts, VFW halls, gun shops, flea markets, police stations, Waffle Houses, Confederate monuments and country clubs on a line from Plano to Dothan to Jacksonville to Columbia to Charlotte to Nashville to Memphis, where it will crash spectacularly into Graceland with a shout of “Allahu Akhbar!”

  11. 11
    Cassidy says:

    Have you heard the one about the “secret” navy base it’s been relocated too? These people are whackjobs.

  12. 12
    NotMax says:

    Missed the second ‘h’ in Cthulhu.

  13. 13
    Punchy says:

    Has Tim F. offered his services to CNN? They’re running out of BREAKING NEWS! and clueless Joes to report it. Aquaman Tim here would add some veracity and a cogent discussion to said shitshow.

  14. 14
    Yatsuno says:

    @Botsplainer: So in other words resolving the Civil War in the correct way?

    I’m kidding. Kind of. Well not really.

  15. 15
    Face says:

    Is Thermo related to Patsy or Kevin?

  16. 16
    feebog says:

    Serious question; I understand that the battery life of the black boxes is about 30 days, which means right now. What are the chances of finding them once the batteries go belly up?

  17. 17
    Botsplainer says:


    Too qualified. You’re talking about a nutwork that managed to hire Glenn Beck, Lou Dobbs, Dana Loesch and Ericka Ericksdottir to paying gigs.

  18. 18
    raven says:

    20 injured in Pittsburgh-area high school stabbings, authorities say.

    (CNN) — At least 20 students were injured Wednesday morning in a stabbing incident at Franklin Regional Senior High School in the Pittsburgh area, said Dan Stevens, an emergency management agency spok

  19. 19
    Botsplainer says:


    Really low. At that point, you’re blindly sweeping with towed magnetometers and side scan sonars in an area the size of Texas.

  20. 20
    Gin & Tonic says:

    How far below the surface is the thermocline, generally? This morning’s teevee “news” was saying that the Australian ship towing that USN microphone thingy was towing it at a depth of thousands of feet (I don’t remember how many thousands, I was half-listening while exercising.) That’s part of the reason why it takes so long between towing passes – they have to reel it in, turn around, and drop it back down to its working depth. But I was wondering if its working depth is below the thermocline.

  21. 21
    Gin & Tonic says:

    @feebog: They found the wreckage of AF447 at a depth of 15k feet two years after it went down, based on having a general area to look in, never having heard the pings (to the best of my knowledge.)

  22. 22
  23. 23
    Mike in NC says:

    @raven: So who’ll be the first nut to insist more guns in the school would have been the solution?

  24. 24
    Belafon says:

    @raven: Here at my job in Texas, I’ve already heard the “see, it’s not just guns.” I had to ask “so, how many have died?”

  25. 25
    raven says:

    @Mike in NC: Asparugus

  26. 26
    JPL says:

    @Mike in NC: The school did not have metal detectors so it’s fortunate the student didn’t have two guns rather than two knives.

  27. 27
    Amir Khalid says:

    Why wasn’t there a good guy with a knife available to stop this bad guy with a knife?

  28. 28
    JPL says:

    @raven: Holder will once again have to answer Congress on why he didn’t prevent this.

  29. 29
    Tim F. says:

    @Gin & Tonic: The thermocline is a lot shallower than that. So they probably have a box with dying batteries that is (was) only making intermittent sounds.

    @feebog: Let’s see. In terms of area to cover, think of it as if you buried a safe deposit box somewhere in Wyoming, maybe, but you have no idea where. You cannot just walk around with metal detectors because Wyoming is under more than two miles of water and you have almost nothing that can go anywhere near that deep, so you have to find the buried safe deposit box from a plane. And the visibility conditions are terrible. That should about cover it.

  30. 30
    WaterGirl says:

    Some whales like to find places where a note can hit the thermocline at just the right angle and reach friends half the world away by bouncing back and forth forever between the warm(ish) and cold zones.

    I know this isn’t supposed to be the takeaway from this very interesting post, but reading that made me happy.

  31. 31

    Serious question; I understand that the battery life of the black boxes is about 30 days, which means right now. What are the chances of finding them once the batteries go belly up?

    Pretty much the same as the odds are right now, next week, and a thousand years in the future – zero.

  32. 32
    Roger Moore says:


    What are the chances of finding them once the batteries go belly up?

    Depends on how well they can get its position nailed down before the pings stop. The close they can get while it’s still pinging, the better their chance of finding it once the pinging stops.

  33. 33
    MomSense says:


    No, no, no. Diego Garcia where it is being repainted and fitted with chemical weapons for a false flag attack that will allow Obama to seize our gunz and relocate us to FEMA camps!!

    /crazy for cocoa puffs

  34. 34
    Violet says:

    With all the advances in technology in the last decade or two, it seems like planes could be fitted with some sort of communications device that broadcasts the exact location in short, regular intervals. Like every ten to fifteen minutes or so. And batteries that only last 30 days seem kind of stupid because when a plane disappears, you very well may need longer than that. Also, a flight recorder that loops after two hours? Why? Storage is cheap. Make the recording last ten or twelve hours after the stated length of the flight before override.

    Seems like so many relatively easy upgrades are out there that, had they been done, would have made this search so much simpler.

  35. 35
    Gin & Tonic says:

    @Violet: But you don’t really design and build for the edge cases. How often does an airplane disappear without a trace over an unmapped area of 3-mile-deep ocean after flying incommunicado for 6 hours? Hindsight is easy, but these boxes are built to be recoverable, primarily. I doubt that sitting three miles down for two years was an engineering criterion, but the data were recovered from the AF447 recorders after two years. That is remarkable in and of itself. They sat in salt water at pressures around three *tons* per *square inch*, and were brought up and read two years later.

  36. 36
    coin operated says:


    /crazy for cocoa puffs

    That’s cuckoo for cocoa puffs, I’ll have you know…

  37. 37
    WaterGirl says:

    @Gin & Tonic:

    But you don’t really design and build for the edge cases. How often does an airplane disappear without a trace over an unmapped area of 3-mile-deep ocean after flying incommunicado for 6 hours?

    Yeah, but the likelihood or probability that a loss will occur isn’t the only factor in risk assessment; calculation of risk assessment also involves the magnitude of the loss, if it does occur.

    I would argue that in the case of knowing or not knowing about the fate of all those souls on a plane – people with families and loved ones – and the safety of future passengers on planes, the magnitude of the loss is very high.

    So my conclusion is that it would be very worth it.

  38. 38
    Violet says:

    @Gin & Tonic: I understand the not designing for edge cases, but it seems like they’ve designed for 1980, not 2014. And at this point the costs need to be part of the analysis. The cost for this search and recovery are massive. It would have been a lot cheaper to fit the aircraft with more modern communication equipment so at least they knew where it was when it went missing and had a much smaller area to search in the beginning.

    This issue has been discussed on other blogs and forums with an aircraft/pilot focus and everyone seems to agree that the airlines don’t want to spend the money. In this case everyone is picking up the tab and the reputation hit the airline is taking is big. If Malaysian Airways had spent the money to better communications equipment, their costs and the costs to multiple governments would be much lower.

  39. 39
    Gin & Tonic says:

    @WaterGirl: Without getting too Rumsfeldian about unknown unknowns, risk assessment is often predicated on the assumption that the future will be something like the past. I’m sure things will change (slowly) based on lessons learned from MH370, just as things changed after 9/11, because all prior assumptions about highjackers excluded the possibility that they wanted to die. I’d suggest that anyone who posited a scenario just like MH370’s in a requirements design meeting during the building of the flight data recorders would have been laughed out of the room.

  40. 40
    Roger Moore says:

    As I understand it, the 777 is capable of sending back a lot more information than MH370 did, but the data was not sent because it’s an add-on that requires paying extra. I would not be surprised if one outcome of this incident is that the extra information becomes mandatory, or at least something that has to be deliberately disabled, rather than an extra. And technology for CVRs moves slowly because they have to make some very tough promises about reliability and survivability that require extensive testing.

  41. 41
    The Pale Scot says:

    I’m sorta surprised/wondering why/if the US has/hasn’t sent a couple of attack subs to the area. they can measure the pper thermoclines, and would not be affected by weather or refueling, this seems like a novel way to train the crew and calibrate the equipment. It is off the beaten path but there must be a couple in the Taiwan area.

  42. 42
    WaterGirl says:

    @Gin & Tonic: Curse you for bringing up the vile Rumsfeld, but I don’t disagree with what you’ve written here.

    @Violet: Completely agree with what you have written!

  43. 43
    MomSense says:

    @coin operated:

    Don’t cast aspersions on my asparagus!

  44. 44
    Gin & Tonic says:

    @The Pale Scot: I’d be really surprised if there weren’t already at least one in that area. The USN has a policy of never disclosing anything about their whereabouts or capabilities (I have both a close friend and a son-in-law who have served,) but I’d bet next month’s pay that the searchers are being fed information from sources, both Chinese and American, which are being deliberately vague about their capabilities.

  45. 45
    Aaron says:

    I really do think that if your commercial airline flying a plane over a remote area such as open ocean or uninhabited jungle, you should have a satellite transmitter that uploads a text message of your gps coordinates every minute. Inmarsat includes 10mb for 100$/month. Thats plenty of position text messages for any plane.

  46. 46
    Trollhattan says:

    @Roger Moore:

    Yup. If the “pings” they’ve received over the last few days are in fact from the recorders, then they perhaps now have a narrow enough search area to use sonar to locate significant wreckage and with the wreckage, the recorder(s). If the pings are anomolous then they truly don’t know where to look and it would be very unlikely they’ll find anything. Per the ABC, “The planned search area has been reduced over the past few days and now covers 75,000 square kilometres.” Confess that still sounds like a very large “reduced” area to me, but it beats a million.

    Worth noting not one scrap of debris has been recovered at the surface from either the plane and its contents. This makes the task many orders of magnitude more difficult than Air France, since they knew with far more specificity where it went down and recovered significant floating debris within a day, yet it still took two years to find the wreckage on the bottom, including the recorders. A side note: the data recorder was recovered without its memory module, but that was also found later.

  47. 47
    Mobile Grumpy Code Monkey says:

    @Violet: Cheap storage isn’t robust; robust storage isn’t cheap. It does no good to save 10 hours of data on something that won’t survive a crash.

  48. 48
    The Pale Scot says:

    @Gin & Tonic: After I sent that, Another idea popped up which was why would the USN send a new tech sub to search and let itself be measured by every other sonar out there.

  49. 49
    Trollhattan says:

    @Mobile Grumpy Code Monkey:

    The Curiosity Mars rover’s main camera: 2 MP. IIUC NASA locks down parts specs early in the design process and this whopping 2 megs represents the state of the art for digital imagers that could handle the mission’s radiation, vibration, shock, heat, cold, etc. at that time (2004). While airplane data recorders are theoretically upgradable, the airlines wouldn’t spend the money and with this model we’re talking a twenty-year old design. I’ll wager the 787 and A380 recorders are far more advanced.

  50. 50
    StringOnAStick says:

    Airlines are barely profitable businesses in general, and most people shop for tickets based on who is cheapest, not who has the most up to date equipment. Plus, which airline wants to brag “hey, we spent a bunch on upgrading cockpit voice recorders so your grieving relatives will know how/why your plane crashed!” Eh; not a big marketing tool.

    What likely happened was a hijacking that went bad. First a forced change in flight direction and the crew being forced to turn off the tracking devices we’re already heard tons of speculation about, followed by gunshots that caused catastrophic depressurization, at which point everyone goes to sleep, dies, and eventually the plane runs out of fuel and crashes into the ocean. We’ll never know this for sure because finding the data recorders is so far beyond ‘needle in a haystack’ it is hard to contemplate, and the 2 hour looping for the voice recorder means all the audio evidence of such an event was recorded over long before the plane crashed.

    Why this stays in the news is partly because they need to fill airtime with sensational stuff that draws viewers, and it draws viewers because humans respond to any scary tragedy by trying to gather enough information so they can assure themselves “this will never happen to me”. Random tragedies are extra likely to draw this subconscious response. It’s why we gawk at car crashes or spend time going over and over events when death was close but the scythe missed us. At it’s core it is part of our survival programming.

  51. 51
    Bailey says:

    @Mobile Grumpy Code Monkey:

    It seems like there should be some viable business model out there that eliminates a non-destructible black box entirely and instead just loads all flight data directly to cloud storage.

  52. 52
    Robert Sneddon says:

    @Trollhattan: Actually no. Flight Data Recorders (FDR) are extremely conservative in design but incredibly rugged, fire resistant, water resistant etc. and the designs don’t change much if at all from decade to decade because a new design isn’t as trustworthy. The data from the AF447 recorders, when they were found, was easily transcribable and that was after they had spent almost two years several kilometres down in the ocean at crush pressures of thousands of atmospheres. A newer design might not have survived due to a flaw or something someone overlooked when laying it out.

    The location pinger is another matter. It wouldn’t affect the FDR or the voice recorder structure to rejig the pinger to, say, increase the ping interval over a period of days to allow the battery to last longer than the thirty days most pingers are guaranteed to operate for.

  53. 53
    Trollhattan says:

    @Robert Sneddon:
    Not so much the same old box..

    Boeing’s 787 will be the first aircraft equipped with enhanced airborne flight recorders (EAFR), combined “black-box” cockpit voice and flight data recorders (CVR/FDR) with crash-protected memory and the capability to record datalink messages and cockpit imagery.

    Smiths Aerospace is developing the 787 EAFR to meet the Arinc 767 standard now being finalised by the Airlines Electronic Engineering Committee (AEEC). The solid-state EAFR will meet current mandates for CVR and FDR performance, but with capability “well beyond” that, says Daniel Martinec, director of industry activities for Arinc, which administers the AEEC. Arinc 767 includes the capability to record 4-6 frames/s of cockpit imagery.


    These technologies do progress, just very slowly, which is the point I was trying to make.

  54. 54
    Elie says:

    I dunno — there is just something transfixing about the eternity of this depth and that these poor people who were just living their lives, being with their families and working — end here.

    I have been going through a process of bearing witness to my mother’s slow leave taking from this life — how long and deep this process is turning out to be — how mysterious and profound….

    Yes, the media have exploited this — but I must also confess to a persistent curiosity about this mystery — even as I live through the mystery of my mother’s separation from this life

  55. 55
    Mobile Grumpy Code Monkey says:

    @Bailey: This is way out of my comfort zone, but I think there wouldn’t be sufficient bandwidth to replace FDRs entirely. Not to mention you’d need rock solid comms.

Comments are closed.