Malaysia Airlines 370 and Reporters Who Have No Idea What They Are Talking About

March 21st, 2014
submit to reddit Share

The disappearance

of Malaysia Airlines flight 370 is, without doubt, one of the strangest episodes in recent aviation history.  Though it has not been found, the current evidence seems to indicate that someone on board the plane, most likely a crew member, shut down most of the on board communications systems and then flew the plane in a direction away from its flight plan.  Given that the 777 aircraft has exceptionally long range capabilities and that it appears to have been headed toward a large area of open ocean, with no radar coverage, the search has been very difficult.

The reporting on this event has ranged the gamut from pretty good to absolutely horrible.  One of the worst things seen is the numerous glaring errors in major publications about basic technical facts regarding aviation and the aircraft in question.

Reporters, of course, don’t generally know a lot about commercial aviation, aerospace technology, search and rescue or any of the other specialized topics involved.  Degrees in journalism don’t usually requite training in basic aircraft systems.   That’s a given, as it is with most highly technical topics.  However, it’s not exactly difficult to find people who are real experts in the area.  So if you are reporting on a story for a newspaper or other publication, why not track down an actual expert before writing about transponders or ACARS or ETOPS requirements or anything of that kind?   In fact, I’d advise tracking down more than one, just to make sure the one you find first is not BS’ing you.

Here are some examples:



Via The Associated Press:

Part of the problem, said Andrew Thomas, the editor-in-chief of the Journal of Transportation Security, is that airline systems are not as sophisticated as many people might think. A case in point, he said, is that airports and airplanes around the world use antiquated radar tracking technology, first developed in the 1950s, rather than modern GPS systems.

A GPS system might not have solved the mystery of Flight 370, which disappeared March 8 while flying from Kuala Lumpur, Malaysia, to Beijing. But it would probably have given searchers a better read on the plane’s last known location, Thomas said.

“There are lots of reasons why they haven’t changed, but the major one is cost,” he said. “The next-generation technology would cost $70 to $80 billion in the U.S.”

I don’t know where they dug up this guy Andrew Thomas, but the statement is simply not true. For one thing, the radar technology used for air traffic control is not vintage 1950′s, at least not most of it. In the US the radar system used for most air traffic control is the Joint Surveillance System, which came online in 1983. A commonly deployed radar system is the ARSR-4, which was first developed in the late 1980′s and has since been upgraded. There may be some older radar systems that have either been upgraded or integrated into newer systems, but it is not 1950′s technology. Europe and most other industrial nations are similar in their radar systems.

The reason that radar is still used is that it does not require an active transmitter from the aircraft. Even if the aircraft loses power, does not have the proper equipment or the equipment does not function properly, radar will still be reflected back.

In addition to the “old fashioned” radar, air traffic control also utilizes a number of other tracking technologies, many based on GPS. Modern air traffic control includes secondary surveillance radar, which uses a transponder on the aircraft to provide additional data. ADS-B is another method of locating aircraft by means of GPS data transmitted back to ground stations. Aircraft which travel outside the coverage area of ADS-B receivers can report their position by ACARS, which uses VHF, HF or satellite transmitters.   There are other data systems that are non-standard but are used by some airlines to transmit data such as location.

These systems are widely deployed and MH370 had all of them. It seems they were turned off when contact was lost.  I have no idea where the 70 to 80 billion dollar figure comes from.


Via RTV6:

Tuesday marked 10 days since Malaysia Airlines Flight 370 (MH370) disappeared en route from Kuala Lumpur International Airport to Beijing Capital International Airport.

But Cincinnati IT worker and hobby aviation enthusiast Keith Ledgerwood says he may have figured out what happened.

Ledgerwood said a major roadblock to that theory, however, is the fact that India and Pakistan have reported no unidentified aircraft entering their airspace and it would be highly unlikely for a Boeing 777 to slip through undetected.

But Ledgerwood believes the plane may have “shadowed” another plane to appear undetected – and his theory was picked up Tuesday by The Washington Post as one of the “best” on the Internet.

“After being unable to escape the idea that it may have happened, I began to do some analysis and research and what I discovered was very troubling to me,” Ledgerwood wrote in his Tumblr blog, which has now had more than 1,000 shares.

After days of research and plotting out points in elaborate diagrams, Ledgerwood claimed the Malaysia Airlines jet could have flown through India and Afghanistan airspace by closely following a Singapore Airlines jet flying a similar path.

By turning off its communications systems and following the other jet at a certain distance, the two “would have shown up as one single blip on the radar,” Ledgerwood said.

“SIA68 (the Singapore Airlines jet) would have had no knowledge that MH370 was anywhere around and as it entered Indian airspace, it would have shown… with only the transponder information of SIA68 lighting up military radar screens,” he said.

Ledgerwood said the collision avoidance systems on modern planes use the transponder, which someone on the Malaysia Airlines flight could have turned off.

“Once MH370 had cleared the volatile airspaces and was safe from being detected by military radar sites in India, Pakistan and Afghanistan, it would have been free to break off from the shadow of SIA68 and could have then flown a path to its final landing site,” he speculated in his blog.

But Ed Bridgeman, terrorism expert and head of the criminal justice program at the University of Cincinnati Clermont College, called Ledgerwood’s theory “highly, highly unlikely.”

“From what I know about aircraft systems, I don’t think it’s possible for flights to be close enough to blend radar signatures,” Bridgeman said. “This isn’t like World War II movies where submarines sneak under a destroyer. The Malaysia flight would have given some indication to the other flight that somebody else was in the neighborhood.”

This one local news report has become highly cited and the hypothesis that the plane shadowed another is now showing up all over the newsmedia. Many now leave off the part about it being “highly highly unlikely.” This idea was developed by an amateur with no official credentials. It’s not news. It’s one guys idea.

The problem is that radar systems are indeed pretty good at telling one aircraft from two. It is possible this could be done if the two were extremely close, but that’s unlikely. The 777 is a very large aircraft and it would be all but impossible to actually get close enough to another aircraft without anyone noticing and stay in a stable flight pattern, despite the wake turbulence of the other plane.

It’s idle speculation that is not worth reporting.


Via International Business Times:

Malaysia Airlines Flight MH370, which disappeared March 8, lacked a simple computer upgrade that could have made a world of difference by providing crucial satellite data to the unprecedented international search effort, which entered its thirteenth day Thursday.

The upgrade, which Malaysia Airlines decided not to purchase, has a wholesale price of only $10.90 for each flight. The upgrade for a system called Swift could have continued to send flight data by satellite even after MH370’s transponder and Aircraft Communications Addressing and Reporting System, or ACARS, were switched off, The Washington Post reported on Thursday, citing a satellite industry official with knowledge about the equipment.

According to investigators, the transponder and the ACARS on the plane — which went missing with 239 people on board on a flight from Kuala Lumpur to Beijing — were deliberately turned off before the plane is suspected to have continued flying for another seven hours. However, if MH370’s Swift system had been upgraded, it could have sent information on the plane’s engine performance, fuel consumption, speed, altitude and direction, even if the transponder and ACARS were disabled, the Post reported.

“When ACARS is turned off, Swift continues on. If you configure Swift to track engine data, that data will be streamed off the plane. It continues to be powered up while the aircraft is powered up,” the official, who compared the Swift system to a mobile phone sending data to a satellite, told the Post.

In 2009, satellite data from a similar computer upgrade helped investigators find an Air France jet, which had crashed into the Atlantic Ocean. Thanks to data provided by the upgrade, investigators were able to quickly narrow their search area to a radius of about 40 miles in the Atlantic Ocean, the Post reported.

The full package of the Swift system is used by many major airlines, and the details it provides are authorized by international aviation guidelines and required equipment for airlines that frequently fly the North Atlantic corridor between the U.S. and Europe. However, according to the satellite industry official, there are no such requirements for planes flying other corridors elsewhere in the world.


The author of this article does not seem to know what Swift is or what it means or would help.  In fact, the author is just plain wrong about the whole thing, although it might have helped in the situation, that would be coincidental.

The aircraft transmits data back from the ACARS via the Inmarsat satellite system.  When the ACARS system was shut down, the Inmarsat transceiver continued to function.  It sent out an hourly ping which told the satellite it was still online.  That was the first clue that the flight was still in the air.   But the satellite does not track position and the data included no information on location.  Eventually, the satellite firm was able to deduce the probable distance from the satellite based on things like the signal strength and the timing.

Inmarsat also offers a service known as SwiftBroadband, which operates like their conventional satellite data system but is much faster.  It also uses spot beams, rather than one global beam.  So if the data is transmitted back via the SwiftBroadband system, it might have been possible to know what spot beam the aircraft was in.

The airline didn’t really skimp on SwiftBroadband because they are cheap.  It’s simply not necessary to have that much bandwidth for a lows-peed data protocol like ACARS.  The fact that it uses spot beams is not really a selling point of the service.  It just turns out that aspect could have been useful in this unusual and generally unforseeable case.


A few other common errors, reported frequently:

  1. Disabling some of the communications systems such as ADS-B and ACARS would not have required expert knowledge.  The manuals could be found online.  Transponders were disabled by the 9/11 hijackers too, and they had only rudimentary knowledge of aircraft systems.  Anyone with enough planning could find out how.
  2. The arcs shown on maps of the search areas do not show the possible flight paths.  As explained above, that is the probable distance from the satellite at the last transmission.
  3. The system that continued to transmit is not a secondary communications system that is separate from ACARS, it is the system that ACARS uses, but which not operational at the time.  It basically was in a kind of standby mode.
  4. The plane did not need to fly extremely low to evade radar.  There’s no air traffic control radar at sea.   The only radar to avoid might be if naval vessels were in the area, in which case, flying low would have been a lot more suspicious than flying at altitude.  There’s no evidence it flew extremely low in altitude.
  5. It’s probably not possible to make a transponder or transmitter that “can’t be switched off.”  This has been suggested by many in the news media.  What they seem to forget is that all these pieces of equipment tie into the plane’s electrical system through a system of circuit breakers.  These breakers can be turned off.  It’s a necessary safety feature in case there is a short circuit or fire.


This entry was posted on Friday, March 21st, 2014 at 9:08 pm and is filed under Bad Science, media, Misc, Not Even Wrong. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.
View blog reactions

7 Responses to “Malaysia Airlines 370 and Reporters Who Have No Idea What They Are Talking About”

  1. 1
    Jeebus Says:

    There is an article making the rounds under the title “a surprisingly simple theory of what happened to AH370″ or something like that. It has been shown to be pretty bunk, but still making the rounds.

    Quote Comment
  2. 2
    DV82XL Says:

            Jeebus said:

    There is an article making the rounds under the title “a surprisingly simple theory of what happened to AH370″ or something like that. It has been shown to be pretty bunk, but still making the rounds.

    I spent over forty years in aviation, and while I won’t say that the criticisms leveled at that article are off the wall in any way, I will say I’ve seen and heard and been involved in investigations of incidents where some very strange looking events wound up having rather pedestrian explanations, when all was said and done, than what the initial evidence seemed to show. That in and of itself is not proof that this is the case here, but only that I certainly know where this guy is coming from. In my experience, good accident investigation starts with looking at things like how a pilot would likely be reacting to manage an in-flight situation right from the beginning, and that is essentially what this “surprisingly simple theory” is suggesting be done.

    Quote Comment
  3. 3
    James Greenidge Says:

    Doesn’t take technical acumen. Just common sense and a little research. Why is it so hard for reporters to, instead of saying “streaming information” and “streaming engine information links” like it’s a internet thing, to just use the old aeronautical standby “telemetry”? To see reporters trying to play amateur techies is something to behold. I remember on WCBS-TV here in the ’90s a reporter stating that the “space shuttle just returned from a flight of over 4 million miles ‘way out’ in space, and what speculator views of Earth we got!” It wasn’t the mileage she got wrong, just the direction. With precision reporting like this, what reporter needs an anti-nuclear bias?

    James Greenidge
    Queens NY

    Quote Comment
  4. 4
    Sigivald Says:

    “A case in point, he said, is that airports and airplanes around the world use antiquated radar tracking technology, first developed in the 1950s, rather than modern GPS systems.”, he said.

    Notice “around the world”.

    I think he was really – poorly – trying to specify “in the not-so-developed world” rather than “everywhere in the world including the US and Western Europe”.

    From what I’ve heard from pilot types talking about this, Malaysia and Thailand and Vietnam don’t have great first-world air tracking systems, or anything close.

    I think he’s probably at least partially right, read closely and with “around the world” meaning “outside of the industrialized/modern world”, rather than “everywhere in the world”.

    Quote Comment
  5. 5
    PsihoKekec Says:

    My guess is that something like Helios Airways Flight 522 happened, pilot turned the plane around before losing consciousness and it flew in that direction until it ran out of fuel and crashed.

    Quote Comment
  6. 6
    drbuzz0 Says:

            PsihoKekec said:

    My guess is that something like Helios Airways Flight 522 happened, pilot turned the plane around before losing consciousness and it flew in that direction until it ran out of fuel and crashed.

    When this first happened that is also what I was thinking. But after more data came out, that seems less likely.

    First, if a depressurization were to happen, the absolute first thing the pilot would do is put on an oxygen mask. If they had time to turn around, they should have done that too.

    Also it does not explain why the transponder, ADS-B and ACARS were powered down. These are different systems, so it’s hard to imagine an electrical fault that could kill all these independent systems but at the same time not kill all the electrical power, since clearly the inmarsat transponder was powered on and it seems like some of the flight dynamics controls had power as well.

    Quote Comment
  7. 7
    Josh Says:

    Another issue is a lot of reporters are talking out their asses saying that all airplanes should have black boxes that transmit the data and the voice to the ground continuously, so if you lose the box, no big deal and there is 100% constant monitoring.

    Then they lament that it must be that we’re not willing to spend the money because the technology exists.

    Do they have any idea how hard that would be?

    I used to sell communications stuff for ships, so I have some idea of the problems here.

    Okay, first of all, you will need lots of ground stations or do it based on satellite (probably better, for coverage over all areas)

    Now lets see how much data we need. The audio comes from a few microhones (I think there are four independent channels. One per headset and two ****pit mics) and it needs to be good quality so you can hear voices even in noisy background. So lets say a bitrate of maybe 64 kbps * 4 = 256 kbps. Plus maybe 32 kbps for simple systems data = about 288 kbps. Plus overhead and error correction, lets say 350 kbps.

    Does not sound like much, right? Wrong. Mobile satellites charge an arm and a leg for bandwidth because they have limited bandwidth capability. I was looking at InMarsat which is who the Malysia airline used and they charge like a dollar a minute for voice and a dollar for every 2 megabytes data. Wow

    This is why ACARS only sends a small string of text every few minutes, btw. That saves bandwidth.

    30,000 flights in the air at any time! Could our mobile satellites even handle that? I don’t know, but any way it works, it would be mega-expensive.

    Either that or put in lots of ground stations.

    BTW: Internet satellites are cheaper (like for satellite internet in remote areas) but not designed for in motion stuff, limited coverage and not reliable enough for critical data on an airplane.

    The other issue is it would not be cheap equipment. Yeah, your iphone can transmit at that rate, but not to a satellite. Plus, consumer grade transmitters would not cut it. If it is used for this it needs to be aviation grade and thoroughly tested.

    Quote Comment

Leave a Reply

Current month ye@r day *

Please copy the string pykR7i to the field below:

Protected by WP Anti Spam