Would it have hurt too much to have linked to a real civil aviation news site like the Aviation Herald for your running realtime summary of developments, rather than a generic mass media news site run by non-specialists?
(It's about as useful as linking to the BBC for news of a new OS vulnerability, rather than CERT, or even The Register.)
Personal observation: I flew through KUL on Malaysian last year, and was very unimpressed by security: it was cursory in the extreme and if I remember correctly out passports weren't electronically captured on entry or departure. Reports of someone using a stolen Italian passport being among the passenger manifest are a bit worrying (although some sort of mechanical failure is far more likely than terrorism).
the threat models vary. I haven't flown on MAS from KUL, but I have flown Lion and Air Asia from there. The security is a bit odd from American perspective but I didn't think (at least on the Lion terminal) that it was any different than what I have seen in Singapore or Taipei.
The last time I read a news title like this (where they use words like Flight Vanishes) was in 2009 about the Air France Flight 447 from Brazil to France.
I still find this story extremely puzzling for many reasons:
1-Boeing 777 is one of the most technologically advanced and safe planes.
2- Malaysian Airlines are one of the best airline companies in the world.
3- The weather seemed to be clear with no storms or strong wind (unlike the case with the French flight)
4- communication was lost completely with the aircraft.
I'm not an airline expert, but shouldn't be there at least a fallback for the communication protocol being used e.g: radio/satellite/GSM etc .. ?
On #4, that's one thing I noticed too. No mayday. No emergency declared. No contact. The flight just failed to show up.
I can think of three cases where this could be. The first is where the pilots tried to communicate that something was wrong and were unsuccessful, but I would expect that someone might have heard something.
But the other two are more catastrophic. The plane could have disintegrated in the air, giving no time for communications. Midflight breakups are uncommon but they do happen.
Finally it is possible that the pilots for some reason didn't know they were descending and could have crashed into the ocean without ever knowing they weren't at altitude. This apparently happened with AF447 to some extent. It has happened before as well (static ports get clogged, pilots get disoriented at night, etc). In this case the first sign that anything was wrong could have been impact.
They were talking to a retired pilot on the news and they asked him how it was possible that no distress call was made from the plane. The pilot said that pilots are trained to keep the plane in the air, if something starts to go wrong they start troubleshooting and focus all their energy on fixing the problem. If they then have time to send out a call, they do, but their main priority is to try and keep the plane in the air.
So it's possible that if the problem they experienced was complicated or happened too quickly, they never got a chance to send out a distress signal.
As I think about it, with the growing complexity of automated systems on the aircraft, the things that can go wrong multiply, and the chance of getting a distress call out diminish :-P
I don't know about avionics much either, but I'm wondering why, even at the time of AF447 if the planes do broadcast information about their current position (including altitude and speed), there seems to be no obvious "Your plane is dropping from the sky!" warning generated by any systems and remotely sent. Even if it's not a continuous stream of data, planes don't fall infinitely fast; it took a few minutes for AF447 to reach sea level.
The A380 does this. If you watch the Air Crash Investigations episode (also known as "Mayday) Season 13 Episode 10 - it talks about the Qantas A380 that had an engine explosion.
All the data from the plane was being relayed back to Qantas HQ - and the engineers at Qantas HQ were able to see in realtime all the issues facing the plane.
Ironically there were so many issues, they actually initally thought the system was faulty (i.e. there were too many faults, so the fault must be the alert system).
That's another case of the automation paradox though, is it not? The engineers are suddenly inundated with all kinds of information and it isn't clear what the actual cause is. When you look at AF447, part of the problem was that the pilots appear to have been unaware that all the errors they were seeing were caused solely by loss of airspeed indicators.
So most of the time you aren't looking at stuff, then suddenly something happens and you get tons of information on what exactly isn't working right now, all of which may or may not be useful in determining what the real problem is. And in this case humans have to step in quickly and react flawlessly based on perfect knowledge of the failure.
Many modern planes do have satellite uplinks, but they're mostly used for sending short text messages. Primary voice communications are still over VHF or HF radio. If you're having a mechanical problem, the pilot is going to be more focused on not crashing than composing an email.
(There are some automated logs that are sent too -- and these were actually used as part of the Air France investigation -- but this is primarily for maintenance rather than real-time tracking.)
Even ignoring fancy telemetry systems, planes are normally equipped with an emergency transponder (ELT) which activates if there's a sudden impact.
That signal would normally be detected by a Cospas-Sarsat satellite within 1-2 hours. In this case, it's possible that the transponder didn't activate, or the signal is too weak to be detected due to being underwater.
Your comment demonstrates how monumentally "The War on Terror" has failed, now that people have been so thoroughly terrorized that the most available explanation for a plane disappearing has become "terrorism".
That depends on how you view the war on terror. If you view it as political construct of Govts. wishing to grab extraordinary amounts of power and oversight over a largely disenfranchised populace then I would say it has been remarkably successful.
A large modern aircraft completely disappears suddenly, in good weather, without a trace. Terrorism could certainly be a logical explanation. It wouldn't be the first time. Mid-air collision, shot down, etc. The suddenness does raise extreme conclusions.
in this day and age? Sure. With the technology behind the newer planes flying to day it would take an extraordinary event to down one without word.
So the thoughts that come to me are, flew into terrain (mountain), door opened causing decompression and deformation of airframe, airframe failure, bomb, or someone took over fast enough to silence communications.
What is the availability of communications to passengers in that area? Cell has to be pretty much far between?
I don't think anybody can disable communications from inside the plane (but then, that's a guess, if anybody knows better, please tell so).
It would be compatible with some group kidnapping the plane, and changing the route to go over the ocean. But the plane would have already appeared somewhere, so I don't think that's likely.
Also, if it was a bomb, it would need to be quite a big one, to completely destroy the plane. Also unlikely.
He said it had a non-zero probability and should be considered, not that it was the most available explanation.
Considering that just a few days ago, China suffered a terrorist attack on a railway station by separatists that left nearly three dozen dead, it would certainly not be surprising to see an attack against an airplane flying to China and carrying a large number of Chinese passengers.
IMO, that's the most likely scenario. This highly advanced airplane was in clear weather in a spot where any distress signals it sent would have been received. The plane itself can send automated signals even if the pilots were incapacitated. No signals went out, and according to media reports, the plane simply "vanished" from radar (i.e. the data didn't indicate the plane descending before it disappeared).
All of the data indicates a mid-air explosion. There are some possible causes of such an explosion that are not terrorism related, such as mislabeled/improperly stored explosive cargo, but IMO those causes are at least as likely as terrorism in this case. It will be interesting to see if any groups come forward claiming responsibility.
I can think of two likely possibilities. The plane might have disintegrated in midair (that is not necessarily intentional btw, and most midair breakups are bad maintenance), or something could have gone drastically wrong. For example this plane had a previously damaged wing. If something went south, it is possible that the plane could have ended up in a flat spin and crashed.
Another possibility is AIRDU failures leaving the plane to fly into the sea.
> The fault-tolerant ADIRU was designed to operate with a failed accelerometer (it has six). The redundant design of the ADIRU also meant that it wasn’t mandatory to replace the unit when an accelerometer failed.
This seems unwise to me. I could understand if the rule was that you could put off replacing a failed accelerometer until the next regular maintenance. If it is OK to simply never replace it, then why not just ship the plane with five and save money?
> I could understand if the rule was that you could put off replacing a failed accelerometer until the next regular maintenance.
I think that's the idea. Whether or not it is followed in practice is another matter, and testing further failures was not the priority it should have been.
Of course I may be proven wrong, but I feel that it is quite premature to make such an assumption. A plane can rapidly disappear from radar for a number of reasons: not including your suggestions of terrorism or explosive payload.
The thrust of my comment was that the data indicates a very rapid breakup of the plane (explosion), and I indeed said that there are other possible causes of mid-air explosions besides terrorism. My mention of explosive cargo wasn't intended to be an exhaustive list of alternative causes. I just think that terrorism is certainly within the realm of possibility given the totality of the circumstances.
There is also a long stop in the flight data, is it possible this was one of the pilots trying fool a hijacker, by secretly change something so that data was sent again?
Hope they find it quickly.
That's not really how we use probability, how about we start by considering the most likely scenario and, if needed, progress to the "non-zero" probability events.
In 2013 there were 12 accidents concerning aircraft, none of which were related to terrorism.
In 2012 there were 12 accidents relating to aircraft, none of which were related to terrorism - there was also on attempted hijacking
In 2011 there were 25 accidents concerning aircraft, none of which were related to terrorism.
In 2010 there were 26 accidents concerning aircraft, none of which were related to terrorism.
- One of these bore some similarity to this incident, in that the aircraft was lost on radar. - Pamir Airways Flight 112, vanished from radar ten minutes after takeoff. It was eventually found 12 miles from Kabul.
In 2009 there were 21 accidents related to aircraft.
- There were also two terrorist incidents, one hostage taking and one attempted bombing.
Air France Flight 447 was one of the planes lost that year, it was lost over the Atlantic Ocean and it took two years for the wreckage to be found and the black boxes recovered.
It vanished 'without a Mayday call, a witness, or even a trace on radar [...] Flight 447 seemed to disappear from the sky'
Going by those figures, a little over 3% of the incidents involving aircraft were terrorism-related. And none of those, to my knowledge, resulted in deaths, or even significant injury.
1) Accidents are frequent (and I suspect the list I've got is out by one or two since I didn't count it automatically, and somewhat incomplete in any case) - the number of planes in the sky pretty much ensures that.
2) Terrorism related to flights seems relatively rare, and most commonly relatively ineffectual as compared to things like messed up landings.
3) Planes have vanished from radar before without foul play, (and seemingly in similar conditions in the case of 447.)
And that's roughly in line with what we'd expect before seeing evidence. Unless you know that the two are strongly related, the way to bet is that the intersection of two low probability events is going to be vanishingly small.
So, while there's no solid proof that it isn't terrorism in this case, that's the way I'd bet if I were a betting woman ^_^
You're not going back far enough! Why would you think 5 years is enough to draw conclusions? If you look back 15 years (long enough to include 9/11), then 50% of commercial aviation fatalities are due to terrorism.
50% of commercial aviation fatalities being due to terrorism, well maybe if you're counting the people on the ground, (I'm not interested enough in fatalities to do the maths on that.) Disregarding, the ground fatalities however:
There were for planes in 9/11. According to the list I last quoted, there were:
92 people on a Boeing 767200-ER
65 people on a Boeing 767-200
64 people on a Boeing 757-200
44 people on a Boeing 757-200
=> 265
If the fatalities are 100% in this most recent crash, it almost exceeds the 9/11 aviation fatalities by itself. Excluding this most recent incident - and just going from the wikilist again - you only have to go back to May 14th 2012 to exceed those casualties:
So, I find it highly unlikely that 9/11 does account for 50% of the commercial aviation fatalities in the last 15 years since it would account for less than 50% in the last 2 if it had occurred in those years.
-------------
As for the other points -
I'm interested in incidents concerning aircraft, and not the number of fatalities in general. The question isn't what is your chance of dying in a terrorist attack, the question is what sort of figure can we hang off this plane vanishing being due to a terrorist attack. 9/11 is four aircraft - from that point of view it doesn't change the figures much.
I could go further back and increase the confidence level but I don't really see the point. This isn't the sort of scenario where a more precise number seems likely to matter, even if I were out by an entire order of magnitude, it would make little difference. It would still be the thing to bet against it being terrorism.
I might bother to get a more precise number if I were putting money on it, but that would largely be a matter of knowing how much money to bet rather than the direction of the bet.
Basically, if you want a more precise number, you'll have to do the work yourself.
Governments have developed so much surveillance technology that they can easily invade the privacy of many citizens, yet tracking the exact whereabouts of planes flying in the sky - something that for scheduled, commercial flights whose routes should very much be publicly known information - still seems out of reach? I find that fact a little unsettling.
Alternatively, such information is available, but the system that produces it is classified so the data is not released. The surface area of the Earth is 1.7e14 square metres, so it would be feasible to cover the earth with pixels to a resolution of a square metre. Each uncompressed whole earth image would be less than a Petabyte at 32 bits per pixel. Add compression and deep pockets and it would be possible to create and store whole Earth movies.
You can't be practically tracking every co-ordinate at every second, not yet. Once you know what to track, you can do it effectively. For airplanes, it's already late. Maybe, if they are lucky, based on satellite imagery, they can gather traces.
I appreciate that radar only works above a certain altitude, but can someone explain to me why a vehicle worth > $200m doesn't have a GPS unit broadcasting over longwave radio,satellite connection or some other remote communication means. I feel like I'm missing something fundamental.
All modern aircraft are equipped with Mode-S transponders [1] which broadcast several parameters such as altitude, lat/longitude and aicraft ID info. There are also specific messages that signal "emergency status"[2].
You can 'listen' to these messages yourself with very basic radio equipment tuned to 1090 MHz[3].
Even though these signals can reach long distances (I remember getting ~300 Km with my home made antenna) it's possible that the plane was going through an area where no base stations exist.
Sure. I find that I easily lose track of planes if they are below a certain altitude. Still, it is very strange that the last recorded altitude is 35000 ft [1].
What looked odd to me at first was not the deviation but the fact that it seemed almost mechanical. As I looked at other flight paths vs actual flights however it became clear that this sort of deviation was relatively common.
All those kind of systems can fail or give inconclusive data. What you need to know is the crash site, not the last GPS trace. It would be better to include substances in aircraft that would be released in the event of breakup and be easily visible to remote sensing and radar satellites.
Apparently an Italian passenger on the passenger list was in fact not on the plane[1] (and did not buy a ticket, passport was stolen), could be another sign for a terrorist background.
For more information/speculation the airliners thread[2] is also a good source.
What we know is that two apparent passengers were on board with stolen or forged passports. Obviously they meant mischief, since you don't travel internationally on a stolen passport if everything is above board.
But when you think about it the sorts of mischief that could be involved are quite many and varied. One can't rule out terrorism, but it's one possibility among sufficiently many that this is not where I personally would start. Far more likely you are dealing with organized criminals doing more typical organized criminal things (like smuggling stuff).
Additionally usually if the plane is intentionally downed, someone will want to take credit/responsibility. That this hasn't happened is an indication, particularly absent patterns of attacks here in the past, that this is just criminal business as usual and not related to the accident.
Finally there are early unconfirmed reports that the Vietnamese navy may have spotted smoke columns yesterday. if that turns out to be the case, it would be a strong indication that the plane exploded on impact with the water, not in the air.
At present I am so far from being able to endorse the T-word explanation that it doesn't seem worth looking further into at this point.
there is a 250-400 km range for FlightRadar, it does not cover oceans very well.
http://www.flightradar24.com/how-it-works
"Flightradar24 has a network of more than 3,000 ADS-B receivers around the world that receives plane and flight information from aircraft with ADS-B transponders and sends this information to our servers. Due to the high frequency used (1090 MHz) the coverage from each receiver is limited to about 250-400 km (150-250 miles) in all directions depending on location. The farther away from the receiver an aircraft is flying, the higher it must fly to be covered by the receiver. The distance limit makes it very hard to get ADS-B coverage over oceans."
Not really, no. Immediately after that statement it says "When contacted, Malaysia Airlines declined to confirm or deny the reports", and nor has it been confirmed in the six hours of updates to that story since.
How difficult is it to scan the entire area using satellites? There are civilian satellites that offer resolutions better than 30cm today, surely enough to spot wreckage in the middle of the sea. DigitalGlobe alone has over 3 active satellites that can scan an area of over 10,000 km^2 in a single pass, and the combined revisit times should be good enough. This should also be possible using SAR with a good enough resolution, as the sea would appear mostly black and anything non-smooth on top of it should stand out.
Edit: made a mistake concerning the area. Fixed now.
The whole area between Malaysia and Ca Mau is bounded within a square of less than 500 km^2, and even that is a vast exaggeration of where the plane is most likely to be.
Either I am completely confused or you are misreading something. The linear distance is more than 300 kilometers. 500 square kilometers is only 22 kilometers on a side. It's a tiny little patch of the sea there.
(It's about as useful as linking to the BBC for news of a new OS vulnerability, rather than CERT, or even The Register.)
The (detailed) AvHerald coverage is here: http://avherald.com/h?article=4710c69b&opt=0
Personal observation: I flew through KUL on Malaysian last year, and was very unimpressed by security: it was cursory in the extreme and if I remember correctly out passports weren't electronically captured on entry or departure. Reports of someone using a stolen Italian passport being among the passenger manifest are a bit worrying (although some sort of mechanical failure is far more likely than terrorism).