Slashdot Mirror
US Navy Strategists Have a Long History of Finding the Lost
Hugh Pickens DOT Com writes "Benedict Carey reports at the NYT that the uncertainties surrounding Malaysia Airlines Flight 370's disappearance are enormous, but naval strategists have been unraveling lost-at-sea mysteries as far back as the U-boat battles of World War II, and perhaps most dramatically in 1968, when an intelligence team found the submarine Scorpion, which sank in the North Atlantic after losing contact under equally baffling circumstances. "The same approach we used with Scorpion could be applied in this case and should be," says John P. Craven who helped pioneer the use of Bayesian search techniques to locate objects lost at sea. "But you need to begin with the right people." The approach is a kind of crowdsourcing, but not one in which volunteers pored over satellite images, like they have in search of Flight 370. "That effort is akin to good Samaritans combing a forest for a lost child without knowing for certain that the child is there," writes Carey.
Instead, forecasters draw on expertise from diverse but relevant areas — in the case of finding a submarine, say, submarine command, ocean salvage, and oceanography experts, as well as physicists and engineers. Each would make an educated guess as to where the ship is, based on different scenarios: the sub was attacked; a torpedo activated onboard; a battery exploded. Craven's work was instrumental in the Navy's search for the missing hydrogen bomb that had been lost in the Mediterranean Sea, off the coast of Spain in 1966 and this is how Craven located the Scorpion. "I knew these guys and I gave probability scores to each scenario they came up with," says Craven. The men bet bottles of Chivas Regal to keep matters interesting, and after some statistical analysis, Craven zeroed in on a point about 400 miles from the Azores, near the Sargasso Sea. The sub was found about 200 yards away.
In the case of the downed Malaysian plane, forecasters might bring in climate and ocean scientists, engineers who worked on building the plane's components and commercial pilots familiar with the route. Those specialists would then make judgments based on the scenarios already discussed as possible causes for the disappearance of Flight 370: terrorism, pilot error, sudden depressurization and engine failure. Sound-detection technology in and around the Indian Ocean may aid this forecasting. The sound of the airliner's fall — if it hit the water — might already have been picked up by submarines watching each other. "In that case the information would be classified," says former submarine commander Alfred Scott McLare, "and we wouldn't know anything until it was released through back channels somehow.""
Instead, forecasters draw on expertise from diverse but relevant areas — in the case of finding a submarine, say, submarine command, ocean salvage, and oceanography experts, as well as physicists and engineers. Each would make an educated guess as to where the ship is, based on different scenarios: the sub was attacked; a torpedo activated onboard; a battery exploded. Craven's work was instrumental in the Navy's search for the missing hydrogen bomb that had been lost in the Mediterranean Sea, off the coast of Spain in 1966 and this is how Craven located the Scorpion. "I knew these guys and I gave probability scores to each scenario they came up with," says Craven. The men bet bottles of Chivas Regal to keep matters interesting, and after some statistical analysis, Craven zeroed in on a point about 400 miles from the Azores, near the Sargasso Sea. The sub was found about 200 yards away.
In the case of the downed Malaysian plane, forecasters might bring in climate and ocean scientists, engineers who worked on building the plane's components and commercial pilots familiar with the route. Those specialists would then make judgments based on the scenarios already discussed as possible causes for the disappearance of Flight 370: terrorism, pilot error, sudden depressurization and engine failure. Sound-detection technology in and around the Indian Ocean may aid this forecasting. The sound of the airliner's fall — if it hit the water — might already have been picked up by submarines watching each other. "In that case the information would be classified," says former submarine commander Alfred Scott McLare, "and we wouldn't know anything until it was released through back channels somehow.""
Navy guys will need more data.
Those much hyped arcs from Inmarsat are pretty much bogus. The trouble is that the problem is badly conditioned - because satellite is way too far (geosynchonous orbit - not your friendly neighborhood gps) and it's right on top of the search area. In other words - small errors in time/distance measurements, satellite position, etc. produce huge errors in estimation. They're lucky they placed the airplane on earth.
We've seen maps of where MH770 could be based on the angle of last ping received from the engines. Here's one: http://static01.nyt.com/images/2014/03/16/world/asia/16flight-map/16flight-map-superJumbo.jpg
We have a Last Known Position (indicated on that map). We know how fast 777s can fly. If we had the ping arc data as shown in red on the above map for every ping received, we could determine MH770's course, and narrow down where it ended up significantly.
The following numbers are wrong, but a concrete example is easier to follow. Say the first ping occurs 15 minutes after the Last Known Position, and we think the 777 is flying at 500 mph. Set your compass for 125 miles (scale), put the pointy end on the last known position, and draw a circle. That circle will intersect the First Ping Arc in two places (we hope). If it doesn't, we need to rethink assumptions. Anyway, the plane was in one of those positions (more or less) at the time of tyhe first ping.
Do it again for the second ping arc. And again. Some of these potential courses will make no sense and no longer need to be followed. With any luck. though, there will emerge a Most Probably Course for the aircraft.
It may be necessary to rerun this analysis for different speeds - if MH770 was flying low to avoid radar it would travel more slowly. Do it. Hell, throw the entire problem to a computer and let if grind out possibilities.
Has the satellite angle data, or the location arcs at particular ping times, been released? Can it be released?
The plane was stolen. Forget about failures that there are no reason to think happened, about explosions or mechanical failures, about suicides or searching the ocean for debris. Just figure out where a stolen 777 was taken and you'll find the plane.
I'm an American. I love this country and the freedoms that we used to have.
Actually, it is tricky; it sounds like they know more than they do. They talk about 2 flight paths, but actually it is a giant arc from Pakistan to Thailand to the Indian Ocean, and they don't even have a direction. Just a range from the satellite based on the signal strength, which produces an arc that it probably was in when the ping transmitted. There are 2 obvious "flight corridors" in that arc, so those are the best guesses. Sounds clearer than it is.
Also nothing has been released about if they stopped for fuel, or if it is known. The US keeps saying they think they crashed into the sea in 1 of 2 areas, which implies that they don't know that they DID refuel; but the way they phrase the combination of statements, I think they don't have information to negate refueling, they don't have indication of it. And without refueling, and assuming it was in one of those two corridors, then it would have likely crashed.
Also they're assuming that the fuel supply is based on having been properly fueled for a flight to Beijing, but no public information has said anything at all about having verified on the ground how much fuel was loaded, or if that can be accurately checked up on with certainty. Seems like airport corruption would have to be 0% in order for them to even know. I'm under the impression that airports in Malaysia actually have a significant corruption problem, and so it is probably impossible to go back and check in the past how much fuel was really taken on. Maximum range at maximum load for the 777-200ER is 7,725 nmi (14,310 km, 8,892 mi), a whole lot more than the 2500 nmi circles the media is drawing on the screens.
So if they re-fueled OR if they loaded extra fuel, they could be anywhere, and the Indian Ocean flight corridor that is speculated on would lead to waypoints to the middle east. I'm guessing Iran, but it could just as easily be in Sudan or Pakistan.
The delays in turning off the transponder and the data stream to the modem, flying between way points on a well known path etc might be explained by confused and disabled pilots too.
Hypoxia can set in as little as 90 seconds of oxygen deprivation and will severely incapacitate and confuse people. Cabin pressure loss is the most common theory for hypoxia. But cabin pressure loss would deploy oxygen masks, sound alarms and the pilot would have been alert in the first few seconds to declare emergency and radio out. The captain seems to be nerd with home made flight simulator, he would have reacted correctly to oxygen masks dropping from the ceilings.
Carbon monoxide is a way for hypoxia to set in. If there was a slow smoldering fire in the cockpit, not hot enough to trigger fire alarms it could result in incapacitated confused pilots. Again there are CO detectors, and warnings and associated with it.
I am not sure how regularly these systems that detect cabin pressure loss and CO detectors are tested. It is quite expensive to actually deploy all those oxygen masks. So even the regular testing protocol would require the maintenance crew to disable the actual deployment of the oxygen masks and test the detection and deployment signals. They could forget to turn them back on, like the did in the Helios flight disaster I mentioned in another thread. CO detector is chemical based. They have to be replaced regularly and this is an old plane.
Once the pilots flip switches on and off in confused state lose their consciousness completely, the plane would fly on autopilot following the way points that happened to be programmed.
If there is foul play involved, it would be worthwhile exercise to make sure every flight plan that was file in that duration and every flight directed by the control towers in that time is legit and locate those planes. The pilot(s) could easily turn off the transponder, drop out of radar, pop back in and start using a different call sign. Without a transponder, air traffic control completely trusts the pilot to self identify the plane correctly. If the malefactors had filed a fake flight plan, the plane could change its identity mid flight without attracting attention.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
1 The last fix from Inmarsat gave a Line of Position (LOP) which is a very broad arc.
They had a ping every hour, each of which should have resulted in an LOP.
Is there a way to combine these LOP's to get a better idea of the flight path?
(Old school marine folks would walk the old LOP's forward in time and combine them.)
One would have to guess a direction and speed to do this which makes the logic somewhat circular.
Still, there should be more information in the rest of the LOP's.
2) Who benefits from all this?
This has focused attention on the flight and not on what's happening in with Russia.
This seems an unlikely motivation, but it is a definite consequence.
I certainly hope this is not the motivation behind this.
After each search result comes in, you recompute and find the next best place to search.
davecb@spamcop.net
Nope. That NYT piece seems to have forgotten about the initial statement from Malaysia Airlines, which said the last time of contact with flight 370 was at 2:40am:
https://www.facebook.com/my.ma...
That was before the Malaysian authorities went into full denial mode and claimed last contact was at 01:20am. The 02:40 time was inconsistent with their estimated "crash site" in the Gulf of Thailand, which was one of the initial sources of confusion. However, 02:40am turned out to be the exact time of last military radar contact which they were forced to confirm more than 5 days later. Additionally, there were the "small" details that two transponder systems were turned off one after another more than 10 minutes apart, and that the ACARS system was turned off before the last voice contact with the pilots.
In order to fit all these facts into a theory of stupidity, you'd have to accept that: 1. an unidentified flying object the size of a 777 can just fly across the width of Malaysian airspace (more than 1 hour of flight time) at cruising altitude without being noticed by the Malaysian military 2. that 02:40am time from Malaysian Airline's initial statement just turned out to match the time of last military radar contact by complete coincidence 3. nobody noticed the time descrepancies between the two transponder systems turning off.
This is clearly beyond the realm of incompetance, and can only be explained with a touch of malice. The Malaysian authorities knew from the beginning what was going on, but was more concerned with the possible liabilities and damages to their "image" resulting from a rogue pilot, than with actually finding the plane. With wanton disregard for the 239 lives on board and their relatives on the ground, they knowingly misled the international community on a wild witch hunt across the Gulf of Thailand, delaying the search for at least five crucial days, thereby eliminating any possiblity of finding survivors (if the plane had ultimately crashed), and quite possibly lowering the likelihood of finding the cockpit recorders to near zero.