Join 3,552 readers in helping fund MetaFilter (Hide)


"We've been hit."
October 3, 2006 4:56 AM   Subscribe

In a corporate jet flying 37,000 feet above the Amazon rainforest, I heard the three words I will never forget: “We’ve been hit.”
posted by NotMyselfRightNow (70 comments total) 2 users marked this as a favorite

 
I'm trying to think of any other midair at FL300 or above where a plane survived, and I can't. Both are going to be flying at Mach .7 or higher. I think the speculation about a wingstrike will prove to be correct -- it is about the only way I can see the E600 losing a fin and a bit of tail, while the much larger and heavier B737 loses enough to spiral out of control and crash.

Of course, I have lots of questions? Who was controlling -- at that level, ATC is responsible for seperation, and we had two planes at the same altitude in the same place. TCAS should have been installed on the 737, I don't know if the E600 would have required it. I have to assume no, because TCAS warnings aren't In the US, Europe, Australia and China, a plane of that weight in commerical service would be required to have TCAS II, (the US required such by 1993, the rest by 2000, EASA now requires ACAS II, a superset of TCAS II, since 2005.quiet, esp. resolution advisories.

Here is the Embrare Legacy 600 and the Boeing 737. As a simple size comparison, the E600 has a MTWO of ~50,000lbs/ 22,500Kg. The 737-600, smallest of the currently built 737s, has a MTOW of ~145,000lbs/66,000Kg. (I don't know what model of 737 was involved, but I suspect an 800 or 900, given the number of souls aboard.)

I can't find anything about Brazil's requirements for TCAS. In the US, Europe, Australia and China, both planes in commerical service would be required to have TCAS II, (the US required such by 1993, the rest by 2000, EASA now requires ACAS II, a superset of TCAS II, since 2005. Most demand TCAS II in planes with more than 15,000Kg MTOW, in some nations, that number is lower.
posted by eriko at 5:32 AM on October 3, 2006


Aside from describing the damage to the business jet’s wing and tail, investigators have provided few details supporting their theory of a midair crash between the aircraft, both of them new and equipped with systems designed to prevent such a collision.
posted by srboisvert at 5:37 AM on October 3, 2006


Holy shit. Hard to figure out whether people in that situation are lucky or unlucky.
posted by dg at 5:49 AM on October 3, 2006


Well, to me, that damage is pretty compelling that something hit something. It's not utterly concusive, but the chances that it just happened to have lost the wingtip fin, which also damaged the stabliizer, in the same airspace as a 737 that crashed, on the same night, near the same time, are really, really low.

It isn't impossible, mind you, but the fact that the pilots said "we were hit" implies that there was some sort of collision, and the damage pattern argues against explosive damage.
posted by eriko at 5:58 AM on October 3, 2006


Mental note: always fly on Embraer jets whenever possible. Interesting that they're made in Brazil, and that's where this happened.
posted by zsazsa at 6:01 AM on October 3, 2006


maybe they both hit some third object that we haven't found.
posted by lester's sock puppet at 6:29 AM on October 3, 2006


Apparently this was ATC error:
posted by Osmanthus at 6:30 AM on October 3, 2006


Thanks for the link. Kudos to Joe Sharkey, the author, that was the most straightforward, unembellished account of an event like that I've ever read. I think my hands would still be shaking to the point that I couldn't type.
posted by HuronBob at 6:38 AM on October 3, 2006


What are the chances that a travel writer was on the plane that survived?

I used to fly frequently on Embraer jets when I lived in the midwest. I knew in the back of my mind they were a Brazillian company and always wanted to find out more about them.
posted by sohcahtoa at 6:53 AM on October 3, 2006


Permanent link from the NYT link generator.

Sharkey's account is chilling. I am not sure if being surrounded by aviation experts like he was would make it more or less scary.

I somehow doubt the midair collision theory and find the parts dropping off the 737 theory more plausible. In any event, nice link.
posted by caddis at 6:57 AM on October 3, 2006


No way there was a midair collision as described.
posted by dead_ at 7:02 AM on October 3, 2006


No way there was a midair collision as described.

Do you have a more plausible explanation? I thought the rationale made perfect sense - the 737 banked to avoid the smaller jet, took off the portions of the plane described, and went down. What else could have caused the damage?
posted by docpops at 7:04 AM on October 3, 2006


maybe they both hit some third object that we haven't found.

Occam's razor, dude. They were "at the same altitude, in the same airspace". (you did read the article, right?) To be hit by the same object (meteoroid from the air, or missile from the ground) they would have either:
Had to be at different altitudes (to be struck by the same object in the same part of the sky)
or
have been close enough to be struck by the same object (which would have to be awfully darn close). If they were that close, how can you think a third object is more likely than the probability that they touched?

Perhaps airliners striking the WTCs and burning didn't bring them down. Maybe it was some third object we haven't found yet.
posted by spock at 7:15 AM on October 3, 2006


No way there was a midair collision as described.

I think you have to be high or stupid not to believe the account, considering it's been confirmed by everyone, their mother, Jesus Christ, Vishnu, and the late Bob Hope. Keep smoking that crack.
posted by cellphone at 7:19 AM on October 3, 2006


spck: yeah, i read the article ... it was a good read. didn't see the quote you referenced there, though. did you read the article? or did you just make up quotes that sounded good to you?

and i didn't say it was more likely ... i just raised the question. don't assume my comment has anything to do with those bizarre 9/11 conspiracy theories.
posted by lester's sock puppet at 7:42 AM on October 3, 2006


Uh, I don't think parts just "fell off" that 737. It was a freshly-delivered plane. Like, less than a month old.
But the collision seems so David vs. Goliath. The little plane loses a bit of wing and lands safely, but the big jet takes a fatal hit.
posted by ninjew at 7:54 AM on October 3, 2006


Controllers in the two towers failed to discuss that the planes were entering the same airspace and instead assigned similar flight altitudes to their respective planes, leading to the collision, O Globo reported.

lol conspiracy.
posted by Uther Bentrazor at 7:58 AM on October 3, 2006


There's been much speculation on the local news, almost everything coming from unidentified sources.

The E600 departed from Sao Jose do Rio Preto (in Sao Paulo state) headed to Brasilia and Manaus, where it would be refuelled before heading to the US.

The 737-800 departed from Manaus heading to Brasilia and then Rio de Janeiro.

The following link (Portuguese only, sorry) has some unidentified source from the Brazilian Air Force stating that the E600 was on the wrong altitude.

http://noticias.terra.com.br/brasil/interna/0,,OI1171097-EI7792,00.html

Basically, the source says that when arriving at Brasilia's air space, the E600 should lower from 37,000 feet to 36,000 feet, in order to avoid the "air corridor" used by airline jets coming from Manaus (which was the one used by the 737), said corridor being at 37,000 feet.

If that's true, then I guess that it would be easy to prove it. Just comparing the flight plan (that must be recorded with ATC) and the data from the flight recorder. Of course, if there was an attempt from ATC to change the flight plan during the flight, than it cannot be the pilot's fault.

After that, there's a lot of unsubstanciated rumours (at least, for the moment) regarding why the colision course was not detected.

It's been said that the E600's TCA equipment was turned off by the pilot, in order to test the airplane (personally, I don't believe it).

It's been said that there's a communication "black hole" on the area and that it would be the reason why the pilot could not be warned.

It's been said that it could be ATC failure, with reports on the heavy work conditions that AT controllers are submited here (long shifts, intense work, military subordination increasing the level of stress, etc).

It's all rumours, for now.

The flight recording devices for the 737 were found yesterday, but it seems that the one that records the flight specs might me damaged. They will be sent to the US for analysis and later will be compared to the recordings from the E600.

I seriously doubt that the E600 pilots would be doing something stupid such as turning off safety equipments or changing the flight plan without consulting with the ATC. Specially to "test" the airplane.

First of all, there were guest passengers and executives from both Embraer and ExcelAir aboard. They were not alone.

Second, the region they were flying over is very dangerous: it's basically dense forest, with no signs of civilization nearby. This means that if anything went wrong with the "test", they would crash on the worst possible site (that was the case with the 737), with no help at all. I wouldn't pick a spot like this to perform tricks.

Third, there used to be a lot of "illegal flights" over that region, coming from the borders of Brazil (a lot of drug smugling, small airplanes from mineral digging sites, etc). Knowing this, I would stick to the flight plan.

Both airplanes were brand new (the 737 had 234 flight hours on it), Gol's reputation here is for a toughtfull company, with brand new airplanes, experienced pilots and good manteinance. Embraer's reputation is of an excel builder with a solid presence on the market (I'm not sure if the fact that the pilots were able to land the damaged plane is a statement of the quality of the airplane or not, I guess that there are lots of things to factor here).

I believe that there was some kind of ATC fault and maybe that there could be something wrong with the flying of the E600 (i.e., the pilot could have made a mistake, but I don't think that he would do it on purpose or by way of reckless behaviour).

Anyway, it's a sad affair. The rescue will take a lot of time, it's almost sure that not all of the bodies will be recovered and the identification might take up to 2 years in some cases, due to the state of the bodies.

My heart goes out to the families and friends (I have an acquaintance that lost a friend on the accident).
posted by rexgregbr at 8:04 AM on October 3, 2006 [2 favorites]


No way there was a midair collision as described.

Not if we can come up with a more tasty conspiracy theory!

CIA drug runners with surface to air missiles and itchy trigger fingers is mine.
posted by Artw at 8:06 AM on October 3, 2006


Uh, I don't think parts just "fell off" that 737. It was a freshly-delivered plane. Like, less than a month old.

I thought the Embraer was newly delivered. Where was it mentioned that the 737 was also a new plane? Maybe the jetliner started malfunctioning/breaking up and a piece of it hit the the smaller jet.
posted by reformedjerk at 8:07 AM on October 3, 2006


Wow.
posted by cavalier at 8:17 AM on October 3, 2006


reformedjerk: the company (Gol) stated that the 737 was received September, 12 and that it had 234 hours of flight.

I can't evaluate the amount of reports that you guys are getting, but it's big news down here and this information was widely reported on the first hours after the accident (when the plane was still listed as missing).

The official site for the communications from Gol is http://www.voegol.com.br/comunicado/

The content is in Portuguese. There's a menu that allows selecting English or Spanish, but I couldn't figure how it would work and I'm not sure that there's content in English available.
posted by rexgregbr at 8:23 AM on October 3, 2006


What a great article. The whole event seems strange but it's hard to believe in another scenario other than the one given. I guess we'll find out more in the weeks to come.
posted by ob at 8:23 AM on October 3, 2006


I thought the Embraer was newly delivered. Where was it mentioned that the 737 was also a new plane? Maybe the jetliner started malfunctioning/breaking up and a piece of it hit the the smaller jet.
posted by reformedjerk


More info on the 737.

"The 737-800 that crashed Friday in the heart of the Amazon jungle was delivered to the Brazilian airline GOL last month. GOL, a discount airline that has modeled itself after Southwest Airlines in the United States, is one of the fastest-growing airlines in Latin America. Like Southwest, it operates 737s."
posted by ninjew at 8:26 AM on October 3, 2006


Brings to mind an old Thin White Rope song:

The pulsing sun reverses the propellers
On the Cessna plane
It falls into the jungle and is swallowed by the rain

Week after week the monkeys watch the fallen sons decay
Trying to get thorough the windows to the
Chocolate on their face

Thin White Rope - Sack full of silver
posted by oh pollo! at 8:27 AM on October 3, 2006


spck: yeah, i read the article ... it was a good read. didn't see the quote you referenced there, though. did you read the article? or did you just make up quotes that sounded good to you?

Reading ≠ comprehending what was read. The quote I paraphrased is on page two and attributed to the military inspector:

"Both planes were, inexplicably, at the same altitude in the same space in the sky."

Regarding those who think it weird that the BIG plane crashed, I would imagine that the crash had more to do with the combination of the impact and the simultaneous hard bank that is postulated (and which probably saved the smaller plane from a direct impact with the larger plane).
posted by spock at 8:27 AM on October 3, 2006


As an aside, how often do NYtimes travel writers fly on corporate jets?
posted by Staggering Jack at 8:29 AM on October 3, 2006


Well, the bbc is saying it was a collision.
posted by ob at 8:30 AM on October 3, 2006


That's some very, very good writing there. Chilling, as someone said.

(Mind you, it made me immediately think about the Challengers Of The Unknown, especially the version of their origin presented in DC: The New Frontier.)

As an aside, how often do NYtimes travel writers fly on corporate jets?

The author mentions specifically that this wasn't for the NYTimes, but for Business Jet Traveller magazine.
posted by beaucoupkevin at 8:38 AM on October 3, 2006


As an aside, how often do NYtimes travel writers fly on corporate jets?

From the article, he was on a freelance gig for a magazine column and he had been a travel writer for the NY Times for 7 years which likely landed him the gig.
posted by cavalier at 8:38 AM on October 3, 2006


spock: thanks for your quote clarification. that guy was speculating. of course, he doesn't take into account the malfunctioning (?) anticollision devices. I wouldn't accept his accounting of events at this point.

but yeah, the scenario he described, other then the anticollision equipment failure, does sound plausable. but it seems to me that this accident would have required two such failures, not one.
posted by lester's sock puppet at 8:39 AM on October 3, 2006


Thanks cavalier - I read the article but missed that paragraph.
posted by Staggering Jack at 8:46 AM on October 3, 2006


Occam's razor, dude

I hate this use of Occam's razor. Once you have the data you can argue theoretical aesthetics all you like but you shouldn't be shaving off the beard of reality until you at least have reached the level of investigative puberty.

I think the prudent scientific philosophy to employ here is "Let's see the data before we decide"
posted by srboisvert at 8:52 AM on October 3, 2006 [1 favorite]


Would both anti-collision systems have to fail to cause a collision? I would think that if even one was functioning then a collision would be avoided. The odds of both failing or being turned off seem pretty remote. A mid-air collision seems unlikely in such circumstance.
posted by caddis at 9:09 AM on October 3, 2006


Though no one can say for certain yet how the accident occurred, three other Brazilian officers told me they had been informed that both planes were at the same altitude.

Yeah, I would imagine two colliding planes would be at the same altitude.

I don't disbelieve the account at all, but just as a general point: a lot of mechanical and electronic devices (and humans, for that matter) are more likely to fail when brand news as basic flaws become exposed; if they survive they can look forward to a relatively trouble free use-life given proper maintenance, followed by increasing numbers of problems as they age. So, to me,the fact both planes were brand new suggests it is more likely one of them had some kind of electronic or mechanical failure.
posted by Rumple at 9:13 AM on October 3, 2006


caddis writes "Would both anti-collision systems have to fail to cause a collision? "

TCAS is a two way system, if one plane was not equiped or it was malfunctioning then TCAS doesn't function.
posted by Mitheral at 9:22 AM on October 3, 2006


Acutally, one TCAS can issue some warnings, provided that the other plane has the proper transponder on board (and that's been required far longer than TCAS has.)

Yes. TCAS talks to the other box, to figure out where the planes are going, and most importantly, to make sure that a resolution advisory doesn't conflict. This is a "coordinated resolution advisory." If a solo TCAS told you to dive, and the other plane chose to dive, you lose. TCAS talks between to make sure that doesn't happen, and Coordinated RAs are, by most countries rules, to be followed instantly and exactly.

However, if the other plane has a Mode A, C or S transponder, your plane can still issue Traffic Advisories, and with the latter two, uncoordinated resolution advisories, but it won't be able to coordinated them. For those wondering, TA means "Pay attention, I'm thinking we're get too close", and RA is "Oh fuck, we are too close, DODGE!"

In the US, despite false alarms, pilots are instructed to treat TCAS coordinated resolution advisories as the second highest priority warning they have, and in case of conflict between ATC and TCAS, they are to follow TCAS. This has been true since the incident over Switzerland, where one craft followed TCAS, the other the controller. The reason for that is ATC doesn't know what TCAS is telling the pilots, whereas two TCAS II boxes do know.

For those of you wondering, GPOW (Ground Proximity Warning) is the only alert with a higher priority, and should you find yourself in hell with a GPOW climb order and a TCAS dive order, you climb, because you might miss the plane if you climb, but you won't miss the ground if you dive. You also utter "Oh, Shit!" -- that's nigh on traditional in such situations.
posted by eriko at 10:17 AM on October 3, 2006 [2 favorites]


They probably ought to put radar on those planes as well, just to be on the safe side.
posted by delmoi at 10:34 AM on October 3, 2006


Here's something I need some clarification on: wouldn't there be a tremendous amount of turbulence from jet wash as the two jets pass? How is it possible for two planes traveling at such high speed to collide and not both be blasted off course or lose altitude very quickly due to turbulence?

Also, how did the pilot of the corporate jet not see the other plane at all?
posted by spicynuts at 10:57 AM on October 3, 2006


If a pilot faced with a GPOW climb order and a TCAS dive order can manage to limit himself to a mere "Oh shit," then those guys are a hell of a lot cooler under pressure than me. I don't know what I'd say, but "Jesus holy fucking shit" would be merely a very small part of it.
posted by rusty at 11:00 AM on October 3, 2006


Ockham's razor isn't always a useful tool when you're investigating airline crashes. Usually the simplest explanation would have prevented the crash of the plane, and the only explanation for it going down is a really complex one indeed. In 2002, Bashkirian Airlines Flight 2937 and DHL Flight 611 collided over Germany. It turned out that:

1) one ATC was handling the whole airspace
2) the second ATC on duty was sleeping against regulation but with the permission of management
3) a ground-based collision warning system which would have alerted the controller to collisions had been switched off for maintenance, and
4) the one ATC didn't know it
5) the phones were down, preventing other controllers from calling him and warning him
6) there was a delayed flight on a second workstation that was taking the majority of his attention
7) when he finally figured it out, he told Flight 2937 to descend, and then
8) the collision warning systems on the planes activated, telling Flight 2937 to ascend and the DSL flight to descend
9) Both pilots decided to descend

Ockham was on vacation that night.
posted by Plutor at 11:05 AM on October 3, 2006 [1 favorite]


Also, how did the pilot of the corporate jet not see the other plane at all?

Because at a closing speed of 1500 feet per second he could look at his instruments, look out the other window, or even blink and miss it?
posted by weapons-grade pandemonium at 11:06 AM on October 3, 2006 [1 favorite]


That accident, in fact, is the accident that eriko mentioned as the reason that TCAS has been given higher priority than the ATC.
posted by Plutor at 11:09 AM on October 3, 2006


That's not the first mid-air miracle with an small jet.

http://www.acig.org/artman/uploads/botswana_bae.jpg

On 6 August 1988, the Botswana Air Wing British Aerospace 125-800, flown by Col. Albert Scheffers (CO Botswana Air Wing) and British Aerospace transport pilot Arthur J. Ricketts, carrying the president of Botswana, J. K. Quett Masire and eight other passengers, was underway at 35.000ft over Angola. All of a sudden, those on board heard a loud bang and there was an explosive decompression of the cabin. The aircraft yawed, rolled and dived almost at once and then a shower of fan blades from the starboard Garrett TFF.731-5 engine was observed as it separated away from the aircraft and went in forward direction. The aircraft was hit by one of two R-60/AA-8 Aphid missiles fired at it from an Angolan MiG-23ML interceptor. The first missile hit the starboard engine and ripped it together with the entire pod off the aircraft. The second missile then hit the same engine after it was already off the aircraft. Col. Scheffers regained control of the aircraft at 28.500ft, completed an emergency shutdown and thus preserved the other aircraft systems. 2.000lb of fuel in the right wing and one engine were lost, and there was extensive damage to the right wing and flaps. The aircraft spiraled down and then landed at Cuito Bie, in Angola. Subsequent inspection reveald that the hydraulic pump, starter/generator and alternator were still attached by hoses and cable looms. Angola apologized to Botswana for the missile firing, especially as usual air traffic control clearances had been received before the presidential trip. Nobody on board the aircraft suffered serious injuries, and the aircraft was subsequently repaired.
posted by centerpunch at 11:17 AM on October 3, 2006


Probably dumb question:
How do I show a photo in my post?
posted by centerpunch at 11:18 AM on October 3, 2006


Don't.
posted by dead_ at 11:24 AM on October 3, 2006


centerpunch: "Probably dumb question:
How do I show a photo in my post?"


You could probably find out by reading this. But it's a bad idea in general, believe me. You're better off just llinking to any images, rather than posting them.

posted by koeselitz at 11:24 AM on October 3, 2006


Yes -- the problem was also classic Soviet mentality. Soviet bloc air pilots were taught that ground control was to be followed absolutely. Western pilots were taught that they're ultimately responsible for their aircraft, and that if they felt that an ATC order would put them in danger, they were to ignore it.

Both pilots did exactly as they were trained to do. Murphy won.
posted by eriko at 11:25 AM on October 3, 2006


It kind of amazes me that mid-air collisions happen at all. When you think of how small airplanes are compared to how big the sky is, it just seems like something that shouldn't occur.

I realize that air traffic lanes are responsible for this, but it still seems odd.
posted by quin at 11:33 AM on October 3, 2006


eriko writes "Acutally, one TCAS can issue some warnings, provided that the other plane has the proper transponder on board (and that's been required far longer than TCAS has.)"

What I was getting at is TCAS isn't a radar or other similiar system it depends on the inputs from each plane. If one of the planes transponders is either not broadcasting or broadcasting incorrect information then TCAS won't work as planned. It's one of the identified failure modes. Plane A is flying at 24K. Plane B is flying at 25K but reporting it's altitude as 24K. TCAS thinks the planes are on a collision course and issues a dive command to B and a rise command to A thereby causing a collision of two planes that would otherwise have safely passed.
posted by Mitheral at 11:49 AM on October 3, 2006


Plutor: As srboisvert points out, Ockham's Razor should be applied in light of all available data, and not willy nilly. The best explanation for your example may indeed be complicated, but it is still the simplest one which fits the known facts.
posted by metaplectic at 12:35 PM on October 3, 2006


Regarding those who think it weird that the BIG plane crashed, I would imagine that the crash had more to do with the combination of the impact and the simultaneous hard bank that is postulated (and which probably saved the smaller plane from a direct impact with the larger plane).

That's what I was thinking too. If the 737 initiated a hard banking maneuver to avoid the 600 and still had contact with the aircraft, it could very easily have done enough damage to equipment that's crucial to re-stabilizing the aircraft, at which point a hard bank very nicely follows the laws of physics into a precipitous nose dive crash.

It seems likely that the life savers in this instance were the pilots of the 737.
posted by Brak at 12:53 PM on October 3, 2006


Because at a closing speed of 1500 feet per second he could look at his instruments, look out the other window, or even blink and miss it?

Actually we don't know the closing speed unless we know the headings of both planes at the time of the impact. Note that this may not have been the same as their overall flightpath, particularly if the 737 was trying to swerve. They could have snagged each other at an acute angle in which case the speed of impact could have been a small fraction of their airspeed. Which might in turn explain the survivable level of damage taken by the smaller craft.
posted by George_Spiggott at 1:28 PM on October 3, 2006


spicynuts asks: "...wouldn't there be a tremendous amount of turbulence from jet wash as the two jets pass?"

Sharkey writes: "... I felt a terrific jolt and heard a loud bang, followed by an eerie silence, save for the hum of the engines."

I'd imagine some very violent turbulence after impact if the planes were on opposing headings, and the Embraer flew into the 737 jetwash. But apparently there was silence and calm. So perhaps they were on offset headings. But then it's harder to imagine the 737 banking to evade -- a climb or dive seems more likely.

Or maybe my understanding of the physics is inadequate.

I hope they're able to piece together what really happened.
posted by Tubes at 2:32 PM on October 3, 2006


Might the fact that the 600 survived have something to do with the where contact was made on either plane? The 600 appears to have been hit on a small winglet on its tail. The winglet probably isn't too strong and because it is fairly close to the fuselage there isn't enough leverage to pull the entire tail off the airframe. The 737 was probably hit on the end of a wing. That means that the force would have been multiplied by an approximately 15m wing acting as a lever - my guess would be that this was enough to cause fatal damage to the wing itself or the airframe.
posted by r1ch at 2:58 PM on October 3, 2006


I hate this use of Occam's razor.

Well, considering that Occam's razor is conventionally translated as "Entities should not be multiplied unnecessarily" it seems singularly apposite to apply it to the question of a third object...
posted by Luddite at 3:22 PM on October 3, 2006


r1ch: The Embraer was hit in the winglet and the tail. Winglets are little fins that you can find on the wingtips of some airplanes to decrease drag. So there would have been a large lever there too (although winglets are lightly built, non-structural parts that can break off relatively easily).
Otherwise, where the Boeing was struck is indeed pretty crucial: for instance, if the impact had cut all three hydraulical circuits (which would be extremely unlucky, but not unheard of) the pilots would have been left without control of the aircraft.
posted by Skeptic at 3:28 PM on October 3, 2006


"It kind of amazes me that mid-air collisions happen at all. When you think of how small airplanes are compared to how big the sky is, it just seems like something that shouldn't occur."

Seems like a good reason to have planes fly at 37080 feet, 36150, 28410, etc... Why stick with just every thousand feet as options for cruising altitude? Doesn't that just unecessarily increase the chances of a mid-air collision?
posted by gummo at 4:17 PM on October 3, 2006


Speculation should not be multiplied unnecessarily.
posted by furtive at 4:17 PM on October 3, 2006


gummo, I was wondering the same thing. Do pilots stay EXACTLY at 37,000 feet? You'd think going up or down a couple hundred feet would make crashes a lot less likely.

And, as quin points out, there's a LOT of sky up there, and not much plane. You'd think this would never, ever happen.
posted by Malor at 6:58 PM on October 3, 2006


What are the chances that a travel writer was on the plane that survived?

clearly, the new york times knew about the collision ahead of time, and allowed it to happen.
posted by quonsar at 7:25 PM on October 3, 2006


More than 50 bodies were found today, in spite of the bad weather.

The task of finding the bodies and removing them is very hard, due to the difficulties presented by the site, which is basically a very dense forest.

The bodies are to be removed by helicopter from the crash site to a farm (which is the nearest outpost of civilization) located more than 200 km away. In the farm, the bodies are stored in a special truck and will be transported on it to the air base where the Legacy jet landed. From there, an air force carrier will take them to Brasilia.

The bodies of the pilot and the co-pilot have already been found.

Native-Brazilians are helping the search parties, by walking on the suposed course of the 737. The airforce believes that the bodies and the parts of the plane might be spread around a 10 km radius from the main crash site.

The air force also believes that the 737 might have desintegrated while still on air, since there are no clearings on the forest indicating some sort of forced landing.

The pilot of the Legacy jet stated that he was at 37000 ft and that the TCAS didn't issue any warnings on the collision. Both pilots from the Legacy jet will stay in Brazil pending the investigations, in order to help provide information.

Here is the main site for the accident coverage from a major internet portal from Brazil (sorry, Portuguese only):
http://noticias.terra.com.br/brasil/voo1907

On this site, you'll find links to pictures from the crash site and even a Google Earth link for the site (tough the images are not very good, you can understand the difficulties that the rescue team faces).

There are 75 men from the military working on the crash site. The first men that arrived on the site had to open a clearing on the forest in order to allow the helicopters to land.
posted by rexgregbr at 8:58 PM on October 3, 2006


That's not the first mid-air miracle with an small jet.

See also this photo of a wingless F-15 which successfully landed after a mid-air collision with an A-4N Skyhawk during a simulated dogfight (May 1, 1983).
posted by mazola at 9:54 PM on October 3, 2006


That's not the first mid-air miracle with an small jet.

See also this photo of a wingless F-15 which successfully landed after a mid-air collision with an A-4N Skyhawk during a simulated dogfight (May 1, 1983).
posted by mazola at 9:54 PM on October 3, 2006


It kind of amazes me that mid-air collisions happen at all. When you think of how small airplanes are compared to how big the sky is, it just seems like something that shouldn't occur.

There are tens of thousands of flights commercial every day over the United States alone, and many more semi-commercial or private flights. That mid-air collisions don't happen more is a testament to the effectiveness of modern air traffic control.
posted by randomstriker at 12:09 AM on October 4, 2006


Well, considering that Occam's razor is conventionally translated as "Entities should not be multiplied unnecessarily" it seems singularly apposite to apply it to the question of a third object...

And if the planes don't have colliding paths revealed by their flight data recorders? Do you apply the apposite Ockham's bandage to recover the too hastily cut out alternatives?
posted by srboisvert at 2:52 AM on October 4, 2006


Why stick with just every thousand feet as options for cruising altitude?

Because altimeters, historically, weren't that precise. Having them seperated by some distance meant that you could have some error in the altimeters.

Do pilots stay EXACTLY at 37,000 feet?

In the sense of the question you're really trying to ask? Yes, they try to keep as close to the assigned altitude as they can. One, it is a matter of saftey, two, a matter of pride. A competent pilot flys his clearance correctly.

In the literal sense, it is almost certainly not 37,000 mean sea level. The problem is most altimeters are pressure altimeters. However, changing air pressure and changing altitude both change the readout of the altimeter. What do you do? Depends on where you are.

On the ground, or flying low, you call the airport you are flying to (or, more likely, listen to ATIS) One of the bits of info you get is the current barometric pressure at the field. You then set that pressure into your altimeter, which, for that area, now reads feet over MSL correctly. When you're on the ground, you can do this without know the pressure -- you look at your chart, see the field elevation, then change the altimeter's pressure until the altimeter reads correctly. That gets you close, most runway charts will have exact elevation points at the runways, so before you takeoff, you can get it exact, and, as you fly, you are cleared to altitudes in feet over MSL. "AA47 clear to take offf runway 30R, climb to 3000."

Here, AA47 is being told to takeoff and fly to 3000 feet above mean sea level, flying along the runway, which is going to be about 2300 feet above the ground.

Problem. I take off from LA, you take off from Chicago. LA is basking under a high pressure area, Chicago is under a low. Our atimeters are correct only when we took off, if we keep flying like that, we might run into each other real problems.

So, you only fly feet MSL to 18,000 feet. After that, you change the altimeter presure setting to 29.92in/Hg, or 1013.2mbar. You are now no longer flying at 18,000MSL, you are now flying at Flight Level (FL) 180, and everything from there to near 60,000 feet is exrpressed as Flight Levels. So, AA47 is now talking to Gander Radio, the call sign of Gander Oceanic Control, and is cleared to follow Track V across the Atlantic. Since AA 47 has already flown nearly 1000 miles getting to the entry points, she's flying lighter than the east-coast planes, so they clear her up high -- and she's given clearance to cross on NATV (North Atlantic Track V) at FL390.

FL390 is not 39,000 feet MSL. It may not be even close to 39,000 . In hurricanes, the pressure and actual altitude have been off by over 3,000 feet. But she will be 1000 feet about a plane flying track V at FL380, and that's the point of flight levels. Up high, we all set our pressure altimeters exactly the same, thus, when we are close to each other, they read the same.

If you've flown a newer 767 or 777 across an ocean, you'll note that the fancy tracking screen says "37,000". It's reading off the pressure altimeter, and indeed, the plane is flying at 37,000 if the air pressure at the surface happens to be 29.92in/hg. Most likely, it isn't, and what you are really flying at is FL370.
posted by eriko at 5:49 AM on October 4, 2006 [4 favorites]


I am marvelling at the knowledge behind eriko. Thanks eriko!
posted by cavalier at 6:53 AM on October 4, 2006


U.S. pilots' passports seized as Brazil crash probed
posted by Krrrlson at 10:05 AM on October 4, 2006


eriko, thanks for the detailed information. I've just read an article on the Brazilian press that explained how the TCAS works and it was a simple version of your explanation. In fact, they even used the same example from this thread (the one in which one of the pilots followed the ATC and the other followed the TCAS).

If possible, could you enlighten us about the planning of the flight?

After the flight is planned, how is it registered? Is it recorded somewhere?

I understand that the flight plan might suffer changes during the flight itself (it seems plausible that any alteration on conditions that assure a safe flight should end up on a change of plans), but can the pilot change the flight plan without communicating with ATC?

One of the things that was stated in an article is that the flight plan for the Legacy was to fly at FL370 from Sao Paulo to Brasilia and then change to FL360 from Brasilia to Manaus (because the flights from Manaus to Brasilia would be at FL370). But the pilot stated that he was flying at 37000 ft as planned.
posted by rexgregbr at 4:33 PM on October 4, 2006


« Older Serious plans are now being formulated and carried...  |  Music for Funerals... Newer »


This thread has been archived and is closed to new comments