As I type this, the Brazilians have finally released the two pilots involved in the mid-air collision of their Embraer Legacy jet with a GOL Airlines 737 on Sept. 29. They have spent more than two months under "house arrest," with their passports confiscated, preventing them from leaving the country, with no charges filed against them. I doubt many would consider that much of a vacation. I consider it outrageous.
Under the glare of intense publicity and worldwide attention, a Brazilian judge finally ordered their passports returned to them, and they have now left Brazil. Welcome home, guys! Take a couple of days off, and get back to work.
Within hours of the judge's order, "the authorities" filed charges, and forced an agreement requiring them to return for trial, if one is ever held. I'm a little surprised it took them so long, and that the pilots were allowed to leave. Possibly just a face-saving measure. "Face" is important to officialdom everywhere, and especially south of the border.
I am not even slightly surprised that two aircraft collided while under "control" of Brazilian ATC, but I am very surprised we don't see more such mid-airs.
I flew in Central and South America, including Brazil, in the late '50s, mostly cargo and ferry flights. In 1994, while working for JAL, I began flying three trips a month between Los Angeles and Sao Paulo until my "first retirement" in 2001. Not much had changed in the intervening four decades.
Communications are still horrible to non-existent. HF is still being used routinely, even when VHF is available. It is somewhat anachronistic to be flying near enough to Porto Velho to see the lights of the city, and still have to talk to them on HF. Call them on the VHF frequency and they may answer, but they will often ask to switch to HF for the position report, or for the next call. There is no question they prefer using HF, but I still don't understand why. As far as I know, all ATC services are provided by the military, and by rather low-paid and poorly trained personnel. The results of that are inevitable, and many times I've flown through an ATC sector without being able to raise anyone, HF or VHF. If someone does respond, it is sometimes obvious they've just awakened. There are several sectors (Porto Velho being one) where any transmission from the ground is overwhelmed by loud music in the same room as the mike, and it sounds like the controller is across the room, yelling in the general direction of the mike. Party time, I guess, or maybe just trying to stay awake.
Even when the radio works, all communications are in Portuguese, unless no one on the aircraft can speak it. Then English will be used, but it's very hard to understand. Of course, any transmission in English that is not absolutely standard and very common will not be understood on the ground at all, leading to "Say again?" or, worse, they will ignore further calls of any kind. The vast majority of flights over Brazil are flown by crews who do not speak either Portuguese or English as a native language, so it is the Tower of Babel all over again. It is dangerous, but heck, the same thing happens in France, Quebec, Russia and many other countries, too. We are very fortunate in the good-old United States, where we can push a button and talk to someone in English. Most of the time, anyway.
There is essentially no radar coverage in South America, except around large cities. Where there is radar, they don't use it en route, because aircraft will soon be out of coverage again, so they are forced to fall back on timed separation at all times, and the old-fashion position reports (which most American pilots have never done). To be fair, arrivals and departures are sometimes vectored in the terminal area at low altitude.
In seven years of my operating on that route, there were five incidents where other aircraft were definitely in "my airspace" by any standards. This is made worse by those countries who consider a national airline a matter of pride, and whose crews take short-cuts. In one of those, I watched a Lan Chile aircraft cross our track a mile or two ahead, at our altitude, close enough to identify the logo at night. Both Lan Chile and ATC denied it, for the aircraft was supposed to be crossing at a VOR about 60 miles behind us. They were giving phony position reports (in Spanish), and simply taking a big shortcut. I felt it prudent to climb a few hundred feet to avoid a huge bump from the wake. File a report, and it would never see the light of day. I did anyway, and never heard a thing.
In my opinion, it would be much safer to do away with ATC entirely in areas like this. In trying to "control" aircraft with the equipment they have, and the "skill" they demonstrate, they create danger. I'd feel much more comfortable going with random routes and altitudes and using TCAS for my own separation. Wonderful stuff, that TCAS.
In bygone days, navigation was a sloppy matter, with most aircraft being well off-course by modern standards. It was quite common and acceptable to arrive at the Oceanic Gateway after the crossing (usually where radar coverage begins) 10 nm or more off course, and I've been as much as 20 nm off after the long Pacific crossing via the "Mid-Pac" routes (no established tracks in those days). This was using the old Loran A, with an oscilloscope, and charting Lines of Position. In those days, it was very rare to see another airplane, even when on the same filed track.
As inertial systems came into common use (1970s), that tightened up a lot, and when the INSs were able to talk to each other and come up with better "average" positions, we started to see a lot more airplanes en route! Now, with inertial systems constantly updated by GPS, the accuracy is phenomenal, and everyone is dead-on, not even a wingspan apart, or so it seems to me. This has created a new danger of mid-air collisions, just exactly what we saw in Brazil. Before GPS, even if two airplanes were at the same altitude, and ostensibly on the same track, the odds of colliding were remote, simply due to inaccuracies in navigation. We didn't fully realize how good a thing that was!
With this in mind while flying the South American routes, I experimented with setting up waypoints with an offset of 3 nm. Didn't work. My fellow crewmembers were so uncomfortable with the idea of being "off-course," they just couldn't stand it. JAL has such rigid procedures that they thought they'd be fired if they were ever caught doing this. I also tried flying 300 to 500 feet above or below the assigned altitude. They didn't like that either, even with 2,000 feet between levels. I finally caved in, and just flew as assigned, hoping the TCAS would work.
Very recently, this problem has gained some attention, and now pilots have the "approved option" of flying 2 nm off-track (to the right) on the North Atlantic routes. That's where you'll find me. Pity it takes a formal government regulation to make the obvious "safe."
Much is now being made over the hemispheric altitude regulations, and whether the Legacy crew should have changed altitudes automatically on the leg where the collision occurred. Unfortunately, their route was very close to "North," and this causes a dilemma as to which altitude is appropriate. Mercifully, I never had this problem, as the route between Los Angeles and Sao Paulo is Southeast/Northwest.
But wait. The NTSB failed utterly to mention that the vast majority of radio calls to ATC go unanswered normally in South America. It often takes a dozen calls to get a response, and very often we'd go through an entire sector with no response at all. Is this radio failure, when I'm talking to other aircraft normally? How many unanswered calls to ATC does it take to define "radio failure"? Further, if we do manage to come up with some magic number, do we climb or descend to the "proper" altitude? What if the opposing traffic does the same thing? With the lousy communications down there, we would have airplanes climbing and descending all over the place, none under "control" of ATC!
There is a partial NTSB report available.
One paragraph jumps out at me:
Beginning at 4:48 pm, the crew of N600XL made a series of 12 radio calls to ATC attempting to make contact. At 4:53, the crew heard the call instructing them to change frequencies, but the pilot did not understand all of the digits, and requested a repeat. No reply from ATC was received. The pilot made 7 more attempts to establish contact.
I'd be willing to bet a chocolate-chip cookie the ATC tapes (if they exist at all) would show all those calls. Even if such tapes were made, I doubt they survive today.
This was a few minutes before the mid-air. This would not make me consider "Loss of Communications" procedures, or a change in hemispheric altitude. It would make me think the ground controllers were asleep, inept, or simply ignoring me, as usual. Of course it could also be old, antiquated equipment. In that part of the world, it is common for money allocated for upgrading such things to end up, shall we say, somewhere other than where it was intended.
The Legacy crew would have been further assured that they did not have a case of communications failure by hearing other aircraft on the frequency. It will be interesting to see if that's the case. Their best bet would have been to follow their latest clearance, which is apparently exactly what they did. That's what I would have done.
Another partial quote:
Primary (non-transponder) radar returns were received corresponding to the estimated position of N600XL until about 4:30 pm. For 2 minutes, no returns were received, then returns reappeared until 4:38 pm. After that time, radar returns were sporadic.
I'd bet another cookie the problem wasn't the transponders in either aircraft, except for the fact that the collision did occur, and both aircraft were supposedly equipped with TCAS. The head-on case is the most difficult for TCAS, and requires very, very prompt action to follow the cues, but it is doable. It's very hard to miss the TCAS roaring, "CLIMB, CLIMB, CLIMB" even if asleep. I wonder what happened there.
All over the world, it is common to ask for and receive "wrong-way" altitudes. ATC does not routinely change an aircraft's altitude because the course changes from 359 to 001, or 179 to 181. Perfectly legitimate and safe, provided ATC keeps it all straight, and coordinates with the sectors ahead. Pilots are not expected to make such altitude changes on their own because of a minor course change, and even in the case of complete communications failure (meaning they can't talk to anyone), an altitude change is fraught with danger. It is probably best to take a look at the overall route (departure to destination), and which hemisphere it falls in before considering a unilateral altitude change. Consider the case where one leg between VORs might be 10- or 20-nm long, and just that one leg calls for altitude change with loss of communications. Would you climb or descend that 2,000 feet, then descend or climb back? I wouldn't. If I had to change altitudes, I'd only change it 1,000 feet. Even before RVSM, that would assure separation.
Even in the U.S., the loss of communications rules are antiquated, and outdated. They were written in a time when aviation was moving from no communications at all to the sheer luxury of having just one HF transmitter and receiver, often set for just one frequency. Old-timers may remember the old ART-13B HF radio, with a multitude of dials, lock rings, and power needles that made a good radio operator a treasure. It was usually hooked to a "trailing antenna" with a lead weight on the end that had to be reeled out to a specific length according to frequency. Woe unto the crew who failed to reel it back in before landing! Oh, and morse code (CW) worked a lot better than voice at cutting through the static.
Later we moved to the ARC-1, a five-channel VHF transceiver. Pure luxury! No more frying bacon in the ears and clear communications, even if limited by line-of-sight.
Those old radios often failed, and loss of communications rules came into being. They haven't changed much, and as we moved to "dual everything" they became less and less important, much less used, and less known, too. Best now to just get it on the ground VFR if you can, and if you can't, go shoot an approach anywhere you want, as soon as you can. ATC will assume that's what you'll do, and will clear the way. A working transponder will make it all very easy and safe, and if your transponder is also out, there is always primary radar. If that fails, we're back to the "Big Sky Theory," which is really what we depend on in Brazil and other such countries.
I sure hope I don't get any more trips to that part of the world.
Be careful up there!
More from AVweb's Pelican is available here.