Ueberlingen Trial Continues

Pity the late Peter Nielsen. Being dead and all, he’s not available to defend himself, and so everybody from the press to his old employers at Skyguide have taken advantage of that fact to demonize him.

You’ll recall that Nielsen was the lone Skyguide air traffic controller on position when the Lake Constance/Ueberlingen/Bodensee midair took place on 1 July 2002, and he was brutally murdered outside his Zurich home by Vitali Kaloyev on 24 February 2004. Kaloyev, Nielsen’s killer, lost his wife, son, and daughter in the midair.

Here’s an example in a copyrighted Associated Press story from the International Herald Tribune, published 22 May 2007:

The controller gave the two aircraft only 44 seconds’ warning that they were getting too close to each other. He also mistakenly told the Russian plane to descend — sending the jetliner straight into the cargo jet.

Let’s dispense with the easy stuff first: at this point the Swiss court hasn’t blamed anyone. Therefore, the only official findings on this accident are in the Bundesstelle für Flugunfalluntersuchung report which lists these, and only these Causes:

The following immediate causes have been identified:

• The imminent separation infringement was not noticed by ATC in time. The instruction for the TU154M to descend was given at a time when the prescribed separation to the B757-200 could not be ensured anymore.
• The TU154M crew followed the ATC instruction to descend and continued to do so even  after TCAS advised them to climb. This manoeuvre was performed contrary to the generated TCAS RA.

The following systemic causes have been identified:

• The integration of ACAS/TCAS II into the system aviation was insufficient and did not correspond in all points with the system philosophy. The regulations concerning ACAS/TCAS published by ICAO and as a result the regulations of national aviation authorities, operational and procedural instructions of the TCAS manufacturer and the operators were not standardised, incomplete and partially contradictory.
• Management and quality assurance of the air navigation service company did not ensure that during the night all open workstations were continuously staffed by controllers.
• Management and quality assurance of the air navigation service company tolerated for years that during times of low traffic flow at night only one controller worked and the other one retired to rest.

So, back to what the Associated Press story seems to be implying are the causes of the accident:

1. The controller gave the two aircraft only 44 seconds’ warning that they were getting too close to each other.

Yeah. Ok. The report says that Nielsen noticed the conflict too late and thus, his resolution came too late. But,

2. He also mistakenly told the Russian plane to descend — sending the jetliner straight into the cargo jet.

Oh please.

The article is implying that, because the Russian crew was getting a TCAS climb RA from its onboard equipment, an instruction from the controller to descend was a mistake on that controller’s part. This idea is utter nonsense. Here’s why.

TCAS-equipped aircraft coordinate between themselves to negotiate a solution to a potential collision and Nielsen had no way of knowing that this was happening. He saw targets merging and made an effort to separate them. Again, he had no way of knowing that the TCAS computers aboard the two aircraft had already come up with their own solution and that at least one of the aircraft was complying with TCAS RA commands. That’s why pilots and controllers have come to an agreement on how this sort of thing needs to be handled. In short, when a TCAS RA orders a pilot to take a particular action and the controller orders a different action, the pilot must always follow the TCAS. The reason for this is simple: as I noted above, the controller has no way to know that the pilots are following at TCAS RA until one or both of the pilots mention it. Of course, technological advances are beginning to make that information available to controllers, but that night in July 2002, Nielson had no way to know that a TCAS RA had been generated or whether either crew were following that RA.

Germany’s Aeronautical Information Publication (ENR 1.8.22 in this case) addresses TCAS RAs and Paragraph 2.2.2, Resolution Advisories, says,

a) All Resolution Advisories (corrective or preventive) should be followed unless the pilot can visually identify the conflicting traffic and decide that no deviation from the current flight path is needed. When subsequently the Resolution Advisory changes the pilot should again respond promptly in compliance with the indications. Failure to comply with a Resolution Advisory can result in a less desire vertical miss distance at the closest point of approach.
c) If a decision is made not to follow a Resolution advisory (corrective or preventive) a manoeuvre should never be made in a direction opposite to the one indicated by the Resolution Advisory. This is particularly important as the system may coordinate, unknown to the pilot, with another equipped aircraft.
d) Recovery manoeuvres to resume to assigned ATC clearance should be initiated immediately after the system announces “clear of conflict.”

Recall that the AP story says that Nielsen “…mistakenly told the Russian plane to descend — sending the jetliner straight into the cargo jet.” This is nonsensical. Sure, Nielsen was late, but his control instruction was exactly right. Not knowing (and not being expected to know) of the TCAS RA transpiring between these two aircraft, he picked one aircraft and issued a control instruction. While I don’t know why he selected the TU instead of the Boeing, it wouldn’t be unreasonable to start the TU down in that he was scheduled to be descended anyway. Moreover, when we make these sorts of instantaneous decisions, all other things being equal (for instance, targets have already merged), descending the aircraft who was planning for lower in the first place might just be the most expeditious thing to do.

Nope. Nielsen didn’t make a mistake. Given the proximity of the targets he used all the info he had to sort it out. The mistake came when the TU crew followed Nielsen’s control instruction rather than the TCAS RA. Read that German AIP excerpt above again. If you’re not going to follow the TCAS RA, whatever you do, don’t do something opposite to the TCAS RA. Again, pilots and controllers have come to a formal agreement on this point: If I (as a controller) give you a control instruction that conflicts with a TCAS RA, you are to follow the TCAS RA every time. If I don’t know you’re following the TCAS RA I may keep jabbering, giving you control instructions, but just ignore me. Follow the RA.

So where does the AP get the idea that Nielsen made a mistake? Doesn’t that word mean something completely different? Is it just Nielsen-bashing?

Given all of this, the paucity of information coming out of the Skyguide trial is especially irritating. Here’s part of a copyrighted report from the Russian RIA Novosti  dated 22 May 2007:

At Tuesday’s court hearing, [Zurich Prosecutor Bernhard] Hecht also dismissed suggestions that errors by the Russian crew could have been a factor in the crash.

An independent Austrian expert who gave evidence to the Buelach court earlier this week also said the Tu-154 pilots had been correct to follow the human controller’s instructions rather than those given by the on-board Traffic Collision Avoidance System (TCAS).

Say again? That’s nuts.

But among all of the trial reports the real kicker (and the reason for this rant in the first place) can be found in the same RIA Novosti article:

The eight Skyguide personnel who are on trial in the case denied any personal involvement, saying Nielsen, slain in 2004 by a Russian who lost his family in the accident, was the only one to blame.

Yeah. That was easy. Stay tuned.

-Dave