In the end, no harm done. But certainly an example of an air traffic controller having a bad day. Note: at 32 seconds the controller assigns the Challenger a heading of 180.
Air traffic controllers fall asleep on the job. At least they do occasionally. That came as big news in 2011, when two airliners landed at Washington’s Reagan National Airport without ATC help because the lone controller was snoozing. No injuries there, but in 2006 a Comair regional jet crashed while taking off in Kentucky, killing 49 of the 50 people on board. The air traffic controller who cleared the plane for takeoff didn’t notice the plane was taking off from the wrong runway. He had slept only two hours in the previous 24.
The question then was whether the problem of sleeping or sleep-deprived controllers was an isolated one or instead a significant, pervasive risk to aviation safety.
The FAA paid NASA $1.2 million to find out. The NASA study's findings: the schedules controllers work lead to chronic fatigue and pressure to fall asleep. In fact, a third of the controllers in the study reported that fatigue was a “high” or “extreme” safety risk. More alarming was that 6 out of 10 controllers report that they had fallen asleep or experienced a lapse in attention while driving to or from a midnight shift. If sleepiness is making controllers unsafe to drive, it’s certainly making them unsafe to work.
The study’s conclusion was clear:
Chronic fatigue may be considered to pose a significant risk to controller alertness, and hence to the safety of the ATC (air traffic control) system.
Perhaps most interesting is that, according to an Associated Press report, the FAA has kept the FAA’s report secret for three years. Though completed in 2012, the study was released just this week. No explanation from the FAA as to why it has kept the study from the public for all this time.
There's not a lot of air traffic at night. So some air traffic control towers close altogether. Any landing aircraft is on its own. Other air traffic control towers are staffed with just one controller. Not surprisingly, lone controllers working the night shift tend to doze off.
That little secret is now out. That led to the resignation of the head of the Air Traffic Organization. And then, just yesterday, the FAA announced that a second controller will be added to the overnight shift at 27 airports.
Sounds like moves in the right direction. But what do you get when you put a second controller into a dark, quiet control tower in the middle of the night?
Two sleeping air traffic controllers.
It's not a matter of just adding staff. It's a matter of dealing with the somewhat complicated issue of how night shifts disrupt a workers' circadian rhythms. At least so says Dr. Mark R. Rosekind, the newest member of the the National Transportation Safety Board.
I had the pleasure of hearing Dr. Rosekind speak a couple of weeks ago at Menlo High School in Atherton, California. Dr. Rosekind is one smart guy. And he happens to be a sleep expert. In fact, Dr. Rosekind was the Director of the Center for Human Sleep Research at Stanford University. So he knows a thing or two about "fatigue management."
Unfortunately, the FAA isn't required to listen to the NTSB, and frequently doesn't. In the past, when it comes to fatigue risk management, an act of congress was required to get the FAA to do something.
Not to worry. This time the FAA, or at least the US Transportation Secretary, Ray LaHood, is on top of it. He is outraged. LaHood says he "will not sleep" until there's good safety in the control towers. (Yes, he really said that.)
NTSB Chairman Deborah Hersman's recent testimony before congress concerning the mid-air collision over the Hudson raises more questions than it answers. She stated that the Teterboro controller instructed the Piper pilot to switch to frequency 127.85 to contact the Newark controller. But before leaving the Teterboro frequency, according to Hersman, the pilot read back to the controller "127.87," which was wrong. Thereafter, the pilot was in contact with neither Teterboro nor Newark, and so neither facility could warn him of the impending collision. Hersman's remarks are here.
Hersman's implication is that the Teterboro controller failed to correct the pilot, and so the controller contributed to the pilot's getting "lost in the hertz" (out of radio contact) at a crucial moment. However, the animation that the NTSB released on the same day that Hersman testified does not appear to back Hersman up. It just doesn't sound as though the pilot read back "127.87" as Hersman states. You can listen to the audio yourself beginning at minute 2:25.
The NTSB has now given us further reason to question whether it deserves the confidence we place in it. On Friday, the NTSB came out with a block-buster press release condemning the Teterboro air traffic controller who had cleared the Piper airplane for takeoff. According to the NTSB's report, the Teterboro controller could see on his radar screen that the Piper pilot was on a possible collision course with the Liberty Tours helicopter. In fact, according to the NTSB, the controller could see the conflict before the Piper pilot switched off from the Teterboro controller’s frequency. Yet, according to the NTSB, the controller failed to warn the Piper pilot.
At 1152:20 the Teterboro controller instructed the pilot to contact Newark on a frequency of 127.85. . . At that time there were several aircraft detected by radar in the area immediately ahead of the airplane, including the accident helicopter, all of which were potential traffic conflicts for the airplane. The Teterboro tower controller, who was engaged in a phone call at the time, did not advise the pilot of the potential traffic conflicts.
That was wrong. True, the controller was on the phone when he should not have been. But the helicopter did not appear on the controller’s radar screen until after the Piper pilot was supposed to have switched to a new frequency. Of course, by then it was too late for the controller to advise the pilot of anything. In other words, it appears that there was nothing the controller could have done -- whether he was on the phone or not.
Over the weekend, the air traffic controllers’ union privately asked the NTSB to correct its error. The NTSB refused. So today the union issued its own press release setting the record straight. The press release claims that the NTSB's account, which implies that the controller should have prevented the accident, is "outright false" and "misleading." Worse, it charges that the NTSB knows it, but refuses to correct its error.
This afternoon, after the controllers' union went to the press, the NTSB finally conceded that it was, in fact, wrong. It thus issued a new press release, explaining that the controller could not have seen the helicopter after all.
The accident helicopter was not visible on the Teterboro controller's radar scope at 1152:20 [when the controller instructed the Piper to change frequencies]; it did appear on radar 7 seconds later - at approximately 400 feet.
The NTSB offered no apology for its error. Nor did it offer an explanation. Rather, despite that the union was right, and the NTSB was wrong, the NTSB’s only reaction was to kick the union off the investigation.
The NTSB’s blunder was a whopper. It laid blame for the accident where it does not appear to belong. The NTSB's only interest is supposed to be in getting the facts right. If that’s so, why did it not correct its error when the union asked it to? Why did it require the union to force the issue?
Compared to pilots in other countries, pilots in the US have extraordinary freedom. Of course, to keep commercial airliners safe from collisions, pilots of small aircraft are excluded from certain airspace near major airports unless they have first obtained a clearance from air traffic controllers. If a pilot obtains the necessary clearance, controllers will dictate the pilot's path and use radar to monitor the pilot's every move.
But that still leaves many places where pilots are permitted to fly without being supervised or controlled in any way. One such area, appropriately enough, is near the Statue of Liberty. As long as the pilot stays below 1100 feet -- outside the airspace used by airliners -- the pilot doesn't need a clearance, doesn't need to have filed a flight plan, and doesn't need to communicate with any tower or other air traffic control facility. The pilot is totally on his own.
Many non-pilots are surprised to learn that the method used to prevent collisions in such uncontrolled areas is called "see and avoid." The pilot is supposed to look out his window, "see" the other aircraft, and "avoid" them. Pilots talk about having to "keep their head on a swivel" when flying in uncontrolled airspace. Though this method of collision avoidance may sound primitive, over the years it has worked well.
There is one problem. Helicopters and airplanes don't mix well in a "see and avoid" environment. Helicopters fly slower than airplanes. And because they have a small cross section, they are hard to spot -- especially when viewed from directly behind. That puts them at risk of being rear-ended. It doesn't help matters that helicopters tend to manuever in a fashion that most airplane pilots find to be unpredictable.
Because of all that, helicopter pilots are supposed to "avoid the flow" of airplane traffic. In other words, as best they can, they are supposed to stay out of the way. Unfortunately, when both a helicopter and airplane are headed to the same spot, or are both looking at the same feature on the ground, that can be difficult to do.
We don't know what factors combined to result in the midair over the Hudson. But the NTSB has long recognized that when it comes to uncontrolled airspace, helicopters -- especially tour helicopters -- don't mix well with airplanes.
Air traffic controllers work within the guidelines set forth in the Controller's Handbook (pdf), which they often call "the Bible." The Handbook is hundreds of pages long, and controllers must follow it to the letter. If they deviate and an accident results, the Federal Tort Claims Act permits the victim to sue the FAA for negligence.
Sometimes, the Handbook doesn't cover a particular air traffic situation. In those cases, the controller is supposed to simply use his best judgment. But this would seem to present a problem for the victim of the controller's error. That's because one of the Federal Tort Claims Act's most important limitations is the "Discretionary Function Exception." The Discretionary Function Exception states that a victim can’t sue the federal government for bad decisions that the government left to the federal employee's best judgment. Regardless of how careless the employee was, the government is immune from suit.
Does that mean that, if a controller makes an error in a situation not covered by the Controller's Handbook, the victim can't sue?
No. Courts have ruled that an air traffic control error never falls within the Discretionary Function Exception. It doesn't matter whether the air traffic situation was covered in the Handbook, or was one left to the controller's judgment. If a controller's error caused the accident, the victim can sue the FAA for negligence, just as though the FAA were a private party.
However, certain other rules will apply to the victim's lawsuit:
- Before starting the suit, the victim must file a claim against the government on a Form 95:
- The lawsuit must be filed in Federal Court, not State Court;
- The judge -- not a jury -- decides the case;
- No punitive damages can be awarded;
- The victim's attorney can charge a contingency fee of no more than 25% of any judgment that the court renders;
- If the FAA settles out of court, the attorney can charge a contingency fee of no more than 20%.