Everything Else

I've just gone over most of what the NTSB has put up on the website regarding American Airlines Flight 587, which you may recall crashed into a residential neighborhood of New York about two months after 9/11, killing all 260 people on board and an additional 5 people on the ground. The long and short of what happened was this: the first officer, who was controlling the plane, performed control operations to attempt to contend with wake turbulence which caused the vertical stabilizer (the tail of the plane to most folks) to be torn completely off. The plane then became uncontrollable and subsequently crashed.

The NTSB report basically says all these people died from human error - their finding is as follows:

The first officers initial control wheel input in response to the second wake turbulence encounter was too aggressive, and his initial rudder pedal input response was unnecessary to control the airplane.

While this is certainly a contributing factor to this disaster, there's another part that the NTSB glosses over which I find distressing.

[More:]

It's useful to read the summary linked above before viewing the related videos, as you can see a pretty clear disconnect between the final cause of the accident, and the initiating factor: wake turbulence. If you read what pilots say, wake turbulence is a pretty nasty thing to encounter, particularly as it's at its worst during takeoff and landing. There are a couple of interesting quotes in that link (googled) - my favorite being "I've been through wake turbulence before, and I will tell you, it isn't fun. If you aren't buckled in, you WILL be injured..." as well as some very impressive pictures... I highly recommend reading the entire thread through.

Here's another snippet from that executive summary:

The local controller complied with Federal Aviation Administration wake turbulence spacing requirements when handling flight 587 and Japan Air Lines flight 47, which departed immediately before flight 587.

All well and good. However, take into account the two quotes from the Cockpit Voice Recorder (CVR), as given from the Flight Path Animation submitted with the NTSB report:

(09:13:35, right after clearance given for takeoff) F/O: you happy with that distance?
(09:13:39) CAPT: ahh, he's... we'll be all right once we get rollin'. he's supposed to be five miles by the time we're airborne, that's the idea.

Note, this is the first officer voicing a concern about distance while the plane is still safely on the ground. At 09:15:35, the control wheel indicator shows a series of sharp corrections, indicating an upset of the aircraft that the pilot is recovering from. Pay attention to the timestamps as we rejoin the transcript a short time later:

(09:15:44)little wake turbulence, huh?
(09:15:46) F/O: ..yeah

At 09:15:51, the pilot initiates a final series of very severe control inputs, which continue until the Cockpit Data Recorder stops, followed shortly after by the end of the CVR data.

Taken completely out of context, the NTSB finding is completely correct - the pilot essentially over-corrected for turbulence, ultimately causing the vertical stabilizer to exceed its structural capacity, and tear off the plane. Additionally, the design of the A300 would appear to be such that it is too easy to cause this sort of thing, an issue also duly noted by the NTSB. So why am I distressed?

The key problem lies in this sentence, previously quoted: "...complied with Federal Aviation Administration wake turbulence spacing requirements..." Simply put, the FAA has been under increasing pressure for the past decade or more to decrease congestion at the nation's airports. Since the construction of new airports cannot possibly ramp up to meet the increase in passenger and freight demand which really began to spike in the early '90s, the solution has been to try and figure out how to get more planes in the same airspace.

While better technology has the capability to optimize crowded airspace, particularly around airports, no computer program (nor phalanx of airline lobbyists papering the election funds of lawmakers) will ever be able to overcome the simple laws of physics. While the FAA is certainly aware of the problem, it would appear that it is putting more emphasis on "a potentially negative impact on the ATC system flexibility and capacity" than it is putting on safety.

To conclude, I think the NTSB was incorrect in that it too-narrowly defined the cause of the accident. While all of the factors it lists are in their essense correct, by ignoring the pressure put upon the FAA to loosen spacing requirements, and the subsequent caving to industry pressure in thes regard (look up the meaning of "RVSM airspace" if you don't believe this, the NTSB report missed a chance to directly address a core vulnerability in passenger flight safety.