On March 27th, 1977 a chain of events occurred that resulted in gross misunderstandings between an air traffic controller and the flight crew of a 747.  This miscommunication and misunderstanding ultimately resulted in the Captain of the 747 starting his take-off run when he was not cleared to do so because of traffic on the runway.  Subsequently, the 747 collided with another 747 that was taxing the opposite direction down the runway.  The resulting collision was the largest aviation accident ever, excluding terrorist acts.

Robert Ginnett, in chapter 3 of Cockpit Resource Management talks about groups in terms of norms, status, authority, roles, and boundaries.  One of the most difficult aspects of analyzing this accident, is understanding how the norms, roles, boundaries, and authority concepts of 1977 differed from those of the present day.  As a flight instructor and commercial pilot I’m fairly well versed in the established roles, authority, norms and boundaries as they apply to aviation in 2004, but how were these ideas different in 1977.  I believe, to truly understand what transpired to cause this accident to occur, one must put on virtual 1977 aviation glasses.

The Spanish and Dutch reports point the finger of responsibility at different roles.  It is my belief that neither report illustrates a clearly accurate cause for this accident, however; both reports make several good points.  I believe the causes of this accident are fivefold.  First, the structure of the communications between both aircraft and the tower were not adequate, and could even be called sloppy.  Second, the ATC procedures used by the tower were sloppy, and the controller perhaps did not have control over the situation.  Third, the captain of the KLM 747 did not go out of his way to research an inquiries by his crew.  Fourth, the PanAm and the controller should have retransmitted and requested a response to their blocked communications.  An finally fifth, I believe the KLM, PanAm, and the controller were all fatigued and suffering from a bad case of get-there-itis; or in the case of the controller, he probably wanted all these aircraft to depart so he could go back to running a quiet field.

The first cause, being the nature of the ATC communications was the single greatest cause of this accident.  By today’s standards, the vocabulary and structure of these transmissions could only be called sloppy and unprofessional, but what about in 1977?  This is why I earlier stated that to understand this accident one would need to look at it from the perspective of aviation in 1977.  Had the crews been using today’s terminology “position and hold” or “cleared for takeoff runway xx” or “negative – you are not cleared for takeoff” then there is little chance this accident would have occurred.  The use of the work “OK” by the tower controller played a huge role in the final seconds – had the controller instead responded “negative do not take-off – hold position” then the accident would have been avoided.  It’s hard to judge the conduct of 1977 aviation professionals using today’s standards, since certainly many of today’s ideas and constructs are based on this particular accident.

The second key issue is the traffic control situation.  Aviation then, just as it is now is a high-stakes game, lots of lives at stake; the controller did not appreciate this idea.  I personally believe having two 747s on the runway at once is a sloppy procedure, and the controller, especially since he could not see the aircraft should have held PanAm on the ramp until KLM was airborne.  There were several indications in the report about the controller possibly watching a ball game.  Whether the controller was watching it or not, this does illustrate the level of professionalism of the controller.

Element three, the KLM captain should have done a better job to research the possibility of the other aircraft on the runway.  The Spanish report paints the KLM captain as a rouge captain with his own agenda that the first and second officers were afraid of.  I very much believe this was not the case.  I do not believe those officers were afraid of him, I believe that rather all three in the cockpit on the KLM aircraft were not aware of what was going on, the FO and SO thought the other guy had it straight, and the FO and SO each didn’t want to be wrong.  They both raised the issue, but because they were unsure they didn’t raise it forcefully enough.  The issue for the FO and SO was actually their own uncertainty, and not fear of the captain.  I think the captain thought he was aware of what was going on.  Basically, all three had a piece of the picture, but none fully understood the situation.  To a certain extent the KLM captain did ignore the concerns of his crew, and should have researched the possibility of the other aircraft further.

The fourth element was the fact that the tower controller and the PanAm aircraft both transmitted a message to the KLM aircraft, neither got a response, and neither chose to retry the transmission.  Blocked communications are a fact of life in aviation, we shouldn’t blame the blocked transmission on the accident as both reports did, but rather on the people who should have transmitted a subsequent message asking for response.  The PanAm crew should have started a conversation with the controller stating “verify KLM is holding in position”, “were still on the runway.”  The controller should have called back to the KLM aircraft and said “verify you’re holding in position.”

The fifth and final element is a combination fatigue and desire to get out of Tenerife, not only for the KLM and PanAm crew, but also for the controller.  The controller probably wanted all these planes to leave so he could go back to running a quiet field.  The KLM and PanAm crews wanted to get to their destination and end the day.  This may have caused the controller to take shortcuts like having 2 planes on the runway at once.  This could have also contributed to the generally sloppy communications that occurred.

This accident, just like many others, illustrates that aviation is an unforgiving business.  It demands the highest level of professionalism and precision, not only in flying and technical skills, but also in communications between people.  Even the most remote possibilities of problems still need to be researched if the possibility would affect the safety of flight.
 
References
Wiener, Earl L. et al., (1993). Cockpit Resource Management.  New York, NY: Academic Press.

Google BlogSearch Communicating with Air Traffic Control

Several days ago a new student pilot posed the question: "Why does is the communications with ATC so structured, and is it necessary that we follow that structure?  Using proper terminology and read-back procedures increases safety and keeps everyone in the know as to your aircraft's intentions and position.

The structure of ATC communications as we know it today is the result of development and research prompted by various accidents caused by confusion resulting from informal communications.  These informal communications resulted in misunderstandings between aircraft and controllers and other aircraft.

One such accident that resulted in the development of today's ATC/aircraft communication structure occurred in Tenerife Spain in 1977.  On the 27th of March, 1977, two Boeing 747s collided on the runway on the island of Tenerife Spain.  One aircraft was engaged in a back-taxi, the second was on take-off roll.  The collision resulted in the largest aviation disaster ever excluding the acts of terrorism committed on September 11th, 2001.

I have written a narrative analysis of this accident and provided a link below.  I have also provided a link to the accident report.

So to answer my student's question - the formal communication structure is for safety, and yes it is required.

Most people commonly believe there are only two factors related to determining density altitude: field elevation and temperature.  The belief that these are the only factors stems from the age-old use of the “density altitude chart” that every pilot encounters in textbooks, on tests, and in aircraft performance manuals. 

In reality, humidity and air pressure also impact density altitude, however not to the extent of temperature and field elevation.  Combined however the impact of these two factors can be significant.

Today, many of the electronic flight calculators include functions for calculating the actual density altitude.  Additionally, there are several web sites with complete density altitude calculators such as the one created by Richard Shelquist that can be found at: 

http://wahiduddin.net/calc/calc_da.htm

Imagine a paved runway located at the top of Mount Elbert, Colorado’s highest peak, at 14,433’.  Would you or could you take off from that runway?  On a summer day, the density altitude at Leadville or Telluride Airport can exceed 14,000 feet, and it's not unheard of for the density altitude at airports like Eagle, Grandby, or Steamboat to exceed 10,000 feet.  The difference however is that unlike Mount Elbert that is surrounded by lower terrain and a treeless landscape on all sides, many Colorado’s mountain airports are surrounded by higher terrain on all sides as well as trees and other obstacles.

Flying in the Colorado Rockies in the summer is a very rewarding experience.  As with all General Aviation flying there are risks.  The key to mountain flying is learning the risks, and how to manage or avoid the risks.  Mountain CFI offers a 2 day course designed to do just this.

Our 2 day course starts with several hours of one on one ground instruction and review on the variables of mountain flight, high altitude aerodynamics, and weather.  We then practice techniques specific to mountain flying and specific to the pilots make/model of aircraft.  Towards the end of the first day we visit several mountain airports located in the vicinity of Eagle Airport.  Our second day is spent entirely in the air visiting a variety of mountain airports throughout the state that each present different problems and challenges for takeoff, approach and landing.

This course can be taught in the Pilot’s own aircraft, or rental aicraft that Mountain CFI will supply.  Additional days of training are available for pilots that want more practice.

This course (both ground and flight training) qualifies as completion of a full WINGS phase.  Participants will be provided with WINGS flight and ground signoff, flight review endorsement as well as certificate of completion for Mountain Flight Training.