“How did it get so late so soon?” This quote, credited to the late Theodor Geisel, aka Doctor Seuss, can be the lament of most accidents of old. As mentioned before, the true tragedy of an aircraft disaster is the lessons we do not learn, those we forget or … take too long to correct. Such is the lesson of many smaller accidents, where accident findings move down the list of problems to amend. In that case, it would be better to quote J.R.R. Tolkien when he wrote, “All we have to decide is what to do with the time that is given us.”
In aviation safety, we must do all we can in ‘the time that is given to us’.
Eastern Airlines Flight 66 was a Boeing B727-225 (-200) that impacted Runway 22-Left’s approach lights at JFK Airport on June 24, 1975. The airliner’s flight crew were more than qualified and properly trained, with each pilot having the necessary hours in the B727. The aircraft was sound and airworthy; there was no conceivable reason for the airliner to crash. However, the accident was a result of unfortunate circumstances in the form of windshear.
In those days we thought of windshear as a cruel tempest, illusive and misunderstood; an event that defied the logic of how air should move. However, windshear is not irrational at all. At the time of Eastern 66, it was next to impossible to study windshear because it was difficult to capture a windshear event with Seventies technology.
To understand the catastrophic effects of windshear, one point needs to be made clear: fixed-wing aircraft take off and land into the wind; coupled with the aircraft’s velocity, facing the wind adds to the lift that builds atop an aircraft’s wings. The loss of the rapid air moving from forward to aft on an aircraft’s wing equals a loss of lift, resulting in the aircraft surrendering to gravity.
According to the Federal Aviation Administration, ‘Windshear is a change in wind speed and/or direction over a short distance.’ It can be horizontal or vertical, happen at high or low altitude; it is associated ‘with strong temperature inversions or density gradients’, as are found in thunderstorms. There are also four types of original low-level windshear: Frontal Activity, Thunderstorms, Temperature Inversions and Surface Obstructions.
The science of windshear is more involved than can be related here; we’ll look at thunderstorm-caused low-level windshear at the fundamental level. At an airport, a down draft towards the ground results in horizontal winds moving out and away from the storm’s center. As the aircraft enters the storm cell’s perimeter, head winds increase, thus lift increases. This false head wind forces the aircraft above the glide slope (approach angle); the pilot reduces engine power to recapture the approach angle, while altitude continually decreases. When the aircraft recovers and dips below the approach angle, it is flying ‘lower and slower’; the wings are ‘dirty’, meaning the landing flaps and slats are extended; landing gear are down and locked; all inducing drag. Closer to the ground, the aircraft enters the storm cell’s inner envelope where dwells the vertical down draft winds. The pilot tries to power up, but the engines will take time to spool up to regain forward momentum. With a decreased head wind, increased drag and low speed, the aircraft drops from a loss of lift.
In the day when Eastern 66 crashed, the only warning to a windshear event was from a Pilot Report, known as PIREP, or a verbal reporting from other pilots. On June 24, 1975, Eastern Flight 902, an L-1011 aborted its landing shortly before Eastern 66’s approach; the L-1011 pilot responded to air traffic’s query, “Would you classify that (the event) as severe wind shift … correction, shear?” Eastern 902 responded, “Affirmative.”
Ahead of Eastern 902, Flying Tiger Flight 161 “highly recommend that you change the runway …” and was more adamant when questioned by the air traffic controller, “I don’t care what you’re indicating; I’m telling you that there’s such a windshear on the final on that runway, you should change it to the northwest.” Eastern 66 followed National Airlines Flight 1004 into JFK; National 1004 landed safely, but Eastern 66 was not so fortunate.
Eastern 66’s pilots were aware of the conditions from the radio communications; the First Officer, who was flying, anticipated any problems, even stating he would make the approach by keeping “a pretty healthy margin”. Whether Eastern 66 continued the approach in error or Air Traffic Control (ATC) faulted by not closing the runway, the accident was a consequence of slow and uncertain information, plus a base-level of training for pilots and ATC. In hindsight, what did we learn?
Between 1985 and 2000, after successfully tracking a tornado in 1973, Doppler radar joined improved weather radar networks to aid in the prediction of storm cells and their danger. However, more sensitive tracking of strong winds and precipitation from thunderstorms, plus the suddenness of their appearance, was still in development.
On August 2, 1985, Delta Airlines Flight 191 was on final approach into Dallas/Fort Worth Airport, when it experienced a windshear just north of the airport; it crashed within sight of the landing runway, 17-Left. The National Transportation Safety Board (NTSB) transcript stated that the Captain anticipated the problem thirty seconds before impact and told the First Officer, as the flying pilot, to “Push it up; push it way up”; this was followed by the sound of engines going to high rotations per minute (RPM). Less than ten seconds before impact, a ground proximity warning goes off (due to a quicker than normal descent rate) and the Captain orders a Go-Around.
During interviews, the radar specialist in ATC, relying on weather information from Meteorology, said the storm reached a level 4: Very Strong with severe turbulence that stayed stationary. Although the means to detect were more advanced, the reporting still relied on ‘ground truth’ reports: observances by people on the ground or in the air to the tracking reports. The technology was still not fool-proof, yet pilots and ATC still put trust in it.
The NTSB Findings included ‘a lack of guidance, procedures, and training for avoiding and escaping from low-altitude windshear; and the lack of real-time windshear information.’ The technology, though improved since Eastern 66, still relied heavily on human observations instead of technological accuracy. However, it’s the lack of training that stands out most from the time since Eastern 66. After the accident, studying Delta 191’s flight data recorder allowed each airline’s training department to make use of real-time information for flight simulators across the industry.
On July 2, 1994, US Air Flight 1016, a DC-9-31, crashed during a missed approach go-around at Charlotte/Douglas Airport. The NTSB determined that the flight crew continued with an approach into conditions that were conducive to a microburst; the flight crew’s recognition of the conditions and to act in time to avoid the accident. Also blamed was ATC’s lack of procedures to inform and display the weather threat.
An interesting footnote to this accident is the windshear alerting system in the aircraft failed to detect the threat. It did not fail as a detection device, but as it was programmed to detect. The conditions with which it senses versus the settings the aircraft was operating at were not lined up, e.g. the flaps were retracted out of landing configuration – the windshear alerting sensors ‘turned off’ as they were programmed to do. This equipment glitch, however, does not relieve the airline from providing the flight crew with the training needed to anticipate and avoid windshear.
On March 23, 2009, FedEx Flight 80, an MD-11F aircraft, crashed on the runway in Narita, Japan. Although the MD-11’s windshear sensing equipment did not give an alert to windshear conditions, a Boeing B747 landed eight minutes earlier reporting that it had encountered a windshear. The attitude of FedEx 80 was not a normal landing attitude (nose up); it then ‘bounced’ before touching down again, hard on the left gear. The left wing struck the ground and broke free, resulting in the aircraft flipping. NOTE: A bounce doesn’t guarantee the cause was windshear. However, Delta 191 bounced off State Road 114; its flight attitude was similar to FedEx 80’s, before the final impact.
The referral to windshear might be merely coincidental in FedEx 80’s situation, but another driving factor is not: Training, or the lack thereof. The Captain was on medical leave from October 2008 through February 2009; he received a successful proficiency training in February 2009. One thing he did not receive was ‘bounce’ training for over two and half years. As the accident unfolded, it is during the first touch down that he ‘bounces’ and the second touch down the wing breaks. Was this ‘bounce’ brought on by windshear? Possibly, but not provable.
The ‘bounce’ training was specific to the MD-11 and was being taught by FedEx after another FedEx MD-11, flipped in Newark/Liberty airport in 1997, an accident that was not a result of windshear. However, training is still the common denominator. Could enhanced training have saved any of these aircraft? In at least three of the four, the enhanced training should have made a difference in decision making, whether to enter or not. Proficiency training is not just a way to make pilots aware of situations, but to ensure they consistently refer to what they’ve learned to fly safely. Otherwise, it’s just checking a box.
Doctor Seuss is also quoted as saying, “They say I’m old fashioned and live in the past, but sometimes I think progress progresses too fast!” Progress will continue to progress too fast; humans will continue to fall behind the progress. The technology and realism of flight simulators is incredible; they can simulate, with impressive accuracy, many inflight emergency events. The correct employment of technology is necessary, it helps us to be safer. But we, the humans, have to make the final decision in any circumstance; we still can’t rely on imperfect technology to save us. When entering a situation, can we trust our training or should we trust the technology to pull us out? It’s leaning towards trusting the technology. I feel, if we can’t trust the training, maybe we should just go around.