Making the choice

October 1, 1999

The media frenzy surrounding the John F. Kennedy Jr. accident in July was predictable even if the outcome of the accident investigation is not. The coverage weighed the pros and cons of light-aircraft transportation and focused on the uncertainty of VFR flight at night in marginal conditions. It will be months before the NTSB completes the full investigation. Without passing judgment on this accident, some general comments about night, weather-related decision making, and passenger perception are in order, even if the probable cause turns out to be something other than the pundits' guesses.

East Coast pilots contend with haze from May through September. Moist air bakes in the summer sun and mixes with natural and man-made pollution. Bermuda high-pressure systems are a frequent summer visitor. Nothing changes much from day to day, except that the haze gets thicker. Hazy conditions can last for two weeks or longer. A gruel of dust particles and pollution causes water droplets to condense more readily than they would in cleaner air. The Atlantic Ocean only increases the moisture supply. Dewpoint is another critical factor, as every private pilot knows. Or do they?

Why is all this weather mumbo-jumbo important? Because it can drastically diminish visibility and turn a simple flight into a fatal one.

Kennedy's accident in a Piper Saratoga occurred when the East was suffering from hot, hazy, and humid weather conditions. These common, apparently benign conditions are something that VFR pilots should respect. Weather may not be a contributing factor here, but we'll use this as an opportunity to discuss some cautions regarding nighttime VFR flight.

The 1998 Nall Report, the AOPA Air Safety Foundation's annual review of accident trends from the prior year, shows some very sobering statistics relative to accidents that occur in instrument conditions after dark. Loss of visual reference continues to be a leading cause of fatal accidents in general aviation. In 1997, more than 80 percent of fatal GA weather-related accidents were caused by visual flight into instrument meteorological conditions (IMC). The fatal accident rate is almost three times higher at night in IMC than during day VFR. In 1997, however, the night VFR fatal accident rate was about the same as the daytime rate. If the weather is decent VFR, then there should be relatively little trouble. The reported weather and the forecasts did not indicate that there should have been any problems during Kennedy's flight. A very important caveat here is that the weather is reality — not the forecast. What is noted at a reporting point is not what may be encountered en route, and pilots must be prepared to adjust their flights accordingly — sometimes making the decision in a matter of moments.

There have been some calls to eliminate nighttime VFR. One justification is that other countries allow only IFR flights at night. Despite the Kennedy tragedy, maintaining perspective is essential. On that Friday, the Saratoga was the only aircraft that was involved in an accident after dark — out of the hundreds or possibly thousands that were flying night VFR.

With all mishaps, multiple factors comprise the accident chain. The chain is a sequence of events leading up to an accident. Alter only one item or one decision, and the accident does not occur. Let's look at some of those factors. Kennedy was a relatively new private pilot with an estimated 300 hours. He was not instrument-rated, although there is speculation that he may have taken some instrument training.

The weather was reported as good VFR at the Marthas Vineyard Airport. The 9:53 p.m. METAR from the airport was wind 240 degrees at 10 knots, gusting to 15; visibility 10 miles; sky clear; temperature 24 degrees Celsius; dewpoint 18. The temperature-dewpoint spread was 6 degrees, outside the normal range for fog. However, it could easily have been foggy at the crash site. Boaters in the area reported poor visibility, and other pilots reported reduced in-flight visibility. Even with visibility of more than three miles, that may not have been enough to maintain visual orientation over the ocean. A moonless night, or one with high cloud cover to block the moon or stars, makes it extremely difficult to distinguish sky from ocean. Flight instruments become a necessity to stay oriented. In daylight hours, the odds would have improved significantly. If a localized fog bank formed, it would have been visible and more easily avoided in daylight.

The physiology of night flight contributes to its difficulty. Visual illusions under those prevailing conditions are well known to accident investigators. The flying public should understand them as well. According to the Aeronautical Information Manual, featureless terrain, sparse ground lights, and certain atmospheric conditions can lead pilots to think they are higher than they actually are. The absence of ground lights or haze makes objects appear more distant, so the natural inclination is to descend. That is why reference to instruments is essential. It sometimes happens that the aircraft descends into the ground or surface under control. That is controlled flight into terrain or CFIT. That does not appear to be the case in the Kennedy accident. The high rate of descent shown by radar indicates that the Saratoga was out of control well before impact.

Fatigue after a long day at the office can cloud judgment. Pilots miss cues that would otherwise trigger a prompt response. It leaves no signature for investigators to follow. There is nothing in the autopsy, in the physical wreckage, or on the radio/radar transcript to provide a clue that decision-making skills were below the pilot's normal standard. There is circumstantial evidence that human performance begins to deteriorate after a full day.

"Spatial disorientation," "vertigo," and "unable to tell which way is up" are all descriptions of what usually happens when VFR pilots get into instrument conditions. ASF continues to remind pilots in this column and in an ongoing series of free national seminars offered over the past five years that weather and decision making go hand-in-hand. Practice and wariness are the best defenses. Keeping options open is the only salvation. The only guarantee about weather is that it will change, and it may not be what is reported or forecast. Much of the Kennedy accident investigation will hinge on what the computerized briefing that someone, using Kennedy's subscriber log-in code, received over the Internet at 6:34 p.m. There were no airmets or sigmets advising of reduced visibility.

Training is part of the equation as well. Before 1974, no nighttime flight training was required. After that, three hours of night instruction, with at least 10 takeoffs and landings, was required to earn night privileges as a private pilot. In 1997, the requirement for a night cross-country flight of at least 100 nautical miles total distance was added, and ASF supported those changes. We also think that at some point during training, or just after a new pilot is certificated, he or she should be exposed to marginal weather with low ceilings or visibility. A hazy night is a good time to go up with an instructor and an IFR flight plan if necessary. Three miles at night, while still technically VFR, will make a believer out of most skeptics that this is not good cross-country flying weather.

New pilots must demonstrate competency in basic maneuvers by reference to instruments and recovery from unusual attitudes. ASF recommends that those skills be practiced during every flight review, if not more frequently. Transition to complex and high-performance aircraft should cover the above, along with a clear understanding on how to use the autopilot. We will never know why Kennedy apparently failed to use the autopilot or if it failed. When encountering unexpected weather conditions, the ability to select the heading mode and to make a 180-degree turn — along with altitude hold — can be a lifesaving procedure for VFR pilots. If you are about to check out in an aircraft with an autopilot, insist that the instructor show you how to use it. Fly the airplane in various modes to learn what the equipment can and cannot do. This is not to be used in place of an instrument rating, of course, but it can save your life. The rating is recommended if you plan to use the aircraft for cross-country flights on any sort of schedule.

There is always the possibility of an instrument or vacuum failure that compounds the problem significantly. At this writing, the investigation into the Kennedy accident has not uncovered anything to support or eliminate that possibility.

As pilots in command, we have to set the parameters for our passengers. Since the risk in marginal conditions goes up at night, we can improve safety by mandating a daytime arrival. Kennedy's sister-in-law reportedly could not get off work so the choices were to: leave her behind; hire a CFI to fly with Kennedy out to the islands and buy a return airline ticket; make the flight in the early morning light and hope there is no fog; or cancel the flight and make alternative arrangements. There are always choices and perhaps most important, there is nowhere you have to be. This is a hard decision. There is also commitment to passengers and pressure, either external or internal, to deliver them to the destination on time. The ego insists that other pilots are doing this, so I can do it too.

What do we say to passengers and friends who say GA is unsafe? The thoughtful answer is that under certain circumstances, it can be, but that is true of any activity involving technology, speed, and personal skill. The accident chain is a uniquely personal thing — the links vary from pilot to pilot, day to day, aircraft to aircraft, weather situation to weather situation. Do everything right and the odds of a successful trip go up tremendously — the huge numbers of successful flights prove that. Do it wrong and there is still no guarantee of a crash — we all know pilots who have been tempting fate for years. As in all of life, there is an element of timing and luck. The benefits of flight outweigh the negatives for most of us — if the risks are managed well. No form of transportation has ever eliminated them.

This accident is a mandate for safety organizations like ASF — and for pilots — to redouble their safety efforts. It is not a mandate to bash GA, although there are opportunists who are trying. The only positive outcome of this loss is a new awareness, and I think that John F. Kennedy Jr., the aviation enthusiast, would have wanted it that way.


See also the index of "Safety Pilot" articles, organized by subject. Bruce Landsberg is executive director of the AOPA Air Safety Foundation.