It is often pointed out that although people commonly imagine aircraft accidents as arising from a single big cause — engine failure, thunderstorm — they very often follow a series of small ones. Any small thing slightly out of order may, under sufficiently unlucky circumstances, become the cause of an accident. It is the inherently precarious nature of flying that makes this true. As pilots, we shut our eyes to what is obvious to uneasy passengers: If things go sufficiently wrong, we cannot pull over to the side of the road and telephone for help. There is no road; there is no shoulder. Things, including the human components, need to work properly.
The small problem that lights the fuse, so to speak, of an aircraft accident sequence is often something that is not dangerous in itself. The classic case is a door or canopy popping open after takeoff. It is in no way a life-threatening event in itself, and yet it has caused a surprising number of plane crashes. The unexpectedness, the noise, the unfamiliarity of the sensations — they throw the pilot off so badly that he forgets to fly.
But the initial small problem need not be noisy or dramatic. Here is an example from the summer of 2011 of a tiny error whose snowballing consequences led, hours later, to the deaths of two men and the destruction of their airplane.
A Cessna 182 left Meridian, Mississippi, at 4:21 p.m. CDT, bound for Columbus, Georgia, with two aboard. The 200 nm flight was to be VFR, no flight plan, at 5,500 feet. The pilot, 79, accepted the local controller’s proffer of flight following and a discrete squawk. He connected with the departure controller a couple of minutes after takeoff and received some vectors around areas of rain. Twenty minutes airborne, the pilot tried to get some weather information from the controller, who now told him his radio was unreadable and asked that he try a different transmitter. He did, and the quality of the transmission was better.
Some exchanges about the weather ahead followed, and the departure controller then gave the pilot a handoff frequency for Atlanta Center. The pilot read the frequency back incorrectly at first, but that confusion was straightened out, and the pilot left the Meridian departure frequency.
He did not, however, come up on Atlanta’s frequency. Controllers continued to track the Cessna, which went on squawking its flight-following code. They tried over and over to contact it on various frequencies, including 121.5, without apparent success. They repeatedly requested acknowledgement in the form of an ident squawk and tried to relay through another aircraft, all without result.
The Cessna 182 passed by Columbus, its announced destination, at 7 p.m. EDT and turned northeastward. The last attempt by controllers to contact the airplane, no more successful than the others, was made at 8:43 near Sumter, South Carolina.
At 9:05, the Cessna’s transponder switched from its discrete code to 1200. At 9:37, the airplane began descending from 5,500 feet. It made a sidestep maneuver to the east to line up with the Runway 5 localizer for Harnett Regional Jetport (HRJ) in Erwin, North Carolina, near Fayetteville. This was the pilot’s home base. Still in radio silence, the Cessna 182 flew through the upper tier of Pope Air Force Base Class C and proceeded toward HRJ at 1,200 feet.
The pilot-activated runway lights at HRJ were dark. A mile or so short of the runway, the Cessna turned right, flew for several miles, then reversed course, crossed the localizer and continued for several miles in the other direction before making another 180 and returning to cross the localizer again. During this bowtie-shaped maneuver, its altitude varied randomly between 1,200 and 2,100 feet and its groundspeed between 90 and 100 knots. Immediately after crossing the localizer southeastbound, it began a 270-degree right turn over dark, wooded terrain to rejoin the final approach course. Halfway around, it began to lose altitude and pick up speed. It intercepted the localizer all right — but at ground level, plowing through trees at high speed, shedding a trail of wings and struts before plunging into the Cape Fear River half a mile from the runway threshold.
During the long time the flight was without communication, it appeared to controllers that the Cessna 182 must have suffered a complete communications failure and could neither hear nor transmit. Though the transponder continued to operate, and therefore this was clearly not a case of a total electrical failure, it never squawked 7600, as would have been expected for lost comm. Nor did the pilot land at Columbus, though the customary lost-comm procedure, even under VFR, is either to land at the first opportunity or else to continue to where you said you were going. Often, however, pilots with minor problems like to get back to their home base rather than be stuck away during repairs.
National Transportation Safety Board investigators studying the wreckage found a likely explanation for the long radio silence. The transmit selector on the airplane’s King audio panel was set to PA — the passenger address system which, as the Cessna manual explains, is “unused” in the Cessna 182 — rather than to one of the two valid COM options. (The KMA 26’s COM3 option was presumably also unused.) The NTSB concluded that in the process of switching to Atlanta Center frequency the pilot somehow disabled his transmitters by inadvertently selecting PA. Arriving at his home field, he found himself unable to activate the runway lights, which required that the radio transmit multiple short blips on the common traffic advisory frequency. At this point, the instrument-rated pilot, who had been in the air for more than four hours, probably tried to position himself for landing using his panel-mounted GPS and the runway’s approach path indicator lights but lost control when he became disoriented during his steep turn over completely dark terrain.
It is not easy to understand how the mistake occurred in the first place. The pilot switched transmitters early in the flight at the request of the controller. If he was in the habit of using one transmitter rather than the other and had switched to the one less often used, then he might, from force of habit, have made the frequency change to Atlanta in the first radio, then accidentally switched to PA rather than back to the radio in use. He may have been predisposed to believe he had radio trouble by the controller’s complaint about the radio he started with. In any case, the NTSB’s theory that the position of the transmit selector led to the events of the flight seems generally plausible, especially in the absence of any other explanation.
The finding of probable cause — simply that the pilot became disoriented over dark terrain — is minimalist. The NTSB could also have mentioned the possible role of fatigue, and there was also a question of currency, or lack of it. In the year preceding the aircraft accident, the pilot had logged only 12 hours. No flights whatever had been recorded in the previous eight months. His last biennial had taken place five years earlier.
The pilot’s en route decision-making also deserved a comment. His choosing to continue past Columbus and return home was understandable. He may have failed to anticipate that he would be unable to turn on the runway lights at Harnett; after all, he could have had carrier but no modulation. But the decision to attempt a night landing on an unlighted runway was bold, to put it mildly. Turning inbound from a tight circle less than a mile from the runway wouldn’t make the trick any easier, and beginning the turn at 1,700 feet agl made it nearly acrobatic. Even without the disorientation, to align with a dark runway and then touch down under control, using only one’s landing lights, would be an achievement, especially for a pilot who flew very seldom, and almost never at night. A localizer was available — why did he not at least use it?
A single misstep, relatively innocuous in itself, led, through a series of minor decisions and errors, to a final, fateful choice. The moral of the story is that it is better to retreat from difficulty, if we can, rather than to plunge deeper into it. The trouble with improvisation in flying is that it may uncover our weaknesses.
This article is based on the NTSB’s report of the aircraft accident and is intended to bring the issues raised to our readers’ attention. It is not intended to judge or to reach any definitive conclusions about the ability or capacity of any person, living or dead, or any aircraft or accessory.