Although pilots do their best to minimize the risk they assume every time they switch on the magnetos, somewhere in the back of most pilots heads is a tiny voice urging them to watch out for the unexpected. Some may consider this self-doubt and banish it from their consciousness. Others embark on a near-manic attempt to silence it through endless training, hardware and conservative decision-making.
Though most pilots generally fall somewhere in the middle, there are those occasional flights where almost anyone may be tempted to shout down the doubt or defer the flight because something doesnt feel right. Usually those flights pass without incident. Occasionally they dont.
The voices may sing because the weathers less than ideal, or maybe the airplanes mechanical condition is suspect. The pilot may feel rusty on the gauges, stressed out, or otherwise not up to par.
Its part of the pilots job to listen to the voices and factor what they say into the go/no-go decision. Sometimes a pilot stays when he or she could have gone. Other times its just the opposite.
One October Friday afternoon a 59-year-old private pilot called Burlington Flight Service to get a briefing for a flight from Albany, N.Y., to Provincetown, Mass., two hours later. He told the briefer right off the bat that he wanted to go VFR, but could go IFR if conditions demanded.
Without hesitation, the briefer asked, Is low IFR going to be a problem?
Uh, I dont think so. It depends on how low.
Low IFR is ceilings less than 500, the briefer replied.
No, that wont be a problem, the pilot said confidently.
The briefer told of a low pressure system off the coast of New Jersey and a stationary front through coastal Cape Cod trailing south/southwest. The low pressure was forecast to move over Cape Cod that evening, and three other stationary fronts converged in the area.
An airmet for moderate icing had been issued at altitudes above the cruising altitude the pilot planned. Radar showed widespread rain showers, some of which were strong, over a wide area. The pilots intended destination reported ceilings of 200 scattered, with the temperature 18 and the dewpoint 17.
The pilot apparently understood the implication of fog, and asked for a reasonable alternate with a ceiling of 1,000 feet or more. After scrolling through a handful of airports that would not have suitable weather, they settled on Concord, N.H., and the pilot filed an IFR flight plan for the flight.
Two hours later, the pilot loaded his airplane with some items to take to his house on Cape Cod. Just as he had done almost every week for the previous eight months, his dog jumped aboard through the airplanes sliding canopy and curled up in the cargo area. He fired up the Grumman AA-5 Traveler and called for his clearance, then took off on runway 1. He climbed to 7,000 feet and set out on his planned 1 hour, 50 minute flight.
He shouldnt have been flying the airplane at all, according to the FAA. When the airplane was parked at Provincetown a month earlier, an FAA inspector spotted several flaws he considered serious enough to ground the airplane. The left aileron had a -inch hole on its trailing edge and the leading edge of the elevator had a hole about 2.5 inches by 1.5 inches in it. In addition, the landing light was not secured to the cowling; it rattled around inside the bottom of the cowling.
The inspector attached an Aircraft Condition Notice to the plane, grounding it until repairs were made or a Special Flight Permit was issued. The notice was followed by a certified letter to the pilot, who owned the airplane, but it was returned unclaimed. The FAA inspector reported that he talked to the previous owner of the airplane, who said the pilot had admitted to him that he had received the notice.
The pilots reason for ignoring the order to repair the airplane is uncertain, but he should have known better. He was, afterall, a lawyer.
At any rate, the pilot continued to fly the airplane, and as he flew into the gathering dusk he called Cape Approach with a request. He had tuned in the AWOS at Provincetown and it was reporting below minimums. Rather than diverting immediately to his planned alternate of Concord, he asked the controller for alternates on the Cape.
The controller advised Hyannis was available and told the pilot to get the ATIS information. The pilot responded that he already had it. Clearly the pilot was several steps ahead in the flight.
He told the controller he would make one approach into Provincetown and, if he missed, he would divert to Hyannis. The controller agreed, and the pilot began his overwater descent toward the ILS Runway 7 approach at Provincetown. When he was just over 17 miles from the airport, he was cleared for the approach. Seven and a half minutes later, he was told to change to the Unicom frequency and report his IFR cancellation as soon as possible because there was an airplane on the ground at Provincetown waiting to take off.
As the aircraft approached the ILS course, the pilot began his descent. Radar data shows the airplane was within the limits of the localizer at all times. Although he began descending before he intercepted the glideslope, his relative position to the glideslope remained constant for several miles. As the descent continued, however, something changed.
When he was at about 1,100 feet – 900 feet above minimums – the rate of descent increased dramatically and continued to increase until the plane leveled off about 100 feet off the water. At that point, he was also about 2.6 degrees to the left of the localizer. He corrected to the right, holding altitude, and then the radar return was lost.
The pilot of a commuter airline was waiting for the Grumman to land so he could get an IFR release. As the minutes ticked off, there was only silence at the airport.
Im halfway down the taxiway looking. Ill be back to you in a second. Weve been sitting here and havent seen anything come in, so unless he came in with his lights off and sneaking in he wasnt here. But Ill have a firm answer for you in just a second, said the pilot of the Cape Air flight.
Appreciate it, the controller said.
Sir, there is nothing on the ramp. Im going to go ahead and take a run down the runway, but I dont believe there is anything there either. Ill call you back with the runway sweep.
A few minutes later, the Cape Air pilot made the call that pretty much cinched it.
We went down to the end there and coming in the center its a tailwind approach and shutdown. There are no engines out there. Theres no smell of any kind. No nothing.
Five minutes later the Provincetown Unicom operator called approach to say the person at the airport to pick up the pilot hadnt heard from him, either.
The pilots dog, a 50-pound chow, washed up on the western shore of Cape Cod Bay two days later. Six days after the accident, the pilots body was recovered from the Atlantic side of Cape Cod. Fishermen snagged part of the airplane 17 days after the crash. Finally, 23 days after the crash, another fisherman recovered the main wreckage only a few miles from its last reported position. Then the mystery deepened.
When the wreckage was examined, neither navigation radio was tuned to a frequency that corresponded to the approach at Provincetown or any of the potential approaches into Hyannis. The ADF was one click from the Provincetown NDB, which would have been necessary to fly the missed approach.
Nav 1 was set on 114.05, while the localizer frequency was 111.1. The OBS was on about 072, which corresponded to the localizer course. Nav 2 was set on 115.6 with the OBS on 091. The Hyannis alternate had three instrument approaches, an ILS on 108.95, an ILS on 110.5 and a VOR on 114.5. Investigators determined there had been no internal slippage of the tuning controls. The frequency of 114.05 could not be correlated to any navaid in the area. Providence, R.I., matched the nav 2 frequency, but the OBS on 091 would not have made sense on the pilots planned or actual route.
After being immersed in salt water for a month, the wreckage stubbornly held onto its secrets. However, investigators were able to determine that the vacuum pump was intact and the aircraft had not run out of fuel, and there was no evidence of any kind of mechanical failure.
The pilots logbook reported a total time of 1,371 hours as PIC, with 262 hours of night flight and 253 hours of actual instrument time, along with 67 hours of simulated instrument time. He had passed an instrument proficiency check a few months earlier and had logged 36 total hours, 9.4 night hours and 6 hours in IMC in the previous 90 days.
The NTSBs final ruling blamed the accident on the pilots failure to follow the published instrument approach, with the dark night and low ceilings as contributing factors. While that may be true, there are too many unanswered questions to firmly conclude that this was simply a case of a poorly flown ILS.
The pilot had flown the route repeatedly in the recent past, perhaps giving him undue confidence in his ability to handle the weather. The fact that he ignored the order to ground the airplane says something of his willingness to fly with known deficiencies.
And what of the dog? Could something have spooked it at a critical time, leading to a loss of control at low altitude? The pilots wife said the dog routinely flew in the airplane without any kind of restraints. Then theres the matter of the mystery radio frequencies. The violence of the crash could have led to a knob or two getting jostled into another position, as in the ADF, but the VOR receivers werent even close.
Mysteries such as these feed that little voice that questions a pilots preparation. Poorly flown approach, maybe. But that fact that it could have been something much more mysterious demands pilots pay attention to the potential troubles that surround them.
-by Ken Ibold
