NTSB Factual Report

HISTORY OF FLIGHT

On August 27, 2015, about 0200 eastern daylight time, a Piper PA-44-180, N553MD, was destroyed by impact following a collision with trees and terrain during an instrument approach procedure to Houlton International Airport (HUL), Houlton, Maine. The airline transport pilot was fatally injured. The flight departed Goose Bay Airport (CYYR), Goose Bay, Labrador, Canada, about 2115. Instrument meteorological conditions prevailed, and an instrument flight rules flight plan was filed for the flight, which was conducted under the provisions of 14 Code of Federal Regulations Part 91.

The accident flight was the third leg of a ferry flight that originated from Keflavik International Airport (BIKF), Keflavik, Iceland, on August 26, 2015, about 0428. The airplane was scheduled to be delivered by the pilot to a customer in Merritt Island, Florida, on the day of the accident.

As the airplane approached HUL, the pilot contacted the Boston Air Route Traffic Control Center controller and requested vectors to AHOLA, which was the final approach fix (FAF) for the RNAV (GPS) RWY 5 approach at HUL. Over the next several minutes, the controller explained that the pilot could not proceed directly to the FAF, but must choose between two different initial approach fixes (WATAN or CIYEP) in order to establish the airplane on the approach.

Ultimately, the airplane was instructed to maintain 3,500 feet mean sea level (msl) until it reached CIYEP, and was cleared for the approach. The airplane descended to about 3,000 feet msl, crossed CIYEP, reversed course, and tracked approximately along the final approach course of 049 degrees. The intermediate minimum descent altitude prior to CIYEP was 3,000 feet msl, however; the airplane passed CIYEP about 2,500 feet msl. The airplane continued its descent on the inbound course, and the final radar target was at 1,500 feet msl about 2.5 nautical miles prior to the landing runway.

Data recovered from a Garmin GPSMAP 696 receiver on board the airplane depicted a descending, "S"-shaped track along the final approach course. A three-dimensional depiction of the airplane's ground track along the final approach course showed the last data point at 869 feet msl, 103 knots groundspeed, on the right side of the published approach course, and in the vicinity of the accident site.

United States Customs and Border Patrol agents awaiting the airplane's arrival heard the airplane approach. In a telephone interview, one agent said the engine noise was smooth and continuous, until the sounds of impact were heard. He said the runway lights were never illuminated, and that it was foggy, "misty," and that visibility was poor. According to the agent, "It was so foggy you couldn't see 200 feet. It was wicked low."

PILOT INFORMATION

According to Federal Aviation Administration (FAA) records, the pilot held an airline transport pilot certificate with ratings for airplane single-engine and multiengine land. His most recent FAA first-class medical certificate was issued on April 15, 2014. The pilot reported 1,460 total hours of flight experience on that date.

On an insurance application dated August 8, 2015, the pilot declared 3,000 total hours of flight experience; of which 1,400 hours were in multiengine airplanes. He reported 45 total hours of flight experience in the year prior to the application, and 15 hours in the 90 days previous to the application.

AIRCRAFT INFORMATION

The four-seat, twin-engine, low-wing, retractable-gear airplane was manufactured in 1980 and was equipped with two Lycoming 180-horsepower reciprocating engines, equipped with Hartzell controllable-pitch propellers. The airplane's maintenance records were not recovered. The hobbs meter indicated 4766.6 total aircraft hours at the accident site.

METEOROLOGICAL INFORMATION

The 0153 recorded weather at HUL included wind from 230 degrees at 3 knots, 10 statute miles visibility, and an overcast ceiling at 400 feet. The temperature was 18 degrees C, the dew point was 17 degrees C, and the altimeter setting was 29.89 inches of mercury.

At 0217, visibility had diminished to 1-3/4 miles in fog, and the ceiling had lowered to 300 feet overcast.

AIRPORT INFORMATION

HUL was a non-towered airport which operated under class G airspace. The field elevation for the airport was 490 feet msl and the traffic pattern altitude was 1,035 feet msl. The airport was equipped with one asphalt runway. Runway 5/23 was 5,015 feet in length and 100-feet wide.

The weather planning minimums for the RNAV (GPS) RWY 5 approach were ceiling 700 feet and 1-mile visibility, and the minimum descent altitude was 960 feet. According to the procedure notes section of the approach chart, "Procedure NA (not authorized) at night."

WRECKAGE INFORMATION

The accident site was located in a wooded, rural area about 570 feet elevation. The wreckage path was approximately 360 feet long and oriented on a magnetic heading of about 014 degrees. The airplane struck trees and terrain and broke into several pieces before it came to rest submerged in a pond. Pieces of angularly-cut wood consistent with propeller contact were found along the wreckage path.

Prior to the arrival of investigators, the wreckage was recovered by local law enforcement and first responders and moved to HUL. During recovery, 15 gallons of fuel was recovered from the airplane.

The wreckage was examined in a hangar on August 29, 2015. All major components were accounted for with the exception of the left flap. The fractured hinge points from both ends of the flap remained attached to their respective mounts, and the fracture surfaces displayed features consistent with overload.

Examination of the airframe revealed the engines were separated from their nacelles, but each three-bladed propeller remained attached to its respective engine. Four of the six blades were fractured from their hubs, and each displayed similar twisting, bending, leading edge gouging, and chordwise scratching. One blade tip was fractured and not recovered. The fracture surface was consistent with overload failure.

The nose and right main landing gear were separated. The right main landing gear and the left and nose gear actuators were in the "down" position.

The left wing was separated at its root, and the right wing was separated outboard of the engine nacelle. Each wing displayed concave dents oriented perpendicular to the spar. The radius of the dents was consistent with the size and shape of the fractured tree trunks at the accident site. Control cable continuity was established from the cockpit flight controls, through several breaks, to the flight control surfaces. The cable breaks displayed features consistent with overload failure or cutting by first responders.

Examination of the cockpit revealed both fuel selectors were in the "on" position, and the flap handle was in the 40-degree position.

The left engine displayed impact damage to the No. 1 intake pushrod tube. Impact-damaged accessories were removed, and the crankshaft was rotated by hand at the propeller. Continuity was established from the powertrain through the valvetrain to the accessory section. The dual magneto remained attached and contained pond water. Two towers sparked when the magneto was rotated by hand. The carburetor was attached and intact, and disassembly revealed no anomalies. The float bowl contained fuel.

The right engine displayed impact damage to the No. 4 intake and exhaust pushrod tubes. Impact-damaged accessories were removed and the crankshaft was rotated by the vacuum pump drive pad. Continuity was established from the accessory section through the valvetrain to the powertrain. The dual magneto was impact-damaged and could not be rotated. The carburetor was impact-damaged and separated from the engine.

MEDICAL AND PATHOLOGICAL INFORMATION

The Office of Chief Medical Examiner, Augusta, Maine, performed the autopsy on the pilot.

Toxicological testing performed on specimens from the pilot by the FAA Bioaeronautical Sciences Research Laboratory, Oklahoma City, Oklahoma, were negative for drugs and ethanol.

ADDITIONAL INFORMATION

Fatigue

The pilot departed BIKF and arrived in Greenland 5.3 hours later, where he remained on the ground for 1.7 hours before departing for Newfoundland. Required operational duties (fueling/flight planning/customs) would have precluded or limited any opportunity for restorative sleep during that time.

The pilot departed Greenland and arrived in Newfoundland 5.8 hours later, where he was on the ground for 3.8 hours. Such a length of time may have allowed for sleep outside of operational duties; however, the investigation could not determine the pilot's activity during this time.

The pilot's total duty time after departing BIKF until the accident occurred was 21.6 hours, during which the pilot accumulated 16.6 hours of flight time.

FAA Advisory Circular 120-100, "Basics of Aviation Fatigue," describes the fundamental concepts of human cognitive fatigue, the conditions that contribute to cognitive fatigue, and how individuals can reduce or mitigate its effects. According to the advisory circular, the main factors leading to fatigue include, time since awakening, cumulative sleep debt, and circadian rhythm of attention. The publication states:

Fatigue due to extended work hours, time of day, and shift work induces reductions in vigilance and reaction time and increases in risk of poor decisions, human error, incidents, and accidents. As described earlier, cognitive fatigue results from the interaction of sleep limitations and circadian drives for sleepiness with time-on-task and cumulative duty time effects. Human error due to fatigue is the result of sporadic losses of brain function and attention that increase in frequency with reductions in sleep and circadian drive for sleep. Lapses in attention are random in time, making it difficult to demonstrate the role of fatigue in specific accident cases. Self-reported, subjective sleepiness cannot be relied upon in this context, because it has a low correlation with actual performance impairment. One way to understand the role of fatigue in accident risk is to consider that fatigue causes random periods of inattention that occasionally coincide with operational conditions that demand attention to avert a serious event or incident. Hence, accident risk increases when either: 1) Lapses of attention increase in frequency due to sleep loss or circadian factors; or 2) Work demands increase the need for attention.

NTSB Probable Cause

The pilot's descent below the published minimum descent altitude in night instrument meteorological conditions without visual contact with the runway environment, which resulted in a collision with trees and terrain short of the runway. Contributing to the accident were the pilot's acute fatigue and his decision to attempt an instrument approach procedure that was not authorized at night.