de Havilland DHC-2 accident at Ketchikan, Alaska, 13 May 2019

Accident narrative

On May 13, 2019, about 1221 Alaska daylight time, a float-equipped de Havilland DHC-2 and a float-equipped de Havilland DHC-3 collided in midair about 8 miles northeast of Ketchikan, Alaska. The DHC-2 pilot and four passengers sustained fatal injuries, and the airplane was destroyed. The DHC-3 pilot sustained minor injuries, nine passengers sustained serious injuries, and one passenger sustained fatal injuries; the airplane sustained substantial damage. Both airplanes were operating on-demand sightseeing flights under visual flight rules.

### The flight

The DHC-2 was operated by Mountain Air Service LLC, and the DHC-3 was operated by Venture Travel, LLC, doing business as Taquan Air. Both airplanes had been conducting sightseeing flights to the Misty Fjords National Monument area and were converging on a scenic waterfall in the Mahoney Lakes area before returning to the Ketchikan Harbor Seaplane Base. Visual meteorological conditions prevailed.

The DHC-2 pilot, age 46, had about 11,000 hours of total aeronautical experience. The DHC-3 pilot, age 60, had about 25,000 hours of total flight experience.

The DHC-3 departed from an inlet about 1203. Automatic dependent surveillance-broadcast (ADS-B) tracking data showed that at 1217:15, the DHC-3 was about level at 4,000 feet mean sea level on a track of 225 degrees. At that time, the DHC-2 was 4.2 nautical miles south of the DHC-3, climbing through 2,800 feet on a track of 255 degrees. The DHC-3 pilot stated that about this time, he checked his traffic display and saw two groups of targets to the left of his path, but he did not observe the DHC-2 on his display before the collision.

About 1219, the DHC-3 started a descent from 4,000 feet, while the DHC-2 was climbing from 3,175 feet. Over the next 1 minute and 21 seconds, the DHC-3 continued to descend on a track between 224 and 237 degrees, and the DHC-2 leveled out at 3,350 feet on a track of about 255 degrees. Between 1220:21 and 1221:14, the DHC-3 made a shallow left turn, then a shallow right turn.

The airplanes collided at 1221:14 at an altitude of 3,350 feet. The DHC-3 experienced a brief upset in vertical load factor, entered a right bank, and pitched nose down. The pilot maintained control as the airplane descended, completed a 180-degree turn, and impacted George Inlet at 1222:15. The DHC-2 fractured into multiple pieces and impacted the water and terrain northeast of Mahoney Lake.

### What the investigation found

**Wreckage and impact** The DHC-2 right wing was in multiple pieces, with an initial impact point on the trailing edge of the right aileron and a series of mechanical cuts progressing inboard and forward into the wing structure. The DHC-3 engine and propeller remained attached to the engine mount, and all three propeller blades displayed leading edge chunking and gouging.

The Board found no evidence of any structural, engine, or system failures with either airplane before the collision. The Board also ruled out pilot qualification deficiencies and pilot impairment.

**Cockpit visibility** The National Transportation Safety Board conducted a cockpit visibility study to determine if the pilots could have seen each other. The study revealed that during the 3 minutes before the collision, the DHC-3 would have been obscured from the DHC-2 pilot by the DHC-2's cockpit structure, right wing, and the passenger in the right front seat. Therefore, the DHC-2 pilot would not have had the opportunity to see the DHC-3.

For the DHC-3 pilot, the DHC-2 would have appeared slightly ahead of the left window post as a relatively small, slow-moving object below the horizon against a complex background of terrain and water. The DHC-2's apparent size would have grown slowly, but it would have remained mostly obscured by the window post until 1 second before the collision. The Board concluded that the circumstances underscored the inherent limitations of the see-and-avoid collision avoidance concept, including how collision geometry and obscuration by aircraft structures make it difficult to see nearby aircraft.

**Avionics and traffic alerting** Both airplanes were equipped with cockpit displays of traffic information capable of displaying ADS-B data, but neither pilot received visual or aural alerts of the impending collision.

The DHC-2 was equipped with a FreeFlight Systems RANGR 978 transceiver and an iPad running the ForeFlight mobile application. ForeFlight can generate visual and aural traffic alerts if a target approaches within a specific distance and altitude. However, the DHC-3 was not transmitting pressure altitude data. Because the DHC-3 lacked altitude data, the ForeFlight application could not determine its relative altitude and treated it as distant traffic, precluding any alerts.

The DHC-3 was equipped with a RANGR 978 transceiver, a Garmin GSL 71 control panel, and Chelton electronic flight instrument system displays. The investigation found that the Garmin GSL 71 control head selector knob was in the OFF position. When this knob is off, the transceiver does not transmit pressure altitude or a transponder code. Recorded data showed the DHC-3 had not transmitted pressure altitude since April 29, 2019. The DHC-3 pilot stated he was not aware the unit was off and did not touch it during the flight. The Board noted that the checklist found in the DHC-3, which was created for another company, did not contain an item to ensure the GSL 71 was turned on before takeoff.

Furthermore, even if the DHC-2 had been broadcasting a collision threat, the DHC-3's Chelton display could only generate alerts if the transceiver sent "alert status" messages. The RANGR 978 transceiver installed in the DHC-3 was not designed to send alert status messages.

The Board noted that the DHC-3 had originally been equipped with a Garmin GDL 90 transceiver during the Federal Aviation Administration's Capstone Program. That original unit featured a conflict situational awareness function that provided visual and aural alerting. In 2015, the FAA upgraded the Capstone aircraft to the FreeFlight RANGR 978 transceiver to meet new ADS-B Out performance requirements. The new transceivers did not maintain the traffic alerting function provided by the originally installed equipment. The Board concluded that this reduction in capability resulted from the FAA's upgrade program.

**Safety Management Systems** Neither operator had a safety management system (SMS) in place, as it was not required for Part 135 operators. The Board concluded that if Taquan Air had been required to have an SMS, the safety risk management element would have provided better opportunities to discover and mitigate the increased risk posed by the changes to the Capstone-affiliated avionics and the potential for broadcasting incomplete data if the GSL 71 was turned off.

### Probable cause

The National Transportation Safety Board determined that the probable cause of this accident was the inherent limitations of the see-and-avoid concept, which prevented the two pilots from seeing the other airplane before the collision, and the absence of visual and aural alerts from both airplanes' traffic display systems, while operating in a geographic area with a high concentration of air tour activity.

Contributing to the accident were (1) the Federal Aviation Administration's provision of new transceivers that lacked alerting capability to Capstone Program operators without adequately mitigating the increased risk associated with the consequent loss of the previously available alerting capability and (2) the absence of a requirement for airborne traffic advisory systems with aural alerting among operators who carry passengers for hire.