National Transportation Safety Board
Aviation Accident Final Report




Tyonek, AK

Accident Number:


Date & Time:

08/23/2017, 2245 AKD





Aircraft Damage:


Defining Event:

Loss of control in flight


1 Fatal

Flight Conducted Under:

Part 91: General Aviation - Personal



The noninstrument-rated private pilot had spent the previous 4 days and nights conducting a solo sheep hunting trip and was returning home when the accident occurred. Before departing the remote, mountain airstrip, the pilot contacted a Federal Aviation Administration (FAA) Flight Service Station (FSS) on his satellite phone for a weather briefing and asked whether visual flight rules (VFR) flight was recommended for his route that night.

The flight service specialist (FS-S) provided the pilot with the terminal area forecast for the destination airport; information from the current Area Forecast, which indicated cloud ceilings around 5,000 ft mean sea level (msl), marginal VFR conditions, and rain; and a pilot report from a nearby mountain pass, which advised that VFR was not recommended. The pilot asked several times about a specific mountain pass, and although the FS-S described the conditions at that pass based on FAA weather camera images and expressed pessimism about the prospect of the pilot attempting VFR flight, she did not provide the pilot with the information contained in the Area Forecast for the pass, which indicated instrument flight rules conditions, rain, and mist, as well as isolated moderate turbulence in the area below 6,000 ft msl. The FS-S also did not provide weather radar information, which showed the intended route of flight under an extensive area of precipitation, including areas of moderate or greater intensity. Despite the pessimism of the FS-S on the prospect of the pilot attempting a VFR flight to his intended destination that night, had the FS-S provided the current forecast information for the specific pass and the regional turbulence found in the Area Forecast, as well as a description of the current weather radar depiction for his intended route of flight, the pilot would have had a more accurate picture of the weather over the route of flight.

Data obtained from an onboard GPS unit for the last several minutes of flight showed the airplane conduct two descending, spiraling turns. The airplane continued to descend before the data terminated. The airspeed during the last several minutes of the flight ranged from 49 knots to 82 knots.

The pilot departed on the flight about 10 minutes before sunset. Twilight conditions during much of the flight would have provided the pilot with some illumination from which to see; however, cloud cover would have decreased the amount of illumination during this period, as would the end of civil twilight, which occurred about an hour after takeoff. Thus, dark night conditions would have existed for at least the last 14 minutes of the flight, as the flight proceeded over a remote area devoid of cultural lighting. The dark night conditions, lack of available ground lighting, and possible instrument meteorological conditions present at the time were conducive to the development of spatial disorientation, and the airplane's flight track is consistent with the known effects of spatial disorientation. Whether the pilot may have been experiencing fatigue before and/or during the flight given his hunting activities of the previous 4 days, it could not be determined based on the available information.

Based on the available radar weather and airplane track data, it is likely that the airplane flew into or came very close to an area of moderate or greater precipitation just before the accident; however, the extent to which the weather contributed to the pilot's spatial disorientation could not be determined.

The pilot chose to depart at a time that would have required him to operate in dark night conditions with a lack of cultural lighting, eliminating his reference to a visual horizon and requiring reliance on the airplane's flight instruments for attitude control. The dark night conditions would also have precluded the detection and avoidance of weather.


Probable Cause and Findings

The National Transportation Safety Board determines the probable cause(s) of this accident to be:

The noninstrument-rated pilot's decision to initiate a visual flight rules flight into dark night, marginal visual flight rules to instrument flight rules conditions, which resulted in a loss of control due to spatial disorientation.



Personnel issues

Spatial disorientation - Pilot (Cause)

Aircraft control - Pilot (Cause)

Environmental issues

Dark - Decision related to condition (Cause)

Dark - Effect on operation (Cause)

Below VFR minima - Effect on operation (Cause)

Factual Information



On August 23, 2017, about 2245 Alaska daylight time (AKDT), a wheel-equipped Piper PA-18AS-125 airplane, N1905A, impacted remote, tree-covered terrain about 31 miles northwest of Tyonek, Alaska. The private pilot sustained fatal injuries and the airplane was destroyed. The airplane was registered to and operated by the pilot as a Title 14 Code of Federal Regulations Part 91 visual flight rules (VFR) personal flight. Night instrument meteorological conditions were present along the airplane's route of flight at the time of the accident and no flight plan was filed. The flight originated from a remote airstrip near Telaquana Lake, located in the Lake Clark National Park and Preserve, about 2130, and was destined for Merrill Field (MRI), Anchorage, Alaska.

The pilot's wife reported that the pilot was returning from a solo Dall sheep hunting trip in the Alaska Range. The pilot departed from MRI on August 19 about 1630 and arrived at the remote airstrip about 1930. She reported that the length of the hunting trip was open-ended with no set return date.

The pilot placed three separate phone calls via satellite phone to the Federal Aviation Administration (FAA) Kenai Flight Service Station (FSS), Kenai, Alaska, on the night of the accident requesting weather information for his route of flight. The Kenai FSS is located at the Kenai Municipal Airport (PAEN), Kenai. Each call was dropped due to interruptions of satellite coverage. The same flight service specialist (FS-S) spoke with the pilot on all three occasions. The public docket contains the transcripts for each phone call.

The first phone call was at 1950 and lasted about 1 minute. The pilot reported that he was south of Merrill Pass and requested an outlook briefing for flight via Merrill Pass to Anchorage for the night of the accident and for the following day. He asked, "whether or not it's going to remain VFR from Merrill Pass back into Anchorage tonight and into tomorrow." The FS-S provided the terminal forecast for Anchorage before the call was dropped.

The second phone call, at 1956, lasted about 3 minutes. The pilot reported that he was near Merrill Pass on the west side and requested a "quick weather brief" asking if VFR flight from Merrill Pass or Telaquana Pass to Anchorage would be recommended for that night. The FS-S provided the Merrill Pass forecast, which called for marginal VFR (MVFR) conditions, the area forecast (FA), and the Anchorage terminal forecast before the call was dropped.

About 2100, the pilot contacted his wife via a satellite phone and asked her to retrieve weather information for a flight to MRI. She instructed the pilot to call her back in about 5 minutes; however, the pilot never called back.

The third and final phone call to the FSS was at 2108 and lasted about 10 minutes. The pilot asked if VFR flight across the Alaska Range or in the Merrill Pass area was recommended that night. The FS-S responded by saying, "…I don't have anything that's giving me a good report…" and provided the pilot with information from the current FA, which included the current airman's meteorological information (AIRMET) and regional ceiling, visibility, precipitation information, and information from a pilot report (PIREP) advising that VFR was not recommended on the east side of Lake Clark Pass (about 34 miles southwest of the accident site).

The pilot again asked about conditions at Merrill Pass, to which the FS-S responded with a comprehensive description of what she was seeing from the FAA Merrill Pass "low" weather camera. The FS-S indicated the FAA Merrill Pass "high" weather camera was out of service. Following a request from the pilot, she also provided a comprehensive description of what she was seeing on the FAA Beluga weather camera. Despite being asked several times about Merrill Pass, the FS-S did not provide the pilot with the forecast for instrument flight rules (IFR) ceilings, rain, and mist that was indicated for the pass in the current FA.

Despite being asked directly for turbulence information for the Cook Inlet and Matanuska-Susitna Valley area, the FS-S did not provide the pilot with the forecast for isolated moderate turbulence below 6,000 ft above mean sea level (msl) for the region found in the current FA. Additionally, the FS-S did not provide the pilot with the current Kenai weather radar information.

The pilot indicated that he was not in a hurry to get home, and then asked about conditions the next day. The FS-S provided a synoptic summary of frontal systems and expected precipitation and wind and indicated that she was, "…not seeing anything too much for the passes." The pilot requested forecast information valid for 24-48 hours from that time. The FS-S stated "…that occluded front basically right now still over Kodiak Island there pushing into uh Cook Inlet/Susitna Valley definitely got your rain looking at that low coming right over by four o'clock in the afternoon tomorrow right over us. I'm showing IFR conditions right in your area where you're coming from so that's around four o'clock in the afternoon now tomorrow okay that occluded front is going to be right over the top of us into Prince William Sound there's going to be isolated areas where it's going to be in the clear but they're talking about the drying of those winds the Turnagain Arm winds there drying out that moisture at the same time you got those winds cause you got turbulence going to be in the forecast there…"

The pilot asked, "So it's going to get worse is what you're saying?" to which the FS-S responded, "yeah basically…" The FS-S then stated, "…I didn't know how else to relay that to you I just cannot tell you yeah go ahead and try to come home through Merrill Pass it's not looking too good especially what they said about the Alaskan range through the Aleutian areas there so I would not chance that." The pilot then indicated his intent to, "probably go tonight" and started to file a VFR flight plan, but the call was lost before the flight plan was completed.

About 2220, the pilot's wife received a text message from the pilot stating that he was flying over Kenibuna Lake and that he should be home around 2300. She did not receive any further communications from the pilot. She reported that he was not rushed to return home nor was there any sort of emergency that required him to depart when he did.

Upon departing from the airstrip and heading to the northeast, the route of flight was over mountainous terrain. Continuing east, the route was overwater as the airplane crossed Kenibuna Lake and Chakachamna Lake. Once past the mountainous terrain and continuing to the east, the area consisted mainly of hills before reaching the Cook Inlet and the Matanuska-Susitna Valley. An overview of the route of flight, along with the remote airstrip the pilot departed from, the accident site, and the destination airport is shown in figure 1. Figure 2 shows the last portion of the flight.

Figure 1 – Map of the route of flight, obtained from the onboard GPS unit (courtesy of the NTSB).