The U.S. aircraft register, updated daily
Accidents · NTSB WPR21FA143 · Final report

Airbus Helicopters AS350-B3 accident near Palmer, Alaska, March 28, 2021

On March 28, 2021 at about 2:36 am local time, a 2008 Airbus Helicopters AS350-B3, registered N351SH, was substantially damaged in an accident during landing near Palmer, Alaska. It was an other work-use flight under charter and air-taxi rules (Part 135). 5 people were killed and 1 person was seriously injured. The weather was visual conditions (good weather).

The NTSB's probable cause their words, unchanged

The pilot’s failure to adequately respond to an encounter with whiteout conditions, which resulted in the helicopter’s collision with terrain. Contributing to the accident was the (1) operator’s inadequate pilot training program and pilot competency checks, which failed to evaluate pilot skill during an encounter with inadvertent instrument meteorological conditions, and (2) the Federal Aviation Administration principal operations inspector’s insufficient oversight of the operator, including their approval of the operator’s pilot training program without ensuring that it met requirements. Contributing to the severity of the surviving passenger’s injuries was the delayed notification of search and rescue organizations.

Source: NTSB aviation accident database, copy made October 5, 2026. Docket and reports at the NTSB.

What the record shows

Date
March 28, 2021 · about 2:36 am local time
Place
Palmer, Alaska · map
Type
Accident
Injuries
5 people were killed and 1 person was seriously injured.
Weather
visual conditions (good weather)
Aircraft
Airbus Helicopters AS350-B3, built 2008 · all AS350-B3s on the register
Registration
N351SH · no longer on the register · serial 4598
Damage
Substantial damage
Flight
Other work-use flight · charter and air-taxi rules (Part 135)

The NTSB's narrative final · quoted from the NTSB record

A local lodge had contracted with the helicopter operator to transport passengers from a private residence to a heli-ski area at a nearby mountain. The surviving passenger stated that, before the last ski run of the day, the pilot attempted to land on a ridgeline but that the helicopter lifted off for an attempted second landing. The passenger also stated that, during the second landing attempt, the snow was “real light” but that the helicopter became “engulfed in a fog which made it appear like a little white room.” The helicopter subsequently began “going backward real fast” and impacted the ridgeline and rolled backward down the mountain. Postaccident examination of the airframe and engine revealed no preimpact mechanical malfunctions or failures that would have precluded normal operation of the helicopter. The passenger’s recollection of the conditions just before the accident was consistent with whiteout conditions caused by rotor wash while the helicopter was hovering near the ridgeline. Thus, the pilot likely experienced whiteout conditions during the second landing attempt, which caused him to lose visual reference with the ridgeline and resulted in the helicopter impacting terrain. Title 14 Code of Federal Regulations Part 135 required flight-locating for the helicopter in case it was overdue so that information about the helicopter’s location could be reported to a Federal Aviation Administration (FAA) or a search and rescue facility. The helicopter operator stated that it had delegated the responsibility for flight-locating to the local lodge. However, this delegation was not documented in the company’s FAA operations specifications or general operations manual, as required by Part 135. On the day of the accident, the local lodge was providing flight-following for the accident helicopter, which, unlike flight-locating, was not required by Part 135. The helicopter was expected to depart the heli-ski area for the principal operations base once all the ski runs had been completed. The flight-follower informed his supervisor that 40 minutes had elapsed since the helicopter moved from its last recorded position and that there had been “no positive comms” with the ski guide during the last 1.5 hours; this notification was made 41 minutes after the last “ping” from the helicopter. However, the remote area in which the accident flight was operating had limited communication capabilities, and no clear evidence indicated that an accident had occurred. The flight-follower’s supervisor contacted another heli-ski company to help determine the status of the helicopter. The heli-ski company considered the flight to be “ops normal” and expected that the last lift would occur shortly. The supervisor instructed the flight-follower to “keep an eye on” the accident helicopter; however, the lodge’s emergency response plan stated that a search and rescue facility should be contacted “if communication with the helicopter is not established by the end of the prearranged [time] or 30-minute grace period.” Therefore, it would have been reasonable for the lodge to activate its emergency response plan at this point given that the helicopter’s location was unknown at the time. The flight-follower continued to try to reach the helicopter but was unsuccessful. About 90 minutes after the last flight-following “ping” for the helicopter, the lodge received erroneous information from the heli-ski operator that the accident helicopter was “inbound” for the private residence; however, lodge personnel did not realize that this information was not accurate and that the accident had occurred immediately after the last ping. This incorrect information likely played a role in the lodge’s further delay in activating its emergency response plan. About 1 hour and 50 minutes after the accident (and the last flight-following ping), the heli-ski operator notified the accident helicopter operator about the overdue aircraft. Five minutes later, the lodge notified the helicopter operator that it was activating its emergency response plan. About 2 hours after the accident occurred, the helicopter operator activated its emergency response plan. About 17 minutes later, the helicopter operator notified the Alaska Rescue Coordination Center. The director of operations for the helicopter operator stated that the search and rescue notification did not occur before that time because he had been “working through the information that was provided” about the helicopter. The helicopter wreckage was located about 3.5 hours after the accident. Rescue personnel launched within 1 hour of notification and arrived on scene less than 30 minutes later (about 5 hours 40 minutes after the accident). The surviving passenger was transported via helicopter to a local hospital. Upon arrival at the hospital, the surviving passenger had hypothermia and severe frostbite. A shorter exposure to the cold would likely have decreased the severity of the surviving passenger’s injuries. However, a faster emergency response time (and thus shorter exposure to the cold) could only have occurred if the notification to search and rescue personnel had been timelier. Thus, because the lodge and the helicopter operator did not activate their emergency response plans sooner, the initiation of search and rescue operations was delayed. Given the circumstances of this accident, the investigation considered three types of training that the pilot should have received: inadvertent instrument meteorological conditions (IIMC) training; controlled flight into terrain-avoidance (CFIT-A) training, during which instruction in whiteout conditions would be conducted; and ridgeline training. Review of the operator’s pilot training program showed that ridgeline training was not provided for the make and model of the accident helicopter (or the previous helicopter in which the accident pilot had been trained). Further, IIMC training was a part of CFIT-A training, and the CFIT-A manual stated that pilots were required to complete IIMC training annually. However, the chief pilot for the helicopter operator stated that the related test for pilots (to demonstrate understanding of the subject) was only administered when a pilot was first hired, and the director of operations stated that the company’s only IIMC flight training involved recovery from unusual attitudes. In addition, review of the accident pilot’s flight training records found that he completed IIMC training 14 months before the accident (which was about 1 year after he began working for the operator), but the records did not reflect the specific IIMC training that the pilot received. Based upon the information provided by the operator, it could not be determined if the accident pilot had fulfilled the training requirement. The director of operations reported that the helicopter operator did not accomplish flight training as part of its CFIT-A training; however, flight training was not required for that subject, and the pilot received CFIT-A ground training. The CFIT-A manual stated that, if inadvertent whiteout conditions were encountered, the pilot was to rely on flight instruments and carefully attempt to maneuver the helicopter away from obstacles and terrain. Additional review of training records revealed that, during competency checks, the helicopter operator was not evaluating several requirements of 14 CFR 135.293, including recovery from IIMC, navigation, air traffic control communication, and instrument approach flying. Paragraph (c) of the regulation required a pilot to demonstrate the ability to maneuver the helicopter into visual meteorological conditions after a simulated encounter with IIMC, a skill that was needed during the accident flight. The operator stated that it did not have any instrument-flight-rules (IFR) aircraft capable of IFR approaches, but the GPS model installed on the accident helicopter had IFR capabilities with instrument approach procedures in its database. Thus, given the deficiencies in the operator’s pilot training program and Part 135 checkrides, particularly regarding IIMC, it is likely that the pilot did not meet the qualification standards to be the pilot in-command of the accident flight. The FAA principal operations inspector (POI) for the operator failed to ensure that the company’s operations specifications (specifically, paragraph A008) contained the operational control information required by 14 CFR 119.7 and 135.77. (Flight-locating was part of operational control.) The company’s operations specifications did not describe who would be responsible for the safe operation of company flights and how those flights would be operated to meet requirements; thus, the operations specifications were incomplete. Although the company should have noticed this omission before the operational specifications were signed, the POI was ultimately responsible for ensuring that the operations specifications contained all the required information. FAA Order 8900.1, paragraph 3-1255, discusses helicopter flight training maneuvers that “must be conducted for satisfactory completion of each category of flight training.” The order also stated that all helicopter pilots operating under Part 135 “must be trained on procedures for the avoidance and recovery from IIMC” and that inspectors were required to ensure that operational procedures for recovery from IIMC are incorporated into the certificate holder’s training curriculums. Thus, the POI failed to ensure that the operator’s training program contained all required elements, which also included ridgeline training, before approving the training program. In addition, the POI was also unaware that the operator was not conducting competency checks in accordance with section 135.293(c) and that its checkrides were only assessing a pilot’s recovery from unusual attitudes. From 2011 to 2013, the POI was the chief pilot for the accident operator; from 2001 to 2011, she worked at another helicopter company with the person who later became the president of the accident operator. The POI started her employment with the FAA in 2016 and, 2 years later, became the POI for the accident operator. The available evidence for this investigation was insufficient to determine whether the POI’s previous employment history was a factor in the inadequate oversight of the accident operator. Toxicology testing of specimens from the senior lead ski guide identified two central nervous system stimulant drugs: amphetamine and cocaine. Given the drug levels measured in his blood, the senior lead ski guide was likely impaired by drug effects at the time of the accident. Toxicology testing of specimens from the other lead ski guide identified delta-9-tetrahydrocannabinol (THC), the primary psychoactive component in cannabis. The low THC level measured in his blood indicates that he was not likely experiencing significant impairment from THC effects at the time of the accident. Although ski guides are not considered crewmembers according to the Federal Aviation Regulations, they have safety-related responsibilities during heli-ski flights such as coordinating with pilots about landing and pickup zones and assisting pilots with hazard and pickup zone identification. However, the investigation was unable to determine whether the senior lead guide’s illicit drug use played a role in the accident.

The complete narrative as the NTSB published it. The NTSB's docket holds the report as a PDF and any photographs, statements and other documents from the investigation.

The factual record from the NTSB's investigation tables, in plain English

What happened, in order

  1. Loss of visual reference during landing defining event
  2. Collision with terrain or object (not controlled flight into terrain) during landing

The NTSB's findings

  • Environmental issues › Conditions/weather/phenomena › Ceiling/visibility/precip › Whiteout › Contributed to outcome
  • Personnel issues › Task performance › Use of equip/info › Use of equip/system › Pilot
  • Personnel issues › Experience/knowledge › Training › Total instruct/training recvd › Pilot
  • Organizational issues › Support/oversight/monitoring › Oversight › Oversight of operation › FAA/Regulator
  • Organizational issues › Support/oversight/monitoring › Oversight › Oversight of operation › Operator
  • Personnel issues › Experience/knowledge › Experience/qualifications › Qualification/certification › Pilot
  • Organizational issues › Support/oversight/monitoring › Oversight › Oversight of operation › Other institution/organization
  • Organizational issues › Management › Policy/procedure › Adequacy of policy/proc › Operator
  • Personnel issues › Action/decision › Info processing/decision › Decision making/judgment › Pilot
  • Organizational issues › Support/oversight/monitoring › Training › Initial training › Operator
  • Organizational issues › Support/oversight/monitoring › Training › Recurrent training › Operator
  • Organizational issues › Support/oversight/monitoring › Training › Emergency proc training › Operator

Pilot

  • Certificate: commercial pilot
  • Ratings: instructor: helicopter; instructor: instrument helicopter; instrument: helicopter; rotorcraft: helicopter
  • Flight time: 3,286.4 hours in all; 1,505.8 in this make and model; 78.2 in the last 90 days; 46 in the last 30 days; 3,230.1 as pilot in command
  • Last flight review: January 21, 2021
  • Medical certificate: Class 1 (without waivers/limitations)
  • Seat: rgt
  • Injury: fatal

The aircraft

  • Airframe total time: 5,675.2 hours
  • Last inspection: annual inspection, March 3, 2021; 24.2 hours since
  • Seats: 6
  • Landing gear: fixed
  • Engine: Turbomeca ARRIEL 2B1 (turboshaft); 0 hours total
  • Operator: Soloy Helicopters LLC

The flight

  • Departed from: Wasilla AK at 10:53 pm

Weather at the time

  • Light: daylight
  • Wind: from 330° at 5 knots
  • Visibility: 10 statute miles
  • Sky: clear
  • Temperature: 34°F (1°C), dew point 10°F (-12°C)
  • Altimeter: 29.89 inHg
  • Observation at 6:53 pm from PAAQ, 23 miles away

Weather report (METAR): PAAQ 280253Z AUTO 33005KT 10SM BKN065 01/M12 A2989 RMK AO2 SLP127 T00111117 53031 TSNO

Injuries

FatalSeriousMinorNone
Flight crew1
Passengers41

Documents from the investigation the NTSB's docket: the evidence folder behind the report

32 documents, released by the NTSB on September 27, 2023. View them here, or download them; the NTSB redacts some personal details before release, and this site shows the NTSB's own titles rather than its file names.

#DocumentWhat it is
1 Operations/human Performance Group Factual Report PDF, 54 pages · our copy View Download
2 Operations/human Performance Factual Report Attachment 1 - Interview Summaries PDF, 25 pages · our copy View Download
3 Operations/human Performance Factual Report Attachment 2 A�� Interview Transcripts PDF, 492 pages · our copy View Download
4 Operations/human Performance Factual Report Attachment 3 A��personnel Records PDF, 49 pages · our copy View Download
5 Operations/human Performance Factual Report Attachment 4 A�� Weight and Balance Information PDF, 12 pages · our copy View Download
6 Operations/human Performance Factual Report Attachment 5 A�� General Operations Manual Excerpts PDF, 6 pages · our copy View Download
7 Operations/human Performance Factual Report Attachment 6 A�� Operations Specification Paragraph A008 PDF, 6 pages · our copy View Download
8 Operations/human Performance Factual Report Attachment 7 A�� Flight Locating Definition PDF, 3 pages · our copy View Download
9 Operations/human Performance Factual Report Attachment 8 A�� Timeline of Events PDF, 2 pages · our copy View Download
10 Operational Factors/human Performance Attachment 8 Addendum PDF, 2 pages · our copy View Download
11 Operations/human Performance Factual Report Attachment 9 A�� FAA Follow-up Correspondence PDF, 3 pages · our copy View Download
12 Operations/human Performance Factual Report Attachment 10 A�� Soloy Follow-up Discussions PDF, 6 pages · our copy View Download
13 Operations/human Performance Factual Report Attachment 11 A�� Cfit-a Manual Excerpts PDF, 9 pages · our copy View Download
14 Operations/human Performance Factual Report Attachment 12 A�� SMS Manual Excerpts PDF, 5 pages · our copy View Download
15 Operations/human Performance Factual Report Attachment 13 A��base Radio Operator Training Information PDF, 3 pages · our copy View Download
16 Operations/human Performance Factual Report Attachment 14 A�� Hsus Operating Guidlines PDF, 45 pages · our copy View Download
17 Mfr - FAA Order 8900.1 Legal Interpretation Request PDF, 3 pages · our copy View Download
18 Global Positioning System (GPS) Device - Specialist's Factual Report PDF, 21 pages · our copy View Download
19 Attachment 1 to Global Positioning System (GPS) Device - Specialist's Factual Report data file · our copy Download
20 Survival Factors Specialist's Factual Report PDF, 15 pages · our copy View Download
21 Survival Factors Specialist's Report - Attachment 1 ELT Teardown Report PDF, 3 pages · our copy View Download
22 Survival Factors Specialist's Report - Attachment 2 Sar Timeline PDF, 3 pages · our copy View Download
23 Meteorology Factual Report PDF, 25 pages · our copy View Download
24 Airworthiness Group Factual Report PDF, 29 pages · our copy View Download
25 Airworthiness Group Factual Report - Attachment 1 PDF, 3 pages · our copy View Download
26 Airworthiness Group Factual Report - Supplemental PDF, 10 pages · our copy View Download
27 Medical Factual Report PDF, 6 pages · our copy View Download
28 Injury Factual Report PDF, 6 pages · our copy View Download
29 Toxicological Report - Pilot PDF, 1 page · our copy View Download
30 Statement of Party Representatives to NTSB Investigation - Soloy PDF, 1 page · our copy View Download
31 NTSB Email Communication with Tml PDF, 3 pages · our copy View Download
32 Photo Array PDF, 3 pages · our copy View Download

The same docket at the NTSB.

Other NTSB records under N351SH the same tail number, which may have belonged to a different aircraft at the time

2017-08-03ANC17CA041 · accident near Delta Junction, AK · substantial damage · no injuries

Everything on this page comes from the NTSB's public records. The narrative, probable cause and findings are the NTSB's own words; the coded tables behind the report are written out in plain English, with pilots' ages, home towns and medical details left out. The documents and photographs are the NTSB's docket, shown as the NTSB released them. This site's own text never names anyone involved. A preliminary report can change; the final report usually follows one to two years later, and the page is refreshed when it does.