On September 7, two-and-a-half years after the crash of a Yute Commuter Service single-engine Piper PA-32 near Tuntutuliak, Alaska, the National Transportation Safety Board (NTSB) released its probable cause finding on the five-fatality accident, citing a pilot’s decision to fly into potentially whiteout conditions. But a lengthy investigation also pointed to systemic safety issues and lax procedures that enabled the decision-making.
The PA-32 departed under a special VFR on the morning of Feb. 6, 2020, for the village of Kipnuk, less than 100 miles from the company base in Bethel. It crashed about 30 minutes later, at approximately 11:10 a.m. local time.
Investigators determined the decision of the pilot, Tony Matthews, to continue VFR into “reduced visibility, including likely flat light and/or whiteout conditions,” was the cause. Matthews had 645 hours of total time; the Kipnuk flight was only his fourth revenue trip for the company.
As a contributory factor, the NTSB also found Yute’s operational control procedures were inadequate and “permitted” the pilot to depart below company minimums.
Yute is a VFR-only Part 135 commuter and air taxi. It primarily serves the southwestern Alaska region known as the Yukon-Kuskokwim (YK) Delta. The area is notorious for challenging winter weather conditions, with snow, fog, and winds prevalent throughout the season. Flat light is a common hazard.
When asked to describe what it was like, Yute’s chief pilot Byron Paul was blunt: “…the ground’s the sky and the sky’s the ground,” he told an investigator, “[that] is the only way I can tell you.”
The Kipnuk flight departed Bethel at 10:42 a.m. Twelve minutes earlier, as Matthews was taxiing out, the weather was 600 feet overcast with four miles of visibility in mist. At 10:43 a.m. there was a special observation with the ceiling still at 600 feet overcast but visibility down to one and one-quarter miles.
A special issued two minutes later reported the ceiling at 800 feet overcast and five miles visibility, then at 11:05 a.m., it was back down to 500 feet overcast and three miles.
In Kipnuk, 40 miles from the accident site, conditions were 600 feet overcast with nine miles of visibility in light snow at 10:56 a.m. In the next hour, both Bethel and Kipnuk had worsening conditions. Visibility went down to one-half mile, and the ceiling measured as low as 400 feet with light snow, mist, and freezing fog.
Many of the villages in the YK possess limited weather reporting options. Some have no weather reportage at all, some have no certified weather available, and others suffer from technical difficulties including chronic telecommunication outages rendering the equipment useless.
As stated in the 2021 FAA Alaska Aviation Safety Initiative, “most rural airports do not have weather reporting systems. Without certified weather reporting, or an approved alternative with adequate fidelity, by regulation, Part 135 operators cannot conduct IFR operations into these airports.”
Attempts to reach Yute Commuter Service for comment were unsuccessful.
Operational Control and Risk Assessment
The NTSB interviewed eight current Yute employees in the course of its investigation. In all of these interviews, Wade Renfro, the company owner, president, and general manager was present and permitted to ask questions of the interviewees.
The investigation largely focused on three things: lack of coherent policy establishing company minimums and individual pilot limitations, the flight’s missing risk assessment form, and the manner in which daily operational control functioned.
The two individuals who commonly held operational control were often physically absent from the office. Renfro lived part of the year 360 miles away, in the town of Soldotna, and director of operations (DO) Tony Spangler resided part of the year in Las Vegas. The other two individuals who could be delegated with the authority, Paul and Ernie Turrentine, a senior pilot and former chief pilot, both flew the line and rarely exercised the responsibility.
On the day of the accident, Renfro held operational control and phoned in at about 7 a.m. from Soldotna. Spangler was in Las Vegas; Turrentine and Paul were scheduled to fly and not in the building.
Regardless of management presence, daily flight planning and dispatch were handled by a flight follower and flight coordinator. The flight coordinator on duty the morning of February 6 had been with the company for one year and had been promoted from her position of flight follower six months earlier.
The flight follower had been with Yute for nine months. Before their employment with Yute, they were both customer service agents for other operators; neither had any formal aviation education or knowledge other than what they obtained on the job.
For example, the flight follower could not correctly define special VFR when asked by investigators. He was also unfamiliar with common aviation weather terms including sigmet, airmet, VMC, and IMC.
The flight follower was primarily responsible for obtaining the weather each morning, by viewing FAA weather camera websites and phoning village agents. Meanwhile, the flight coordinator checked with agents about daily passenger “traffic” and created load manifests, including mail and freight. No pilots were officially assigned to flights prior to arriving at work.
The DO assigned pilots to aircraft following their checkrides and the aircraft were assigned to routes by the flight coordinator. Upon arriving at work, pilots learned the routes for their aircraft and discussed the weather with dispatch. (Yute had about 12 pilots total at the time of the accident, but they worked in rotations with about six in Bethel at any one time.)
Dispatch spoke with the person holding operational control before the early departures and any concerns that might result in delay or cancellation were addressed then. (Both Spangler and Renfro said they also checked weather from their homes.)
Before a flight was dispatched, the risk assessment form, as required by the general operations manual, was completed by the pilot. It was checked for accuracy by either the follower or coordinator and if it did not reach a risk figure meriting management attention, it was signed by one of them, then the pilot, and filed. After the accident, the risk assessment form for the Kipnuk flight could not be found.
In her interview, the flight coordinator initially insisted the flight follower reviewed and signed Matthews’s form and had final possession of it. However, under increased questioning, she changed her mind and said she did not remember seeing the form at all that morning. For his part, the flight follower admitted the risk assessment form was required before a flight could be released but did not remember signing it, did not remember seeing Matthews complete it, and did not know where it might have ended up.
Every member of Yute management who was asked about the missing form professed to have no idea how it could have disappeared or why no one recalled filling it out.
Company Minimums and Pilot Limitations
In discussing risk assessment, Matthews’s status as a new pilot, and how that would affect his flight risk calculation, was addressed extensively by investigators. Interviewees mostly agreed that crosswind limits of 15 knots were placed on all new pilots and remained in effect until lifted by the DO. From there, the question of how pilots were individually limited became murky.
Spangler said there were restrictions only for low-time new hires in the Cessna 172 and dismissed the general crosswind limitation, while Turrentine said there were individual limitations as set by Paul or Spangler for every new pilot “until he flew a couple of shifts.” (In this context, shifts refer to scheduled pilot rotations.)
According to Spangler, the company’s overall weather minimums were 500 feet and two miles of visibility. Paul stated they were 1,000 feet and two miles for new pilots and 500 feet and two miles for himself. The flight coordinator said they were 500 feet and three miles, while the flight follower said they were 1,000 feet and two miles.
Renfro avowed that they were 500 feet and two miles but the company did not launch pilots in those conditions. He acknowledged an experienced pilot like Turrentine might depart at 500 feet and three miles for a short flight, but otherwise, “500 and 2 is not a condition that we want to be flying around in.”
Everyone who was asked told investigators they had no knowledge of the limitations placed on Tony Matthews. For a new pilot, Renfro said, “usually 1,000 and five or somewhere in that neighborhood would be more appropriate,” while Paul said new pilots could depart only with a 1,000-foot ceiling and two miles visibility and were never dispatched under a special VFR unless he or Spangler flew with them. (Spangler never flew with Matthews and Paul only flew an evaluation flight in the Cessna 172.)
The flight coordinator said the only limitations for new pilots were those that were told to her on the day of a flight, by whoever had operational control. Spangler, when asked to assess the conditions Matthews departed under, told investigators it was “a typical operation.”
“…In the wintertime,” Spangler told investigators, “if you’re going to fly out here, you normally have to be able to handle some, you know, instrument-type conditions getting in and out of Bethel, usually getting into the point that you can keep the airplane…I mean, this has been something for years and years out here…so no, there were no restrictions.”
The DO also confirmed that none of Yute’s aircraft were IFR certified, and the ability to fly approaches on instruments was “limited because of the equipment in the airplane.”
Paul, the chief pilot, said the company provided only about two hours of annual instrument training, under the hood, during initial training and the checkride. (This was also part of recurrent training.)
Turrentine, who taught the ground school Matthews attended, said if IMC was encountered while flying, “you should go on your instruments and you do a 180 to get out…” In his opinion, the accident happened because Matthews “fixated outside and quit looking at the instruments for too long.”
Multiple Safety Issues
In addition to Yute, Renfro owned and operated a charter-only company, Renfro’s Alaskan Adventures, also based in Bethel. (That company was involved in four accidents in the previous five years, most recently in September 2019.)
He purchased Yute in 2017 and relocated it from Kodiak to Bethel. Spangler served as the company’s DO and Turrentine was initially the chief pilot. Prior to the Kipnuk flight, the company crashed five times under their oversight, including one accident in 2018 when the pilot—who had over 21,000 hours—permitted a cargo handler with no flight experience to sit in the left seat. During takeoff, the pilot lost rudder control and the aircraft veered off the runway and hit a small ridge. The FAA later characterized this accident as allowing a “non-company unqualified pilot to attempt takeoff.”
There were other events and occurrences at Yute following Renfro’s purchase that drew the attention of the FAA. These included multiple issues with not grounding the aircraft during refuel, failure to secure baggage doors during preflight, the failure of a nose gear actuator on approach, collapsed gear resulting in a prop strike, failure of a pilot to maintain control on landing in Nightmute, and, in April 2019, failure of a pilot to maintain control when landing in Bethel, which resulted in the wing and propeller contacting the runway. The company did not report this event to the FAA.
On April 23, 2019, Yute’s principal operations inspector (POI) sent Spangler a letter noting the “large number of reported incidents” over the previous few months. The DO was charged with developing a plan to “correct deviations from company procedures” within the next two weeks. One month later, in response to a request for a conformity inspection to add another PA-32 to the company certificate, the FAA informed Yute a risk analysis had determined that “due to the poor safety record exhibited by the company,” the agency would not permit the addition of other aircraft at that time.
While these issues were ongoing, there was also another problem with Yute’s operation that captured FAA attention: the abilities and professionalism of chief pilot Ernie Turrentine.
Prior to his position at Yute, Turrentine developed an extensive aviation history in Alaska, logging more than 30,000 hours and employment with several Part 135 operators. But in July 2018, during an FAA observed checkride, he permitted a pilot to land at a controlled airport without clearance and then recorded that check flight as satisfactory. Because of this event, in January 2019 the FAA formally notified Renfro that Turrentine’s check pilot approval was withdrawn. Two subsequent observed rides to return that authority were also “recorded as failures.”
In May 2019, the FAA notified Renfro that a review of Yute’s pilot training records found multiple violations. In a subsequent letter, the agency noted that three recently hired pilots did not meet the minimum regulatory qualifications to serve as pilot-in-command.
Combined with his check airman authority issues, the FAA determined Turrentine “failed to exercise his duties and responsibilities with the highest level of public interest.”
Absent effective arguments from Yute, he was to be removed as chief pilot in thirty days. Records obtained via the Freedom of Information Act (FOIA) show that Paul did not obtain chief pilot approval until the end of September, however.
In the interim, according to a company pilot who spoke with AIN on the condition of anonymity, there were several failed attempts to have other pilots approved as chief pilot in the summer of 2019. A FOIA request was submitted to the FAA to determine if Turrentine continued to hold the chief pilot position of chief pilot for Yute in this period but to date, no response has been received.
On August 5, six months before the Kipnuk flight crashed, a Yute pilot made an anonymous hotline complaint alleging the “chief pilot” departed Bethel the day before with visibility under two miles due to fog.
The pilot further alleged the chief pilot bragged about landing at a village with three-quarters of a mile visibility, that a senior pilot encouraged other pilots to depart in conditions below minimums for several days previously, and that on August 3, when the complainant refused to fly below the minimum of 500 feet and two miles, another pilot was dispatched instead.
Upon investigation, the FAA determined that all of the allegations were unsubstantiated. Specifically, it was determined that while eight flights on August 3 were dispatched at either 500 feet and/or two miles visibility, with one departing only six minutes before the ceiling dropped to 400 feet, none officially occurred at conditions under two miles.
The FAA also found the lowest published weather when Turrentine was flying on August 5 was 600 feet overcast and two miles visibility. “Through interview,” the agency concluded that other claims could not be substantiated.
The complaint was discussed with Spangler, and he was told that “every flight released must have a reasonable expectation of being able to maintain the required VFR minimums…” It was recommended he review risk assessment procedures and make sure “these operations have adequate management oversight.”
A memorandum from Yute’s POI on the complaint was provided to the Anchorage FSDO frontline manager on November 5. On November 10, Yute crashed a PA-32 shortly after takeoff from the village of Goodnews Bay. The subsequent investigation revealed the pilot suffered spatial disorientation while operating in night marginal VFR.
On November 14, another letter was sent to Renfro immediately revoking approval of Yute’s training program based on multiple issues, including the unsatisfactory performance of a check pilot candidate on November 8, the observed failures of multiple pilots to perform checklist procedures during a recent inspection, the failure of a pilot to maintain VFR clearance of clouds in an observed flight, and the company’s failure to provide, as requested, proper notice of training. (To date, no notice had been provided.)
The FAA then removed Yute’s approval for all check pilot activities and required the company to resubmit a training program. Tony Matthews participated in the company’s next ground school, in January. Taught by Turrentine, it was not observed by the FAA.
After the Crash
According to Yute’s former assistant business manager, who was interviewed by the NTSB one month after the accident, pressure on company pilots extended beyond regulatory and flight safety issues. The manager left Yute in June 2020 but was present in the office on the day of the accident (and took notes at the time of what happened). He spoke to the NTSB of “a constant push to keep the airplanes flying because they did not make money if the airplanes were not flying.”
He said Renfro talked often “about how they have to ‘fly, fly, fly.’” Pilots complained about flying aircraft that were not airworthy and returned to base for mechanical issues. Turrentine, he noted, “would fly in anything.”
Soon after the Kipnuk flight was reported overdue, he told investigators a Renfro Alaskan Adventures pilot came into the office, noted the weather at the time the flight was dispatched, said it should not have gone, and then destroyed his own risk assessment form. Meanwhile, Turrentine “spent time working” on Matthews’ training records before they were reviewed by the NTSB. (Renfro was not present for this interview.)
In March, the FAA cited the Kipnuk accident, among other accidents, incidents, and occurrences, in a letter proposing the removal of Tony Spangler from his position as DO. “Based on this pattern of increasingly severe aircraft mishaps and events that culminated in the February 6th fatal accident,” the Polaris CMO aviation safety manager wrote to Renfro, “it appears Mr. Spangler did not enact the necessary controls to prevent further aircraft mishap events or effectively restrict operations in light of them, and by failing to do so he has not exercised the standard of care expected…”
Renfro responded with a list of specific actions to prevent future accidents including promising to work with the FAA to ensure the DO and the company’s flight and training programs aligned with “recommendations to promote flight safety and awareness.” He asked that Spangler be permitted to remain in his position. Since 2021, he is no longer Yute’s director of operations. The exact date of his removal from the position is unclear.
During its investigation, one person the NTSB did not interview was the company’s POI. (In fact, no one from the FAA was seemingly interviewed.) When asked about this omission by AIN, a board representative stated that in the course of an investigation, the investigator-in-charge makes “judgment calls” about how best to utilize their “limited resources.” In this case, it was believed that the investigator on the Kipnuk flight likely decided she needed no further documentation to complete the factual record. As the investigator left the NTSB in June 2021 however, the representative could not “definitely determine” if that was the case.
There is no reference to any of the FAA’s documented concerns about Yute’s operational control, corporate safety culture, and pilot training in the accident record. It is unclear if the NTSB’s investigators were aware that they even existed.
Former Pilots Came Forward
Three weeks after the accident, a former pilot sought federal whistleblower protection to provide information on Yute. The pilot, who maintained anonymity in his interactions with the FAA, submitted a report to inspectors detailing how he witnessed the falsification of training records, attended ground school under Turrentine where participants were told to teach the class on their own, and experienced flight training in altitudes at less than fifty feet. He stated that pilots were instructed to change aircraft altimeter settings so passengers were unaware when they were flying below 500 feet and were instructed to fly in the clouds when they were alone, and purposely build up ice on the wings, so they could learn how the aircraft handled it.
In communications with AIN, this pilot noted that Paul especially pushed pilots to fly in bad weather, recalling, “they tell the new pilots, we don’t know the [effing] weather and you fly when we tell you.”
When the FAA counsel refused to extend whistleblower protections to the pilot, a safety inspector attempted to corroborate his account from other Yute pilots but was unsuccessful; the investigation was closed.
But two weeks after the accident, another former pilot, who left Yute in 2019, gave an on-record interview to the NTSB making similar allegations. He told investigators pilots were provided the answers to test questions during ground training and described a harrowing checkride with Turrentine during which they departed with only two hours of fuel into a ceiling that never rose above 600’ throughout the four-leg trip.
One destination airport, he said, was at 200 feet and visibility was less than a mile. When he asked if they should wait on the ground for the weather to improve, Turrentine took off into IMC, handed over the controls, and told him to continue to the next destination. At that village, Toksook Bay, they broke out at 150 feet. By the time they returned to Bethel, they had less than ten gallons of fuel.
This pilot complained that there was little guidance or limitations for new pilots and they were encouraged to hold under poor weather until a special VFR clearance could be obtained out of Bethel, regardless of weather conditions elsewhere. “They want you to go no matter what,” he told investigators, and flying at low altitudes in low weather was common; some pilots would go as low as fifty feet. He estimated the company “lost seven airframes” in accidents during the year he was employed there. To his knowledge, most of them were not reported. (Renfro was not present for this interview.)
In the hours surrounding the Kipnuk accident, weather at Bethel shifted from marginal VFR to IFR conditions. There were overcast and low ceilings in the morning and visibility was restricted “in mist, with snow and unknown freezing precipitation” right before the flight’s departure.
In Kipnuk, the conditions were VFR to marginal VFR in the early morning, with IFR developing at the time Tony Matthews departed. IFR prevailed in the hours after the crash, with “low ceilings and visibility restricted in snow and mist” into the evening.
When Renfro sat for his own interview with the NTSB three days after the accident, he told them he believed Matthews was flying at 1,000 feet when he left Bethel. He said he spoke with a controller at the Bethel tower after the aircraft was reported overdue who told him the weather was 1,000 feet and ten miles visibility when the aircraft departed.
This information was counter to all published weather reports but fit his narrative about the company’s minimums and safety. “…we don’t launch people at 500 and 2 from this business,” he told investigators. “We don’t want you out there flying at 500 and 2.”
The responsibility for making sure those flight minimums were maintained was, according to Renfro, found only in the cockpit. “The pilot,” he told investigators, “ultimately checks his own weather and makes sure it’s good. It doesn’t matter what he’s handed. He is still to check the weather for himself and make sure the flight is good because he is the final authority as to whether the flight goes.”
When they left Bethel, Tony Matthews and his passengers had been delayed for an hour and a half. There was no one present in the building with operational control authority, and the two employees in dispatch had limited aviation knowledge and experience. Matthews should have completed a risk assessment form prior to departure but it could not be found and neither the flight follower nor the coordinator admitted to signing it with him. He required a special VFR clearance to take off, even though company management asserted this was not common practice. The area forecast around the accident site included an airmet for IFR conditions in light snow and mist. Matthews was a VFR-only pilot in a VFR-only aircraft; he had 645 hours and completed his initial operating experience requirements for Yute only the week before he died.
It is unknown if Wade Renfro, Byron Paul, or Ernie Turrentine are still approved to hold operational control with the company. AIN continues to wait for a response from the FAA to a FOIA request for documents concerning this issue.