The Tara Air Flight 197 Final Report
A second look at Nepal's Fatal Air Crash in May of 2022
I previously wrote on this incident based on the release of the preliminary report. Nepal’s Civil Aviation Authority (CAAN) has now released its final report on the incident. This article will reference my previous, but with new detail and recommendations based on the latest findings.
Except as indicated by underlined text, all factual information in this article is from CAAN’s two reports on this incident.
On May 29, 2022, Tara Air Flight 197 killed 22 people when it crashed about 12 minutes from Pokhara’s then-regional airport (VNPR). Operating the flight was a de Havilland Canada (Viking) DHC-6 Twin Otter, registered as 9N-AET. The flight left Pokhara at 9:55 am (all times NPT) and its wreckage was discovered nearly a day later near Sano Sware Bhir in Thasang, Mustang. It was bound for Jomsom Airport (VNJS) in Gandaki Province, Nepal. Two pilots were in control of the flight. One, Captain Prabhakar Ghimire, was the pilot-in-command. His co-pilot was Utsav Pokhrel, who had just about one year of service with Tara Air.
Flight 197 began the day with a weather-related delay. The first flight to depart, a Summit Air Let-410, left Pokhara for Jomsom at 9:27 am, and at 9:50 reported patches of clouds along the route at 12,500 feet. Moments later, the second Summit Air flight (601) departed, and Tara Air Flight 197 took off just one minute after that. CAAN noted that Captain Ghimire indicated reluctance to fly, but someone—either a passenger or grounds crew—advised him “strongly” to conduct the flight. It further concluded that the departure of the Summit Flight 601 induced even more pressure on Ghimire to start the flight. Once airborne, Ghimire again showed further reluctance to continue, but apparently did so based on weather reports given by Summit Flight 601. After switching communications to Jomsom’s tower, Flight 197 encountered increasing cloud cover, which Ghimire tried hard to avoid. At least one report to air traffic control suggested that Ghimire was steering the plane toward “light and brighter” areas while maintaining or climbing out of 12,000 feet. Flight 197 met its end 7.7 nautical miles southwest of Jomsom Airport at 4,050 meters elevation, on Sanusare Mountain. Below is pictured the accident aircraft, taken prior to the incident at an unknown date at Jomsom Airport.
The initially reported probable cause was “the flight crew’s failure to monitor and maintain the proper course while inadvertently flying in IMC conditions with the aircraft Terrain Avoidance and Warning System [TAWS] inhibited which resulted into a CFIT accident.” It added 8 contributing factors:
1. The flight crew’s failure to follow the SOP of the company
2. The aircraft flight operation with TAWS inhibited during deteriorating en-route weather condition
3. Loss of situational awareness of crew
4. Deteriorating en-route weather
5. Less experienced copilot for that sector and high crew gradient
6. Poor CRM during the flight
7. The whole cockpit duties [both PF and PM] were undertaken by the PIC, which likely impaired his performance
8. Underutilization of the available Navigation instruments
The final report listed 31 conclusions, many of them overlapping with the above. Before delving into the details, however, some contextual information is required. CAAN uses acronyms frequently. While I shall do my best to avoid them (even though we love them in aviation), I will explain them here for clarity. “IMC conditions” (Instrument Meteorological Conditions) means that the weather conditions require pilots to fly primarily in reliance on their instruments. Flying “inadvertently” into IMC conditions suggests that either the weather changed quickly and the pilots were not prepared for such changes, or the pilots were flying under “VFR” (visual flight rules) initially, and thus either did not have the equipment, training, or clearance to fly using only their instruments. To put this simply, VFR flying requires the pilots to be able to see certain distances and features in order to safely operate the aircraft. It does not allow sole reliance on instrumentation. IMC describes weather conditions that makes VFR difficult or impossible. If flying under VFR, pilots are expected to avoid flying into areas, or on days, where the weather conditions might substantially inhibit flight by external visual cues. A “CFIT” accident refers to a “Controlled Flight into Terrain.” I have previously explained what that is in my article on the Yeti flight. In short, a CFIT is an avoidable crash into the ground or other obstacle. Another term the CAAN report discusses involves the “TAWS” (Terrain Avoidance and Warning Systems). TAWS uses a radio altimeter reading to calculate the proximity and closure rate of nearby terrain. As obstacles become dangerously close to the aircraft’s trajectory, an alert notifies the pilots so they can take evasive action. There are different classes of TAWS, but all perform roughly the same function.
The Flight Path to Jomsom
It is worth noting that flying to and from Jomsom Airport is particularly challenging. To start, it is located at 8,976 feet elevation, where climb capability is reduced due to thinner air. Its only runways, 06 and 24, are just 2,674 feet long and 66 feet wide (815 X 20 meters). The terrain around it is rugged, high, and unforgiving. Because of the height of the surrounding mountains, pilots fly a portion of the flight through gorges. From 1960 through April 2019, the Jomsom route was the second most lethal in Nepal. Moreover, remote weather coverage in the area is lacking, leaving pilots to rely on other pilots to ascertain the viability of flight. Rescuers searching for the Tara accident flight resorted to tracking ping locations of Captain Ghimire’s phone to help locate the downed aircraft because the weather and terrain made it so difficult to find.
Weather that day along the Pokhara-Jomsom route was intermittently bad. Flight 197—and all other flights to Jomsom that day—waited around as Jomsom Airport operations remained closed in the morning for some time. As the weather seemed to clear up enough, the airport resumed operations at 9:06 am. Summit Air’s flight operating the LET-410 immediately took the opportunity to depart. Flight 197 departed a short time later and encountered its first adverse weather conditions after crossing Ghodepani, which is located approximately 19 nautical miles northwest of Pokhara. Jomsom Tower reported to Flight 197 the weather around that time as “Wind South Westerly up to 30 kts … Temp 18º.” Ghimire inquired twice as to whether 30 knots indicated the maximum. Thirty knots can be a dangerous windspeed for the DHC-6, depending on the direction in relation to its flight path, especially in an area like the route to Jomsom where deviation from the planned path leaves few options. Additionally, CAAN learned later from the Lete climate station, that the area that morning was covered by “dense fog and cloud” accompanied by a light drizzle. In other words, this flight absolutely should not have been operated under VFR. The pilots likely couldn’t see much of anything.
LET-410 (sample); source: https://simpleflying.com
Crash Site Evidence
Most accident investigations start with the wreckage. This aircraft did not have adequate sensors or software for flight data or cockpit voice recording. In other words, there was no “black box” evidence. Nevertheless, nearly all recoverable aircraft wreckage provides clues as to what happened. Flight 197 was found in very bad condition, though some parts remained intact. According to CAAN, “The main wreckage consisted of wings and damaged portions of cabin & cockpit. The left main landing gear, with its wheel assembly intact, was found near the main wreckage.” In addition, the physical layout of the crash site indicated how it first struck the terrain. Basically, the right side of the plane impacted the terrain first, which put the aircraft into a roll causing various other sections of the plane to strike the ground. Evidence did not indicate a pre- or post-collision fire.
The Flight Crew
Captain Ghimire, 62 years of age, was the “PIC” (pilot-in-command). Assuming command is customary for senior officers on potentially challenging flights. Ghimire started as an air traffic controller, then moved to the role of pilot and flight safety inspector for the Civil Aviation Department. Afterwards, he flew for Necon Air, Skyline Airways, and Yeti before switching to Tara. CAAN stated that Captain Ghimire “was very sensitive to flight safety and never tried to violate rules of the flight. Similarly, he seldom operated in deteriorating or challenging weather and used to teach other colleagues and students regarding flight safety during various situations. He always encouraged others to follow rules without any question. He was in person and also regarded as a strong advocate of safety and compliance.” Ghimire worked as an instructor pilot for Tara Air, training new pilots and carrying out various checks on active pilots.
Little more is reported on the co-pilot, Utsav Pokhrel, age 25. All we seem to know is that he had just about one year of service with Tara Air, with 315 hours on this type of aircraft. Pokhrel was in compliance with medical, training, and licensing requirements.
Both pilots appeared to have been well-rested, medically fit to fly and properly licensed and/or certified.
The preliminary report explicitly stated that this crew exhibited poor Crew Resource Management, and that the “whole cockpit duties [both Pilot Flying and Pilot Monitoring] were undertaken by the PIC, which likely impaired his performance.” In plain language, Captain Ghimire seems to have taken on all the responsibility of the flight, perhaps ignoring his co-pilot or possibly because his co-pilot simply remained silent or unhelpful for whatever reasons. Co-pilot Utsav Pokhrel had very little experience, and this flight was only his second into the Jomsom area. Moreover, this flight took place in difficult terrain, with adverse weather where proper crew resource management is critical. Crew Resource Management in a flight crew means that each pilot effectively conducts the duties of flying or monitoring and communicates well between each other. The Pilot Flying is controlling the aircraft. The Pilot Monitoring is ensuring that the pilot flying is following the proper course, while also overseeing critical operational issues such as speed, radio traffic, and instruments. If one pilot is attempting to do all these things at once, while in difficult terrain or conditions, the expected performance will not be good. In the case of Tara Air Flight 197, the complete breakdown of Crew Resource Management was listed in the final report as one of the “major contributory factors in the accident.”
Exhibiting poor Crew Resource Management may be the result of what CAAN’s report referred to as a “high crew gradient.” The gradient of a crew refers to the power disparity between the pilot and co-pilot. In some airlines, there is a culture in which very senior or ranking pilots do not expect—or like—to be questioned by inexperienced or lower ranked co-pilots. Other times, this dynamic is more culturally driven by the society in which an airline operates, even if discouraged by the airline itself. And in still others, this gradient only exists in the minds of specific pilots and is not encouraged or even condoned by either the airline or home culture of the crew.
Whatever happened here, investigators stated that crew gradient is a frequent problem in Nepal. CAAN’s National Aviation Safety Plan for 2023 to 2025 noted that an advisory was circulated recommending airlines implement a Crew Resource Management Training Program.
The Airline History
Tara Air Pvt. Ltd. is a subsidiary of Yeti Airlines, the company whose ATR-72 crashed in Pokhara on January 15, 2023. Formed in 2009, Tara took over operations of Yeti’s STOL (Short Takeoff and Landing) flights in Nepal, using its de Havilland DHC-6 Twin Otters and Dornier 228s. According to its website, Tara Air now operates three Twin Otters (DHC-6-300 and DHC-6-400). No compliance issues were found in the past two years, and maintenance records indicate that the aircraft was properly fit to fly. Of note, Tara previously experienced fatal flights in 2010 and 2016. The 2010 flight was attributed to “the [] unwise decision taken by the crew to descend without taking the harsh mountain terrain into consideration.” And the 2016 incident was attributed to the “crew’s repeated decision to enter into cloud during VFR flight and their deviation from the normal track due to loss of situational awareness aggravated by spatial disorientation leading to CFIT accident.”
Equipment Issues
One problem noted in the report was that Flight 197 exceeded its maximum takeoff weight. It appears that an additional passenger was added at the last minute. The report doesn’t offer much on whether this affected in-flight handling, but I would note that exceeding the maximum takeoff weight can have a detrimental effect on the climb and maneuvering capabilities of this aircraft.
This aircraft was equipped with TAWS, used for avoiding terrain. The investigation indicates that the plane’s Ground Proximity Warning System (“GPWS”) did trigger a warning just seconds before impact. The GPWS is a part of the TAWS system. None of the other features of the TAWS platform seemed to be functioning, thereby leading CAAN to characterize it as “inhibited.” Normal alerts produced by the TAWS are both textual and aural, and often occur a minute or more before potential impacts. CAAN reports that the pilot had purposely put the TAWS in an inhibited state because of issues with altitude measuring equipment, which led to “nuisance” alerts. Pilots will often disable systems producing erroneous alerts to reduce potential confusion or distraction, though it is not always appropriate to do so. Regardless, that appears to have occurred here. Had the TAWS been working correctly, the crew would have received numerous alerts prior to impact perhaps enabling them to save the flight.
The incorrect altitude readouts occurred because the pilots did not enter the regional barometric setting into the Air Data Computer. The regional barometric setting is compared against the measurements from the Pitot Tubes and Outside Air Temperature. Failing to enter the correct setting leads to wildly incorrect readouts of altitude on the on-board instruments. Two results from this included the crew not knowing the airplane’s true altitude, and receiving erroneous alerts (or failures to alert) of the crash avoidance systems.
Otherwise, the aircraft seemed to be properly maintained with routine inspections and equipment updates having been applied in the weeks just before the incident.
Recommendations
CAAN recommended improvements in three main areas—upkeep and reporting on flight-assisting technology—specifically TAWS in STOL (Short Takeoff and Landing) aircraft, improving Crew Resource Management and flight deck gradient disparities, and improving real-time data resources for en-route weather to STOL airports.
Conclusion
Some commentators in the media have shown disappointment with this final report, calling it inadequate or shallow. The reality here is simple. The Pokhara-Jomsom flight is a difficult one. Weather patterns change quickly and pilots do not have real-time access to these changes in all cases. The aircraft operating this route are necessarily small and thus have limited climb characteristics, requiring them to fly sometimes perilously close to terrain. In this case, while I believe the crash was avoidable, the general problems with it are routine in the aviation industry and CAAN appears to be making efforts to improve where it can.
Aircraft crash investigations are not focused on assigning blame, however much some commentators seek it. Rather, the primary purpose is to uncover what happened, identify the deficiencies, and use that information to make improvements to reduce or prevent future incidents.
Captain Ghimire seemed, by all accounts, to have a history of excellent piloting skills and ample experience on this route. Unfortunately, he seems to have succumbed to pressure to conduct this flight, against his better judgment. This, however, occurs far too often in the aviation industry everywhere because economics frequently trumps due diligence. Despite the deep criticisms on the Crew Resource Management in the flight deck during this flight, my suspicion is that Ghimire knew the flight was in trouble, but felt more confident in his experience and skill than his co-pilot, which effectively sidelined Pokhrel. The result was that Ghimire became overwhelmed with the workload. As a longtime pilot in Nepal, it is probable that Ghimire simply had grown accustomed to what CAAN describes as a tradition of the crew gradient effect. In other words, Nepal aviators traditionally deferred to the senior pilot in most cases and it appears that occurred here. CAAN has released several imperatives to change this culture and, like aviation in many other places, it will invariably improve if the drive continues to make it happen.
The issue with the TAWS seems to me to be one of training and regulation. For STOL flights, regulations should require Minimum Equipment Lists (MELs) that include terrain avoidance technology to certify a flight as airworthy. Pilots should have to make strong justifications for “inhibiting” such tools, and failure to report doing so should be a punishable offense. In its report, CAAN advises conducting random spot checks to ensure that TAWS are properly configured and functioning. This is an excellent recommendation. Moreover, it appears more training is needed to ensure pilots receive and enter proper barometric readings to ensure the accuracy of onboard flight data.
Nepal is slowly working to improve its weather coverage in the remote areas. This will also vastly improve safety for STOL flights if implemented, but understandably will take time and investment.
Finally, I wish to emphasize here that domestic flying in Nepal is not as dangerous as its reputation suggests. Short range, small aircraft do not have the sophisticated redundancy systems of commercial airliners, are more susceptible to inclement weather, and routinely fly more complicated routes. As I have noted previously, Nepal’s domestic aviation industry is far, far safer than its comparative in Alaska in the United States. None of this justifies failing to strive to improve. And indeed, CAAN’s National Aviation Safety Plan for 2023 to 2025 indicates that it intends or is already working to tackle safety issues in many areas. Progress appears to be happening. Just two weeks ago, the Kathmandu Post reported that Nepal Airlines finally has plans to conduct flights to Australia and Singapore, four years after signing an air service agreement. This is big news considering Nepal’s aviation sector continues to try to remove the ban on it flying into EU countries. My article on the Yeti Air Flight 691 discusses that in detail.
***
I am a Certified Forensic Computer Examiner, Certified Crime Analyst, Certified Fraud Examiner, and Certified Financial Crimes Investigator with a Juris Doctor and a master’s degree in history. I spent 10 years working in the New York State Division of Criminal Justice as Senior Analyst and Investigator. Today, I teach Cybersecurity, Ethical Hacking, Digital Forensics, and Financial Crime Prevention and Investigation. I conduct research in all of these, and run a non-profit that uses mobile applications and other technologies to create Early Alert Systems for natural disasters for people living in remote or poor areas.
Find more about me on Instagram, Facebook, Twitter, LinkedIn, or Mastodon. Or visit my non-profit’s page here.
