Jeju Air flight 2216
On the 29th of December 2024, a Boeing 737-800 operated by a low-cost South Korean airline crashed after an emergency landing and overshooting the runway at Muan International Airport in South Korea.
Jeju Air flight 2216 was a scheduled passenger flight in a Boeing 737-8AS, registered in the Republic of Korea as HL8088, enroute form Bangkok, Thailand to Muan International Airport. There were 175 passengers, four cabin crew and two flight crew on board.
On approach, the captain made a mayday alert, declaring “bird strike, bird strike”. He offered no further information as to what had happened to the aircraft.
The flight crew announced that they were going around. They turned and attempted to land in the opposite direction of the runway’s operation that day, going north to south. The landing gear failed to deploy and the Boeing landed on its belly, skidding some 250 metres off the runway before crashing into a concrete platform used for the airport’s Instrument Landing System antennae.
The collision killed all 175 passengers and four of the six crew members. The only survivors were the two cabin crew seated in the aft jump seats.
I’d been waiting for enough time to do a review of this crash but with new details coming out every day, it’s been difficult to create a comprehensive summary. Instead, here is a collection of some of the discussions around this. I hope that you all will add links in the comments as new details are released!
Aviation Safety Network
The listing at Aviation Safety Network has all of the basic details and media sources for more information
FlightRadar24
Ian Petchenik at FlightRadar24’s blog has done a short post on the ADS-B (navigation) data reported by the aircraft before it went offline. Bearing in mind ADS-B data is not perfectly accurate, especially at low levels.
Korean Herald
Byun Hye-jin at the Korea Herald reported that both the South Korean 항공·철도사고조사위원회 (Aviation and Railway Accident Investigation Board) will be heading the investiation with support from the US National Transportation Safety Board (NTSB) and Federal Aviation Administration (FAA).
Four minutes missing from recording equipment
Robert Plummer from the BBC reported that both the South Korean and US analysis of the cockpit and flight data recorders showed the recorders were missing the four minutes leading up to the collision. This crucially includes the time of the bird strike and landing attempts.
Thanks to Mike for highlighting that the aircraft went into service in 2009 while the US regulations requiring backup power for data recorders applied to aircraft manufactured after April 2010, which helps to explain how the final minutes could be missing, presuming the power was taken out by the bird strike.
On PPRune, user tdracer, who apparently worked on 737 engine control systems offers a possible explanation for the overrun.
The 737 does not use Weight on Wheels (WoW) to enable the thrust reversers. This was done early in the program (737-100/200) because it was discovered the 737 was prone to float in ground effect enough that it could prevent WoW from going true after touchdown – same with wheel speed. So the 737 T/R uses radio altimeter to enable the reversers. This has been maintained over the various versions of the 737, since the fundamental issue that resulted in using radio altimeter rather than WoW never changed. It could also help explain the long float resulting in a very late touchdown.
While I was in the business long enough to know better than to say “never”, it’s extraordinarily unlikely that a bird strike (or other damage) could cause the reverser to deploy (as in something like 10-12/hr.). It would require the failure of multiple locks – including the ones added after the Lauda 767 crash. It’s clear that the RH reverser is deployed in the available photos/videos (I haven’t seen anything definitive on the LH engine). The 737 thrust levers use a mechanical interlock that requires the lever be at the idle stop before the ‘piggy back’ reverse levers can be lifted to deploy reversers. That same interlock will keep the forward thrust levers locked at idle so long as the reverse levers are not stowed – it’s almost inconceivable that a pilot could think they were commanding forward thrust with the reverse lever raised and the thrust lever at idle. In other words, it’s highly unlikely that pilots were attempting a go-around after touchdown.
I see no evidence of speed brakes deployed in what I’ve seen. Normally, commanding reverse will automatically deploy the speed brakes, but that feature requires the speed brake handle to be placed in an ‘armed’ position prior to landing – so if they had failed to perform the landing checklist (for whatever reason), that would explain the lack of auto-speed brakes.
From the limited available data, I think the speculation that the left engine was shutdown (possibly in error) or failed – and the right engine was damaged and not producing sufficient thrust to maintain flight – is the most plausible explanation.
I’m not convinced that they suffered only a single engine failure — I can’t help but feel that the decision to go around followed by a quick turn back to the runway was motivated by a partial failure of the other engine, but that, at least, should be clear from the physical evidence, even without the flight data.
The preliminary report is expected at the one month anniversary of the crash, which will hopefully offer more information soon on what exactly went wrong in this case.
“I can’t help but feel that the decision to go around followed by a quick turn back to the runway was motivated by a partial failure of the other engine”. I wonder about that given how far beyond the threshold they were; do recent 737’s with full-ish loads float as much as the early models described by tdracer; the pilots could have been giving themselves margin against the other engine completely failing, or they could have misjudged what they needed to do with one working engine, or they might have thought that coming in high would give them a better chance of avoiding a strike from low-flying birds. (There’s a lot of water around that airport, probably with a fair number of waterfowl inured to jet noise.)
One thing not mentioned: the plane might not have smashed nearly so badly if it hadn’t run into a solid wall beyond the far end of the runway: see https://www.bbc.com/news/articles/c0mvynnxzzmo. The ~bunker contained the localizer, which has to be on-axis with the runway, but there’s argument that it should have been in/on a structure of steel girders (that would have broken away when the plane hit them) rather than behind a solid wall that stopped the airplane cold.
I haven’t seen any discussion of whether there was a proper runway end safety area (RESA) that would have slowed the plane after it overran the far end of the runway; the regions of what look like gravel followed by plowed ground (which may not be appropriate for RESA) on Google Maps look shorter than the 300 feet the ICAO calls for — and there’s argument that even 300 feet isn’t enough. (What looks like grass beyond this area might or might not count; Wikipedia doesn’t discuss materials.)
If something went badly wrong enough to cut off the recorders four minutes before the crash, it would almost certainly have been noticeable in the cabin, and there would have been passengers taking photos and video. It may take some time to recover that data but I would expect it to play in important role in the investigation.
Given the extent of the ensuing fire, anything not carried out with survivors or ejected from the aircraft on crashing may well be unusable, sadly.
Several sites have reported that there have been found signs of bird strike in both engines, although it hasn’t been officially confirmed. This may explain the loss of FDR data which seems to correlate with bird strike timing, and also why they attempted to land precipitously rather than do a proper single engine go-around.
There are many mysterious losses in aviation. Sometimes the aircraft is lost. Sometimes the pilots are disoriented at night or in IMC. Sometimes disaster strikes from outside or inside. But rarely do we have such a mystery in broad daylight within sight of an airport. Speculation abounds.
Two posts by “wonkazoo” on pprune impressed me. Here’s an excerpt of the first:
https://www.pprune.org/accidents-close-calls/663324-jeju-737-800-crash-muan-airport-south-korea.html#post11806856
I suppose that after one engine was damaged by the birds the pilots were afraid that the second engine would do the same.
The pilots would have been expected to empty the kerosene tanks before proceeding to a potentially forced landing. The plane would have been lighter by a few tons and therefore easier to stop and the size of the fire would have been perhaps minor. Here, however, I think panic set in, namely the pilots’ fear of losing the other engine. Loss that could occur at any moment amid the flocks of birds seen by the pilots. This fear prompted the rush to land.
But did the pilots know the state of the runway? Meaning the length available for landing as well as the existence of the concrete block at the southern end of the runway? Not from maps or documents, but from the express warning of the control tower when the pilot was given permission to land on runway 19. If the pilot had known what was waiting for him, maybe he would have avoided landing and today the 175 were alive people.
Did the control tower watch the entire landing process?
The control tower knew that there were flocks of birds in the airspace of the airport, it knew that there was only one person in the field to scare the birds, it knew that runway 19 had a concrete block at its southern end, and that the length of the available runway was shorter.
Why is there no obligation for landings to be watched for their entire duration by a person in the control tower? Landing is a critical period, as is take-off.
Why aren’t drones used to repel birds? It would be the most effective solution.
Why don’t airports have emergency runways specially designed for problem landings? Quite long and provided at the ends with specially constructed obstacles, i.e. with nets to ensure a less harsh impact. Or with runways in the form of water channels or lakes for a landing on water, therefore less harsh.
Why are aircraft not equipped with parachutes designed to reduce landing speed? The American space shuttles have demonstrated their effectiveness.
Why don’t pilots have a computer-provided voice breakdown guide instead of flipping through pages of thick, printed manuals looking for solutions to get out of the jam? Especially when the on-board instruments and cockpit lighting run out of power.
Such a guide could assist an aircraft crew in determining the most rational steps to take. Additionally, and most importantly, this guide can be updated very easily. And it should not be neglected: the man in case of panic reacts poorly.
Why are there no ejection seats or parachutes on aircraft to allow the ejection or rescue of passengers before the aircraft hits the ground or water? For the case when the aircraft can no longer be controlled. And also of the two data boxes.
Why can’t pilots see their own engines or controls in the back of the aircraft? If he could see them he could draw the necessary conclusions.
Why can’t pilots see pictures of the passenger cabin? To be able to get information quickly in cases of fires or terrorist attacks.
In cases where the pilots can no longer operate the control levers for various reasons, maybe a joystick would be useful.
If the captain or first officer becomes unavailable there should be a 3rd pilot available on multi-passenger, long-haul aircraft.
The condition of the pilots before boarding should be mandatory tested. A tired, sleep-deprived pilot should not be cleared to fly.
To avoid suicidal actions by the pilots there should be a back-up secret code kept on the ground that allows the cockpit door latch to be opened when needed and the door cannot be locked manually by the pilots
Do not approve taxiing on runways already occupied by an aircraft until after its take-off (see the case of Tenerife).
Traffic controllers must constantly follow the movement of aircraft on the airport runways by GPS and the images must be recorded as evidence.
A runway can be offered by the control tower for a landing or a take-off only after the movement of a control robot/drone on that runway to avoid collision accidents with objects reaching the runway (see the case of Air France’s Concorde).
There are a lot of … questionable … ideas in the above. I’ll take on one of the most obvious:
“Why are there no ejection seats or parachutes on aircraft to allow the ejection or rescue of passengers before the aircraft hits the ground or water?”
Aside from the problem of making the entire top of the aircraft detachable so that the passengers’ skulls wouldn’t be crushed on ejection, there’s the fact that ejection seats require that the ejectee be solidly strapped in, not just wearing a lap belt. There’s also the question of how many random civilians can survive the force of an ejection, and the landing afterward. Add to this the dangers of carrying a couple of hundred solid-fuel rocket motors on every passenger flight, and the fact that many crashes have survivors (cf the plane the Russians shot ~down in last week’s FoL), some of whom might well have died if they ejected.
Also the extra weight and space needed for ejection seats would severely cut into the number of passengers and cargo….and fuel….able to be carried
Point. And there are probably other anti-features I missed….
This is normally done when an aircraft makes an emergency landing shortly after its trip has started and its tanks are still full. Jeju Air was scheduled to land at Muan und thus was not overweight.
The pilots knew the state of the runway, there was nothing wrong with it. Specifically, it was long enough to support an emergency landing. Here, the pilots’ problem was that they could not get the aircraft to actually touch the runway.
The concrete embankment was not “at the southern end of the runway”. Beyond the end of the runway were 300m of clear overrun. This is mandated at airports everywhere and helps with smaller pilot mistakes, like aborting a take-off too late. Most airports anywhere in the world are not prepared for aircraft that are not able to slow down or touch the ground and not able to execute a go-around under these conditions.
There is a strong safety culture in commercial aviation. Rest assured that all of your suggestions have already been evaluated by capable engineers.
More grim news: the person ultimately responsible for the structure the plane smashed into is dead, presumed a suicide: https://www.nytimes.com/2025/01/22/world/asia/south-korea-crash-executive-dead.html
Unfortunately, there have been several cases in the literature wherein a pilot of a twin-engine aircraft, having been notified of an engine failure, has then shut down the wrong engine – the one that’s still working fine.
Ms. Wright would know.
In most cases, that error can be detected and often corrected, but in this case at extremely stressful conditions requiring split-second work that may not have happened.
This could explain the total power failure, as well as hydraulic failure. They have little wind-driven pumps to run the hydraulics (sort of – see the “Gimli Glider”), but as far as I know they take time to spin up and are not perfectly effective even then.
A hydraulic failure would explain the lack of flaps and gear. A total power outage would explain the gaps in the record and the radio silence.
It’s also possible that both engines failed on their own. J.
On 24 January, https://simpleflying.com reported “… investigators found bird feathers and blood in both engines …” Also on 24 January, https://aviation.direct reported “Feathers and blood stains were found in the engines of the crashed plane.”
Aviation.direct also reported—same date: “South Korea’s National Biological Resources Agency (NBRA) confirmed that a total of 17 samples, including feathers and blood stains, were discovered in the plane’s engines.”
From this tragedy that took the lives of some people, conclusions must be drawn to help avoid repetition. Here are my thoughts:
Guilt of the control tower
The flight controller knew of:
the presence of the flock of birds,
he knew that runway 19 was shorter due to repairs
knew of the existence of the concrete block at the end of runway 19 on which the ILS antenna was mounted.
So he was only allowed to clear runway 19 from the end with the ILS antenna. The reason: a bird malfunction like it did would have caused a forced landing, and a forced landing should not have been allowed on a runway known to be dangerous.
Guilt of pilots
Even if the pilots had data they should have asked the air traffic controller for explanations regarding:
the presence of flocks of birds in the airspace whose existence in the area they knew about;
the condition of the runway on which the plane could have landed under normal or forced landing conditions.
Not having this real and weather data, because later it was too late, the pilots could not choose a safe landing option before the collision with the birds.
What the pilots had to do:
to ask the control tower about the presence of birds in the airspace of the airport;
ask the control tower for data on the status of the runway authorized for landing (available length and objects beyond the end of the runway). Such data must have been known by the pilots before landing.
Guilt of the bodies authorized to check the state of the airports
These bodies have allowed the operation of Muam airport without permission despite the existing insecurity conditions, conditions that have also contributed to the tragedy of those who died in terrible conditions (burned alive or mutilated) and to the irreparable suffering of the families.
Perhaps some of us are indifferent to this tragedy, not being directly affected. However, it seems that we should reflect more on the ordeal these people were subjected to. And let’s help with what we can so that something like this doesn’t happen again.
Stefan
People deal with catastrophe differently, and I understand your contribution in that light.
However:
• the air traffic controllers did a good job,
• the crew was very experienced and dealt with very difficult circumstances under a lot of pressure,
• Muan’s airport safety is and was up to international (ICAO) standards.
The blame you are trying to assign has no basis in fact, and runs counter to established facts.
That is exactly how the aircraft was cleared to land. The pilots did not land (go around), and then were forced to turn around and try to land in the opposite direction.
Please note that for a landing to be safe, an aircraft should 1) land against the wind, 2) make use of the ILS in reduced visibility (IMC). For this, the ILS antenna needs to be on the opposite end of the runway. These considerations impact many flights daily that land safely as a result. The fact that there is a low concrete embankment 300m (1000 ft) past the end of the runway only impacts flights that have already gone terribly wrong. Generally, aircraft are able to stop in time, or climb above the obstacle, and airports all over the world are designed with that in mind.
The prelimary report is out, but there’s not a lot in it that we don’t already know.
• The crew had 6000/1500 hours on type.
• Both engines collided with Baikal Teals, a duck species that weighs 1 pound on average.
• NTSB (USA) and BEA (France) are assisting the investigation.
Following the bird strike (and loss of thrust from both engines), why did the pilot(s) choose to go around–from landing on runway 01 to landing on runway 19?
As one of the previous comments said, given the lack of other information, we should assume the pilots had a reason for all of their actions.
It seems most likely that they went around because they didn’t think it would be safe to land directly, then found they had lost power and needed to land as soon as possible.
Are our opinions, made known thanks to the kindness of your site, read by any of society’s decision makers? Or are they just a valve offered to the vulgar for expression in the absence of other concerns, such as mass participation, man next to man, in a football game.
In my youth, I lived through a period when an action called brain storming was initiated where everyone participated freely with an opinion and these opinions were then analyzed to obtain some conclusions. It was then a great idea, a sign of society’s ability to understand the value of the ideas of unknowns, unknown yes but capable.
Personally, this is how I see the point of these opinions received by you. with permission and disseminated to be known by others.
However, the truths regarding this tragic event, especially for the relatives of the deceased who will be marked for the rest of their lives by the suffering of losing loved ones, should be known and used to prevent another tragedy. Interventions of excuses or hiding the truths will not solve the shortcomings that contributed to the accident.
The human factors that contributed to the accident are the pilots, then the controller in the control tower and the Board of Directors managing Muan Airport.
1. Pilots were not allowed to enter the landing procedure before knowing whether or not there was a bird hazard. He had to fly to another airport. Point. If, however, landings were made in the presence of birds without consequences, those in those planes can declare themselves lucky, today they can understand how lucky they were. At this point the air traffic controller did his duty, which was to send the bird alert to the pilots.
2. The flight controller was not allowed to release the landing due to the presence of birds, a presence announced by him through the alert sent to the pilots. The reason is now easy to deduce: the birds could have caused the accident which is now proven.
3. Did any member of the airport board ever go to Runway 19 to see the presence of the concrete block supporting the ILS antennas? If so, then he would have intervened with his authority to dismantle that concrete block in a maximum of two hours, when he would have had to bring in a team of builders to do the job. So, today, the members of this council do their duty, but too late.
I stop here. There are of course other culprits who had a supervisory role so that the failures of others would not go unnoticed.
In conclusion: if at least one of the human factors listed above had done its duty, the tragedy of those who died in terrible agony and the suffering of the living marked for the rest of their lives would have been just a story told to school children.
Finally, I have a question: we people today really can’t find judicious solutions as adults, many of us being parents of children?
Stefan
Can you please stop posting uninformed speculation? This horrid tragedy is under investigation; what we know is that both engines had bird strikes based on the evidence, and that the airport met ICAO standards for RESA (from what I have read).
Speculating that someone from the airport should’ve summarily demolished compliant infrastructure, or that ATC should’ve changed procedures on the fly is not helpful. Changes to these things absolutely need to be considered, but without a good understanding of the circumstances, it is impossible to make appropriately considered and safe changes.
I believe the events leading up to the tragic impact of Jeju Air Flight 2216 went like this:
Immediately after the pilots started the landing procedure an engine was hit by the birds and went out of service, after which the pilots announced My Day. They could no longer continue the landing due to the time spent in finding the engine failure and notifying the control tower, during which time they did not have time to reduce the speed to the limit necessary for a safe landing. In this situation the pilots abandoned the landing and accelerated and banked for a turn to the other end of runway 19 to resume landing on runway 19. I think they were afraid to go to another airport for fear that the birds might have struck and the other engine. During this maneuver the bird strike and the other engine occurred, followed by its out of service, with the consequences of the total loss of the electrical energy source. From this moment the situation of the pilots became desperate. They were only able to get the aircraft level for heading to runway 19 by increasing speed through a steep descent to the runway. There was no time to remove the landing gear, and even if they had been able to manually operate it, there would have been no time to verify that the operation was successful, plus without engines they could no longer count on the brake reversers. The pilot-in-command believed he took control of the aircraft and managed a master landing on the belly of the aircraft which glided down the runway to the killer obstacle without incident. During this entire period of time from the first bird strike to the tragic end, the pilots behaved exceptionally well. However, the friction of the metal fuselage against the concrete runway was not enough to reduce the speed of the aircraft until it collided with the concrete massif of the ILS antenna. The metal I believe heated up to incandescence after which the specific coefficient of friction between metal and concrete decreased greatly.
The above are only assumptions, built on the basis of information gathered from what has been published in the mass media
Pity for the two exceptional pilots and of course for the tragedy of the passengers in the cabin. God rest them in peace!
However, no one can erase the suffering of the relatives of these victims.
In addition to the other causes that contributed to this tragedy, we must not forget the care taken even to the point of ridiculousness to avoid a possible logical risk. When the presence of birds in the airspace of the airport was detected, the access of any plane to this space had to be firmly prohibited.
It would be interesting to know how many passenger aircraft landed in Muan in the year 2024 alone with birds present in the airport’s airspace. All those people could have the fate of those who were on board Jeju Air Flight 2216 on December 29 2024.
Stefan
Please, leave the investigation to the authorities.
We know this:
1) The CVR and FDR lost electrical power.
2) The pilots reported a bird strike, declared an emergency, and announced a go-around.
3) They executed the go-around and turned left as per the “missed approach” procedure.
4) The pilots aborted following the missed approach route, and turned back to the airport.
5) The aircraft was not configured for landing, and touched down very late, fast, and with the gear up.
My speculation:
• The accident started with a bird strike to both engines (and possibly other systems) which impaired the electrical system for reasons that will need to be investigated.
• I expect the pilots felt the engines were still working, and decided to go-around, re-gain altitude, and sort out the electrical problem.
• I expect that, when TOGA power was applied, that caused the engine damage to become apparent, or possibly made it worse.
• The pilots then decided to land immediately.
I am looking forward to reading the conclusions of the accident investigation team in about a year, and finding out whether my guess is close to the facts or not. I am also looking forward to the safety recommendations that the final report will contain.
The airport and the air traffic controller did not cause this accident. Birds damaged the aircraft, posing a problem for the aircrew that they were unable to solve in the time they had, although they did their best to try. May they rest in peace.
Thanks for your thoughtful input. I come here to read the informed commentary; always hard to discuss these things without speculation though!