<strikethrough>There was a similar accident in Hiroshima, Japan several years ago: Airbus A320-200 skidded past the end of the runway at similar speed and struck down ILS. It eventually stopped -- damaged the airplane but no fatality.</strikethrough>
1. Localizer at Yeosu Airport, similar to Muan's, raises safety concerns, 2025.01.02 (23:58), KBS News.
Are you talking about Asiana Airlines 162? It hit localizer on its way to the runway because it came in too low. It then hit the runway, skidded on the runway, and stopped about halfway (after veering off the runway at the last moment).
If the same thing happened in Muan, the plane would have hit the localizer and then touched down, stopping in the runway. The fact that the localizer's base was concrete wouldn't have mattered because that's not where the plane would hit it.
The barrier was 250m away from the end of the runway, the extra 50m if following regulations wouldn’t have changed the outcome. And if the wall wasn’t there, the plane would dive right into a highway anyway. That’s the point.
I think the fact we have dramatic footage of the crash makes it a very attractive topic for engagement. News are spinning it into something it is not. A handful of massive errors happened and we know nothing yet, the focus on the structure for being in the way just makes for good clickbait, gives people an easy target to blame - the airport, engineers, regulators (doesn’t really matter who), and something to get riled up about.
It's not feasible to plan for every possible freak occurrence, an accident like this is only possible after a long list of other safety procedures have failed (as is often the case for aviation).
Planes sometimes overshoot the runway, building an unecessary wall at the end of it might comply with a standard, but that doesn't make it a good idea.
Where did the idea of fortifying Localizer (LOC) come from, was there prior art specifying any degree of fortification (or lack) ? Perhaps these kind of radar (adjacent) installations are traditionally fortified, like maybe the civil air services were influenced from militry air services were things tend to be overengineered?
Perhaps saving the LOC is more important than saving a single aircraft? Some idea like this piece of infrastructure being more important that any single aircraft's safety?
The point is we might not see the reason why fortifying the LOC was obvious or straight forward to the Korean engineers.
Also, I've seen no credible evidence to support the idea the LOC is supposed to be built on a level surface, or built in a destructable way. Addressing the first point, the LOC makes sense to be slightly elevated given how they operate with near field low power radio waves. That said, these kind of slightly elevated instrument might not require an earthen mound, sourounded by concreate walls. Which goes to the second point, I can think of reasons to have these kind of LOC runways homing antenas destructable, and I can see them being robustly durrable. This thing was way past the end of the runway, and as unpo0pular as it might seems... it's a very bad thing to run out of runway. Somebody else wrote that some airports have a lack of open field beyond the end of the runway, and so it's a persuasive argument stuff could be built out there.
<https://en.wikipedia.org/wiki/American_Airlines_Flight_1420>
Partially-effective runway braking and a much greater distance from runway perimeter to the nonfrangible ILS structure likely played a role.
That one touched down short of the runway at a much lower speed and configured for landing.
Nit: I think the word you want is “frangible”, easy to break.
I'm speculating, of course, but pilots made the decision to land there (albeit in a very short amount of time). They probably made the reasonable assumption that they could "safely" (as safe as it can be, of course) overshoot the runway in that direction. They were certainly not expecting to hit a concrete structure that would pulverize their plane.
Having large concrete structures near airports is not unreasonable, hiding them absolutely is. If instead of a hidden piece of concrete it had been a terminal like in SFO, a sea wall, or another known hazardous structure, the pilots could very well have decided to land somewhere else. Including in the very large body of water next to (or beyond) the runway.
You don't know, I don't know, and we might never know depending on what is uncovered by the investigation.
There was also construction work going on at one end of the runway (until March), and the threshold was pushed back 300 metres, shortening the runway by that much:
http://aim.koca.go.kr/eaipPub/Package/2024-10-31-AIRAC/html/...
The runway also is not flat (which is why the localiser beams at that end need to be raised in the first place to intercept the correct glideslope angle).
As the OP mentioned, trying this (a very fast landing, with no gear or flaps, spoilers) at many airports around the world on such a short runway (albeit one which with gear and flaps down is long enough for normal landings with the required 240 m runoff areas), is not going to work well.
Of course, I'm making the assumption that the pilots somehow had to attempt a "a very fast landing, with no gear or flaps, spoilers". The core of the issue is probably there, hopefully the investigation will yield useful results.
But what I am fundamentally questioning is whether the pilots would have attempted that landing if they had been expecting a piece of reinforced concrete at the end of the runway.
To say it differently, it's not the existence of deadly obstacles near an airport that bothers me (after all, some runways are quite literally in the middle of cities), but the fact that the pilots could have reasonably not know about them. That, for me, is a pretty big issue.
There were plenty of concrete structures nearby when US Airways Flight 1549 ditched into the Hudson river: notice the pilot aimed for a path where there weren't any. Maybe that Jeju air pilot could have attempted something similar. Maybe not. But the absurd nature of that deadly piece of reinforced concrete probably didn't help making a good decision.
The airport where it happened doesn’t appear to have any less clearance than usual around the runway [1], if not more when comparing to Jeju Airport for example [2].
[1] https://maps.apple.com/?ll=34.976342,126.382712&q=Marked%20L...
[2] https://maps.apple.com/?ll=33.515456,126.498733&q=Marked%20L...
I dont think the plane would get pulverized hitting a brick wall, and the distance will also slow the speed.
And the highway is not above the plane's travel axis, so the highway is a non sequitur. Not to mention even more distance to slow down.
People should stop doing this. Transport category airplanes are designed to suffer multiple failures and still be controllable. Why the airplane landed where it did, when it did, and how fast it did are the relevant questions.
I do hope the investigation yields results that improve air travel safety.
They are and will negatively impact the final impression of the investigation results, namely an unwarranted focus on The Wall(tm) (people here are calling it the "Murder Wall", which demonstrates my point) which helps precisely noone.
The focus should primarily be on the plane, particularly in the interests of preventing a repeat.
The cause of any fatality in aviation is never a single thing. It’s invariably a chain of events where removing any one thing in the chain would prevent the disaster.
But that doesn't mean debating whether there's a better way to engineer typhoon-resilient localizer antenna arrays isn't also a good use of time. Safety makes it imperative to discuss all of these matters exhaustively.
Re ocean, no, that isn't so great - sea rescue is a lot more difficult to perform than on land.
I am fairly sure this will be one of the findings of the investigation. I hazard a guess that every sane operator of an airport in the world is walking from the end of their runway to the airfield perimeter and taking a look anyway.
https://en.wikipedia.org/wiki/Engineered_materials_arrestor_...
There are cinderblock a cinderblock wall and numerous chin-link (and razor-wire-topped) fences in the area, which do control airport access. Those would likely not have proved fatal to the aircraft to any similar extent however.
Nobody knows yet why they landed in that configuration. Failed go-around? Engine out landing? Cut wrong engine after a bird strike? Loss of hydraulics? Too rushed for landing checklist in an emergency? All of those are possible. More than one may have happened. Wait for the flight data recorder data.
One article says the runway was equipped with EMAS, an Engineered Materials Arresting System.[1] This sits in the area just past the end of the runway, the part marked with painted chevrons. It's a thin layer of concrete over blocks of a material which includes foamed plastic holding pumice-like rocks. If a plane overruns the runway, the wheels break through the thin concrete layer and start pushing through the plastic/rock mixture, grinding the rocks into powder to absorb the energy. This usually damages the landing gear, but the rest of the plane survives. 22 planes saved so far.[2][3]
It didn't help here. The plane seems to have skidded over the EMAS area on its belly, instead of breaking through and getting the braking effect. The surface of the EMAS area has to be tough enough to survive jet blast on takeoffs, so it can't just be a sand pit.
[1] https://focustaiwan.tw/society/202412300010
[2] https://ops.group/blog/swerving-to-avoid-why-arent-we-using-...
[3] https://www.faa.gov/newsroom/engineered-material-arresting-s...
> https://focustaiwan.tw/society/202412300010
Nowhere in that article says Muan Airpot has EMAS, it says a local official confirmed that Songshan Airport in Taipei has EMAS, following local concerns that Songshan Airport has an even shorter runway (2600m vs 2800m in Muan).
A thread full of armchair experts is already bad enough, please don't make it worse with seemingly well-supported misinformation.
Muan does have the chevron markings on the overrun area, but that does not always indicate EMAS.
Seriously, I think the incident it's a hard lesson for airport designer and ICAO. For better civil aviation safety, the next airport runway should have ample room for safer aircraft landing without landing gears. Previously there's no real-time aircraft tracking requirement for passenger aircraft only for cargo, but after MH370 it's mandatory now and even ICAO acknowledged this very reason for the new regulations introduction.
Furthermore (this is pure speculation at the moment) I think chances are the crew were kind of cosplaying PIA PK-8303 - forgot about landing gears in a stress from bird strike, attempted go-around after realising it, but had not enough power from engines due to bird strike or ground hit. It's plausible final investigation report will conclude absence of localizer antennas wouldn't save them.
Most runway overruns occur with no or few fatalities:
That said, based on my observations of the terrain past the runway / airport threshold, it seems to me that absent the Muan Murder Wall, survivability would have been far higher in this case. See: <https://news.ycombinator.com/item?id=42607464>
I've yet to find a source that does a better job of organising incidents by landing speed / profile in a way which might better provide for more direct comparisons.
I don't think so.
It's simply not possible to build airports in useful places and guarantee three-mile runways.
Muan airport runway distance is one the shortest in Korea, less than 3 km and ironically during the incident reportedly there is an ongoing construction to increase the length of the runaway to more than 3 km, but effectively further shorten the runaway to 2.5 km (similar to Yangyang Airport). Strangely South Korea has many shorter runway international airports.
Most of the modern international airport have more than 4 km runway, and new major airports for example Qatar Doha, US Denver and SA Upington has runaway length close to 5 km.
[1] Operational Landing Distances: A new standard for in-flight landing distance assessment.
[2] Muan Airport runway previously shortened, impact under scrutiny:
These are by no means average "new major airports".
Denver airport (from 1989) is the west's largest airport (by land area), and at 5000 ft+ elevation (necessitating longer runways).
Upington in the far North-West of South Africa was built in 1968 to accommodate a full Boeing 747 flying to Europe non-stop during the apartheid regime when sanctions meant that overflight or stops in the rest of Africa were not feasible. It has one of the longest runways in the world due to the use case and hot & high environment at 2800 ft (and was intended as an emergency runway for Space Shuttles, if memory serves correctly). It is hardly used anymore with less than 20 aircraft movements a day.
There is no recent trend for longer runways. The issue is extremely well known and well understood, by and large.
You need the passengers to survive.
And that's eminently possible, even where the hull itself is destroyed (several cases of fuselages splitting in two or three with no or minimal fatalities):
Passenger survivial: Decelleration g-forces kept within a given threshold, evacuation slides operable, passengers cleared within 90 seconds. Hull is sacrificed.
Airframe survival: No significant damage to aircraft structure or systems.
Humans in this case are substantially more robust than aircraft.
You'll find a similar situation in, e.g., earthquake safety construction. The goal isn't for structural reuse, but for inhabitant survivability. Structures may be renovated in some case but are generally demolished and replaced. They did their job in saving lives.
Steel-reinforced concrete buildings can still sustain considerable damage, possibly to the point that they will be unusable after the quake. This has to do with the way governments set building codes, which tell engineers how to design a building to withstand a certain level of earthquake shaking. Codes, including those in the U.S. and Turkey, generally require that a building achieves what is called “life safety” under a given maximum expected earthquake in an area. “Our seismic codes are only a minimum requirement,” says Sissy Nikolaou, research earthquake engineer at the National Institute of Standards and Technology. “You just want these buildings at least to give you the chance to get out of it alive when the big one happens, under the assumption that they may be seriously damaged.” The situation is akin to a car that crumples in a crash: the vehicle absorbs the impact to protect passengers, but it is totaled.
<https://www.scientificamerican.com/article/how-to-engineer-b...>
The automobile-safety example given above is also apt. A car can be destroyed in a crash, what's key is that its occupants survive.
The much more reasonable alternative is to have 1) fuel-dumping systems which rid the craft of excess fuel prior to an emergency landing (another reason, BTW, to remain aloft as long as possible after an emergency has been declared), and 2) to have the means to evacuate the aircraft quickly. Modern standard is 90 seconds, with only half the exits in use. Initial fireballs, whilst impressive, usually occur largely above the aircraft and direct heat upwards. It's the subsequent ground- and cabin-based fires which are most lethal, and those tend not to develop for about two minutes.
<https://simpleflying.com/aircraft-90-second-evacuation/>
After that, you want the actual fuselage and passenger restraint systems (seats, seat belts) to provide maximum protection against injury in the crash itself. They largely do this. At least when aircraft aren't encountering Muan Murder Walls at speeds well in excess of 100 kt and disintegrating fully.
This is not how safety works.
> Something bad is going to happen if you do that at any runway on earth.
Then the question is can we do any sort of engineering to reduce the number of fatalities that might occur when this _inevitably_ happens.
> Stop looking at the wall and look at the minutes before the crash.
The plane hit the wall and exploded. The wall seems pretty important here. I mean, yes, there are also other problems to solve, but solving them does not let you off the hook here.
Accidents would still happen.
There are dozens of ways to solve this problem. From the short term of better structure engineering to the long term of better ILS antenna installations that don't require such large structures in such commonly dangerous positions in the first place. We could even get into better run off engineering to handle the somewhat unusual case of a fully gear up landing.
It's just insane to me to say "it's a red herring" as if this were a mystery novel and not an emergent failure of several safety mechanisms.
The bay is survivable and I don't think you can hit the terminal. You could possibly hit the freeway though. That said, two of the runways at SFO are 1.5 km longer than the one in Korea.
> In Toronto you'll crash into a highway.
That runway is 1.5 kilometers longer than the one in Korea and it's another kilometer to the highway that sits uphill.
We aren't talking about any of your examples in this crash. And it isn't relevant for many other places either. If you have an open field behind a runway and you put a concrete block directly at the end of it, you can't defend your decision with "well, in this other city it doesn't matter because you'll hit the terminal". It's some weird form of whataboutism that I simply don't understand.
It's inexcusable and it's tiring seeing people defend it as if it's okay.
https://www.pbs.org/newshour/show/south-korean-officials-wer...
A terminal beside the runway at roughly the same distance is not against regulations.
Almost every rule in aviation is written in blood, so if there's a rule about something, there's probably a damn good reason why.
Causality isn’t an equivalence relation with blame. A moral aspect has to be established.
Meaning if we can build the same antenna array but with less risk to airplanes and all at an acceptable economical cost, it feels like something we should do. Regardless of whether or not a runway overrun at other airports and in other situations poses more or less risk.
> In SFO you'll end up in the bay or hit the terminal depending on the orientation. In Toronto you'll crash into a highway.
Ending up in a bay or crashing into a highway would likely have resulted in far less loss of life.
Not sure if there's a regulation that requires this kind of arrangement, but I wouldn't be surprised if there are stricter rules about structures that cross a plane's usual trajectory.
Any length of runway you agree on can still fail. As you just said, it's an eventual certainty.
Rather than fixating on what didn't cause the crash how about we spend that energy on finding out why this flight unlike 99.99% of flights couldn't stop in the allotted space.
If they were gonna land halfway down the runway why didn't they just do another go-around? Did the thrust reversers not work? Doesn't the 737-800 have a backup way of dropping the landing gear?
There is no way to make a runway safe if a plane lands halfway down it, even if the brakes and landing gear are actually working. It's just not possible; runways are limited by geography and we tend to run aircraft as heavy as possible.
Aviation accident history definitely disagrees with you.
None of these people are calling for not investigating the other factors.
At the moment after encountering the Muan Murder Wall, there were 2 souls alive, one severely injured, and 179 corpses, most mutilated beyond all recognition.
Multiple things had clearly gone wrong with the flight, and we're going to have to wait for results of investigations to determine what crew and/or ATC actions and decisions contributed. But the principle lethal mechanism was impact with the immovable object of the Muan Murder Wall, and the ensuing instantaneous deceleration, disintegration, and conflagration of the aircraft and the souls aboard.
Even with multiple contributing factors, had the Muan Murder Wall not existed at that location, the aircraft would have overrun the runway and quite possibly airport perimeter, but would have slowed far more gradually and likely encountered structures less substantial than the Muan Murder Wall.
Wikipedia has a category page listing 55 runway overruns: <https://en.wikipedia.org/wiki/Category:Airliner_accidents_an...>
Sampling from that we find that such accidents often result in no or few fatalities, particularly on landing. E.g.:
- Sriwijaya Air Flight 062 (2008): 130 souls, 124 passengers, 6 crew, 1 fatality, 23 injuries, 130 survivors. The aircraft struck a house, 3 of the injured were occupants. The sole fatality occurred some time after the incident. <https://en.wikipedia.org/wiki/Sriwijaya_Air_Flight_062>
- China Eastern Airlines Flight 5398 (1993): 80 souls, 71 passengers, 9 crew, 2 fatalities, 10 injuries, 78 survivors. The aircraft experienced a tailstrike during a go-around attempt in heavy rain / high winds, broke in three, and came to rest in a pond. <https://en.wikipedia.org/wiki/China_Eastern_Airlines_Flight_...>
- Philippine Airlines Flight 137 (1998): 130 souls, 124 passengers, 6 crew, 0 fatalities, 44 injuries, 130 survivors. Ground casualties: 3 dead, 25 injured, as aircraft ploughed through a residential area. <https://en.wikipedia.org/wiki/Philippine_Airlines_Flight_137>
- American Airlines Flight 331 (2009): 154 souls, 148 passengers, 6 crew, 0 fatalities, 85 injuries, 154 survivors. Aircraft landed > 4,000 feet from the threshold with a tailwind in inclement weather. <https://en.wikipedia.org/wiki/American_Airlines_Flight_331>
- TAM Airlines Flight 3054 (2007): 187 occupants, 181 passengers, 6 crew, 187 fatalities, 0 survivors. The exception in my (random) sample, this aircraft had a nonfunctional thrust reverser on the right engine. Lack of grooving on runway, heavy rain, hydroplaning, asymmetric thrust, and a large warehouse directly beyond the runway perimeter all contributed to the fatalities.
I've omitted one link I'd selected, Air France Flight 007 (1962) as that incident occurred on takeoff, not landing, where fuel load and flight profile greatly alter conditions and likely outcome, and isn't directly comparable. <https://en.wikipedia.org/wiki/Air_France_Flight_007>.
If anyone cares to examine the 50 other listings on the Wikipedia category page, I suspect a similar patter of largely survivable overrun incidents prevails. The conspicuous lack of Muan Murder Walls seems significant.
Here's Google Maps view of the area south of Muan airport:
<https://www.google.com/maps/@34.9731352,126.3829389,1299m/da...>
From a road directly outside the airport, looking toward the ILS structure, we see that had the wall itself not been there the plane would have struck a cinderblock wall as it continued on. This would have damaged the aircraft, but less so than a solid concrete wall:
<https://maps.app.goo.gl/mMGqBC9PX6sEF85B9>
Switching directions we can look to the south along the path the aircraft would likely have followed. The terrain is flat and clear, save for further navigation light structures which would likely have given way readily to the aircraft:
<https://maps.app.goo.gl/Retvh9MH48ta5xPS8>
Note a hill in the distance. This could have helped slow the aircraft further, gently:
<https://maps.app.goo.gl/m1D3WrMG5QYz6cvE8> (above image zoomed in).
Vegetation is low trees and shrubs, which again could have provided a fairly gentle stopping force against the airframe:
<https://maps.app.goo.gl/aCBtY9at1Q1oCZXz6>
Approximately 300m or so from the end of the runway are a few rather unwisely-located pensions and hotels. Those would likely contribute to ground casualties if impacted.
Another few metres past those, mudlands and bay waters, which would be more emenable to a survivable overrun.
AA 1420 (1999) is notable for similarities with Jeju 2216:
The aircraft continued past the end of the runway, traveling another 800 feet (240 m; 270 yd), and striking a security fence and an ILS localizer array. The aircraft then collided with a structure built to support the approach lights for Runway 22L, which extended out into the Arkansas River. Such structures are usually frangible, designed to shear off on impact, but because the approach lights were located on the unstable river bank, they were firmly anchored. The collision with the sturdy structure crushed the airplane's nose, and destroyed the left side of the plane's fuselage, from the cockpit back to the first two rows of coach seating. The impact broke the aircraft apart into large sections, which came to a rest short of the river bank.
Captain Buschmann and 8 of the plane's 139 passengers were immediately killed in the crash; another two passengers died in the hospital in the weeks that followed.
145 souls, 139 pax, 6 crew, 11 fatalities, 110 injured, 134 survivors.
<https://en.wikipedia.org/wiki/American_Airlines_Flight_1420>
Even though this aircraft also hit an ILS structure, fatalities were far lower than those of Jeju 2216, likely as AA 1420 had decelerated significantly both on the runway (despite severely limited wheel and air brakes) and its subsequent 240 cross-terrain slide.
Atlantic Airways Flight 670 (2006) literally fell off a cliff. 4 fatalities of 16 souls. <https://en.wikipedia.org/wiki/Atlantic_Airways_Flight_670>
Bangkok Airways Flight 266 (2009) literally struck a (presumably nonfrangible) control tower. 1 fatality, 71 souls. <https://en.wikipedia.org/wiki/Bangkok_Airways_Flight_266>
https://www.ntsb.gov/investigations/AccidentReports/Reports/...
> The calculated ground trajectory indicated that the flight 1420 airplane departed runway 4R at about 97 knots and impacted the runway 22L approach lighting system support structure at about 83 knots.
97 knots is 112 mph. Somewhere below another commenter said Jeju Air 2216 left the runway at about 160 knots (184 mph). It's a pretty big difference.
I'm no expert, but my guess is that the main distinguishing factor of all the accidents where most/all survived is not the lack of killer berms, but the speed of the plane when it left the runway.
That's a more explicit restatement of my own "likely as AA 1420 had decelerated significantly both on the runway ... and its subsequent 240 cross-terrain slide."
Jeju 2216's lack of braking authority may have resulted from a dual engine outage, possibly a consequence of pilot error, noted here:
<https://news.ycombinator.com/item?id=42605837>
Absent the ILS structure, the aircraft would have had ~300-500m to decelerate across largely forgiving terrain before possibly encountering fairly light structures, and ultimately bay waters, detailed here:
<https://news.ycombinator.com/item?id=42607464>
As I've noted already, we're waiting on investigation conclusions to understand further, though it's entertaining to speculate, and IMO somewhat more productive to look at similar events and history.
The pilots didn't have to land there, they could have attempted a US Airways Flight 1549 rather than aiming at a piece of reinforced concrete.