I'd like to approach this in a sort of hard sci-fi way, where the following is grounded in something plausible.

Suppose we want to take a plane down at a reasonable speed where there is a 30% chance or so of surviving, or rather, something that isn't going to be fatal like a straight on nosedive. While I'm sure there are people who defied the odds to be a 0.001%, I am not looking for such low chances because it would likely break immersion.

I'm trying to find a way to damage a craft so that it would crash, but not in a guaranteed fatalistic way. However, it also has to be realistic.

For example, what if something rips off a wing? How realistic is it that it would spiral out of control without going into a nosedive all the time?

Can I rip off the tail of the plane, and it would descend to the planet?

What about damaging any turbines?

While I marked this as reality-check I am okay if we fudge it a little bit. I don't need a journal article saying what will happen (but that would be nice if one existed describing this case!) or even better, if someone has a video of something.

Ideally I'm trying to do the 80/20 rule, or get rather 80% of the way there with something believable, and then make the last 20% sort of "X survived" because I'm writing a fictional story. The last thing I want is for someone to say "there's no way this would happen" or "immersion broken" or "plot armor" (while the primary character will have to have some plot armor for the story to move forward, I want it to be invisible as best as I can make it).

For the sake of this post, let us limit it to current day technology, but if needed, we could tack on 100 years into the future.

  • 2
    $\begingroup$ Hello Water. Thanks for asking your first question on Worldbuilding! Those pesky tags can be a pain. Thanks for removing the hard-science tag, as you probably realized, it doesn't refer to the "hard science" genre, but demands that all answers be backed up with math, citations, etc. You might want to reconsider the reality-check tag, too, as it's meant to be used in this way: you provide a complete scenario based on the rules of your world and we evaluate that scenario to see if it conflicts with the rules of your world. I think science-based would be right for you. $\endgroup$
    – JBH
    May 7, 2021 at 3:49
  • 5
    $\begingroup$ a 30% chance or so of surviving Is that a 30% chance of everyone surviving or a chance that 30% of the passengers and crew will survive ? Big difference. $\endgroup$ May 7, 2021 at 7:26
  • 2
    $\begingroup$ This may at least partially depend on the plane itself. For example, there are numerous anecdotal reports of a damaged A-10 Thunderbolt II limping back to base with only one engine and significant damage to the airframe, but the same cannot generally be said of most civilian aircraft. $\endgroup$ May 7, 2021 at 12:01
  • 2
    $\begingroup$ It looks like your world is already build and you're asking about a scenario in your world. This is probably a better question for aviation.stackexchange.com $\endgroup$
    – sphennings
    May 7, 2021 at 17:00
  • 3
    $\begingroup$ @sphennings I don't think they accept that question there either. I asked a very similar question (the most amount of damage a transatlantic plane could sustain while still being landable without casualties) and it got closed. For context: aviation.stackexchange.com/questions/55966/… $\endgroup$
    – Nzall
    May 7, 2021 at 19:22

8 Answers 8


Damage to the flight control surfaces

(Or the hydraulics / electronics used to control them)

I'm referring to rudder, ailerons, stabilizers, etc. The big hydraulic surfaces that move into and out of the airstream in order to manoeuvre the aircraft.

enter image description here
(China Airlines flight 006 nearly didn't make it)

The damage needed to cause an absolute emergency situation is usually pretty minimal - you need to at minimum break the smooth airflow over it, but can also remove part of it or lock it in place. It sounds like your looking more for long distance attack rather than a mechanical failure - in which case a few tightly packed bullet holes should cause a secondary failure taking out the majority of the panel, causing a partial loss of control. However if that's not the case mechanical failure or pilot error can also cause it.

Manoeuvring becomes very hard, but with skill and raw determination they can be survived. I think it's just the roll of the dice you want.

Examples (sorted from luckiest to deadliest):

  • 100% survival: A Concorde in 1989 lost part of it's rudder, and they didn't notice until landing in Sydney.

  • 100% survival: 2003 Iraq DHL flight. Another amazing outcome. Missile impacted wing. Total loss of all control. Landed safely (in a minefield) on differential engine power alone.

  • 100% survival: Northwest airlines flight 85. Hydraulic failure caused rudder to lock in hard left in 747. Landed successfully.

  • 70% survival: UA232 in Sioux city Iowa: turbine failure took out rudder hydraulics. Tried to land on differential engine thrust. Failed. Explosive fireball. Cartwheeled down the runway. 112 died. 184 people survived this:

enter image description here
UA232 crash and fireball, and cartwheel. Surprisingly survivable.

  • 50% survival: 4 x 737 rudder hard overs in the 1990s. 2 fell out of the sky and everyone died (United Airlines Flight 585 and USAir Flight 427), at least 2 landed safely (Eastwind Airlines Flight 517 and MetroJet Flight 2710 and possibly more where the cause is suspected but never explained.)

  • Floor collapsing taking out controls. 20% survival. Turkish Airlines Flight 981 was 100% fatality with 346 died. But when it happened on American Airlines Flight 96 everyone survived.

  • 1% survival: Japan air flight 123. Bulkhead ripped and damaged hydraulics. Plane limped around for nearly an hour before hitting a mountain. 520 died. 4 survived. At least 50 more would've survived if rescuers got there sooner.

  • 0% survival: American Airlines flight 587. Pilot over stressed the rudder and ripped it right off. Everybody died.

  • 0% survival: American Airlines Flight 191. Detaching engine severed some flight controls, but more importantly disabled the warnings that those controls didn't match. 100% fatality on board (271) + 2 people on the ground.

There's a whole Wikipedia page about this particular failure. Skimming of the list there it looks like the rough odds of survival for this event is about 50%, but it could be argued up (non-notable events aren't mentioned) or down (there are crashes with unknown causes).

  • 1
    $\begingroup$ As I noted above, survivability particularly for this type of failure, is believed to be highly influenced by the flight crew's training and skill, as well as how advanced the flight assist electronics are (autopilot, warning indicators, etc.). $\endgroup$ May 7, 2021 at 13:59
  • 1
    $\begingroup$ The problem here is that control and stability failures tend not to create ergodic situations. For example, in your 737 rudder hardover cases, either everyone dies or everyone survives - you don't get the sort of crashes where 50% survive, even though there was a 50% survival rate for this class of accident as a whole. Problems with engines or navigation are more likely have outcomes with an elevated, but not certain, risk of death. $\endgroup$
    – sdenham
    May 7, 2021 at 20:07

Examine subsystems:

  1. Airframe - if the airframe fails at any appreciable altitude, the chance of anyone surviving is negligible (unless they can eject and parachute to safety). So don't have a wing, tail etc fall off completely or it's game over for everyone on board. Limited failures are potentially survivable - for example, a wingtip could fall off, or the rudder.
  2. Aircraft controls - there have been occasions where pilots have lost some controls. One of the most notable occasions was where the pilots lost most flight controls and were only able to steer and eventually crash land through differential control of the twin engines. Note that loss of significant aircraft controls on an inherently unstable aircraft (ie a fighter relying on fly-by-wire for maximum agility) will not be survivable.
  3. Propulsion and fuel - this is your best bet for a survivable crash. Engine problems (which can be caused by birdstrike, as in the example L.Dutch gave) including running out of fuel when the airline converts from imperial to metric measurements will require a forced landing. The forced landing may be easily survivable (within easy glide range of an airport), potentially survivable (water landing on still water as in the Miracle on the Hudson) or hideously dangerous and hard to survive (mountains, forest), with survivors engage in cannibalism of those who did not survive the crash.
  4. Loss / injury of flight crew (includes depressurisation) - this is a common trope in Hollywood, but exceedingly rare in real life. Mythbusters examined the concept and came to the conclusion that in a sufficiently advanced aircraft in communication with the ground survival is trivial, where in other circumstances an untrained person will catastrophically crash the aircraft the vast majority of the time.
  5. Loss of ground communications and navigation instruments - assuming that alternative means of navigation such as visual observation are unavailable, this will also lead to a forced landing eventually when fuel runs out, as per 3 above.

I would strongly suggest viewing Mayday / Air Crash Investigation episodes and find a real life occurrence that resonates with the story you wish to tell. Truth really is stranger than fiction.

  • 2
    $\begingroup$ I like the recommendation of modeling it after an actual incident. You won’t have to ever question how realistic you have been and would be an interesting bit of research to share later as well. $\endgroup$
    – Eric G
    May 7, 2021 at 17:12
  • $\begingroup$ On #4--that's assuming someone on board realizes the problem in time. With clouds below an incapacitated flight crew could overfly their destination and head out to sea. If too much fuel is burned before it's realized they won't make it back even with communication. $\endgroup$ May 30, 2021 at 0:23

You can give it a try to something like the miracle of the Hudson

On January 15, 2009, US Airways Flight 1549, an Airbus A320 on a flight from New York City's LaGuardia Airport to Charlotte, North Carolina, struck a flock of birds shortly after take-off, losing all engine power. Unable to reach any airport for an emergency landing, pilots Chesley Sullenberger and Jeffrey Skiles glided the plane to a ditching in the Hudson River off Midtown Manhattan. All 155 people on board were rescued by nearby boats, with a few serious injuries.

At 3:27:11 during climbout, the plane struck a flock of Canada geese at an altitude of 2,818 feet (859 m) about 4.5 miles (7.2 km) north-northwest of LaGuardia. The pilots' view was filled with the large birds; passengers and crew heard very loud bangs and saw flames from the engines, followed by silence and an odor of fuel.

at 3:31 pm, the plane made an unpowered ditching, descending southwards at about 125 knots (140 mph; 230 km/h) into the middle of the North River section of the Hudson tidal estuary, at 40.769444°N 74.004444°W on the New York side of the state line, roughly opposite West 50th Street (near the Intrepid Sea, Air & Space Museum) in Manhattan and Port Imperial in Weehawken, New Jersey. Flight attendants compared the ditching to a "hard landing" with "one impact, no bounce, then a gradual deceleration".

The NTSB used flight simulators to test the possibility that the flight could have returned safely to LaGuardia or diverted to Teterboro; only seven of the thirteen simulated returns to La Guardia succeeded, and only one of the two to Teterboro. Furthermore, the NTSB report called these simulations unrealistic: "The immediate turn made by the pilots during the simulations did not reflect or account for real-world considerations, such as the time delay required to recognize the bird strike and decide on a course of action." A further simulation, in which a 35-second delay was inserted to allow for those, crashed. In testimony before the NTSB, Sullenberger maintained that there had been no time to bring the plane to any airport and that attempting to do so would likely have killed those onboard and more on the ground.

  • $\begingroup$ The problem with that is that you have to ensure the pilot is remarkable enough to pull off the stunt. $\endgroup$
    – Mary
    May 7, 2021 at 3:30
  • 3
    $\begingroup$ @Mary, despite what Hollywood likes to show in its movies, I think a trained pilot is needed for any landing, be it routine or emergency. $\endgroup$
    – L.Dutch
    May 7, 2021 at 3:49
  • 2
    $\begingroup$ If the incident occurs where there's flat ground to land on, the ditching part (which is the miracle) isn't necessary. There are lots of incidents of airliners landing without power (Gimli glider, a similar out of fuel incident over the Atlantic, etc.) with minimal or no casualties. All engines out is the classic survivable forced landing scenario. $\endgroup$
    – Zeiss Ikon
    May 7, 2021 at 12:48
  • $\begingroup$ There's a difference between "trained" and "capable of ditching." (Though if "flat land" also works and more easily, that would be easier.) $\endgroup$
    – Mary
    May 7, 2021 at 12:53
  • 2
    $\begingroup$ @Mary: I think you over-estimate the remarkableness necessary. $\endgroup$
    – jamesqf
    May 7, 2021 at 16:03

Depends on the aircraft (there is a huge difference between military, big civil airliner and small civil plane), if you want to make the crash (almost) certain or just possible, and especially if you want to have the damage made in flight or before takeoff (preferably without anyone noticing).

Some ideas, of various severity levels, mostly relevant to big airliners:

  • Damage the landing gear. If it does not descent, pilots have to belly-land. This is a minor incident, without much probability of life loss. Perhaps you can weld a clamp across the landing gear that permanently shuts when the gear is retracted (so that the takeoff is normal). If you can fiddle with the landing gear indicator so that the pilots believe it works, they will slam the aircraft really hard and we move from a minor incident to a potentially very serious crash.
  • Puncture the tires (make a small hole just before the departure, or rig some nails inside the retraction mechanism or something). The landing gear will collapse during landing, making the crash really dangerous (but fairly survivable).
  • Disable the flaps. While it is possible to land without flaps, the landing will be hard and you need longer runway. You have to do this in-flight, otherwise the pilots will notice it during the takeoff and abort.
  • Disable the engines in-flight, e.g. by creating fuel leak. Modern (or even not so modern) aircraft does not glide well at all and the Hudson river or Gimli glider landing need considerable piloting skills, a bit of luck and an accessible airport (or river) to land. See the Ethiopian Airlines Flight 961 for the more expected outcome (though it might have been better without the hijackers). The problem is that the pilots will notice the fuel leak, divert to the nearest available airport and land normally. Or they might not...
  • The best bet is to disable the brakes and reverse thrust (or just the brakes). This might not be noticed during the takeoff, but the airplane will almost certainly overrun the runway during landing and crash into whatever is beyond the runway, with the results ranging from "unpleasant" to "deadly".
  • $\begingroup$ Puncturing the tires won't cause a major crash. There have been a number of incidents where an airplane landed with the brakes locked (leading to immediate blowouts of all tires), with the only damage being to the lower section of the landing gear. $\endgroup$
    – Mark
    May 7, 2021 at 21:14

A plane can land with a single wing.

Israeli Air Force’s F-15D named, ‘Markia Shchakim’ and piloted by Zivi Nedivi collided with an A-4N. One of the wings was ripped off.

One winged plane.

They still managed to land. So long as you have enough of the wings and body and all to keep it up, they can make a landing less rough. You need a very skilled pilot, and enough luck that not too much goes wrong, and it helps to be in a plane with a powerful jet so you can go fast, but it can be done.

There's a good chance a load of the plane will get ripped up on the landing, and you're fucked if you need to fly far, but if you're in a good location and not too much of the wings are ripped off, you might be able to survive.

  • 6
    $\begingroup$ Fighter planes and passenger planes are very different, though. $\endgroup$
    – Philipp
    May 7, 2021 at 11:28
  • 2
    $\begingroup$ Not comparable at all to a commercial aircraft: the wing of an F-15 is terms of area versus size of fuselage is huge compared to any commercial aircraft, the horizontal tailplanes are likewise huge and provide more lift/control on their own, the design of the fuselage and engines on the F-15 provide some lift on their own, and the engines themselves are proportionally so overpowered there's no comparison. NO commercial aircraft could survive loss of a wing. $\endgroup$ May 7, 2021 at 15:38
  • $\begingroup$ And if I'm not mistake, these planes are designed to be unstable from the beginning, only kept stable by advanced software which would then do whatever it can to keep it upright for as long as possible. Not sure if that's true for the F-15 though. $\endgroup$
    – pipe
    May 7, 2021 at 16:09
  • $\begingroup$ Op said a robust military transport plane, so the exact shape hasn't been defined. $\endgroup$
    – Nepene Nep
    May 7, 2021 at 17:40
  • 2
    $\begingroup$ No, it can't. The F-15 is a freak exception because of its unique flight characteristics. The missing wing WAS corkscrewing the plane to to the right... but because it is made for 10 G maneuvering, the F-15 had enough aileron, elevator and rudder authority to corkscrew it back the other way. Any other plane would be dead. $\endgroup$ May 7, 2021 at 18:20

Sabotage the Fuel System

I'd say the most reliable way to do this is to sabotage the fuel system of the aircraft, such that some of the fuel is unusable, but the flight crew doesn't notice this until out of range of a suitable airport. This obviously also depends on the route of flight; if it's over ocean, or uninhabited territory with no airports such as a desert, it'll be much easier to plan.

A flight over mountains would make it significantly harder to arrange a survivable forced landing; either open water or reasonably flat ground is greatly preferable. Landing on a road very often counts as a "survivable crash" since the cleared area either side of the road may be less than the wingspan of the aircraft, and cars using the road have little or no warning of the aircraft's approach, which will be at highway speed or greater.

The flight crew will normally check the fuel loading as part of the pre-flight inspection. This often includes a simple check of fuel quality, so you can't simply fill some of the tanks with water. However, on an aircraft with multiple fuel tanks, it would be feasible to introduce a blockage in the feed pipe or disconnect a required transfer pump, making fuel in a particular tank inaccessible. These auxiliary tanks are often not used until most of the fuel in the main tanks has been used.

Starting with relatively small aircraft, the Beechcraft Bonanza typically has two main tanks (left and right) from the factory, and the handbook instructs the pilot to regularly switch the feed between the two to avoid unbalancing the aircraft, so a blockage in either line would quickly be noticed while there is still plenty of fuel in the good tank to return to the origin airfield. This is not a favourable situation for the prospective saboteur.

Aftermarket additions, however, frequently include wingtip auxiliary tanks and/or an auxiliary tank in the luggage compartment; these require the main tanks to be run low, then small pumps switched on to move fuel from the auxiliary tanks into the main tanks. Disconnecting these pumps would deny access to this extra fuel at a time when the aircraft might already be out of range of an airport. The remaining fuel in the main tanks would allow the pilot a reasonably free choice of forced landing site, and the aircraft is also small enough to use even small gravel or grass airfields for a safe landing; in many countries there are a lot of these scattered randomly for private use.

The Cessna 421 series, a "light twin" often used for corporate transport, has a notoriously complex fuel system which seems particularly susceptible to interference. One fun possibility, specific to this type, is to insert a timer or a geographic (GPS) trigger that forces the fuel pumps into "high boost" mode. At anything less than full power (which is unavailable at cruise altitude), this will flood the engines with an over-rich fuel-air mixture that the engines cannot run on. This simulates a double engine failure and will probably cause a forced landing in an approximate area that you can choose in advance.

Larger aircraft generally have more sophisticated fuel and failure-monitoring systems that might detect or otherwise mitigate sabotage of this type, at least if they are reasonably modern (hint: if they have turbine-based engines and "glass cockpit" instruments, that's modern enough). This might not be a show-stopper for a determined saboteur, but it does increase the difficulty significantly since both the fuel system itself and the monitoring systems must be modified. It might be sufficient, however, to alter a fuel gauge for a secondary tank so that it shows full when the tank is actually empty; this might trick the crew into failing to refuel that tank.

  • $\begingroup$ you make me think of this Canadian Pilot: en.wikipedia.org/wiki/Robert_Pich%C3%A9 In that case, the pilot glided long enough to return to an airport, achieving the record of the longest glide with an Airbus A330 $\endgroup$ May 7, 2021 at 16:50
  • $\begingroup$ @FélixBrunet Yes, incidents like that are a rich source of inspiration. I wouldn't recommend causing a fuel leak however, because that can easily lead to an in-flight fire which is extremely dangerous. $\endgroup$
    – Chromatix
    May 7, 2021 at 16:53

Introduce Navigational Errors

Particularly for trans-oceanic flights, modern aircraft are nowadays highly dependent on two systems to keep them pointed in the right direction: an "inertial reference system" which estimates the aircraft's speed and course over time to derive position, and the Global Positioning System which relies on receiving weak radio signals from satellites. If you can sabotage both of these, then the aircraft can be sent off course such that it won't find the intended airport; from there it could be quite difficult to find any suitable airport to land at.

Light aircraft still often make use of the ground-based NDB, VOR and DME navaids, which are probably harder to cause confusion with. A pilot who notices the loss of IRF and/or GPS data will generally also try to pick up ground-based navaids as a backup method of navigation, if they are within range. Some of the more powerful VORs can be received as much as 200 miles away at cruise altitude.

The more brute-force method would be to cause a complete electrical failure, which will definitely knock out most if not all of the navigation systems. Aircraft are still controllable in this condition, via emergency power sources if the flight controls are all power-operated. But a prudent pilot will have a GPS-equipped phone or tablet to hand, and will at least be able to plot GPS coordinates on a paper chart and thus get within sight of an airport.

So you should probably try the more subtle approach of altering the received GPS data using a jammer, in a way that isn't easily noticed. This requires simulating the satellite signals that would have been received if the aircraft was actually in the distorted position. Since the autopilot will tend to change course to keep the aircraft on the planned route, to force the actual route to deviate to the south, you need to distort the position northward of the true position.


Intimidation will suffice. If a fighter jet shows up, you will do whatever it says.

Attacking control surfaces is a very dangerous game. The line is too thin between "controllable enough to finish mission" and "not controllable enough to even control a crash".

The engines, however, may be more attackable.

Knockout the engines within glide distance to a landable space

That's it.

"All-engine failure and glide-to-airport" is a scenario regularly trained by crews. They will not let you fly a commercial jet unless you can show you can do that in simulator.

They train gliding to a divert airport because it's common:

All of these are 100% survivable if there's a flat place to set down and the pilot can find it. It may even be a gear-down landing where the plane could be fixed on-site and flown out.

Even knocking out one engine may be enough

A plane with an engine down gets worse fuel economy. With less thrust, it cannot sustain as high an altitude, and must descend into thicker air where fuel economy is worse. It may not be able to reach its planned destination with fuel on board. If that works in your favor territory-wise, it may suffice.

Also, on some aircraft, knocking out engines asymmetrically means their rudder isn't strong enough to compensate for asymmetrical thrust, so they throttle back good engines on the opposite side, worsening power further, along with sustainable altitude and fuel economy. If they can't even sustain ground-level flight, then they are a weak motor glider, and will need to look for a landing site.

There could be techno-drama in the latter, as they figure this out.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .