Recently, I was on a flight from Montreal to Vancouver. We were somewhere between Calgary and Vancouver when we started to fly strange manoeuvres:

enter image description here

The whole flight can be seen on FlightAware. We were definitively not yet approaching Vancouver airport and from the entertainement system I could see that we were at around 40000 feets, flying almost full speed.

So what are potential reasons for this flight pattern?

  • 2
    Did you feel that, or it was just on the screens? Mar 9, 2015 at 22:09
  • 2
    I felt it, saw it on the screens and also saw it happening when looking out of the windows. Mar 9, 2015 at 22:10
  • 14
    This might be more apropos to Aviation.SE.
    – choster
    Mar 9, 2015 at 23:01
  • 5
    this has been answered here: aviation.stackexchange.com/questions/12481
    – Federico
    Mar 10, 2015 at 7:27
  • 2
    Did you personally experience just the 360° or also the crazy zigzag at the end? The corners of that look too tight (on a zoomed-in map) to be physically possible at 40,000 feet without flattening everything in the cabin. Mar 10, 2015 at 18:23

5 Answers 5


I called a friend who is a jet airliner pilot and showed him this question, he said there are two possible reasons for this:

  1. There was some traffic or some sort of closure to the airspace, and was asked by the ATC to take a 360 until things are cleared.

  2. He was asked to lower the altitude immediately, and to do that without going further into the route in the wrong altitude the captain did the 360 to have some space to lower the altitude before continuing the original route.

Both are normal actions.

  • FR24 shows the loop took place at a constant 40,000 feet, so it can't have been about adjusting altitude.
    – Mark
    Mar 9, 2015 at 23:04
  • 1
    @Mark most likely you are right, I guess my pilot friend simply told me reasons for 360 maneuvers, I am sure he is not aware of this particular case and did check the altitude. Mar 9, 2015 at 23:06

Loops like that are usually about timing: it's an easy way to delay the airplane's arrival by a few minutes. By doing it in the middle of nowhere rather than the crowded airspace around the airport, it keeps things simple for air-traffic control.

FlightRadar24 shows the loop as well, but not the zig-zag, so I'm guessing the latter is an artifact of FlightAware's tracking system.

  • 2
    The lack of zig-zat at FlightRadar could also be due to a too low sample frequence on FlightRadar's side. Since the author says he saw and felt the pattern, it might be that FlightRadar is at fault here.
    – Pål GD
    Mar 10, 2015 at 2:45
  • 1
    @PålGD: The samples at FR24 look close enough together that if the zig-zag had been flown as shown in the question, it would need to fly about twice as fast as the rest of the loop. Independently of this, the turns in the zig-zag are crazy narrow. They look like a few km in radius (zoom in!), which would pull several g's horizontally if done at a speed that allows the plane to stay aloft at 40,000 feet. The airframe would probably stay in one piece doing that, but it would have felt beyond dramatic in the cabin. Limbs would be broken. Mar 10, 2015 at 18:10
  • Well, he did say he could feel it! :D
    – Pål GD
    Mar 10, 2015 at 20:44

The loop itself is a way to delay the flight - to make it fit into the landing schedule at destinatioin, probably.

But what about the north-south zig-zag after the loop?

Somehow, it looks like drawing one invalid measurement that was shifted north by some error. But looking closely, the part is not just one point - the end of the zig-zag is curved, for example.

Computer scientitsts intuition tells me that it is an artifact of an algorithm trying to fix an outlier point.

What goes wrong is that the algorithm tries to estimate "how it should look like" based on a part before, of some length. But it does not expect loops in the part before - so the correction gets confused. The result of the correction is then affecting more than one point.


After the plane exited the circular holding pattern, the jump in position is probably just an erroneous transponder report (could be data corruption, inertial reckoning vs. GPS source, etc) and the plotting software decided to fit it on the line instead of rejecting it. This is relatively common - for a particularly extreme example, look at this:

Zigzag plane track due to transponder issues

N73259 was actually flying in a relatively straight line - 737s don't make sharp turns every couple miles.


It is a funny move indeed.

When you zoom on the place it happens and display the layer "IFR High En Route" on the map of FlightAware, you can see the airport of Penticton. On its website, we can see that their fees list suggests that large planes can land there. Indeed, its runway is 6000 ft long and the Airbus A330 (operated on that flight) can land on approximately 6000ft long runway. So it looks like this airport is ideally located for emergency landing of planes flying on a transcanadian routes.

This is just a hypothesis, but maybe the pilot noticed something weird at that point and preferred to fly over the airport again in case. It might also be that another plane landing or leaving Penticton airport would disturb your plane by forcing it to follow some corridor, even though the planes arriving or leaving on March 8 seemed on time and not happening between 7 and 8PM PDT, which may be the approximate time it happened.

  • 6
    The FR24 data shows altitude as well as position; the loop took place at 40,000 feet. That high up, Penticton is too close to be available as a diversion airport in an emergency; by the same token, anything taking place at the airport is much too low to be a factor.
    – Mark
    Mar 9, 2015 at 23:03

You must log in to answer this question.

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