On Monday, January 3, 2022, Southwest Airlines Flight 997 from STL to LGA landed with the following flightpath:

FlightAware SWA 997 flightpath (landing)

The flightpath deviates from the planned stabilized approach to Runway 04 and instead lands on Runway 31. This involved executing a right turn to an approximate heading of 070 degrees followed immediately by a sharp left turn to 315 degrees to line up with the runway, all within the last 90 seconds of the flight.

Why would this have been done?

To me, an amateur aviation enthusiast, it seems to me this is very much a non-stabilized approach, and I wondered why a commercial airliner would do such a thing.

Is there a place where one could find recordings of the ATC-cockpit communications during this event?

Is this kind of last-minute switch to a perpendicular runway common in commercial aviation?

Is this maneuver indicative that there was a last-minute issue with the planned approach, such as a runway incursion?

Source: https://flightaware.com/live/flight/SWA997/history/20220103/1955Z/KSTL/KLGA


1 Answer 1


This is a published visual approach:

enter image description here
Source: FAA via skyvector.com [cropped]

Due to the tight airspace because of the neighboring Newark and JFK, and also for noise, the airspace/procedure designers find solutions that have the least impact on the overall airspace and throughput.

From the FAA's YouTube channel, a Jan 2021 video that is worth watching: New York Metro Area Series – Runway Selection.

Note that the dashed blue line that FlightAware uses does not indicate the landing runway (you can zoom in to verify). That line is drawn from the last point in the flight plan (if available), which does not include the runway, to the airport's coordinates (Airfield Reference Point). Otherwise, it will just be a line connecting the departure and arrival airports.

See also for your other questions:

  • 2
    $\begingroup$ And regarding OP's last question: no, a runway incursion would result in a go-around, not a last-minute change-of-runway. It could result in the aircraft going around and then entering the local traffic pattern and landing on the same or a different runway, but the go-around would happen first. $\endgroup$
    – randomhead
    Jan 4, 2022 at 22:21
  • 3
    $\begingroup$ @randomhead: For completeness, if it's early enough a runway switch can happen, more so in the US than the rest of the world, which was a factor in a recent incident. $\endgroup$
    – user14897
    Jan 4, 2022 at 22:43
  • $\begingroup$ Fair enough. I suppose the location (relative to the threshold) at which a controller would stop saying "cancel approach clearance" and start saying "go around" is close to the location where they would stop saying "change to runway..." and start saying "go around." $\endgroup$
    – randomhead
    Jan 4, 2022 at 23:18
  • 2
    $\begingroup$ @TomMcW I reckon it's a literal translation from the French BEA report. For comparison the "moose test" for cars swerving round an obstacle is known as Test de la baïonnette in French. AVH says "the following still based on the AVH translation of the French original" which I reckon was machine-assisted. It doesn't read like native English. The BEA's own English version (linked by AVH) uses "sidestep" in the equivalent sentence as the French original uses "baïonnette" $\endgroup$
    – Chris H
    Jan 5, 2022 at 11:29
  • 2
    $\begingroup$ @TomMcW: French Wikipedia entry for baïonette (aviation). Rough translation: "During a landing in case of an occupied runway, it is possible to divert to a parallel runway by taking a bayonet-shaped path, i.e. two opposing turns, to get to a free runway." $\endgroup$ Jan 5, 2022 at 13:50

You must log in to answer this question.

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