Menu lock

air safety

Jul 27, 2017

SingaporeAir’s low approach to Canberra landing reviewed by ATSB

How not to land a 777-200 at Canberra Airport

Ben Sandilands — Editor of Plane Talking

Ben Sandilands

Editor of Plane Talking

The big hills near Canberra Airport which is in the middle distance

There is more terrain to think about on approaches to Canberra Airport than at most of the nation’s airline gateways, and as the newest to take wide-body international flights, the arrival of a Singapore Airlines 777-200 with 248 people on board on February 22 this year did so while twice at an unsafe altitude for parts of its descent in visual conditions toward Australia’s capital city.

Using its new fondness for gentle but precise language, the ATSB has nailed a series of mistakes made by the crew of the jet as the passed by some very big hills to the south and southwest of the airport before the pilots could see the runway.

In plain English, the crew at first flew 500 feet below a section of the approach with a safe minimum altitude of 7500 feet, and after correcting that error by climbing back to 7500 feet on the advice of tower control, they then descended the jet to 4600 feet above sea level at a closer in part of the approach profile where they should have maintained 5300 feet.

Canberra Airport is 1886 feet or 575 metres above mean sea level. The aircraft was cleared to continue to a safe landing from that lower than required level.

The ATSB report deals with the state of Canberra Airport at the time, with one navigational aid unavailable, and with the changes underway in near airport navigational procedures at Australian airports which are best left to technical forums.

The report also summaries the responses of Singapore Airlines, starting with its briefing the crew to expect an approach procedure which turned out not be be available when its flight was drawing close to the very high hills near Canberra.

This is the sort of report that might best be read by interested lay air travelers over a nice rich pudding accompanied by a desert wine.

We recommend

From around the web

Powered by Taboola

3 comments

Leave a comment

3 thoughts on “SingaporeAir’s low approach to Canberra landing reviewed by ATSB

  1. Dan Dair

    Does the report tell us whether AirServices communicated to all concerned, particularly Singapore airways, that the expected approach procedure would not be be available.?
    Or did SA’s planners fail to actually check their plans properly.?

    One would imagine that the margins on the decent-floors are there precisely to mitigate instances like this,?
    but you would also expect that any airline, not least a ‘legacy’ one, would have pilots who would have ‘held-off’ their approach, until they were certain that they fully understood the new & unexpected approach pattern they were required to adopt.?

    1. JW (aka James Wilson)

      Dan,

      You need to read the report to understand what happened, but here’s a summary:

      1. The Runway 35 ILS was out of service and the crew was obviously aware of that fact. The available approaches for Runway 35 were the RNAV-Z (GNSS) or the VOR. The crew planned on the POLLI 4P arrival for the RNAV-Z (GNSS) approach.

      2. ATC subsequently cleared the aircraft for the POLLI 4B arrival, which links to the VOR approach, but not the RNAV-Z approach.

      3. After being cleared for the POLLI 4B, the crew entered that arrival into the FMC, but left the RNAV-Z approach programmed, with the intention of requesting the RNAV-Z approach from the approach controller, in lieu of the VOR approach.

      4. When they entered the POLLI 4B arrival with the RNAV-Z approach, the crew joined the two procedures by deleting the first waypoint of the RNAV-Z approach (SCBSG) and its associated altitude restriction (7,500ft).

      5. When the crew subsequently contacted the approach controller, they requested and were cleared for the RNAV-Z approach, and were told to track to SCBSG. The captain manually re-entered SCBSG into the FMC without entering the associated altitude restriction and then manually joined SCBSG to the next waypoint (SCBSI) for the continuation of the approach.

      6. Without the altitude restriction at SCBSG, the FMC descended the aircraft below the 7,500ft restriction before reaching SCBSG, infringing the terrain clearance requirements.

      7. The crew subsequently climbed back to 7,500ft after being told they were too low by ATC. The crew then elected to do a visual approach, but did not advise ATC and were therefore not cleared for that type of approach. Consequently, the controller thought the aircraft was still flying the RNAV-Z approach and warned the crew when the aircraft again descended too low on the RNAV-Z approach profile.

      I believe the ATSB overlooked an important point in its investigation. At the time of the incident, NONE of the published STAR procedures linked to the RNAV-Z (GNSS) approach for Runway 35, the crew’s preferred approach. The POLLI 4P arrival the crew initially entered in the FMC is associated with the RNAV-P (RNP) approach, which is only available to operators authorised by CASA (not SQ). The POLLI 4B arrival was associated only with the VOR approach.*

      Strictly speaking, the crew should have entered the POLLY 4B arrival and then left a discontinuity between the arrival and the first waypoint of the RNAV-Z approach (SCBSG), instead of trying to make the nav display picture look ‘pretty’ by joining the arrival to the approach. By joining the two, they deleted an essential waypoint and altitude restriction that was not re-entered when the waypoint was subsequently re-entered.

      The ATSB report does highlight one very important point: when flying an arrival/approach where vertical guidance is provided by the FMC, it is essential that crews enter the procedures from the FMC database and not enter manual waypoints. The crew must scrupulously check the entered waypoints and altitude restrictions against the published procedure.

      * Since this incident, the POLLI STAR procedure and RNAV-Z approach have been re-designed so that the STAR also links to the RNAV-Z approach.

      1. Dan Dair

        James Wilson,
        ……….& that’s precisely why I didn’t read it…. I can’t make head nor tail of it, because it’s not my field of expertise.
        I did appreciate your precis however & now understand far more than I previously did.
        And it’s good to know that the system (& presumably procedures/protocols) has been redesigned to resolve most of the situations which were apparent in this incident.
        Whether it was CASA, ATSB or Canberra airport which led the changes, the fact that it’s a change for the good (& there was no actual accident) is a very positive outcome.?