Facebook Google Menu Linkedin lock Pinterest Search Twitter


air safety

Feb 7, 2016

Passenger video gives inside view of ATC screw up at Melbourne Airport

*Updated with YouTube link ABC News, and possibly the commercial networks, will tonight air an inflight


The head on lights of one Qantas 737 from inside another in Melbourne ATC stuff up
The head on lights of one Qantas 737 from inside another in Melbourne ATC stuff up

*Updated with YouTube link ABC News, and possibly the commercial networks, will tonight air an inflight passenger video of a set of near collisions involving two Qantas 737s and an Emirates 777 at Melbourne Airport last July.

The Qantas jets were both about to land, while the Emirates wide body was making its take off roll for Singapore.

The TV news programs may also show portions of a flight tracking video showing one of the Qantas jets being diverted to fly over the airport’s terminal area 600 feet below the safe minimum altitude by an air traffic controller to avoid the imminent risk of colliding with the other Qantas jet.

At one point in the video taken inside one of the Qantas 737s the lights on the oncoming other Qantas jet flare into head on visibility before it they are obscured by the wing of the climbing jet.

This is not what is expected inside an Australian domestic airliner approaching the country’s second largest airport.

However that collision risk had arisen because the tower had coordinated the movements of the three jets in what appears to have been unprofessional and negligent manner causing a hasty decision to direct one of the Qantas jets to abandon its approach and climb away from the airport to avoid the risk of colliding with the Emirates jet where its position on the runway being used for the takeoff would meet the landing Qantas jet on the other runway at the point where the runways intersected.

In sequence, the tower was confronted with a situation of its own making where a Qantas 737 could have flown into an Emirates 777, followed by a double go-around panic in which the two Qantas 737s were then put at risk of hitting each other.

The tower’s action in telling the Qantas jet descending toward the runway being used by the departing Emirates jet to go around then put it at risk of running into the Qantas jet that had previously initiated a go around to avoid a collision at the intersection of the two runways.

This was a clusterf*ck of the first order, the dangers of which only become apparent when the video taken inside the cabin of one of the Qantas jets, and another showing the flight tracked paths of the three jets are viewed with the appropriate explanatory narrative.

Which of course, might or might not fit in with the news values of the commercial networks, or the attention span of their audiences.

The information about this incident was pulled together by independent SA Senator, Nick Xenophon following the discovery of the passenger video and embargoed until 5 pm eastern daylight time today, allowing ample time for the complexity, and seriousness of the incident to be considered by the news networks.

Senator Xenophon has been pursuing the safety risks of Melbourne’s simultaneous use of intersecting runways for some time, specifically the procedure called LAHSO or land and hold short operations.

LAHSO has been both much criticised, and defended by, a range of safety studies for many years, notably in Canada and the US, where it is used at some airports.

Senator Xenophon has criticised the private owners of Melbourne Airport for clinging to LAHSO as a way of avoiding the costs of constructing a parallel runway which would prevent such conflicts arising.

The lack of judgment exercised by tower control on that night is astonishing and alarming. The tower was manned by an on-the-job trainer and a trainee controller. As the ATSB noted in its don’t scare anyone with the blunt truth interim report, there were two other controllers performing other functions in the tower that night.

In its interim report the ATSB says the jets involved in the incident didn’t infringe safe separation distances.  However given the procedural breakdown it confirms in that preliminary document, this is less relevant than it might seem.

What doesn’t get highlighted by the interim report is the situation where the safe minimum altitude for aircraft to fly over the airport’s passenger terminal area is 2000 feet, but the Qantas 737 that had been about to land behind the departing 777 received urgent advice to divert across the terminal area, crossing it at the unsafe and illegally low altitude of 1400 feet, while it was under increased power to climb away for its go around.

This breaking of this safe minimum altitude rule underlines the lack of preparedness, judgment and exercise of professional control that occurred in the Melbourne tower that night.

It put the lives of large numbers of travellers at risk in the air, and it broke a rule intended to reduce the risk of an airliner that might for example, experience a sudden loss of power or control, and crash into a terminal.

This last Australia Day there was another incident at Melbourne Airport in which a traffic helicopter at nearby Essendon Airport and two other jets using the larger airport infringed safe separation rules in what the ATSB now says involved a failure of communications in relation to a sudden change of runway procedures.


We recommend

From around the web

Powered by Taboola


Leave a comment

20 thoughts on “Passenger video gives inside view of ATC screw up at Melbourne Airport

  1. Fred

    This incident was definitely a “clusterf*ck”. One point though Ben – according to the interim report, the aircraft was issued a heading as it climbed through 1,400ft in the missed approach. It then crossed the intersection of R34 and R27 in the right turn and above 2,000ft. Consequently, it was well above any obstacles in the immediate vicinity of the airport. The instruction issued by the controller in this case wasn’t unsafe, but it was technically outside the rules. I think the bigger issue by far was the fact that two aircraft were conducting missed approaches on converging headings, with considerable potential for a mid-air collision.

  2. comet

    Australian aviation regulation is shocking.

    It’s time the ICAO downgraded Australia’s ranking to match the third world standards that we practice.

    And why does the ATSB think it is compulsory to always release ‘feel good’ reports to make everyone think ‘there’s nothing to see here’? The ATSB wants to protect its corporate buddies in the aviation industry, which demonstrates the heart of the problem.

  3. Ronnie Moore

    There is too much to go wrong, creating high risks in this 3-plane scenario.
    What if the 777 had to abort takeoff and slows down around the intersection? What if the 737 landing on 27 turns left off onto 34 as a runway exit by mistake? What if the 737 VYE lands long (mistake or wind shear) and simply can’t hold short of 27. What if it has a long touch and go-around but develops engine fire and has to land?
    If there’s sufficient time between these aircraft movements it should be safe, but this incident timing looks far too tight especially for the two 737’s
    Melbourne’s intersecting runway configuration works ok for landings on one and takeoff’s on the other, but is much less safe for LAHSO 27 when a landing on 27 also occurs.

  4. George Glass

    It is not possible to overstate how rubbish Australia’s aviation infrastructure is.The minimum standard everywhere else is 2 parallel 4000 meter runways.One for departures,one for arrivals.Politicians for over 50 years have been terrified of infrastructure. It is the ultimate vote loser.And politicians are cowards.Nothing will change until we have a hull loss.

  5. George Glass

    Oh,and I should have,mentioned,as a working Pilot, we put up with this sort of cr*p in Australia every day.

  6. Dave M

    More information and a somewhat different take on this at the Aviation Herald…


  7. Ben Sandilands

    Simon and I have exchanged cordial notes, and I do accept almost all of the points he and his readers have made.

    However it wasn’t media hysterics that caused the suspension of LHASO at night at MEL and ADL.

    The procedural and cultural issues in AirServices are of valid concern, and while Av Herald is correct in its analysis from a technical perspective, I think there is a different perspective to offer on the performance of the ATC provider, and I have been given some very unambiguous support from pilots and airline people since publication.

  8. ghostwhowalksnz

    If this is a problem when they are near the ground, imagine the situation back away from the airport where there is a ‘traffic stream’ associated with each of the landing /takeoff directions,and each stream has maybe 3 or 4 choices of direction plus maybe 3 or 4 planes sorting out which one of those choices is for them.

  9. Grizzly

    On my reading of the Aviation Herald report, this incident reminds me a little of the Tenerife disaster in 1977.

    In each case, the crew of an airliner misunderstood an ATC instruction and commenced a manoeuvre that hadn’t been authorised.

    In Tenerife, the misunderstanding led to a collision, largely because another airliner was on the same runway and had nowhere to go (although the that airliner had previously lost an opportunity to leave the runway due to its crew’s unfamiliarity with the airport). In the present case, another airliner was in the air, ATC gave its crew an instruction to change course, and that crew obeyed that instruction.

    I’m wondering, though: what might have happened if ATC hadn’t given that instruction? If, in the absence of the instruction, the two 737s had been on a collision course, would TCAS have been activated, and, if so, how safe would it have been for the 737 directed by TCAS to descend to have followed that direction at such a low altitude, particularly if that 737 had been the one above the runway on which a third airliner, the 777, was taking off?

  10. Fred


    TCAS ‘descend’ type resolution advisories are inhibited when the aircraft is below 1,000ft AGL in descent, so the TCAS shouldn’t issue an advisory telling the pilot to ‘descend’ or ‘increase descend’ at low altitude. If anything, the TCAS would issue advisories telling the pilots of both aircraft to ‘climb’. In that case, the TCAS systems in each aircraft would (theoretically) coordinate the required rate of climb for each aircraft to avoid a mid-air collision.

    That said, in the case you mentioned, where both aircraft continue their approach to separate, almost perpendicular runways, then the aircraft are still quite a long way apart on final approach. Consequently, they are likely to be below 1,000ft AGL before they get close enough to trigger any advisories. By that stage the advisories are inhibited anyway, so the TCAS wouldn’t be much help.

  11. Grizzly

    Thanks for your interesting response Fred. It confirms my tentative view that TCAS may not have been of much, if any, assistance.

    I get the impression that the problems that led to the incident in Melbourne were that ATC was concerned about the proximity between the 777 and the first 737 (hence the communication indicating that that 737 may need to climb quickly if it had to go around) and about the proximity between the 777 and the second 737 (because of possible wake turbulence – hence the instruction to change course).

    I also get the impression that there was wasn’t really a strong chance that the two 737s were going to collide, because they had been set up to land at staggered times. However, I may be wrong about that impression – I notice, eg, that the first 737 had been asked to slow down as it approached, which may have put it closer to the other 737.

  12. Fred

    I think it all started because the 777 was slow to commence its take-off roll after being cleared for an immediate take-off. That reduced the separation between the 777 and the 737 on final for R27. The trainee controller told the 737 on R27 to reduce speed and then pre-empted a go-around instruction by saying “in the event of a missed approach” they should climb to 4,000ft expeditiously. The 737 on R27 mistook that as an instruction to go-around, so they did. Meanwhile, the 737 on R34 was approaching the runway with the 777 still on the ground. The Coord directed the trainee controller to instruct the 737 on R34 to go-around, even though both the trainee and the trainee’s instructor considered the 737 on R34 had enough separation from the 777. The 737 on R34 commenced a go-around and that’s when the ‘fun’ really started. I guess we’ll have to wait for the final report, but I suspect that having ordered the 737 on R27 to go-around, the 737 on 34 could have landed safely if the Coord hadn’t butted in.

    Reminds me of an incident in Sydney many years ago, at a time when LAHSO were allowed at that airport and foreign aircraft were allowed to participate. A RAAV VIP aircraft carrying the PM was cleared for take-off on R07 at the same time as a foreign B747 was cleared to land on R34. The B747 was instructed to hold short of R07, but it was still at high speed as it approached the intersection and it was obvious that it wouldn’t stop in time. The RAAF aircraft rejected its take-off roll and stopped well short of the intersection. Meanwhile the B747 sailed on through. If memory serves me correctly, foreign operators were banned from LAHSO shortly after that incident.

  13. Grizzly

    I agree with Fred’s analysis. It seems to me that the issue is an overreaction by ATC to the unauthorised manoeuvre I described above. If the system was working properly, there should have been enough separation between the 777 and the second 737 for the second 737 to land safely even if the 777 was a bit slow to take off.

  14. Seth Knoepler

    I agree that Fred’s analysis sounds right, ditto Grizzly’s reference to Tenerife. Unlike Tenerife, this incident also involved “too many cooks” at ATC. With regard to the latter, I’d be interested to learn about how various responsibilities are allocated when a trainee controller is the “PF”

  15. Fred


    If I understand it correctly, the tower controller position (Aerodrome Control, or ADC) was staffed by an On-the-Job Training Instructor and a trainee controller. Although it was the trainee’s fourth shift under training, he or she had previously held ADC ratings at overseas airports, so obviously had some experience in the role. The trainee was “PF” as you put it, but the instructor had overall responsibility for the ADC position.

    The other controller that was involved in the incident held the Coordinator, or Coord position, and also held the Operational Command Authority (OCA) for Melbourne Tower at the time of the incident. According to the interim report, the OCA has overall responsibility for the provision of the control services by Melbourne Tower, but that does “not include the authority to make operational decisions, such as directing a controller to issue an urgent operational control instructions”. The ADC was responsible for traffic separation, but “may accept advice from the OCA holder”.

    On the face of it, it seems the Coord overstepped his or her authority by directing the trainee to instruct the 737 on R34 to go-around. As you said, a case of “too many cooks”, but perhaps there’s also an element of poorly defined responsibilities and authority. It seems odd that the OCA holder has overall responsibility, but doesn’t have the authority to make operational decisions. The report also mentions that Melbourne ADCs “had completed compromised separation training in the visual simulator in visual conditions by day”, but “they had not been trained in compromised separation recovery in the case of simultaneous missed approaches/go-arounds at night”.

  16. Seth Knoepler

    Thanks, Fred – very interesting. “An element of poorly defined responsibilities and authority” sounds right. Compared with the protocols which have been developed and revised for coordination on the flight deck of aircraft, these rules certainly seem sloppy, leaving plenty of room for precisely this sort of sequence to occur.

  17. nobeljnet

    I note Senator Heffernan of NSW referred to the use of crossover runways at MEL’sTullamarine as “fwxyz, that’s risky shit”, as covered on the ABC’s Insiders yesterday.
    Buzzfeed has it too.

  18. 777 Steve

    Fred, I have to take issue with your comment that this was caused by the 777 being slow to take off, or indeed commence its takeoff roll. Being cleared for an immediate takeoff is a different beast in a 777 v a 737 or A320. Have you ever watched a large aircraft accelerate? its the reason RWY 34 was used as the laws of physics require more distance for an aircraft weighing approx 290T to get airborne rather than a 70 tonne light twin. Based on the information available, the error chain began well before the T/O clearance and should have been trapped far earlier. Inadequate separation and an unrealistic expectation of the multiple factors at play lead to this situation. Air Services would do well to send their trainers/trainees to LHR/LGW for a while so that they could see hows it can/should be done.

  19. Fred

    777Steve: Well, with around 15,000 hours on B747/B777/A330 type aircraft, I should certainly hope I have some idea of how long it takes for a large aircraft to accelerate. My comment was based on a report that the B777 was slow to start rolling after it was cleared for take-off.

  20. Fred

    Take the B777 out of the mix, and it seems to me that the two B737s would have landed quite easily with the required separation. ATC certainly set the scene for this incident, by trying to ‘sneak’ in the B777’s departure between the two arrivals.


https://www.crikey.com.au/2016/02/07/passenger-video-gives-inside-view-of-atc-screw-up-at-melbourne-airport/ == https://www.crikey.com.au/free-trial/==https://www.crikey.com.au/subscribe/

Show popup

Telling you what the others don't. FREE for 21 days.