In January 2023, two H130s operated by Sea World Helicopters were involved in a mid-air collision while carrying out sightseeing flights near the Sea World resort. Last month, the Australian Transport Safety Board released its report. Gideon Ewers considers the findings.
This article, “The sound of silence” was published in the June/July 2025 edition of RotorHub International. To read more articles like this, you can apply for your complimentary subscription today.
There is an old saying in aviation: “It’s the one that you don’t see that gets you”. That simple maxim, drummed into many new pilots, can be said to sum up the ATSB’s report into the Gold Coast mid-air collision of 2 January 2023.
It does so on many levels, from the events of the fateful day but also in terms of the company’s procedures and SMS programme.
Put another way, the report demonstrates how easy it is for individuals and organisations to believe there is no threat – when the reality is that the threat is not only close at hand but masked.
But first let’s recall the events of the day. Two helicopters, both H130s operated by Sea World Helicopters, were conducting sightseeing flights around Gold Coast – a popular tourist destination in Queensland in north-eastern Australia.
One aircraft, VH-XH9, which for brevity we’ll call H9, was flying from the operator’s heliport, while the other, VH-XKQ (KQ), was operating from a helipad within the boundary of the Sea World Theme Park.
Both facilities were on the banks of a body of water known as the Broadwater. Flights offered to customers included five- and 10-minute sorties.
The standard routing for the approximately five-minute flights from either facility called for an initial climb over the Broadwater before turning east to over fly another landing zone (LZ) known as “Sea World grass” and used by another sightseeing operator.
This was followed by a turn to the north to fly parallel to the ocean shoreline of the Southport spit barrier island, followed by progressive turn once past the inlet to the Broadwater separating Southport spit from South Stradbroke Island.
The wide turn would take aircraft over another small island known as Wave Break Island, before returning to either the Park Pad or Sea World Heliport.
Final approach – Flight path overlap, downwash compliance and incomplete situational awareness
Generally speaking, because of prevailing winds, final approach to either would be to the east. Critically, the usual departure path for the Park Pad crossed the arrival route for the Heliport at an angle of about 80 degrees.
In order to facilitate traffic separation, the procedure was to make a series of radio calls on the Common Traffic Advisory Frequency (CTAF).
Further complicating the equation was a requirement for pilots to avoid overflying marine traffic at low altitude in order to reduce the impact of downwash.
On the day in question, the pilots of both aircraft had flown in the morning and, following a break for lunch, began an afternoon cycle of flights.
The first flight after lunch for the pilot of H9 was a 10-minute sightseeing flight landing at around 13:50. At around the same time, the other pilot in KQ also flew a 10-minute sortie.
Passengers in both aircraft had been given pre-flight briefings before being fitted with pouch-style constant wear lifejackets.
At 13:51, H9 took off for a five-minute flight from spot 3 at the helipad, with the pilot plus five passengers on board. The pilot reported making a “taxi” call prior to lift-off, departing to the south-west before turning on course over the Sea World grass. He then made a departure call and entered the coastal lane on turning north at around 500 ft AMSL.
At 13:53, ground crew began loading pax into KQ and around the time loading of the front seat pax began, the pilot of H9 at Porpoise Point (at the inlet to Broadwater) was recorded making an inbound call.
Loading of passengers on to KQ was completed at 13:55. Ground crew reported that they checked the aircraft doors were secure and that there were no marine or aircraft in the immediate area.
Video from the Park Pad shows that check was carried out between 11 and 28 seconds past 13:55. At that point, H9 was around 1,220 metres to the north of the Park Pad.
Five-minute scenic flight route and standard radio calls. Source: Google Earth, annotated by the ATSB
The ground crew confirmed with the pilot of KQ that it was a five-minute flight. The pilot acknowledged with a thumbs-up signal and left the pad to begin preparing passengers for the next flight. KQ took off about 23 seconds later.
Meanwhile, those in H9 recalled that their assessment at the time was that there would be no conflict with KQ, and that KQ would pass behind them.
At the same time, two vessels were heading south along the Broadwater between the Park Pad and the heliport. Keep in mind the operating requirement to keep downwash clear of marine traffic.
Witnesses on the southernmost vessel recalled seeing the pilot of KQ looking at them as the aircraft climbed away from the Park Pad. The same vessel was crossing H9’s intended track to spot 3 and the pilot of H9 reported adjusting his course to pass astern of the vessel.
One of the pax on H9 reported that the pilot’s gaze was directed at the vessel and the aim point at the heliport.
Video on H9 shows that one of the passengers on H9 saw the approach of KQ and attempted to alert the pilot by tapping them on the right shoulder (the shoulder furthest away from the approaching KQ).
The pilot was seen to turn their head to the right in response (and away from KQ) and seven seconds later the aircraft collided.
The collision took place at an estimated 130 ft AGL and around 155 metres west of the helipad.
The main rotor, engine, gear box and tail separated from KQ and the aircraft fell on to a sandbar, with the result that the pilot and three of the pax on KQ were fatally injured in the collision and subsequent impact. The remaining three pax were seriously injured.
On H9, which was struck by the main rotor of KQ, the windshield and composite structure of H9 shattered, peppering the pilot and pax with shrapnel.
The pilot reported that his sunglasses saved his eyes from the shrapnel.
While all instruments had been lost and rotor RPM was decaying, the pilot although injured was able to land H9 on the same sandbar where KQ had fallen.
ATSB findings – SMS failures and safety oversights
So how is it possible that on a clear day two aircraft flying regular and known routes came together with an unfortunate and tragic conclusion?
As always, there is no single point of failure but rather a series of events – call them links or slices of Swiss cheese – that came together create the accident.
The ATSB identified a number of issues with procedure, the first and most obvious being that aircraft were operating from two landing zones close together simultaneously and with intersecting approach and departure paths.
Second, there was an absence of protocols that would mitigate that risk by reducing the likelihood of two aircraft occupying the same space at the same time.
Indeed, the ATSB points to the March 2022 reopening of the Park Pad without a full risk analysis taking into account the changes since the last time two separate LZs had been used.
The converging flight paths of XKQ and XH9 leading to the midair collision. Source: Google Earth, annotated by the ATSB
Under a previous ownership, the primary LZ was the Park Pad – with a temporary high season LZ set up in a car park near the present heliport.
At that time, says the report, the lateral separation between the two LZs was about 53 metres. This, says the ATSB, gave pilots on approach to either LZ a view of activities on the other pad within their normal scan.
The report says that “essentially” the new arrangement was “far more hazardous than the original configuration but without the controls previously employed to manage separation”.
While the ATSB doesn’t mention it, prior to 2019 the car park pad was only used for a few weeks a year and was situated in an area cordoned off from the rest of the car park by temporary barriers.
Operation there would have felt “off airport” and served to heighten pilot senses compared with yet another landing – one of many a day to a known, prepared LS like the Park Pad or Heliport, with an associated feeling of security and familiarity.
This revealed a flaw in the operator’s safety management system (SMS), since it appeared that there was no evidence to suggest that an assessment of the operation and procedure had been carried as far as collision risk was concerned, and therefore the operator believed that its existing procedures were sufficient.
Up until 2 January 2023, the operator employed H125s on its services, with H130s having been introduced shortly before the accident.
Thanks to the prevailing winds, the vast bulk of take-offs are to the south (99.5 per cent of them in the month leading up to the accident). In a H125, since the pilot sits on the right-hand side of the cabin, they have an unobstructed view of the Broadwater, while for the H130 the opposite is true.
Again, no SMS risk assessment of the use of the new equipment at the location and for the operation was completed, with the operator arguing that it had believed the general one carried out when the type was introduced was adequate.
The ATSB noted that the point where the inbound and outbound paths converge is at a moment of high workload for pilots, with outbound pilots monitoring power use and systems for engine health at a critical phase.
For their part, the inbound pilots must manage the path they take, ensuring that either the sandbar or the heliport remain in autorotation range.
As already mentioned, both must look to mitigate the impact of downwash on passing vessels. All of which cumulatively reduces the mental capacity available for scanning air traffic.
Further compounding the problem is the difficulty in spotting the other aircraft thanks to factors like the relative angle of approach, the need for pilots to look across the cabin, and passengers to spot traffic – and then in this case the geometry of the design of the aircraft in which the structure of the aircraft obscures the view of an aircraft approaching from that relative position.
Additionally, the report also showed that something as innocuous as wearing a baseball cap by the pilot of KQ proved critical, since the cap’s peak masked H9 from their view at a time when avoidance would still have been possible.
It is difficult to avoid that which you cannot see.
Footage from on board XKQ showing the relative contrast between XH9 and its background. Source: passenger video, annotated by the ATSB
Under strength – Radio transmission failure and communication breakdowns
Another critical factor identified by the ATSB and one area that the consumer media made much of was that the VHF radio in KQ was faulty in transmit thanks to a number of flaws with the connection to the antenna, which rendered radio transmissions unreliable at best and non-functioning at worst.
Recordings taken from the CTAF frequency recordings on the day of the accident and the day before at a nearby airfield showed that while the bulk of transmissions from H9 were strength 4 out of 5 (based on the ICAO scale where 1 is unreadable and 5 is perfectly readable), the majority of recordings from KQ were at strength 2 or below, with a significant number absent.
Relative positions of Heliport Pad 3, vessel and XKQ at six seconds prior to impact. The opacity of the cockpit structure has been reduced to 60 per cent to show the relative positions of both XKQ and the vessel, which were likely partially or fully obstructed from pilot’s eye position. Source: iwiation GmbH, annotated by the ATSB.
Which is to say the expectation is that they were made but nothing was heard of a quality good enough to qualify as unreadable.
Critically, on the day of the accident, none of the 18 transmissions made by KQ were readable and 14 were absent entirely.
There seems to have been no first flight of the day check that called for a radio check, something that might have revealed the failing antenna connection.
In addition to the antenna problem, the ATSB took issue with the number and timing of position reports, noting that the standard inbound call from Porpoise Point was “not a reliable alert for a pilot on the ground while boarding and interacting with passengers”.
Indeed, even before the report was published, Sea World Helicopters had amended their standard operating procedures to add an additional reporting point abeam the Park at the start of final approach to the heliport. (They also added additional sterile cockpit procedures.)
At the time of the accident there was no requirement for ground teams to monitor the sky around the Park Pad for possible threats – once with pax loading and egress the primary function.
Again, action was taken by Sea World Helicopters long before the report was published that addressed these issues – with the creation of the role of “pad boss”.
In addition to passenger duties, the pad boss has a quasi-Flight Information Service Officer (FISO) task, carrying out a check of the air space to the north and south of the facility as well as surface traffic on the Broadwater before advising pilots of any traffic and that they may “take off at their discretion”.
Another area of concern for the ATSB was the potential survivability of the accident and the role in it that improperly fitted four-point restraint harnesses and their interaction with belt type constant wear lifejackets had on the injuries sustained by the occupants of the aircraft.
Consequently, the ATSB has called for an immediate change in the way passengers are briefed and secured in aircraft, as well as – in the longer term – a study in the efficacy of lap worn constant wear lifejackets.
Simulated view from XKQ, 13 seconds before collision (looking straight ahead). Note that the pilot’s cap is represented by the transparent blue section at the top of the image. Source: iwiation GmbH, annotated by the ATSB.
Exercising the safety muscle – Failure to reassess operational risks
As with all accidents, the event occurred because of a series of elements coming into play which crated the conditions that reduced the probability of avoiding the accident to zero.
It is an interesting case in that it was exactly the type of sequence that SMSs are designed to prevent. It also revealed the fundamental flaw in the application of an SMS, which is to fail to constantly reassess an operation and its procedures, especially when anything about that operation changes.
Sea World Helicopters believed that it had an effective SMS in place, yet the failure to carry out a full examination of the changes that reintroducing operations from both heliports would have revealed the flaw in the assumptions about the relative position of the two LZs, now a little further apart.
Similarly, it was wrong to conclude that the SMS carried out when the company first acquired the H130s would not require another look when they were introduced to the Sea World operation a few weeks earlier.
It seems that there had been no review of the routes, reporting points or ground procedures from the time of the initial SMS analysis.
Australia was an early adopter of SMS and took to the concept, perhaps because it appealed to a national liking for self-critique, but it is also the home of the mantra “She’ll be right, mate” – an idea that is the natural enemy of good SMS.
If nothing else, this tragic accident demonstrates that SMS is like physical fitness. If you do not exercise, muscles will atrophy – and without constant review an SMS will lose its effectiveness.
As such, if you don’t regularly ask if an SMS programme is as good as it can be, then in all probability it isn’t.
