Executive summary

What happened

On 5 April 2022, a Saab 340 aircraft, registered VH-ZRK and operated by Regional Express, was being prepared for an air transport flight from Melbourne, Victoria. On board were the captain, first officer, one flight attendant, and 23 passengers.

During the engine start, the ground power unit was disconnected prematurely from the left engine by ground crew, which resulted in an interrupted start. The flight crew initiated the interrupted start checklist which included motoring the engine to purge residual fuel. During the motoring procedure, the ground crew signalled to the flight crew (using gestures) that they observed flames and smoke coming from the left engine and to shut down the engine. In response, the crew shut down both engines and discharged the left engine fire extinguisher and ordered an evacuation. 

During the evacuation, 2 passengers sustained minor injuries. A subsequent inspection found no indication of an engine failure or damage prior to or at the time of the occurrence. 

What the ATSB found

The ATSB found that, due to repeated non-standard signals from the ground crew about smoke and flames from the left engine, the motoring cycles were not completed which resulted in a tailpipe fire. 

The captain, based on the continued signals from the ground crew, and a rising interstage turbine temperature (ITT), decided that an emergency evacuation was required. However, the captain had not communicated their observations or actions to the first officer prior to making this decision which limited their opportunity to contribute to the identification and management of the situation. The captain did not use all available information to positively confirm the severity of the situation prior to ordering the evacuation and did not communicate their intent to the first officer. 

The decision to evacuate occurred while the flight attendant was using the public address system to brief passengers. As a result, the captain did not initially use the primary method (an announcement over the public address system) to alert the cabin that an evacuation was required, instead using a secondary method (4 chimes). The secondary method was not recognised by the flight attendant, so they were unaware that an evacuation was required until the captain opened the flight deck door to communicate directly.

When the flight attendant then ordered passengers to evacuate, there were limited instructions provided to them. As a result, the passengers in the emergency exit row did not open a useable exit which delayed the evacuation of the aircraft.

During the evacuation, some passengers took their baggage which increased the risk of injury and delays exiting the aircraft. The response to the situation by the aviation rescue and firefighting service was initially delayed by 2 minutes due to the nature of the problem not being communicated directly to air traffic control from the aircraft.

On an organisational level, the ATSB found that Regional Express did not provide flight crew or ground crew recurrent training to review the hand signals required to communicate with each other, including those used in an emergency. In addition, although flight crew were required to notify ground crew if an evacuation had occurred as necessary, the operator did not provide awareness to ground crew on the actions to be taken in the event of an evacuation.

What has been done as a result

Regional Express developed an additional hand signal to indicate an interrupted engine start and included this in training content for both ground and flight crew. The operator also provided additional training and guidance to ground and flight crew about marshalling and dispatch procedures (including all hand signals) and reviewed all related operational manuals. To maintain awareness, posters detailing hand signals were placed in crew high traffic areas.

Safety message

The use of all available resources, including seeking input from other crew members, particularly in abnormal or emergency situations, assists in being able to positively identify the nature of a problem. Undertaking an informed and coordinated approach to decision‑making ensures that the most appropriate action can be taken.

This occurrence also highlights the importance of the use of standard communications (including hand signals) to be able to effectively convey information, particularly in a potentially time‑critical situation. 

The occurrence

Interrupted engine start 

On 5 April 2022, a Saab 340 aircraft, registered VH-ZRK and operated by Regional Express, was being prepared for an air transport flight from Melbourne, Victoria to King Island, Tasmania. The scheduled departure time was 1445 local time. On board were the captain, first officer (FO), one flight attendant, and 23 passengers. 

Following door closure, the flight attendant briefed the emergency exit row passengers and commenced the passenger safety briefing using the public address system. In accordance with the normal procedure, the aircraft was connected to a ground power unit (GPU),[1] and a marshaller (ground staff member) was positioned at the front of the aircraft to monitor the engine start. Closed‑circuit television (CCTV) and the cockpit voice recording showed that at 1444:15 the flight crew had completed the engine start checklist and the right engine was started. 

At 1444:53 the flight crew initiated the start sequence for the left engine, and the propeller began to rotate. Shortly after at 1445:11, and before receiving a signal from the flight crew to do so, a second ground staff member went to the rear right side of the aircraft and disconnected the GPU. They recalled not being sure why they disconnected the GPU without the instruction to do so, but recalled being preoccupied with the next aircraft departure.

After realising the GPU had been disconnected (as the flight deck instruments no longer had power, which at that time were selected to be receiving power from the GPU), the captain started to action the interrupted engine start memory items. The procedure included moving the condition lever to the ‘fuel off’ position, the ignition switch to the ‘off’ position, and motoring[2] to remove any residual fuel from inside the engine. Motoring was initiated by the captain firstly selecting the start switch to the left engine, which also returned power to the flight instruments again. By this time the left engine propeller had stopped and the engine had decelerated from the original start attempt about 30 seconds before. 

Flame and smoke

At 1445:28, as the left engine propeller began to rotate again as part of the engine motoring procedure, a tailpipe fire had developed as indicated by flame and smoke coming from the rear of the engine. The marshaller, who was still positioned at the front of the aircraft, saw a ‘burst of flame’ from the back of the engine (Figure 1) and began to signal to the flight crew to stop what they believed to be an additional engine start (as they were unaware of the motoring procedure). As the marshaller and flight crew were not able to communicate verbally (there was no radio and no communications headset for the Saab 340), the marshaller reported they tried to signal to the captain to shut the engine down using a ‘cut-throat’ hand signal. They also tried to indicate that there was a fire coming from the back of the engine, however, the marshaller could not recall the hand signal for fire and instead attempted to communicate it by mouthing the word ‘fire’ and gesturing to the left engine. CCTV recorded the marshaller waving their hands at this time. The captain was responding to the marshaller verbally (noting the marshaller could not hear) advising that they had to continue, however the marshaller continued to try to advise that they had seen fire from the left engine. As a result of the marshaller indicating they observed flames from the engine, the captain ceased motoring the engine and the left propeller stopped rotating.  

Both flight crew recalled that at about that time the left engine interstage turbine temperature (ITT)[3] was still rising and in response the captain decided to make a second attempt at motoring. At 1446:02 the captain began to motor the engine again. Almost immediately, the marshaller again thought the flight crew were trying to start the engine and signalled to them to stop which prompted the captain to check outside their window. The captain could not see any flame or fire (noting that only the front of the engine is visible from the flight deck). Other than the rising ITT and the signals from the marshaller, there were no other indications of fire. There were no master caution warnings or indications on the central warning panel for either an engine fire or tailpipe overtemperature.

Figure 1: Still image of CCTV showing flame during first motoring attempt

Source: Melbourne Airport, annotated by the ATSB

Engine fire checklist 

The captain reported that, based on the signals from the marshaller and the rising ITT which was observed to be 850^oC, they decided to cease motoring the engine and action the engine fire emergency checklist. At 1446:42 the captain stated, ‘alright mate, fire’ and instructed the FO to pull the fire handle for the left engine which discharged the left engine fire bottle. At 1446:51 following the activation of the fire bottle and as the propeller was slowing, flame and smoke could again be seen coming from the left engine tailpipe. The right engine was shut down and the propeller started to slow down. At 1447:17, the captain instructed the FO to discharge the second fire bottle, around 30 seconds after the first. The captain later recalled that they omitted 2 engine fire checklist items relating to the power and condition levers (see Engine fire procedures) due to the stress at the time. 

Evacuation

While still actioning the engine fire checklist, the captain ordered an evacuation at 1447:03 by cycling the seatbelt sign, creating 4 chimes. Although the primary method for the captain to communicate that an evacuation was required was by making a public announcement (PA) using the public address and intercom system, the captain incorrectly believed they would be unable to use it as the flight attendant was already using the system for the passenger safety briefing. The captain had not yet verbalised their intent to evacuate the aircraft or called for the evacuation drills to the FO. 

The flight attendant did not initially recognise the chimes as a signal to evacuate and continued with the briefing.[4] At 1447:33, about 30 seconds after the initial order, the captain opened the flight deck door and instructed the flight attendant to commence an evacuation, specifying the use of the ‘FO side’ (right side) exits only.[5] Prior to the captain opening the flight deck door to order the evacuation, the flight attendant and the FO were still unaware that an evacuation was required. At 1447:36, the captain made a PA to the cabin to advise passengers to evacuate. At 1447:56 the flight attendant began shouting evacuation commands and opened the R1 exit, discarding it outside the aircraft (see Emergency exits). Passengers were instructed to ‘evacuate’, ‘leave everything behind’, and ‘sit and jump’. The passengers seated in the exit row, at row 6, discussed opening their exit, but decided to instead evacuate through R1. 

The first passenger at the R1 exit hesitated due to the height of the door sill (1.68 m without steps or escape slide) but was encouraged by the flight attendant to exit. This passenger received a minor injury. The FO exited the aircraft after the first passenger, and a refueller and ground staff assisted the remainder of passengers to evacuate. The flight attendant continued to instruct passengers to evacuate and to ‘sit and jump’. Some passengers attempted to bring their bags during the evacuation and the flight attendant and the captain (after exiting the flight deck) instructed them to leave everything behind. After receiving these instructions, a small number of passengers still exited the aircraft with their bags.

At 1447:56, as the flight attendant first started issuing the evacuation instructions, the captain radioed a Regional Express duty officer to advise that there was a fire and evacuation. The captain then attempted to directly contact the aviation rescue fire fighting (ARFF) service by radio to advise of the fire. The captain was unsure of the frequency for ARFF and did not receive a response. At this time, it was unknown whether the emergency checklist was completed as no items were verbalised by the captain. The FO can be heard saying ‘torch, fire extinguisher’. The captain then contacted the ground air traffic controller at 1449:58 to request assistance, stating ‘we’ve got a fire on the bay, can we get the fire chief please’. At 14:50:17, the controller advised the flight crew via radio that they had activated the crash alarm,[6] the fire vehicles would be dispatched and asked the captain the nature of the problem. The flight crew did not respond.

During the evacuation, the ground crew attempted to help the passengers exiting the aircraft but also recalled being unsure of how they should assist. By 1451:24 all passengers and crew had evacuated through the R1 exit, with the evacuation lasting about 4 minutes. There were 2 reported minor injuries sustained during the evacuation: one passenger with a knee injury and another passenger with a grazed elbow. Following the evacuation, the passengers were transported by bus to the terminal, with one passenger undergoing medical treatment. At 1452:42 the ARFF vehicles were dispatched and at 1453:44 the first vehicle arrived.

A subsequent examination of the engine and aircraft did not identify any engine defects, or any fire damage.  

Context

Personnel information

Flight crew 

The captain held an air transport pilot’s licence (aeroplane) with 8,852 hours of aeronautical experience, 5,058 of which were on the Saab 340. The captain’s last emergency procedures training (which included passenger evacuation) was completed in September 2021 and their last cyclic simulator check was completed in January 2022.

The first officer (FO) held a commercial pilot’s licence (aeroplane) with 4,725 hours of aeronautical experience, 952 of which were on the Saab 340. The FO’s last cyclic simulator check was completed in February 2022 and last emergency procedures training (which included passenger evacuation) was completed in March 2022.

Flight attendant

The flight attendant had 12 months’ experience with the operator on the Saab 340 aircraft and had completed their emergency procedures training (which included passenger evacuation) in September 2021.

Ground crew

The marshaller completed their marshalling and receipt and dispatch training in September 2019. This training included GPU connection and disconnection and performing hand signals, including the signal for fire. The second ground crew member completed their training in October 2019 and had been with the operator for 2.5 years. 

Aircraft information

General information

VH-ZRK was a Saab 340B, serial number 340B-397. The Saab is a twin turboprop aircraft fitted with 2 General Electric CT7-9B turboprop engines and is capable of carrying up to 38 passengers (including infants). VH-ZRK was configured with 34 passenger seats.  

Engine start

Usually, an engine start is achieved using either the aircraft battery or by using an external GPU. The starting sequence required a flight crew member to engage the engine’s starter, and then to introduce fuel at the appropriate time. The fuel was then ignited (light-off), and the engine would begin to accelerate. When the engine reached self-sustaining speed,[7] the engine’s starter would disconnect and the flight crew would select the generator ‘on’, and the engine would be ready for normal operation.

An interrupted start occurs when, during the start sequence, prior to the engine reaching self‑sustaining speed, power is lost or not supplied to the starter. When this occurs, the engine will begin to decelerate and fuel will continue to be sprayed into the combustion chamber of the engine, until the condition lever is moved into the ’fuel-off’ position. From the available information, the left engine start was interrupted when the GPU was disconnected prematurely. 

Interrupted start 

The steps normally taken to manage an interrupted start would be conducted from memory by the flight crew using the interrupted start procedure. The objective of the procedure was to prevent engine exceedances and subsequent damage (which can include a tailpipe fire). The required actions stopped fuel being fed to the engine removing its ignition source, and purged any residual fuel from the turbine and tailpipe area by motoring the engine.

Engine and tailpipe fires

Difference between engine and tailpipe fires

There are two main types of fire associated with engines in the Saab 340B: an engine fire and a tailpipe fire. An engine fire is an external fire, outside of the engine casing. When this occurs the ENGINE FIRE warning is triggered, and the nacelle temperature rises. The appropriate crew action is to shut off the fuel supply, isolate the engine and, if necessary, discharge the nacelle fire extinguisher. 

A tailpipe fire is an internal fire that is contained within the engine casing, usually when on the ground. It is due to an excess of fuel in the combustion chamber or in the turbine and may be associated with an engine that is not running (self-sustaining). It causes a rise in interstage turbine temperature (ITT) (which measures temperature inside the engine) but does not trigger a fire warning. It can be visually detected when the engines are in line of sight. The nacelle fire extinguishers cannot extinguish a tailpipe fire; ground fire extinguishers can be used if motoring (to purge residual fuel) does not extinguish the fire. A tailpipe overtemperature would be managed by reducing power on the affected engine.

Engine fire and tailpipe overtemperature detection 

Engine fire detection was provided by a continuous loop detection circuit mounted under each engine, outside of the casing when the temperature reaches approximately 300^oC. Proximity of fire to this loop would change its characteristics and trigger multiple types of warnings in the flight deck. These warnings consisted of a fire bell, master warning lights, a caption for each engine on the central warning panel, and lights in the fire handle for the relevant engine (Figure 2).

Each engine also had tailpipe overtemperature sensors that triggered the master warning lights and a caption for each engine on the central warning panel (CWP) in the flight deck if the temperature in the tailpipe reached approximately 235 °C. The lights would extinguish after the temperature had reduced. 

Figure 2: Saab 340B flight deck

Source: Bidgee (Wikimedia commons), annotated by the ATSB

Public address and intercom system

The public address and intercom system has multiple functions. It can be used to make a public announcement (PA) to the cabin from either the flight deck or flight attendant station and is also used for normal or emergency inter-crew communication between the flight crew and the flight attendant.  

The air crew emergency manual stated that a PA from the flight deck will override those being made in the cabin. The aircraft manufacturer conducted testing of the public address system at the request of the ATSB. They reported passenger announcements are prioritised firstly by the flight crew, followed by flight attendants. 

The testing also identified that if the emergency button was pressed in the flight deck while a PA was being made using the cabin interphone, aural and visual emergency indications would be present however the PA would not be interrupted.

Emergency exits

There are 4 emergency exits on the Saab 340 type aircraft. There are 2 exits at the front of the cabin in row 1, and 2 overwing exits in row 6. The sill height of the 2 forward exits is 1.68 m, with no emergency escape slides fitted (Figure 3).

Figure 3: Emergency exit locations

Source: Regional Express, annotated by the ATSB

The flight attendant was responsible for opening the 2 forward exits in an evacuation, one being the entry door, a type I exit at row 1 on the left side of the aircraft (L1), and a type II exit on the right side of the aircraft (R1). In contrast to L1, the R1 exit is not hinged and when operated for an evacuation will fall inwards and be discarded outside the aircraft (Figure 4). The overwing exits in row 6 are type III exits (also referred to as self-help exits). These exits fall inwards when operated and must also be discarded outside the aircraft. The operator’s procedures required that passengers in row 6 be briefed about exit use, with the expectation that they will be able to operate the exits in an evacuation when directed (see Passengers seated in exit rows).

The air crew emergency manual stated that all useable emergency exits must be used in an evacuation. It also stated that any evacuation requiring the utmost speed should be completed in 90 seconds. While an aircraft with less than 44 seats is not required to demonstrate evacuation procedures during certification, the expectation for larger air transport aircraft is that an evacuation can be completed within 90 seconds, under simulated conditions, using only half the available exits.

For this occurrence, the evacuation, from when the evacuation announcement was made until all passengers and crew had evacuated, was completed in about 4 minutes utilising one of the 4 exits. 

Figure 4: R1 emergency exit on the Saab 340B

Source: Melbourne Airport, annotated by the ATSB

Operational information

Flight crew procedures for interrupted start 

Ground power units (GPU)

The flight crew operations manual (FCOM) outlined that ground power units are desirable for engine starts and are to be used whenever possible. A caution was included that the ground power switch must not be intentionally selected to ‘off’ during the start procedures. The manual further outlined if there were problems with the GPU during the start, then the interrupted start sequence procedures/checklists would need to be completed.

Interrupted start procedure

If the start sequence had been interrupted prior to light-off, the memory items required the crew move the condition lever to the ‘fuel off’ indent, the ignition switch to ‘off’ and the engine was required to be motored for at least 10 seconds, but no more than 30 seconds. If the start sequence had been interrupted after light-off, the engine was required to be motored until the ITT had reduced to less than 175 °C.

Engine fire procedures

Flight crew procedures

The air crew emergency manual stated:

A fire is classified as an emergency. If fire or smoke is observed outside the aircraft, or there is heat radiating from behind bulkheads or panels, the Captain must be advised immediately. Continual surveillance by air crew is essential at all times.

Fires managed by flight crew may include an engine fire… The Flight Attendant must be advised of the intended plan of action regardless of the type of fire.

In the event of an engine fire, the flight crew would undertake memory item actions. They included reducing the power lever to 20–30% torque, moving the condition lever to ‘fuel off’, pull either the left or right fire handle, and then discharge the fire bottle in that engine’s nacelle with a switch adjacent to the handle. If an engine fire warning was still present 30 seconds after the first bottle was discharged, the flight crew could discharge a second fire bottle (located in the opposite engine’s nacelle) by selecting the switch for the opposite engine. The extinguishing agent was discharged around the outside of the engine inside the nacelle.

The FCOM stated that memory items require an immediate crew response and therefore must be committed to memory. If an emergency or abnormal situation occurred and memory items was required, the pilot not flying (also referred to as pilot monitoring) would call the item and place their hand on the appropriate lever, switch and call the action to be taken. The pilot flying would confirm the item and then the pilot not flying would action it. The engine fire procedure was applicable when the aircraft was on the ground or airborne. 

The Rex airport services manual included ground crew hand signals for fire (Figure 5). The hand signals were derived from the International Civil Aviation Organization’s Annex 2: Rules of the Air. The airport services manual was provided to Rex ground staff, but not to flight crew. These signals were taught to ground crew during their initial training only. 

Figure 5: Hand signal for fire

Source: Regional Express

The Rex policy and procedures manual, referenced by flight crew, included the standard marshalling signals and a description for the hand signal for aircraft smoke, which was the same for fire, but no image was included. The flight crew recalled that they could not remember the signal for fire and had not covered it specifically in their training. The captain recalled learning the signal during their command upgrade through their own initiative and the FO learnt the signal in their previous employment. 

Aviation Rescue Fire Fighting

The air crew emergency manual stated that in the case of an emergency situation, a national ARFF emergency frequency (131.0) was available for direct communication between the fire commander and flight crew. Before operating on the frequency, air traffic control must be advised.

The fire commander stated in interview that they initially do not respond directly to calls on the frequency. They are contacted by air traffic control and either placed on alert or the crash alarm is activated. The flight crew can then communicate with the fire commander. They also aim to respond within 3 minutes, and on the day of the occurrence arrived at the aircraft about 1 minute after being notified. 

Failure management principles for flight crews

The FCOM included failure management principles if a malfunction occurred. It stipulated:

Prior to actioning any procedures ensure the malfunction is positively identified before any action is taken, and under no circumstances shall control of the aircraft be compromised. To ensure that the correct procedure/drill is performed the [Pilot Not Flying] will identify the malfunction and the [Pilot Flying] will confirm the identification. (Original emphasis included).

Confusion is often a problem area when conducting [Quick Reference Handbook] procedures. Checklist procedures must not be rushed. It is important crews conduct checklists in a careful and controlled manner.

Evacuation procedures 

Flight crew procedures

The air crew emergency manual stated that if smoke is evident from the engines after start up and continues to produce increasing amounts of smoke after shutting down, the captain may order an emergency evacuation. During an emergency evacuation, depending on the location of the fire, smoke or heat, the captain may advise which exits are to be used.  

Further, the manual stated that the captain is ultimately responsible for determining, in consultation with the FO, an appropriate course of action. The FO must therefore ensure that any actual or impending abnormality would be brought to the attention of the captain.

If the captain believed an evacuation was likely, they must command immediately using the PA system: ‘This is the Captain. Flight Attendant to your station’. 

After this PA, the flight crew must perform the relevant emergency procedures, including the appropriate evacuation checklist as required. 

The quick reference handbook (QRH) also included additional information about conducting emergency evacuations:

When aircraft has come to a stop and parking brake set, [pilot in command] shall use all relevant sources – such as visual observations, external communication and [flight attendant] – in judging the situation quickly: Is emergency evacuation required or not? If in doubt, always perform emergency evacuation although high risk of injury.

Once the decision to evacuate had been made, a number of memory items for flight crew were to be completed as shown in Figure 6.

Figure 6: Flight crew emergency checklist 

The left pilot refers to the captain and the right pilot refers to the first officer. Source: Regional Express.

For the evacuation order, the air crew emergency manual explained there was a primary and secondary method to signal to the cabin that an evacuation was required. The primary method was via a PA to the cabin, stating ‘This is the Captain, Evacuate! Evacuate!’ and was used whenever possible. The secondary method, where the FASTEN SEATBELTS sign would be cycled twice to give 4 chimes was only to be used in 2 circumstances:

• it was known the public address system is unserviceable (and the flight attendant had been advised of this), or
• it became known the public address system had not worked correctly or had subsequently become unserviceable.

Once the emergency checklist items were completed and an evacuation commenced flight crew were required to undertake the duties detailed in Table 1 below.

Table 1: Flight crew duties in an evacuation

Captain’s duties	First Officer duties
Don his/her cap	1.  Don his/her cap
Take the torch from the flight deck	2.  Take the torch from the flight deck
Enter the cabin and assist with the evacuation, as per directions from the flight attendant	3. Take the flight deck fire extinguisher
Perform final cabin check shouting final commands	4. Exit through the first available exit
Exit aircraft through the first available exit	5. Once outside, check for fire and extinguish is necessary
Meet upwind of the aircraft with fellow crew members and passengers, and	6. Assist passengers on the ground as required, and
Account for all passengers	7. Ensure passengers remain grouped, proceed upwind from the aircraft and wait for fellow crew members

The FO was to exit the aircraft as soon as they entered the cabin, and would be the first crew member evacuating the aircraft and generally the first or one of the first occupants to evacuate. The flight attendant and captain would both exit once all the passengers had evacuated the aircraft.

Flight attendant duties in evacuations

The air crew emergency manual also detailed flight attendant duties during an evacuation. Key actions and commands to evacuate passengers relevant to the occurrence are detailed in Table 2. Flight attendants were required to review emergency procedures, including signals to the cabin from the flight crew, every 7 days.

Table 2: Flight attendant actions and commands

Key actions	Evacuation commands
Shout initial commands	Evacuate Evacuate Unfasten Seatbelts Leave Everything Behind Get Out Check Outside Conditions If Safe Open Exits
Check outside conditions
Open usable exits/guard unusable exits
Instruct passengers not to open unusable exits (if necessary)
Shout exit commands	Hurry Come This Way
Shout evacuation commands	Sit and Jump Move Well Away
Evacuate passengers
Check cabin
Shout final commands	Evacuating
Exit aircraft through first available exit, move upwind of the aircraft and report to the first officer

The flight attendant said in their interview they were halfway through providing the initial commands when the passengers were in front of the R1 exit so they stopped. They also recalled providing evacuation commands. On the Cockpit Voice Recorder (CVR), the flight attendant can be heard saying ‘evacuate evacuate’, ‘leave everything behind’ and ‘sit and jump’. In addition, they can also be heard to be encouraging passengers to sit, and to ‘go, go, go’.  

Passengers seated in exit rows

Passengers seated in emergency exits were required to assist in the event of an emergency and are asked at both check-in and at boarding if they are willing and able to accept the responsibility. After boarding passengers seated in the exit rows are provided with a specific briefing about the actions they may be required to take. The briefing included: Only if you hear the command ‘Evacuate Evacuate’ should you take action. 

First check outside conditions (point to exits) for fire, water or any other obstruction. 

If conditions are not safe, or there is water over the wing, do not open the exit and move to your nearest usable exit. 

If the exit is safe to open, push these seats forward as far as possible (point to 5A and 5C). 

Remove the cover and pull down on the handle (point to cover). 

Turn the exit on its side, throw it out and evacuate the aircraft. 

You must follow Crew instructions at all times. 

Once again, only if you hear the command evacuate should you take action. 

Do you have any questions?

The flight attendant recalled in their interview they provided the standard exit briefing to the passengers in row 6 prior to departure. In an evacuation, in addition to the command to evacuate, the flight attendant was also required to shout initial commands that instruct passengers to check outside conditions and if safe open the exits. Based on interviews, the passengers in the exit row (row 6) and the CVR, passengers did not receive this instruction, nor were they advised that the left side exits were not available based on the captain’s instructions. In interviews, the passengers in the exit row recalled they considered opening the exit but assessed it was not necessary as they did not observe any ongoing hazards and did not feel the urgency to evacuate.

Flight crew and ground crew communication

Both flight crew and ground crew had radios. The operator advised that many of the airports they used, including Melbourne Airport, did not have access to the ground-to-air VHF frequency to enable radio communication between flight crew and ground crew. Some airports, such as Port Augusta, had the infrastructure to support access to the frequency, and ground crew in those locations could directly communicate with the flight crew by radio. In locations where flight crew and ground crew could not communicate directly, they had the option to use the network operations centre to relay information.

Ground crew involvement in emergency evacuation

The emergency evacuation checklist included an item for the left seat pilot (captain) to notify ground crew. There is no further detail included in the operator’s manuals about the method of contacting the ground crew and/or any specific instructions about the information that should be relayed. The operator advised that it would be at the captain’s discretion which method they would use to contact the ground crew and what information they would relay. In Rex’s air crew emergency manual, it stated that if time is critical, requiring flight crew to proceed to an airport that does not have ARFF services, they must request company ground staff to call for the local fire service. The Rex airport services manual included procedures for brake fires which stated ground crew were to follow instructions from air crew and to keep people away from the wheels. However, the manual did not have details on ground crew involvement in evacuations nor did it refer to the ground crew roles as described in the air crew emergency manual. Further, no evacuation awareness training was provided to ground crew. The operator advised that the ground crew were trained for their specific role and the only regulatory requirements were providing evacuation training for flight crew and cabin crew. 

Guidance on ground crew involvement in evacuations

The International Civil Aviation Organization’s ground handling manual (2019) recommended that ground handling service providers train all relevant personnel in the functions they are to perform in an emergency, including the use of any emergency equipment and their obligations during an emergency evacuation. There was no regulatory requirement for this in Australia, however the Civil Aviation Safety Authority had published Cabin safety bulletin 25‑ Emergency Evacuation and Occupant Survivability that discussed the importance of procedures and training for flight crew, cabin crew, and ground crew in the event of an emergency evacuation. A number of previous investigations worldwide have discussed ground crew involvement in evacuations, as detailed in the related occurrences section below.

Related occurrences

A review of previous investigations in Australia and overseas found the following reports that discuss ground crew training are included in Table 3 below.

Table 3: Related investigations

Investigation number	Country	Summary	Ground crew actions	Relevant finding/safety actions
200302980	Australia	After landing, the pilot in command (PIC) observed a BRAKE TEMP advisory message and a fire ignited on the right wing landing gear. The PIC ordered an evacuation of the aircraft. As a result of the evacuation, one flight crew member and three passengers were seriously injured.	When the ground crew observed passengers congregating at the base of the slides, they acted on their own initiative and directed passengers away from the aircraft. Knowing that the fire had been extinguished, both of the ground engineers moved to the base of slide and, in an attempt to stop the evacuation, waved and called to the cabin crewmember attending that door.	There was no training provided to the ground staff for aircraft emergency evacuation situations at airport terminals.
200502137	Australia	During the starting of the right engine, the aircraft dispatcher informed the flight crew that there was smoke and sparks shooting from the right engine. The PIC called for an emergency evacuation. 11 passengers sustained minor injuries.	Ground crew assisted passengers during the evacuation and directed them towards the terminal. Several passengers exited and ran towards the edge of the apron without the knowledge of ground personnel.  One of the ground personnel assisting passengers tried to call out to the flight attendant to ‘slow down’ the flow of passengers as they had congregated at the end of the slides. However, he could not see the flight attendant and did not establish communication with her.	None of the operator’s ground personnel present had been given any awareness education about policy or procedures during an aircraft evacuation at the terminal.
A13q0186	Canada	A belt loader caught fire under the left aft cargo door and the smell of smoke entered the cabin. The captain ordered an evacuation. Seven passengers sustained minor injuries.	Ground crew helped the passengers coming down the evacuation slides. Passengers who found themselves on the apron without designated staff to help them wandered around looking for instructions and direction. Some employees reacted quickly by redirecting the wandering passengers towards the door leading to the boarding gate.	If ground crew on the apron are not trained to manage passengers following an evacuation, there is risk of injury both for evacuated passengers and ground crew.
fqa130728	France	A burnt smell was detected in the aircraft. In the cockpit, the first officer saw the message ‘smoke rest upper door’ with an aural warning. The chief flight attendant detected a smell of sulphur in the cabin. The captain decided to evacuate the aircraft. One passenger was injured.	The passengers evacuated by the slides were assisted by ground personnel and taken to the terminal.	Training of ground personnel in emergency evacuation.

Safety analysis

Introduction 

During the engine start, the ground power unit was disconnected prematurely. With the engine speed now slowing, burning fuel in the combustion chamber would have accumulated. This would have caused the increase in temperature inside the engine. 

The captain commenced the interrupted engine start procedure, which involved motoring the engine (running the starter motor without supplying fuel). 

However, during this process burning fuel was blown through the tailpipe, appearing as a plume of flame and smoke behind the engine. As the left propeller began to rotate during this motoring attempt, the marshaller (unaware of the motoring required by the procedure) mistakenly thought that the flight crew were attempting to restart the engine. Having observed the flame and smoke, the marshaller signalled to the flight crew, which prompted the captain to stop motoring the engine. Residual fuel would have continued to burn inside the engine. 

After observing the interstage turbine temperature (ITT) increasing shortly after, the captain resumed motoring the engine. However, in response to continuing signals from ground crew indicating smoke and flames, and the rising ITT, the procedure was discontinued. The captain ordered an emergency evacuation. Two passengers sustained a minor injury during the evacuation. 

The following sections discuss the crew decision-making and actions following the interrupted start and subsequent evacuation, as well as the operator’s procedures and training.

Ground power unit disconnection

While the aircraft marshaller was observing the left engine start process, a second ground crew member disconnected the ground power unit (GPU) prior to a signal from the flight crew. Aircraft manufacturer and operator procedures stipulated that the GPU was to be disconnected from the aircraft after both engines had been started and the flight deck ground power switch was selected off. While there was no time pressure, the ground staff member recalled being preoccupied by the next aircraft they had to prepare for departure. The left engine had not reached self‑sustaining speed, and as there was no longer power to the starter, the engine began to decelerate. Fuel continued to be sprayed into the hot engine until the condition lever was moved to the fuel ‘off’ position.

Contributing factor A ground crew member disconnected the ground power unit without having been signalled to do so. This happened while the left engine was starting, which resulted in an interrupted engine start, and the initial development of an engine tailpipe fire.

Ground crew signals

The interrupted start would have been surprising to the flight crew, but the captain quickly deduced the reason for it and appropriately initiated the interrupted start procedure, which would cut the engine fuel supply to the combustion chamber, purge residual fuel from the turbine and tailpipe, reduce internal engine temperature, and prevent an ongoing tailpipe fire.

As described above, the purging of residual fuel would have generated a plume of flame and smoke. Upon noticing this, the marshaller attempted to communicate the problem to the flight crew. 

With the flight crew and ground crew being unable to communicate directly using the radio at this location, the only recognised method of communication available was the use of hand signals. The hand signal for fire, documented in the operator’s airport manual, was to move the right hand in a fanning motion from shoulder to knee, while at the same time pointing with the left hand to the area of fire. The marshaller could not recall the hand signal for fire so used a ‘cut-throat’ gesture to signal to shut the engine down and mouthed ‘smoke’ and ‘flames’ to the flight crew while pointing to the left engine. Although the flight crew were able to interpret to some degree the ground crew’s gestures, the use of non-standard signals can increase the risk of miscommunication.

Other factor that increased risk The marshaller used non-standard signals when communicating indications of fire to the flight crew.

The marshaller recalled they had learned the hand signals in their initial training, but had not reviewed them since. The captain also indicated they could not recall the hand signal for fire, while the first officer (FO) was aware of the signal, but from previous employment. 

For many of the airports the operator used, there was no way for ground crew and flight crew to talk directly to each other for Saab 340 operations. For some, communication could get relayed via their network operations centre, but this would be inefficient during an emergency. As the hand signals were the only method to communicate emergencies between the flight and ground crew at many locations, it is important the signals are reviewed on a regular basis. However, the training received by ground crews was during initial training only and can be forgotten over time. The flight crews did not receive such training and while there was a description in their manual about the signal, there was no visual representation, increasing the risk of misinterpretation if the marshaller had used the correct signal. 

Other factor that increased risk Regional Express did not provide flight crew or ground crew recurrent training to review the hand signals required to communicate with each other, including those used in an emergency. (Safety issue)

Interrupted start procedure

In response to the plume of flame and smoke (generated by the engine being motored), the marshaller signalled to the flight crew with hand signals that included a ‘cut-throat’ gesture, which the captain understood as needing to stop the engine. In the absence of the marshaller being able to communicate that flame and smoke was being observed, the captain would not have known the reason for being signalled to stop the engine. In response, the captain stopped motoring the engine which meant that burning fuel remained in the engine, and probably in the tailpipe. Later, following discharge of the fire extinguisher bottle, additional flame and smoke came from the left engine tailpipe which indicated that the tailpipe fire had not yet been extinguished. 

Contributing factor After observing the marshaller's signals, the captain stopped motoring the engine. As a result, the fire was not extinguished.

Failure management and crew coordination

The operator had failure management principles that stipulated that prior to actioning any checklists, the malfunction must be positively identified. In this occurrence, these failure management procedures were not followed as the captain did not consult with the FO about the nature of the problem and their intended course of action. This meant that the FO was ‘behind’ in terms of the actions that were being taken by the captain and they probably did not share the same understanding of the situation. 

Had the captain included the FO in their management of the situation, there would have been an opportunity to review available information together, recognise the fire as an expected and manageable problem, and formulate a suitable plan. Instead, with limited information being used, and while under stress, the captain’s concern about the engine fire persisted and increased. 

During the second motoring attempt, the ITT was observed to be high, which can indicate that a tailpipe fire was still present. There were no warnings in the flight deck relating to an engine fire or a tailpipe overtemperature, however the flight crew then carried out the required actions for an engine fire. The tailpipe fire would likely have been extinguished at some time during the second motoring attempt. 

Throughout, the FO would have been largely reliant on their own observations to understand what was happening and therefore had limited opportunity to contribute. The FO reported they only became aware of the intent to evacuate once the captain had commenced the evacuation drill. This further limited the opportunity to involve the FO in identifying and managing the problem (also see Evacuation order below). 

Contributing factor The captain did not coordinate with the first officer their identification of, or response to, the interrupted engine start or the later fire signals from the marshaller. This limited the opportunity for the first officer to contribute to the identification and management of the ongoing situation.

Evacuation order

The marshaller was signalling to the flight crew they could observe fire and smoke, and this likely continued for a significant period of time. The captain also observed that the ITT was rising so decided to initiate the fire drill. The signals and rising ITT were the only sources of information used by the captain to evaluate the situation. The flight attendant was not asked whether they had or could observe any problems with the left engine, and there was no consultation with the FO about what information they understood at that time. 

Instead, the captain reacted immediately to what they considered to be an emergency that could endanger passenger lives. Research has shown that people under stress refer to a fewer number of cues prior to making a decision (Wickens et al., 2022). An aircraft fire is a serious situation that needs a timely response, but there is always injury risk associated with an emergency evacuation. Without being able to speak directly with the marshaller, more communication and coordination with the FO and flight attendant might have led the captain to realise that some level of fire and smoke from the tailpipe was to be expected during motoring, and that the appropriate action would have been to continue motoring in accordance with the applicable procedure. Instead, the captain misunderstood the reason for the fire and overestimated its seriousness, leading to the decision to evacuate.

Contributing factor Likely associated with increased stress and an escalating sense of urgency, the captain ordered an evacuation without having used all the available information to positively confirm the severity of the situation.

The captain initiated the evacuation without communicating this intent to the FO, who only became aware of the situation when the captain opened the flight deck door. It was not possible to establish whether the captain had commenced the evacuation drills prior to this, as there is no verbalisation recorded on the CVR of the initiation of the procedure or performance of the individual items, which included shutting down the engines. The captain later reported that they omitted to confirm 2 engine fire checklist items, relating to the power and condition levers, which was due to the stress at the time. These items were more likely to have been completed with FO involvement. If engines are not properly shut down when an evacuation is commenced, the risk of injury when exiting the aircraft is increased. Further, as the FO was initially not aware that an evacuation was going to occur, this delayed the completion of their evacuation drills and exiting of the aircraft where they needed to check for fire and assist passengers. 

The initial method of communicating the evacuation was the chimes as the flight attendant was completing a PA. This method is a secondary method of communicating the evacuation. The primary method is to communicate using the public address system to prepare the flight attendant for their evacuation procedure. However, as the flight attendant was completing an announcement, the captain (incorrectly) believed they were unable to make an announcement. 

Other factor that increased risk Before ordering the evacuation, the captain did not communicate their intent to evacuate the aircraft to the first officer or call for flight deck evacuation drills. In addition, the captain initially did not use the primary method of commanding an evacuation (through the public address system).

Although likely familiar with the use of the seatbelt chime as an evacuation signal, the flight attendant did not recognise the chimes at the time as a signal to evacuate. Flight attendants are trained to listen for the seatbelt chime, but it is not usually associated with an emergency (unlike a siren for example) and may not gain a person’s attention if they are focusing on a task, such as completing the public address. As the flight attendant recalled, it was common to hear chimes during flight preparation. This method of initiating an evacuation also does not directly convey the intended message in the same way that spoken communication does, so it may be misunderstood or overlooked. Further, as the operator’s procedure only required the use of the chimes when the public address system is unserviceable, and in that situation the flight attendant may be advised, it is even more unlikely that during this occurrence the flight attendant would react to the signal.

Other factor that increased risk When the captain signalled to the cabin to evacuate (using 4 [seatbelt sign] chimes) the flight attendant did not recognise the signal and subsequently did not react to the command.

Passenger instructions

Once the captain ordered the evacuation, they informed the flight attendant to use the right side of the aircraft (due to the expected fire on the left side). While passengers evacuated the aircraft on the right side, only the R1 (front right) exit was used. The right side also has an overwing exit, but this exit was not used during the evacuation. Passengers are not well-drilled in emergency evacuations like flight attendants, and despite a pre-flight briefing of passengers in the exit rows about what to do if an evacuation was needed, flight attendant procedures include shouting commands to passengers in the exit row to check outside and open exits. The flight attendant can also provide instructions to passengers not to open unusable exits. These instructions were not provided by the flight attendant, and passengers at the exits in row 6 had decided among themselves not to open the exits, and instead followed the other passengers to the R1 exit. As a result, the evacuation took 4 minutes although an evacuation with larger aircraft with a full passenger load would be expected to be able to be completed from one side of the aircraft within 90 seconds. If there had been a larger developing fire at the time, the 4-minute evacuation could have had negative consequences for the people towards the end of the evacuation.

Other factor that increased risk When ordered to evacuate, there were no instructions provided to passengers to direct them to use all the available exits. As a result, the passengers in the emergency exit row did not open a useable exit which delayed the evacuation of the aircraft.

Emergency response

After ordering an evacuation and advising the operator’s duty officer, the captain tried to directly contact ARFF but was unsuccessful and only then contacted air traffic control (ATC). The ARFF advised that they do not respond directly to calls and their response is initiated through ATC. Not immediately communicating the situation to ATC can delay the arrival of emergency services to the aircraft, which in some circumstances can increase risk or severity of outcome.

Other factor that increased risk The captain did not initially communicate the situation to air traffic control, which delayed the arrival of emergency services at the aircraft by about 2 minutes.

Passengers with cabin baggage

During the evacuation, several passengers attempted to retrieve their baggage as they exited the aircraft, with some passengers refusing to leave items behind. This occurred even though the passengers were commanded to leave everything behind. Given the nature of the incident and that passengers could not see a fire, they may have determined that they were not at risk or felt urgency. However, it has been well established through numerous previous evacuations (for instance, ATSB investigation AO-2019-073), that taking cabin baggage not only increases the risk of injury while evacuating but can also delay an evacuation. Delaying the evacuation affects the individual taking the bag and also puts others at risk. In a situation where there is a potential fire, exiting as expeditiously as possible is vital. This is why, when larger aircraft are certified they must prove that they can be evacuated within 90 seconds, because in the case of smoke and/or fire conditions can worsen rapidly. 

Other factor that increased risk Some passengers took cabin baggage during the evacuation, which increased the risk of injury and delaying the evacuation.

Ground crew role in evacuation

Ground crew had not been provided any awareness of what might occur in the event of an evacuation at an airport, however the flight crew’s evacuation procedures included a checklist item to notify the ground crew in some circumstances and, in regional ports, there was an expectation that ground crew would initiate an emergency response. 

While it was expected that the flight crew and flight attendant would complete an evacuation by themselves, ground crew, who are often in close proximity to an aircraft when at the airport, are in a position where they can potentially assist. More specifically this assistance could include helping passengers to evacuate safely from the emergency exits from outside the aircraft. In this case, had the ground crew been aware of the evacuation, they may have been in a position to assist the passenger who exited the aircraft prior to the FO, reducing the likelihood of injury. 

The FO is the first crew member to exit the aircraft and they are required to check for fire and then assist passengers outside. They are also required to maintain control of the passengers, which for the Saab 340 can be up to 38 passengers (including infants), until the flight attendant and captain exit the aircraft or emergency services arrive. If an evacuation occurs at an airport, ground crew can also assist in maintaining passenger control once passengers have evacuated. Awareness of what occurs in an evacuation is also important as their proximity to the aircraft can expose them to hazards, such as exits being discarded outside the aircraft. 

Although evacuations are relatively rare and the likelihood of serious injuries or fatalities as a result of ground crew not assisting passengers is low, a number of previous investigations conducted in Australia and overseas have identified the importance of ground crew awareness of evacuations. Guidance available from the Civil Aviation Safety Authority and the International Civil Aviation Organization also recommend training for ground crew for evacuations. While it did not worsen the outcome in any way, ground crew being aware of what happens during an evacuation could be beneficial in other circumstances, particularly when they may be the only personnel around to assist such as in some regional locations.

Other factor that increased risk While flight crew were required to notify ground crew of an aircraft evacuation in some situations, the operator did not provide awareness to ground crew on the actions to be taken in the event of an evacuation.

Findings

ATSB investigation report findings focus on safety factors (that is, events and conditions that increase risk). Safety factors include ‘contributing factors’ and ‘other factors that increased risk’ (that is, factors that did not meet the definition of a contributing factor for this occurrence but were still considered important to include in the report for the purpose of increasing awareness and enhancing safety). In addition ‘other findings’ may be included to provide important information about topics other than safety factors.  Safety issues are highlighted in bold to emphasise their importance. A safety issue is a safety factor that (a) can reasonably be regarded as having the potential to adversely affect the safety of future operations, and (b) is a characteristic of an organisation or a system, rather than a characteristic of a specific individual, or characteristic of an operating environment at a specific point in time. These findings should not be read as apportioning blame or liability to any particular organisation or individual.

From the evidence available, the following findings are made with respect to the interrupted engine start and evacuation involving Saab 340B, VH-ZRK, at Melbourne Airport, Victoria, on 5 April 2022.

Contributing factors

• A ground crew member disconnected the ground power unit without having been signalled to do so. This happened while the left engine was starting, which resulted in an interrupted engine start, and the initial development of an engine tailpipe fire.
• After observing the marshaller's signals, the captain stopped motoring the engine. As a result, the fire was not extinguished.
• The captain did not coordinate with the first officer their identification of, or response to, the interrupted engine start or the later fire signals from the marshaller. This limited the opportunity for the first officer to contribute to the identification and management of the ongoing situation.
• Likely associated with increased stress and an escalating sense of urgency, the captain ordered an evacuation without having used all the available information to positively confirm the severity of the situation.

Other factors that increased risk

• The marshaller used non-standard signals when communicating indications of fire to the flight crew.
• Regional Express did not provide flight crew or ground crew recurrent training to review the hand signals required to communicate with each other, including those used in an emergency. (Safety issue)
• Before ordering the evacuation, the captain did not communicate their intent to evacuate the aircraft to the first officer or call for flight deck evacuation drills. In addition, the captain initially did not use the primary method of commanding an evacuation (through the public address system).
• When the captain signalled to the cabin to evacuate (using 4 [seatbelt sign] chimes) the flight attendant did not recognise the signal and subsequently did not react to the command.
• When ordered to evacuate, there were no instructions provided to passengers to direct them to use all the available exits. As a result, the passengers in the emergency exit row did not open a useable exit which delayed the evacuation of the aircraft.
• The captain did not initially communicate the situation to air traffic control, which delayed the arrival of emergency services at the aircraft by about 2 minutes.
• Some passengers took cabin baggage during the evacuation, which increased the risk of injury and delaying the evacuation.
• While flight crew were required to notify ground crew of an aircraft evacuation in some situations, the operator did not provide awareness to ground crew on the actions to be taken in the event of an evacuation. 

Submissions were received from:

• Regional Express
• Saab.

The submissions were reviewed and, where considered appropriate, the text of the report was amended accordingly.

Purpose of safety investigations The objective of a safety investigation is to enhance transport safety. This is done through:  identifying safety issues and facilitating safety action to address those issues providing information about occurrences and their associated safety factors to facilitate learning within the transport industry. It is not a function of the ATSB to apportion blame or provide a means for determining liability. At the same time, an investigation report must include factual material of sufficient weight to support the analysis and findings. At all times the ATSB endeavours to balance the use of material that could imply adverse comment with the need to properly explain what happened, and why, in a fair and unbiased manner. The ATSB does not investigate for the purpose of taking administrative, regulatory or criminal action. Terminology An explanation of terminology used in ATSB investigation reports is available here. This includes terms such as occurrence, contributing factor, other factor that increased risk, and safety issue. Publishing information Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2024 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Commonwealth Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this report is licensed under a Creative Commons Attribution 4.0 International licence. The CC BY 4.0 licence enables you to distribute, remix, adapt, and build upon our material in any medium or format, so long as attribution is given to the Australian Transport Safety Bureau.  Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

Executive summary

What happened

On the morning of 3 April 2022, a Bell Helicopter 206L-4, registered VH-PRW, departed with a pilot and passenger on board, for a visual flight rules (VFR) flight from a private property at Majura, Australian Capital Territory to Mangalore, Victoria, with a planned refuelling stop in Tumut, New South Wales (NSW). The helicopter was one of 7 helicopters taking part in a flying tour that morning and the weather forecast indicated low cloud, rain and associated reduced visibility on the planned route.

Two of the 7 helicopters diverted to Wagga Wagga, NSW due to weather while 4 others landed near Wee Jasper, NSW. The pilot of VH-PRW elected to continue until they encountered poor weather conditions and landed the helicopter in the Brindabella region shortly before noon. At 1453 local time, the helicopter departed once again at low level, in overcast conditions with low cloud and light rain. At about 1525, the helicopter commenced a rapid climb and shortly after, entered a steep left descending turn which continued until the helicopter impacted terrain at an elevation of 4,501 ft. A search was initiated the next day with the accident site located later that evening. The helicopter was destroyed, and both occupants were fatally injured.

What the ATSB found

The ATSB found that, having encountered the forecast low cloud and reduced visibility conditions, the pilot landed the helicopter at an interim landing site. Later that day, the helicopter then departed into cloud and visibility conditions unsuitable for visual flight. It is highly likely these cloud and visibility conditions resulted in the pilot experiencing a loss of visual reference and probably becoming spatially disoriented. This led to a loss of control and an unsurvivable collision with terrain.

Safety message

Weather-related accidents remain one of the most significant causes of fatal accidents in general aviation. The ATSB publication Avoidable Accidents No. 4, Accidents involving Visual Flight Rules Pilots in instrument Meteorological Conditions found that in the decade from 1 July 2009 to 30 June 2019, 101 VFR into IMC occurrences in Australian airspace were reported to the ATSB. Of those, 9 were accidents resulting in 21 fatalities.

In relation to visual flight rules pilots flying into areas of reduced visibility, some key messages to manage risk are:

• Know your limits. VFR pilots should use a ‘personal minimums’ checklist to help control and manage flight risks through identifying risk factors that include marginal weather conditions. Only fly in environments that do not exceed your capabilities. For visual flight at night, ensure you are both current and proficient with disciplined instrument flight.
• Plan ahead. Avoid deteriorating weather by conducting thorough pre-flight planning. Ensure you have alternate plans in case of an unexpected deterioration in the weather and making timely decisions to turn back or divert.
• Don’t press on! Pressing on into instrument meteorological conditions with no instrument rating carries a significant risk of severe spatial disorientation due to powerful and misleading orientation sensations with no visual cues. Disorientation can affect any pilot, no matter what their level of experience.

The occurrence

On the morning of 3 April 2022, a Bell Helicopter 206L-4, registered VH-PRW, was to conduct a visual flight rules[1] (VFR) flight from a private property at Majura, Australian Capital Territory to Mangalore, Victoria with a planned refuelling stop in Tumut, New South Wales (NSW). The helicopter was one of 7 helicopters taking part in a flying tour, following a common itinerary but operating independently.

The weather forecast indicated that the planned route could be affected by low cloud, rain and associated reduced visibility. At about 0900 Eastern Standard Time,[2] two of the helicopters departed Majura. These helicopters encountered low cloud and elected to divert over lower terrain north of the Brindabella Ranges to Wagga Wagga, NSW.

At 1021, VH-PRW, with the pilot and one passenger on board, departed the property at Majura along with the remaining 4 helicopters. Recorded tracking data showed that VH-PRW initially tracked south‑east before turning west toward the Brindabella Ranges (Figure 1). The flight then proceeded south over Corin Dam before heading north to Wee Jasper. As the 5 helicopters approached Wee Jasper, they also encountered deteriorating cloud and visibility conditions. Four of the pilots elected to land on a property near Wee Jasper. The pilot of VH-PRW did not land but continued south to ‘attempt to find a way through to Tumut’.

Figure 1: Flight from Majura to Long Plain

Source: Google Earth and OzRunways, annotated by ATSB

At 1129, the pilot of VH-PRW encountered poor weather conditions and landed the helicopter alongside Long Plain Road in the Brindabella region, outside of mobile phone coverage.

Shortly after the helicopter landed, a passing motorist stopped and approached the helicopter. The pilot advised the motorist that they had landed to wait for better weather conditions before continuing the flight. The motorist arranged to check on the pilot and passenger during the motorist’s return journey later in the day if the helicopter had not already departed by that time.

At about 1230, when VH-PRW did not return to Wee Jasper and the pilot had not contacted other members of the tour, authorities were notified and a search for the helicopter was commenced.

At about 1415, the motorist returned along Long Plain Road and found that the helicopter had not departed. The motorist transported the pilot to a location that enabled phone contact and the pilot contacted the office of the tour organiser to advise of the safe landing. The pilot also stated that they intended to continue the fight following powerlines at about 50 ft above ground level (AGL). The tour organiser advised against this plan and then notified the other members of the tour and authorities of the landing. The motorist and pilot then returned to the helicopter.

Recorded flight tracking data showed that at 1453, the helicopter departed Long Plain Road, with the pilot and passenger on board. Two minutes later, one of the other pilots in the tour noted the helicopter tracking south on a flight tracking application.

Police officers dispatched to locate the helicopter arrived at the landing site just after it became airborne. The motorist and police officers observed the helicopter depart to the south at low level, in overcast conditions with low cloud and light rain. The police officers stated that the helicopter passed ‘at a similar height or slightly above the powerlines’ before being obscured by low cloud.

The flight then progressed at heights below 500 ft AGL following geographical features along lower lying terrain (Figure 2). At 1504, the flight turned north-west and took up a track that corresponded with a direct track to Tumut. Two minutes later, the helicopter encountered higher terrain and turned around to head southward, again following lower lying terrain. At 1517, in the vicinity of Anglers Reach, the flight turned north. Two minutes later, the helicopter turned to the north-west, again along a flightpath that corresponded with a direct track to Tumut and commenced a climb to about 7,000 ft above mean sea level (AMSL).

Figure 2: Accident flight

Source: Google Earth and OzRunways, annotated by ATSB

The helicopter continued along that track at groundspeeds of 105‑115 kt until 1525 when it descended to 6,800 ft. The helicopter then almost immediately commenced a rapid climb. As the helicopter climbed, the recorded groundspeed reduced. After reaching 7,400 ft at a groundspeed of 60 kt, the helicopter entered a steep left descending turn.

The descending turn continued until the helicopter impacted terrain at an elevation of 4,501 ft at about 1526. The helicopter was destroyed, and both occupants were fatally injured.

On 4 April, in response to the helicopter not re-joining the tour as expected, a second search was initiated. At about 2355, a ground search assisted by helicopter tracking data located the accident site.

Context

Pilot information

The pilot held a valid class 2 medical certificate and a private pilot licence (helicopter).

At the time of the accident, the pilot had about 837 hours of aeronautical experience and did not hold an instrument rating. The pilot’s total flying experience on the Bell 206 was about 532 hours of which about 355 were in the L-4 variant and the remainder in the B-3 variant.

The ATSB found no indicators that increased the risk of the pilot experiencing a level of fatigue known to affect performance.

The post-mortem examination and a review of the pilot’s medical history identified no evidence of a medical event or pre-existing condition that likely contributed to the accident.

Aircraft information

The Bell Helicopter 206L-4 is a 7‑seat, single‑turboshaft engine helicopter equipped with 2-bladed main and tail rotors. VH-PRW was built in 2008 and first registered in Australia in 2016. At the time of the accident, the helicopter had completed about 830 hours in service and was certified for day VFR flight only. The helicopter was fitted with an emergency locator transmitter.

The helicopter was also fitted with the HeliSAS stability augmentation system. This used attitude data and electro-mechanical servo actuators connected to the flight controls rods to apply small corrections to the cyclic as required to maintain a reference attitude. The reference attitude could be set as required by the pilot. The system also incorporated a two-axis (pitch and roll) autopilot.

Tour coordination

The flight from Majura to Mangalore was part of an informal multi-day flying tour involving 7 helicopters. This tour was mostly coordinated by a helicopter operator who provided the itinerary and organised logistic details such as accommodation and fuel availability.

The tour organiser also operated a helicopter flying training and transport operation, but this tour was conducted outside of that operation. The tour organiser held no authority or responsibility for the operation of each involved helicopter, this responsibility was held by each pilot in command.

Terrain

The helicopter departed an interim landing site along Long Plain Road in the Brindabella Ranges. The flight then proceeded over rugged alpine areas of the Snowy Mountains with terrain elevations generally higher than 4,000 ft AMSL. Peaks of 5,854 ft AMSL and 5,726 ft AMSL were located near the final flight track (Figure 3).

Figure 3: Visual navigation chart extract showing terrain in the vicinity of the accident

Source: Airservices Australia and OzRunways, annotated by ATSB

Meteorology

Forecast

The graphical area forecast for the accident region provided a forecast icing level of 10,000 ft AMSL and the following cloud and visibility conditions for the time of the accident (all heights AMSL):

• Generally greater than 10 km visibility with broken[3] cumulus/stratocumulus cloud between 2,500 ft and 10,000 ft.
• Visibility reducing to 4,000 m in scattered rain with broken stratus cloud between 1,500 ft and 6,000 ft. Overlying this, broken altocumulus and altostratus cloud could be expected extending from 6,000 ft to above 10,000 ft.
• Visibility reducing to 3,000 m in scattered rain showers with broken stratus cloud between 1,500 ft and 3,000 ft. Overlying this, broken cumulus and stratocumulus could be expected extending from 3,000 ft to above 10,000 ft.

Photograph

A photograph taken 2 minutes prior to the helicopter departing Long Plain Road showed the cloud conditions at that time (Figure 4). The elevation of the landing site was about 4,429 ft AMSL. The peak of the terrain visible behind the helicopter is 4,573 ft AMSL. This peak was obscured by broken cloud indicating that the cloud base was less than 144 ft AGL.

Figure 4: Cloud conditions 2 minutes prior to departure from interim landing site

Source: motorist via NSW Police Force

Recorded observations

At 1530 (4 minutes after the accident), Bureau of Meteorology weather stations at Cabramurra (14 km south-west of the accident site, elevation 4,864 ft) and Mount Ginini (43 km north‑east of the accident site, elevation 5,774 ft) recorded no rainfall and no separation between the dew point temperature and air temperature. This indicated the presence of very low-level cloud, likely down to ground level at both stations. Neither station was equipped to provide more detailed cloud information.

Visual flight rules

Visual meteorological conditions

The Civil Aviation Safety Regulation (CASR) 91.280 outlined that flight under the visual flight rules (VFR) can only be conducted in visual meteorological conditions (VMC). The criteria are provided in the CASR Part 91 Manual of Standards Table 2.07 (3) and the CASA Visual Flight Rules Guide:

The flight, and the location of the accident, were in Class G (non-controlled) airspace. The following VMC were stipulated for flight under the VFR in Class G airspace when below 10,000 ft and above 3,000 ft or 1,000 ft above ground level (whichever is higher):

• a minimum vertical distance of 1,000 ft and horizontal distance of 1,500 m from cloud
• a flight visibility of 5,000 m.

For helicopter operations in Class G airspace at or below 3,000 ft or 1,000 ft above ground level (whichever is higher), the following minimum conditions were stipulated:

• clear of cloud and in sight of the ground or water
• a flight visibility of 5,000 m or, if operated by day at a speed that allows the pilot to see obstructions or other traffic in sufficient time to avoid collision, 800m.

Minimum height

In addition to minimum visibility and distance from cloud requirements, a pilot is also required to maintain a minimum height above the ground. Unless during take-off, landing or other approved low-flying operation, CASR 91.265 and 91.267 detail that a pilot in command must not fly a helicopter over:

• any city, town, or populous area at a height lower than 1,000 ft above the highest feature or obstacle within a horizontal radius of 300 m of the point on the ground or water immediately below the helicopter; or
• any other area at a height lower than 500 ft above the highest feature or obstacle within a horizontal radius of 300 m of the point on the ground or water immediately below the helicopter.

The investigation identified no evidence to indicate that the pilot intended to undertake an approved low-flying operation.

Recorded data

On-board the helicopter was a mobile device with the OzRunways electronic flight bag application installed. The application had an option for live flight tracking enabled that transmitted the device’s position and altitude. This data was also obtained by the ATSB.

The data showed that at 1519:22 the helicopter turned to a track that corresponded with the direct track to Tumut and 15 seconds later commenced a climb from low level to about 6,600 ft. At 1522:02, a further climb commenced, reaching 7,000 ft at 1522:42. The helicopter continued to track generally toward Tumut at groundspeeds of 105‑115 kt, corresponding to a normal cruise speed for the helicopter. During this segment of the flight, track variations of up to 21° and altitude variations of up to 300 ft were recorded.

At 1525:22, the helicopter descended from 7,000 ft, reaching 6,800 ft about 20 seconds later. From 6,800 ft, a climb was commenced and a further 20 seconds later, the helicopter reached 7,400 ft (at climb rate of 1,800 ft per minute) with a groundspeed of 60 kt. The helicopter then entered a steep left descending turn. During the turn, the groundspeed increased to 137 kt and the descent rate exceeded 3,800 feet per minute (Figure 5).

Figure 5: Aircraft flight path leading up to the accident

Source: Google Earth and OzRunways, annotated by ATSB

Site and wreckage information

The accident site was located within the Kosciuszko National Park in an area of tussock grass, interspersed by bare protruding rock (Figure 6).

Figure 6: Accident site

Source: ATSB

The helicopter collided with terrain between two rock formations in a descending tight left turn and right side-slip in a northerly direction with a westerly heading. At initial impact, a main rotor blade struck the ground and the tail boom separated. The fuselage then turned left to about a southerly heading. Most of the wreckage was located within 8 m of the impact, but the main transmission, mounts and supporting airframe structure continued a further 70 m up an incline. On-site examination indicated that the engine was providing power at impact. There was no evidence of an in-flight break-up or a pre-existing defect with the drive train or flight controls.

The emergency locator transmitter (ELT)[4] antenna separated from the unit during the impact sequence. The ELT was examined at the ATSB’s technical facilities in Canberra and was found to have activated during the accident. However, the separation of the antenna prevented a signal from being broadcast. While this delayed search and rescue efforts, it did not alter the outcome as the accident was not survivable.

Risks of flying in areas of reduced visual cues

The safety risks of VFR pilots flying from VMC conditions into instrument meteorological conditions[5] are well documented. This has been the focus of numerous ATSB reports and publications, as VFR pilots flying into IMC represents a significant cause of aircraft accidents and fatalities. In 2013, the ATSB Avoidable Accidents series was re-published. Of these publications, the booklet titled Accidents involving pilots in Instrument Meteorological Conditions outlined that:

In the 10 years to July 2019, 101 VFR into IMC occurrences in Australian airspace were reported to the ATSB. Of those, 9 were accidents resulting in 21 fatalities. That is, about 1 in 10 VFR into IMC events result in a fatal outcome.

Spatial disorientation

Spatial disorientation is a type of loss of situation awareness, and is different to geographical disorientation, or incorrectly perceiving the aircraft’s distance or bearing from a fixed location.  Spatial disorientation occurs when pilots do not correctly sense their aircraft’s attitude, airspeed, or altitude in relation to the earth’s surface. In terms of an aircraft’s attitude, spatial disorientation is often described simply as the inability to determine ‘which way is up’, although the effects can often be more subtle than implied by that description.  

Spatial disorientation occurs when the brain receives conflicting or ambiguous information from the sensory systems. It is likely to happen in conditions in which visual cues are poor or absent, such as in adverse weather or at night. Spatial disorientation presents a danger to pilots, as the resulting confusion can often lead to incorrect control inputs and resultant loss of aircraft control. The flight control sensitivity and relative instability of helicopters compared to aeroplanes increases the risk of such a control loss.

Related occurrences

There have been many accidents relating to VFR pilots flying into reduced visibility conditions. The ATSB publication listed above identifies a number of similar occurrences. Of note is investigation AO-2015-131.

ATSB investigation AO-2015-131

At about 5.30 pm on 7 November 2015, the owner-pilot of an Airbus Helicopters (Eurocopter) EC135 departed Breeza, NSW, on a VFR private flight with two passengers on board to Terrey Hills, NSW.

Witnesses observed the helicopter land in a cleared area in a valley. After 40 minutes on the ground, the pilot, who did not hold an instrument rating and was limited to visual flight operations, departed to the east towards rising terrain in marginal weather conditions. About seven minutes later, and approximately 9 km east of the interim landing site, the helicopter collided with terrain resulting in fatal injuries to all occupants.

The ATSB found that the pilot likely encountered reduced visibility conditions leading to loss of visual reference leading to the collision with terrain.

Safety analysis

While en-route from Majura, Australian Capital Territory, to Tumut, New South Wales, a group of 5 helicopters including a Bell Helicopter 206L-4, registered VH-PRW, encountered deteriorating cloud and visibility conditions. The pilots of 4 of the helicopters landed at Wee Jasper, but the pilot of VH-PRW continued south into the Brindabella Ranges. After encountering further low cloud and poor visibility, the pilot of VH-PRW landed the helicopter alongside Long Plain Road in the ranges.

At 1453, the pilot decided to depart the interim landing site and continue the flight. The flight progressed for 32 minutes until the helicopter commenced a rapid climb and then a descending left turn which continued until the helicopter collided with terrain.

Site and wreckage examination did not identify any defects or anomalies that might have contributed to the accident. Additionally, there was no evidence to support the pilot being incapacitated. Therefore, this analysis will focus on the examination of the factors that led to a visual flight rules (VFR) pilot operating in an area of reduced visibility and losing control of the helicopter.

Departure into unsuitable conditions

Low cloud and poor visibility conditions were forecast across the Brindabella Ranges on the day of the accident. The pilot, having encountered these conditions, landed the helicopter alongside Long Plain Road at 1129. The pilot and passenger then waited for conditions to improve sufficiently to depart the interim landing site.

Despite no such improvement eventuating, after about 3 hours and 24 minutes, the pilot elected to depart and continue to Tumut at about 50 ft above ground level. Photographs, along with police and witness reports, showed that at the time of the departure the cloud and visibility conditions were unsuitable for visual flight. The broken cloud base of less than 144 ft did not allow the pilot to maintain the helicopter both, clear of cloud as required by visual meteorological conditions (VMC) and at the minimum height above terrain of 500 ft.

Departing into unsuitable cloud and visibility conditions, particularly in the vicinity of mountainous terrain at very low level carried significant risk of both losing visual reference and of collision with terrain.

Loss of control

After departing the interim landing site, the flight proceeded at very low level for 26 minutes until the pilot turned the helicopter to a more direct track toward Tumut and climbed to about 7,000 ft above mean sea level (AMSL).

Observations recorded at meteorological stations in the vicinity of the flight indicated that it was highly likely that there was low cloud in the area of the accident. In addition, significant cloud was forecast and observed from ground level to above 10,000 ft AMSL along the flown track. In these conditions it was highly likely that VMC could not be maintained and that reduced visual cues were encountered by the pilot.

Over the next 6 minutes, minor tracking and altitude variations were recorded. It is possible these variations resulted from attempts to manoeuvre around, or in, cloud or rain showers with associated reduced visibility. Additionally, this manoeuvring indicates that the autopilot was not being used during this part of the flight. It could not be determined if the stability augmentation system was being used.

The flight at about 7,000 ft continued until 1525:42 when a rapid climb of about 1,800 ft per minute was commenced. This was immediately followed by a steep descending left turn. This manoeuvring was inconsistent with normal helicopter operation and was indicative of a loss of control.

The pilot did not hold an instrument rating and the helicopter was certified for day visual flight only. This greatly increased the risk of the pilot being affected by special disorientation and it is unlikely the pilot could have maintained control without visual reference for an extended period. Given the forecast and observed conditions, it is likely that during the 6 minutes the helicopter was operating at the higher level, it encountered poor weather. This likely led to the pilot experiencing spatial disorientation which resulted in a loss of control and the collision with terrain.

Findings

From the evidence available, the following findings are made with respect to the collision with terrain involving Bell Helicopter 206L-4, VH-PRW, on 3 April 2022.

Contributing factors

• Having landed the helicopter at an interim landing site due to encountering forecast low cloud and reduced visibility conditions, the pilot subsequently departed into cloud and visibility conditions unsuitable for visual flight.
• It is highly likely that cloud and visibility conditions resulted in the pilot experiencing a loss of visual reference and probably becoming spatially disoriented. This led to a loss of control and an unsurvivable collision with terrain.

Sources and submissions

Sources of information

The sources of information during the investigation included the:

• Civil Aviation Safety Authority
• aircraft manufacturer and maintainer
• New South Wales Police Force
• Bureau of Meteorology
• OzRunways
• tour organiser.

References

Australian Transport Safety Bureau, 2019, Avoidable Accidents No. 4 Accidents involving Visual Flight Rules pilots in Instrument Meteorological Conditions, Aviation Research and Analysis publication AR-2011-050.

Submissions

Under section 26 of the Transport Safety Investigation Act 2003, the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. That section allows a person receiving a draft report to make submissions to the ATSB about the draft report.

A draft of this report was provided to the following directly involved parties:

• NSW Police Force
• the occupants’ next of kin
• Civil Aviation Safety Authority
• tour organiser.

Submissions were received from:

• Civil Aviation Safety Authority
• the passenger’s next of kin
• tour organiser.

The submissions were reviewed and, where considered appropriate, the text of the report was amended accordingly.

Purpose of safety investigations The objective of a safety investigation is to enhance transport safety. This is done through: identifying safety issues and facilitating safety action to address those issues providing information about occurrences and their associated safety factors to facilitate learning within the transport industry. It is not a function of the ATSB to apportion blame or provide a means for determining liability. At the same time, an investigation report must include factual material of sufficient weight to support the analysis and findings. At all times the ATSB endeavours to balance the use of material that could imply adverse comment with the need to properly explain what happened, and why, in a fair and unbiased manner. The ATSB does not investigate for the purpose of taking administrative, regulatory or criminal action. Terminology An explanation of terminology used in ATSB investigation reports is available here. This includes terms such as occurrence, contributing factor, other factor that increased risk, and safety issue. Publishing information  Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2022 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this publication is licensed under a Creative Commons Attribution 3.0 Australia licence. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided that you attribute the work. The ATSB’s preference is that you attribute this publication (and any material sourced from it) using the following wording: Source: Australian Transport Safety Bureau Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

[1]     Visual flight rules (VFR): a set of regulations that permit a pilot to operate an aircraft only in weather conditions generally clear enough to allow the pilot to see where the aircraft is going.

[2]     Eastern Standard Time (EST): Coordinated Universal Time (UTC) + 10 hours.

[3]     Cloud cover: in aviation, cloud cover is reported using words that denote the extent of the cover – ‘scattered’ indicates that cloud is covering between a quarter and a half of the sky, ‘broken’ indicates that more than half to almost all the sky is covered.

[4]     Emergency locator transmitter (ELT): a radio beacon that transmits an emergency signal that may include the position of a crashed aircraft, activated either manually or in the crash.

[5]     Instrument meteorological conditions (IMC): weather conditions that require pilots to fly primarily by reference to instruments, and therefore under Instrument Flight Rules (IFR), rather than by outside visual reference. Typically, this means flying in cloud or limited visibility.

Overview of the investigation

On 28 March 2022, the ATSB commenced an investigation into a hard landing involving a E-240 Kavanagh balloon, registration VH-ZON, which occurred at Moorabbin Airport, Victoria, on 27 March 2022.

The balloon was being operated on a scenic passenger flight between Reservoir and Moorabbin Airport with a pilot and 10 passengers on board. The pilot provided the operator’s standard safety briefing to the passengers before take-off, while the passengers were in the basket, to explain and demonstrate the position to adopt during normal landings and emergencies. The position was facing opposite to the landing direction, standing with slightly bent knees, holding on to the rope handholds in front, and back rested against padding on the basket. Passengers were to remain in that position until the basket stopped.

There were several other balloons operating the same scenic flight route, and the pilots had collectively decided the departure point and that the weather was suitable for their respective flights (all along the same route). The pilots reviewed several sources of weather information, including the Bureau of Meteorology aerodrome forecasts (TAF) for Melbourne, Essendon and Moorabbin. The Moorabbin TAF forecast the wind to be 11 kt from the north-east at the time the balloons were due to land.

The pilots had also obtained information about the actual wind conditions prior to departure by releasing 2 piballs[1] in different locations to assess the speed and direction of the wind at different levels. Following this assessment, the balloons departed at about 0700 local time.  

The pilot had about 30 years experience as a balloon pilot and had been operating balloons in the area for about 18 years, with extensive knowledge of the planned route.

On approach to Moorabbin Airport, VH-ZON was travelling in a south-easterly direction and was about 500 m to the west of the other balloons. The pilot obtained automatic terminal information by radio for Moorabbin, and it advised of a 4-kt north­-easterly surface wind.

The pilot reported that they had commenced the descent into Moorabbin Airport after the other balloons and found the wind to be faster at the lower levels than expected. Data collected via another balloon pilot’s navigation equipment post-accident showed the wind was about 38 kt at 1,400 ft and 35 kt at 1,000 feet, which was significantly faster than the pilots had anticipated. This meant that the pilot had to conduct a faster than usual descent to ensure they could land the balloon in a suitable area.

When the balloon reached 300–­­400 ft, it travelled through a temperature inversion (where temperature increases with altitude, which is a reversal of typical atmospheric conditions) and the balloon rotated 120°. Although the passengers had been instructed to adopt the landing position, the pilot did not have time to rotate the balloon to the correct orientation (with the passengers facing opposite to the landing direction) before reaching the ground. On landing, the balloon impacted the front right corner of the basket and bounced. The basket was then dragged for a short distance, coming to rest in a culvert at the end of a runway within the airport boundary.     

As a result of the hard landing and the orientation of the basket, 1 passenger was seriously injured and 2 passengers received minor injuries.

As part of its investigation, the ATSB interviewed the pilot and passengers and reviewed:

• weather information including observations and forecasts used by the pilot
• air traffic control recordings
• recorded navigation information used in-flight by one of the other balloon pilots (data could not be retrieved from the equipment used on the accident balloon)
• the operator’s procedures for passenger briefings
• photographs taken in-flight by the balloon operator and others that were provided by passengers and a witness on the ground.

The ATSB notes that, due to unexpected wind speed on descent (which was much higher than the surface wind information that the pilot had previously obtained), and the limited landing site options, the pilot decided to land as soon as possible. This resulted in a faster and harder landing than normal. The balloon’s abnormal orientation after passing through the temperature inversion meant that although the passengers were in the correct position for landing, there was a greater risk of injury.

Reasons for the discontinuation

Based on a review of the available evidence, the ATSB considered it was unlikely that further investigation would identify any systemic safety issues or important safety lessons from this specific occurrence. Consequently, the ATSB has discontinued this investigation.

However, the ATSB is concerned about the number of accidents that have been occurring in commercial balloon operations and has listed the reduction of passenger injuries in commercial ballooning operations as one of its Safety Watch items. The evidence collected during the investigation involving VH-ZON will be used in a safety study further examining these types of accidents.  

Overview of the investigation

At about 2230 on 17 March 2022, an axle in the rear part of train 5WB7 broke as the train entered the yard at Casino, derailing the wagon. As the train passed through the yard it travelled over a set of points for the closed Murwillumbah Branch (set for the mainline). At this point, the derailment worsened with the rear four wagons derailing and becoming detached from the rest of the train. The train then passed through a second set of points (also set for the mainline) and came to a stop. A further four wagons at the rear of the train were derailed. A significant amount of track and signal infrastructure was damaged within Casino yard, however there were no injuries.

During the investigation, the ATSB found:

• The derailment was due to a broken axle on the 55th wagon (RKPF30311).
• The metallurgical analysis of axle concluded that the failure was due to a fatigue crack on the axle barrel. The fracture face exhibited corrosion indicating the crack had been present for an extended period. The initial cause of the fatigue crack and exact duration could not be determined with the evidence available.
• The wagon was operated and maintained by Pacific National until it was transferred to Qube on 1 January 2022.
• The journey on 17 March 2022 was the first time that the wagon (RKPF30311) had been used by Qube since the transfer. The wagon underwent a full train inspection (FX) prior to departing Port Kembla.  
• The train, wagon and axle had been inspected and maintained in accordance with both operators’ procedures.
• The in-service inspections of wheels and axles were by visual inspection as part of the full train inspection and roll-by inspection. However, the inspections were limited to the detection of large or obvious defects.
• When wheels and axles were removed from the wagon they were subjected to visual and non‑destructive testing in accordance with the operators’ procedures and relevant standards. Where damage was detected, the defect could be treated if it was below the maximum permissible defect size. Non-destructive testing was conducted to ensure the defect was removed prior to the axles being returned to service.

During the investigation, the ATSB also reviewed the following:

• the axle inspection and defect data for class E axles held by the operator
• the axle inspection process and engineering standards
• the visual and non-destructive testing process for class E axles and witnessed the inspection process
• asset handover process from Pacific National to Qube.

Reasons for the discontinuation

Based on a review of the available evidence, the ATSB considered it was unlikely that further investigation would identify any systemic safety issues or important safety lessons. Consequently, the ATSB has discontinued this investigation.

In response to the incident, Qube has advised the following safety actions have been taken:

• Visual inspections and ultrasonic testing of the axles on the wagons received from Pacific National.
• Updated rolling stock maintenance manual using data from the on-wagon axle testing to provide clear standard for future axle maintenance.
• Created an axle register to ensure that all Qube axles have unique identifying numbers.
• Made representations to the Rail Industry Safety and Standards Board (RISSB), to consider strengthening the requirements of AS 7515 Axles to mandate the marking of axles with dates and sources of manufacture to align with traceability requirements of AS 7514 Wheels.

The evidence collected during this investigation remains available to be used in future investigations or safety studies. The ATSB will also monitor for any similar occurrences that may indicate a need to undertake a further safety investigation.

During a maintenance inspection of a Boeing 737-300 aircraft, registered VH-YNU, the failure of a wing spoiler cable was identified. The Civil Aviation Safety Authority (CASA) requested technical assistance from the ATSB in the examination of the cable. To facilitate this assistance, the ATSB initiated an external investigation under the provisions of the Transport Safety Investigation Act 2003.

The ATSB has concluded the examination of the cable, and provided the results of that work to CASA on 6 May 2022.

The Civil Aviation Safety Authority is responsible for and will administer the release of any information from the maintenance inspection. Any enquires relating to the matter should be directed to CASA at casa.gov.au.

Executive summary

What happened

On 26 February 2022, at around 1030 local time, a student pilot and instructor were conducting a low-level hover training exercise at Gold Coast Airport, Queensland, in a Robinson Helicopter R22, registered VH-THM, operated by Professional Helicopter Services. 

After around 30 seconds of stable hover just above the ground, there was a ‘bang’ through the airframe. The helicopter then pitched nose up, developed a large vibration and then yawed to the right. The instructor took over immediately and landed. There were no injuries. Subsequent inspection found that the tail rotor gearbox and empennage assembly had separated from the tailcone (boom). 

What the ATSB found

The tail rotor tip cap detached due to moisture-induced failure of the adhesive bond. Tap testing of the tail rotor blades during the most recent scheduled inspection, for the purpose of detecting adhesive failure, did not identify any disbonds that were very likely present at the time. 

The tailcone aft casting then fractured due to the detachment of the tail rotor tip cap, which led to a tail rotor imbalance and severe abnormal loading event on the casting.

What has been done as a result

Robinson released updated service bulletins and service letters in response to seeing tail rotor blades that were allowed to corrode to an unserviceable condition and receiving reports of tail rotor blade tips coming loose in service due to corrosion. Notably, Robinson released Revision A of R22 service letter SL-93 in June 2022, which required blade replacement with any evidence of bond line corrosion. The updated service letter also included inspection criteria and a minimum 10x magnification inspection of the bond line to increase the likelihood of detection. Other recommended tail rotor blade maintenance practices were also updated. 

Robinson has also identified certain part and serial number blades as being susceptible to corrosion. These blades required additional visual inspection prior to the first flight of the day, until replacement with blades that were less susceptible to corrosion. The United States Federal Aviation Administration released an airworthiness directive regarding this matter. 

Safety message

In-service detachment of a tail rotor tip cap due to adhesive failure is an uncommon event, but still presents a significant hazard to the helicopter. The resultant destructive imbalance could result in a loss of tail rotor drive and/or tail assembly. Pilots and maintainers should recognise that tap testing alone may be insufficient to detect adhesive disbonds, and that detailed visual inspection, especially for signs of corrosion around the bond line, will assist in identifying blade degradation. 

Meanwhile, pilots should ensure they are familiar with the immediate corrective actions required for the loss of tail rotor effectiveness, in accordance with the pilot operating handbook, to reduce the likelihood of a loss of helicopter control.

The occurrence

On 26 February 2022, at around 1030 local time, a student pilot and instructor were conducting a low-level hover training exercise at Gold Coast Airport, Queensland, in a Robinson Helicopter R22 registered VH-THM, operated by Professional Helicopter Services. 

Prior to commencement of the exercise, the instructor conducted a daily check of the helicopter, with no issues found. The instructor then taxied the helicopter from the helipad to the western grassed training area. The instructor reported having flown the helicopter frequently and that the helicopter felt normal, with no unusual sounds or vibrations.

After reaching the training area, the instructor handed control of the helicopter to the student to commence the hover exercise at a skid height of approximately 3–5 feet above the ground. This was the student’s fourth hover exercise and the instructor was closely guarding the flight controls against any abrupt inputs form the student. 

After around 30 seconds of stable hover, the instructor described feeling a significant jolt[1] through the airframe. The helicopter then pitched nose up, developed a large vibration and then yawed to the right. The instructor immediately took control of the helicopter, lowered the collective and closed the throttle to control the yaw. The helicopter was landed safely and there were no injuries.

Once on the ground, with the engine still running, the vibration ceased. The instructor shut down the helicopter and on inspection, identified that the tail rotor gearbox and empennage assembly had each separated from the tailcone (boom) and were reportedly located within 10 metres from the final position of the helicopter (Figure 1). The ATSB did not attend the site.

Figure 1: Proximity of separated major tail components from the helicopter

Tail rotor blades (not shown) had separated from the gearbox assembly

Source: Helicopter operator, annotated by the ATSB

Context

Aircraft information

The accident helicopter was a Robinson R22 Beta, serial number 4057, manufactured in 2006 and registered with the current operator since November 2019. The airframe had 3,355 hours at the time of the occurrence. The helicopter’s most recent maintenance was a scheduled 100-hour / annual airframe inspection that was conducted on 24 January 2022 at 3,334 airframe hours, which was 21 flight hours and 33 days prior to the occurrence. 

The tail section of the R22 comprises a tailcone assembly, through which the tail rotor drive shaft and pitch control rod passes. An aluminium casting is riveted to the skin at the end of the tailcone, and supports the tail rotor gearbox and empennage assembly. An example of the tail arrangement on another R22 helicopter is shown in Figure 2.

Figure 2: Exemplar R22 tail arrangement

Source: Archangel12, via Wikimedia Commons, annotated by the ATSB

Component history

The tailcone assembly (part number A023-23, serial number 5986), including the tailcone aft casting, was removed from another helicopter in December 2016 due to it having been dented. A repair was completed by Robinson in April 2017 and fitted to VH-THM in November 2018. The assembly had accrued 1,155 hours since the repair and fitment to VH-THM, and 2,787 hours total time in service. The tailcone assembly had a 4,400-hour service life.

The tail rotor blades (part number A029-2 Rev R, serial numbers 5802 and 5803) fitted to the helicopter had a 2,200 hour operational, or 12 year calendar, life limit. The tail rotor blades were installed as a set in November 2018, having accrued 1,155 hours on another helicopter. At the time of the occurrence the blades had 2,147 hours since new (53 hours of life remaining), and had a calendar life expiry of April 2028. The tail rotor blades had most recently been repainted in November 2020.

Component examinations

Empennage and gearbox

The separated tail rotor gearbox and empennage assembly, and tail rotor blades, were sent to the ATSB’s technical facilities in Canberra for detailed examination. 

The empennage assembly was recovered in one piece and had separated from the tailcone and gearbox due to fracture of the aft casting, part of which remained attached to the horizontal stabiliser. The inboard end of the horizontal stabiliser had sustained impact damage (Figure 3). There was no damage to any of the stabiliser extremities.

Figure 3: Empennage overview and damage

Source: ATSB

The tail rotor gearbox was firmly fastened to the remaining part of the fractured tailcone casting (Figure 4). On disassembly, there was no indication of movement or fretting[2] between the mating surfaces. The gearbox separated from the tailcone due to tearing of the skin immediately forward of the casting rivet line. 

Figure 4: Tail rotor gearbox assembly as recovered

Source: ATSB

The gearbox output shaft was bent. Both tail rotor blades were fractured and separated close to the blade root. One of the tail rotor blades was missing the tip cap, which was not found. The pitch change bellcrank support arm was fractured due to rotation of the bent output shaft.

The tail rotor drive shaft was fractured at the aft flex coupling and the tailcone push-pull tube was fractured and twisted around the driveshaft. The aft flex coupling was determined to have been assembled correctly and the flex plate had deformed and fractured due to overstress. The driveshaft flange had deformation consistent with resistance on the tail rotor side of the drive system. 

Figure 5: Fractured tail rotor aft flex coupling

Source: ATSB

Tail rotor blades

Description and examination

The tail rotor blades were Robinson part number A029-2 Rev R with serial numbers (SN) 5802 and 5803 (Figure 6). After ATSB examination, the blades were forwarded to Robinson for examination under the supervision of the United States National Transportation Safety Board (NTSB). 

Figure 6: Tail rotor blades SN 5802 and SN 5803

Source: ATSB

Deformation and scoring of the tail rotor blades matched the damage sustained to the inboard end of the horizontal stabiliser, indicating they had come into contact during the occurrence sequence (Figure 7). 

Figure 7: Rotor blade contact with horizontal stabiliser

Source: ATSB

The aluminium tip cap was bonded to the internal skin surface of the blade using an epoxy film adhesive (Solvay FM 94K) described in the datasheet as follows:

FM® 94 adhesive offers a unique combination of high temperature performance, toughness and moisture resistance as demonstrated by its ability to bond to wet Nomex® honeycomb and retention of shear properties after pre- and post-bond humidity exposure.

The tip cap of blade SN 5803 had separated. Examination and measurements of the original bonded area identified that approximately 85% was smooth, consistent with adhesive failure,[3] with some discolouration and material transfer from the missing tip cap (Figure 8). The remaining 15% exhibited a rough fracture surface indicative of cohesive failure. The NTSB reported ‘traces of yellow patches were observed across the entire surface … and under microscopic examination, when pressed, the patches expressed fluid.’ Robinson and the NTSB concluded that the tip cap detached from the blade as a result of moisture-induced adhesive failure.

Figure 8: Blade SN 5803 post-occurrence tip cap adhesive condition 

Source: ATSB

Both blades had significant erosion and thinning of the leading edge skin, particularly at the tip (Figure 9). The erosion wore completely through the skin on a small section of the leading edge on Blade SN 5802. Blade SN 5803 had a crack in the leading edge, across the tip cap bond. The heavily oxidised condition of the crack surface indicated that it was not a recent fracture and therefore pre-dated the tip cap separation. Both blades had small areas of corrosion, indicated by paint bubbles, on the blade skin adjacent to the tip cap.

During the examination, Robinson tap-tested[4] SN 5802 with the tip cap in situ and identified likely indications of voids or disbonding ‘mid-span’ along the tip cap bond line.[5] Disassembly of the blade revealed some regions of adhesive failure, appearing to be predominantly progressing towards the blade root from around the circumference of the bond line, amounting to about 30% of the original bond area (Figure 10). 

Figure 9: Tail rotor blade leading edges, showing paint loss, erosion and skin cracking 

Source: ATSB

Figure 10: Blade SN 5802 adhesive failure

The other side of the bond exhibited similar visual characteristics.

Source: Robinson Helicopter Company, annotated by the ATSB

Maintenance requirements

Chapter 2 of the R22 maintenance manual contained a 100-hourly inspection procedure and checklist, which stated:

Tail Rotor Blades: Inspect condition. Refer to § 9.220 for damage limitations…Verify blade tip drain holes (2 per A029-1 blade, 1 per A029-2 blade) are unobstructed. Verify no corrosion. Tap test bonded areas per § 26-44.

Tap testing structures involves lightly tapping an item of specific size, weight and material (often a coin or washer) along the length of a bonded structure and listening for changes in the acoustic response. Areas of delamination or disbond will sound hollow in comparison to a solid structure. A video illustrating a tap test was published by Robinson, at https://www.robinsonheli.com/media.

The 100-hourly inspection procedure and checklist was completed at the 24 January 2022 inspection, with no defects recorded.

Chapter 9, section 9.220 of the helicopter’s maintenance manual detailed inspection and repair limits for tail rotor dents, nicks, scratches and corrosion. Blades were required to be replaced where erosion caused deformation or ripples in the leading edge. The section contained a procedure for re-painting the blades, but not a specific limit or requirement for repainting due to general wear. The tail rotor blade description under section 9.000 stated that ‘Maintaining the paint finish will reduce corrosion and erosion.’ The service letter, SL-93, similarly stated: ‘Paint offers the best protection against leading edge corrosion’ and the ‘preferred blade condition is with a fully painted leading edge’.

Section 26-44 of the maintenance manual detailed void limits during detection by tap testing of main rotor blades and referenced ‘critical’, ‘semi-critical’ and ‘non-critical’ blade bond areas. There were no tail rotor blade void limits given. The maintainer of the occurrence tail rotor blades indicated they had extensive experience in tap testing and had previously identified disbonds in other manufacturer’s rotor blades.

Robinson also published R22 service letter SL-93 in June 2021, the background for which stated:

RHC [Robinson] has recently seen tail rotor blades that were allowed to corrode to an unserviceable condition including severe leading edge pitting and degradation of the bond at the tip cap. Regular preventive maintenance is imperative for continued safe operation and additional care may be required in corrosive environments such as coastal or shipboard operations. 

The service letter recommended practices included removal of any corrosion by hand sanding and instructed to ‘remove only material necessary to eliminate corrosion; any hole that completely penetrates blade skin requires blade replacement.’ The service letter also reinforced the maintenance manual requirement for tap testing of the blades to ‘verify bond integrity’.

Tailcone aft casting 

The two separated pieces of the fractured tailcone casting remained secured to the horizontal stabiliser and tailcone skin/gearbox respectively (Figure 11). There were no visible part or serial numbers. 

Figure 11: Aft tailcone casting fracture, empennage side (left) and tailcone side (right)

Darkened regions on the fracture surface were considered to be due to gearbox oil migration.

Source: ATSB

The casting was manufactured from A356.0-T6[6] aluminium alloy. Chemical analysis and hardness of the occurrence casting material found these characteristics to be in accordance with the alloy standard.

Visually and under low magnification optical microscopy, the casting fracture surface had irregular and faceted, brittle fracture features. There was evidence of localised deformation from contact between the two halves of the fractured bulkhead and as a result of impact with the adjacent pitch‑change push-pull tube. Microstructural analysis showed that the casting was unmodified,[7] and the faceted fracture surface was the result of the presence of coarse, plate-like silicon eutectic constituents in the alloy. Unmodified A356 crack growth studies have found that, at higher stress intensities, crack progression tends to follow the silicon particle orientation, which therefore dominates the fracture surface, producing the faceted fracture features observed in the occurrence sample.

At high magnification using a scanning electron microscope (SEM), there were several areas of minor shrinkage porosity and non-metallic inclusions that were generally to be expected for this type of casting. There were also some very minor striated areas that appeared consistent with high-cycle cracking. However, none of the defects observed were considered significant enough to have meaningfully contributed to the casting fracture. 

To more fully characterise the fracture surface of this type of unmodified cast aluminium alloy, the ATSB conducted some laboratory fractures and the Defence Science and Technology Group (DSTG) conducted fatigue testing on the casting alloy for comparison (Appendix A).

Adhesive bond failures

Failure types comprise adhesive, cohesive, and mixed failures (Petrie 2000):

Adhesive failure is an interfacial bond failure between the adhesive and the adherend [bonded surface]. Cohesive failure could exist within either the adhesive material or the adherend. Cohesive failure of the adhesive occurs when stress fracture within the adhesive material allows a layer of adhesive to remain on both substrates (i.e., the attachment of the adhesive to the substrate is stronger than the internal strength of the adhesive itself, and the adhesive fails within its bulk). When the adherend fails before the adhesive and the joint area remains intact, it is known as a cohesive failure of the adherend.

An illustration of cohesive, adhesive, and mixed failure types is provided in Figure 12. 

Figure 12: Cohesive and adhesive failures

Source: Petrie (2000), modified by the ATSB

Moisture can affect the strength of an adhesive bond. This failure mechanism involves water diffusing through the adhesive, through the interface between the adhesive and adherend and/or through cracks or defects in the adhesive or adherend layers. The absorbed water molecules degrade the interface by deteriorating secondary bonds between the adhesive and adherend, or by weakening the oxide layer of the aluminium substrate.

Related occurrences

R44 in flight break-up

On 4 July 2020 a Robinson R44 Raven I helicopter, VH-NBY, experienced an in-flight break-up shortly after takeoff, 3 km north of Broome Airport, Western Australia. As with VH-THM, the tailcone aft casting failed, but there was no tail rotor blade failure (ATSB investigation AO‑2020‑033). 

R22 tail rotor failure

The US National Transportation Safety Board investigated (ERA22LA340) a July 2022 accident involving an instructional flight in a Robinson R22 helicopter that collided with terrain. The instructor reported hearing a popping and banging sound, followed by an uncommanded right yaw, and subsequent collision with terrain.

The investigation report stated:

Postaccident examination of the helicopter revealed that the tail rotor gearbox had fractured and that the tail rotor assembly had separated from the helicopter. Additional examination of the tail rotor blades, which had remained attached to the tail rotor drive shaft and gearbox, revealed corrosion and interior delamination of the blades. There was also erosion present on the blade leading edges, which likely provided a path for moisture to ingress, thereby resulting in the observed corrosion as well the failure of the bonding adhesive within the blade. It is likely that this condition resulted in an imbalance of the blades that imparted a vibratory loading onto the tail rotor gearbox that ultimately resulted in its failure during the accident flight. Review of partial maintenance records provided by the operator revealed that the tail rotor blades installed on the accident helicopter were not the blades that were noted in the maintenance logbooks, and the service history of the installed blades could not be determined.

Robinson tail rotor tip cap separations

R44 tail rotor tip cap separation

A March 2022 occurrence involving an R44 tail rotor tip cap separation was reported through the Civil Aviation Safety Authority’s (CASA’s) defect reporting system (the location was not provided). 

The tip cap separated from rotating tail rotor blade shortly after landing and the resulting vibration fractured the tail rotor gearbox input cartridge mounting flange, as shown in Figure 13. Internal corrosion at the tip bond area was attributed to a disbond (per the mechanism described in R22 SB-120 and the R44 equivalent, SB-112). The R44 part number C029-3 tail rotor blades were similar in construction to the R22 A029-2 blades, and these reportedly had 1,806 hours since new.

The submitted report advised that tap testing of the bonded area at the most recent inspection gave no indication of disbonding. During subsequent experiments the separated tip was loosely re-fitted to the blade and another tap test carried out. The reporter stated that there was no discernible difference between the sound response of the blade with the refitted, but not bonded, tip and that of a new blade. Additionally, while there was evidence of leading-edge erosion on the blade, the reporter indicated there was no corrosion discernible under external visual examination. 

The ATSB did not examine the subject blade or tip cap. The supplied images indicated minor rough or bubbled paint on the exterior surface, mid-span, adjacent to the bond line, which may have been indicative of underlying corrosion.

Figure 13: R44 tip cap separation and resultant gearbox mount fracture

Source: Civil Aviation Safety Authority, modified by the ATSB

Other tail rotor tip cap separation occurrences 

A 2024 United States Federal Aviation Administration airworthiness directive (AD 2024–04–02) was prompted by 3 reports[8] of tail rotor blade tip caps coming loose due to corrosion at the bond in R44 helicopters (which have similar tip caps). A search of the Civil Aviation Safety Authority and Federal Aviation Administration defect reporting systems did not identify any other reports of tip caps separating on Robinson R22 or R44 helicopters.

Safety analysis

Failure sequence

The normal proximity of the tail rotor blades to the point of impact with the horizontal stabiliser meant that contact was only possible if the empennage assembly separated. The remaining damage to the tail rotor drive system was the result of the tail rotor strike on the stabiliser. Separation of the empennage assembly in this manner was due to fracture of the tailcone casting.

Detailed fractographic analysis of the tailcone casting found that the fracture was the result of a high applied stress intensity, and identified no significant defects that would have resulted in fracture during normal operation. As such, the casting was most likely subjected to abnormal loading conditions around the time of the occurrence. 

Aside from the tail rotor tip cap detachment, there was no evidence of an operational event or issue with the helicopter tail or tail rotor drive system that might have resulted in abnormal loading of the tail assembly. Consistent with Robinson service bulletin advice, in-flight detachment of the tip cap would have caused severe vibration. This would have been in the form of immediate, and relatively high frequency and high amplitude loading leading to rapid fracture of the low-ductility casting. That is, the tip cap detachment and resulting imbalance led to the empennage separation. 

The failures occurred with the helicopter in a low hover. Nevertheless, the instructor’s quick and appropriate actions were effective in preventing a more serious outcome.

Tail rotor tip cap adhesive failure

The condition of the residual adhesive at the tail rotor tip cap was consistent with adhesive failure over approximately 85% of the original bonded area. With normal forces acting on it at the time of the occurrence, reduced bond effectiveness is the only reasonable explanation for the tip cap to separate in this case.

Given the nature of the failure and the presence of the fluid patches (likely water) in the bond of blade 5803 it is likely that this adhesive failure was associated with the presence of moisture. While both blades exhibited adhesive failure, there was a marked difference in the residual cohesive failure area, indicating a more rapid degradation in blade 5803. The tail rotor blades were installed as a set and, to the extent that could be determined, had experienced the same operational life and environmental conditions. This was consistent with the blades exhibiting a similar degree of erosion and pitting of the leading edge. 

The leading-edge skin crack was almost certainly related to the significant erosion and pitting, and the condition of the crack showed that it existed prior to the tip cap separation. As such, it would have provided a moisture pathway to the adhesive bond, in addition to attack around the circumference of the bond line. This probably contributed to the differing extent of disbonding between the cracked blade and the opposing blade. However, it is noted that the reported tip cap separation involving an R44 had no evidence of blade skin cracking, showing that cracking was not necessarily a prerequisite for adhesive failure of the tip cap bond.

The resistance of adhesive bonds to moisture attack relies primarily on appropriate preparation of the substrate. While there were no identified manufacturing anomalies, it was possible that there were small differences in the bonding preparation between the two blades. However, given the very low incidence of this type of occurrence, it is unlikely that the manufacturing process was problematic or that the life limit was inappropriate.  

Maintenance inspections

Defence against detachment of the tail rotor tip cap involved visual inspections of external blade condition and tap testing of the bond area. The extent of adhesive failure at the detached blade tip indicated that a significant disbond likely existed at the time of the most recent maintenance inspection, 21 flight hours prior to the occurrence. However, there were no recorded defects or indications of disbond from tap testing that, per the R22 service letter SL-93, was for the purpose of verifying bond integrity.  

Tap testing is perhaps the simplest and most inexpensive method for non-destructive inspection of adhesive bond integrity. However, the test has limitations. Tap testing relies on operator judgement in identifying differences in acoustic responses while consistently tapping along a bond area. To produce a change in pitch, there needs to be an underlying structural difference associated with the bond. Unbonded substrates maintaining good contact with the adhesive, or as in this case, bonds with extensive adhesive failure may not produce a clear acoustic difference.

The reason disbond indications were not detected by tap testing in this occurrence was not explored in detail, however the limited examples in this report highlight some of the variability and limitations of tap testing to accurately assess adhesive bond integrity: 

• The maintainer of the occurrence blade indicated they had extensive experience in tap testing and had previously identified disbonds in other manufacturer’s rotor blades. This indicates a degree of success with the test method, but not necessarily with the subject blades.
• Tap testing of the opposite blade by Robinson, under ideal conditions during their blade inspection, identified some likely disbonds ‘mid-span’ (although Robinson did not attempt to determine the extent of the disbonds that were later found to be present around most of the tip circumference).
• The reporter of an R44 tip cap detachment occurrence stated that their tap testing during the most recent maintenance inspection did not produce indications of disbonding. They could not subsequently identify an acoustic difference between the blade with the detached tip (after refitment) and a new blade. 

Visual examination of the external blade condition for damage and corrosion cannot provide a direct indication of adhesive bond integrity, but may provide an indication of potential underlying conditions that could lead to in-service failure. For example, the occurrence blades, the blade from the NTSB tail rotor failure investigation, and potentially the R44 occurrence mentioned above had indications of corrosion adjacent to the tip cap bond line. The visible bond line corrosion could extend to the adjacent bond or be indicative of exposure to conditions that increase the risk of adhesive failure. 

At the time of the occurrence, detection of blade corrosion did not require blade replacement, unless it exceeded the depth limits in the R22 maintenance manual, or per SL93, for ‘any hole that completely penetrates blade skin’. Had the corrosion been identified and removed in accordance with this procedure, the tip cap separation would likely still have occurred. In contrast, identification of the leading-edge crack would have necessitated blade replacement. However, the blade skin had deformed and the crack opened up during the tip separation. As such, it wasn’t possible to establish how the crack would have initially presented and whether it would have been readily identifiable under unaided visual inspection.  

Findings

From the evidence available, the following findings are made with respect to the in-flight tail rotor blade failure and tail assembly separation involving Robinson R22, VH-THM, at Gold Coast Airport, Queensland on 26 February 2022. 

Contributing factors

• In-flight detachment of a tail rotor blade tip cap led to abnormal dynamic loading within the tail rotor assembly, fracture of the aft tailcone bulkhead and separation of the empennage.
• The tail rotor blade tip cap bond had a significant proportion of moisture-induced adhesive failure that reduced the bond strength resulting in the in-flight detachment from the tail rotor blade.
• The scheduled maintenance inspection, including tap testing of the tail rotor blades, 21 flight hours prior to the occurrence, did not identify adhesive disbonding that was very likely present at the tip cap bond line of both blades.

Safety issues and actions

Safety action not associated with an identified safety issue

Safety action by Robinson Helicopter Company

On 3 June 2022, RHC published revision ‘A’ of R22 Service Letter SL‑93, R44 Service Letter SL‑82,and R66 Service Letter SL‑40, on the subject of tail rotor blade condition and care. The original service letter was published on 30 June 2021. The background to the service letter stated:

RHC has recently seen tail rotor blades that were allowed to corrode to an unserviceable condition including severe leading edge pitting and degradation of the bond at the tip cap. Regular preventive maintenance is imperative for continued safe operation and additional care may be required in corrosive environments such as coastal or shipboard operations. Recommended practices to prevent and mitigate the effects of corrosion are provided below. Revision A adds inspection criteria.

Revision A corrosion inspection criteria included using 10x magnification to inspect adjacent to the tip cap bond line for the presence of adhesive and absence of corrosion. Blades were required to be replaced if there was any corrosion at the bond line or if there were any gaps in the bond line adhesive.

On 22 December 2022 RHC published revision A of R22 Service Bulletin SB–120, R44 Service bulletin SB–112, and R66 Service Bulletin SB–41. The background to the service bulletin stated:

RHC has received reports of tail rotor blade tips coming loose due to corrosion at the bond. Helicopters operating near saltwater are particularly susceptible to corrosion, especially if stored outdoors. A debonded tip can cause severe vibration and possible failure of the tail rotor gearbox housing. 

The service bulletin listed the affected blade part and serial numbers that have been deemed susceptible to corrosion and provided a recurrent visual inspection to be inserted into the pilot’s operating handbook and carried out before the first flight of the day. Any indications of corrosion or exposure at or adjected to the tip cap bond line required compliance with SL-93 before further flight. The susceptible blades were required to be replaced no later than 31 December 2024, noting that:

Helicopters operated and/or stored outdoors in corrosive environments such as salt water coastlines should replace affected blades as soon as practical.

The bulletin indicated that newer revision blades had tip caps manufactured from an alternate alloy to reduce the likelihood of corrosion. 

Safety action by United States Federal Aviation Administration

The United States Federal Aviation Administration (FAA) issued airworthiness directive AD 2024‑04‑02, effective from 2 April 2024, relating to tail rotor blades on Robinson helicopters, due to 3 reports of tip caps coming loose due to corrosion at the bond in R44 helicopters (which have similar tail rotor blade tip caps). The summary statement stated:

The FAA is adopting a new airworthiness directive (AD) for certain Robinson Helicopter Company Model R22, R22 Alpha, R22 Beta, R22 Mariner, R44, R44 II, and R66 helicopters. This AD was prompted by reports of helicopters losing a tail rotor blade (TRB) tip cap. This AD requires visually checking and inspecting certain part-numbered and serial-numbered TRB tip caps for evidence of corrosion and, depending on the results, removing the corrosion. This AD also requires removing all affected TRBs from service and prohibits installing them on any helicopter. The FAA is issuing this AD to address the unsafe condition on these products.

Glossary

Sources and submissions

Sources of information

The sources of information during the investigation included:

• the instructor for the occurrence flight
• Professional Helicopter Services
• Robinson Helicopter Company and
• the maintenance organisation for VH-THM. 

References

Petrie, Edward M. (2000). Handbook of Adhesives and Sealants. McGraw-Hill, New York.

Submissions

Under section 26 of the Transport Safety Investigation Act 2003, the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. That section allows a person receiving a draft report to make submissions to the ATSB about the draft report. 

A draft of this report was provided to the following directly involved parties:

• the instructor for the occurrence flight
• Professional Helicopter Services
• Robinson Helicopter Company
• the maintenance organisation for VH-THM
• the Civil Aviation Safety Authority

Submissions were received from the Robinson Helicopter Company.

The submissions were reviewed and, where considered appropriate, the text of the report was amended accordingly.

Appendix A – Metallurgical examination and testing of the fractured tailcone aft casting

Casting material

The as-cast A356 microstructure primarily consists of a hard, brittle, plate-like silicon eutectic[9] constituent in an aluminium matrix. In the unmodified[10], as-cast state, the silicon phase is brittle and has a coarse, plate-like morphology, leading to a low ductility material. Alloy modification and heat treatment is often employed to improve the ductility and toughness of the material.

Fractography

General fracture characteristics

The scanning electron microscope (SEM) examination of the tailcone aft casting fracture identified a general brittle fracture surface dominated by irregularly shaped silicon plates within an aluminium matrix (Figure 14). The brittle fracture features were consistent with a metallographic section perpendicular that showed the jagged fracture path preferentially followed the line of silicon plates in the alloy microstructure. A metallurgical examination identified that the microstructure of the aft bulkhead was consistent with unmodified A356 (Figure 15).

Figure 14: SEM micrograph showing the general casting fracture surface morphology 

Source: ATSB

Figure 15: Microstructural cross section of unmodified A356, showing silicon eutectic (grey) in an aluminium matrix (white) 

Source: ATSB

Fracture surface striations

Isolated areas of aluminium matrix with vague parallel features, or striations, were present across the fracture surface (Figure 16). Some of the features were determined to be mechanical damage due to local contact between the crack or fracture halves and/or related to fracture or decohesion[11] of the silicon eutectic particles. 

For comparison, a section of the casting was subjected to a monotonic (single impact) fracture. At low magnification, the fracture surface appeared similar in appearance to that of the occurrence fracture. However, at high magnification there were areas of well-defined, micro-dimpled rupture in the matrix (Figure 17) that were not observed on the occurrence fracture. The dimples were determined to be characteristic of unstable crack growth or ‘fast fracture’ under tensile loading. 

Further examination of the occurrence casting identified localised areas of very fine striations near the part surface or an area of shrinkage porosity, that were more representative of high cycle fatigue crack progression (Figure 18). There were no gross material or manufacturing defects observed on the fracture surface and the total combined defect area was relatively low in comparison to similar castings examined (from other Robinson helicopters). The larger of these defects were located immediately below the part surface and were in the order of 1 mm across the principal axis on the fracture surface (Figure 19). The fracture surface immediately surrounding some of these defects was noticeably flatter and also exhibited similar striation-like features in area of aluminium matrix. The reason for the striations or cracking was not determined, however, the areas affected were too small to have had any meaningful influence on the casting fracture. 

Figure 16: Occurrence fracture surface ‘striated’ matrix

Source: ATSB

Figure 17: Monotonic fracture surface showing micro-dimpled rupture

Source: ATSB

Figure 18: Discrete area of striations identified on the fracture surface 

Source: ATSB

Figure 19: SEM micrograph of an example defect in the aft bulkhead fracture surface where subsurface shrinkage porosity was identified

Source: ATSB

Fatigue testing of the casting alloy

High-cycle and low-cycle fatigue tests were conducted on a representative sample[12] of A356 material by the Defence Science and Technology Group (DSTG) at the request of the ATSB to allow further fractographic characterisation and comparison with the occurrence casting. The specimens were tested to failure in a tension-tension test regime (Table 1, Table 2, Table 3, and Table 4).

Table 1: High cycle fatigue testing specimen geometry

Table 2: High cycle fatigue testing parameters

Table 3: Low cycle fatigue testing specimen geometry

Table 4: Low cycle fatigue testing parameters

Cracks initiated in both specimens at subsurface porosity at one corner of the specimen. Both specimens showed a defined, relatively flat, cleavage-type crack origin and initial crack progression region, with clear chevrons, river lines and striation features, clearest in the high-cycle fatigue specimen (Figure 20). This transitioned to an increased prevalence of the faceted, silicon particle-dominated, fracture surface and micro-dimpled rupture through the unstable crack growth and overstress regions. There was no similar significant area of cleavage fracture on the occurrence fracture surface.

Figure 20: High-cycle fatigue specimen fracture origin

Source: ATSB

Discussion

A review of fatigue crack growth studies showed that that the mechanism of crack progression through unmodified A356 alloy was dependent on the particle morphology and orientation, as well as the applied stress intensity. At low stress intensities, the plate-like silicon particles debond from the matrix unless their principal axis was oriented perpendicular to the crack plane. The result was a flatter, cleavage-type fracture, such as that exhibited by the initiation region of the high cycle fatigue test specimen. At higher stress intensities, particle fracture was found to dominate and the fracture surface produced was more faceted, with the crack path following the orientation of the silicon particles. Based on this, while there were discrete areas representative of the former mechanism, the dominant appearance of the occurrence fracture was consistent with a higher applied stress intensity. 

While there were some minor casting defects in this occurrence, there were no significant material, manufacturing or operational anomalies identified that would have abnormally predisposed the part to fracture, beyond other similar castings. 

Purpose of safety investigations The objective of a safety investigation is to enhance transport safety. This is done through:  identifying safety issues and facilitating safety action to address those issues providing information about occurrences and their associated safety factors to facilitate learning within the transport industry. It is not a function of the ATSB to apportion blame or provide a means for determining liability. At the same time, an investigation report must include factual material of sufficient weight to support the analysis and findings. At all times the ATSB endeavours to balance the use of material that could imply adverse comment with the need to properly explain what happened, and why, in a fair and unbiased manner. The ATSB does not investigate for the purpose of taking administrative, regulatory or criminal action. Terminology An explanation of terminology used in ATSB investigation reports is available here. This includes terms such as occurrence, contributing factor, other factor that increased risk, and safety issue. Publishing information Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2024 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this publication is licensed under a Creative Commons Attribution 3.0 Australia licence. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided that you attribute the work. The ATSB’s preference is that you attribute this publication (and any material sourced from it) using the following wording: Source: Australian Transport Safety Bureau Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

Executive summary

What happened

On 11 January 2022, a Robinson Helicopter Company R22 Beta helicopter with registration VH‑VHE, was participating in a mustering operation south-west of Mitchell, Queensland. At approximately 1015, 25 minutes into the second mustering flight of the day, the pilot heard a ‘pop’ and a ‘crack’ and experienced a significant uncommanded yaw and abnormal vibration. The pilot completed a forced landing from an altitude of approximately 3 to 5 metres, shut down the engine and inspected the helicopter; identifying that the tail rotor drive shaft had fractured.

What the ATSB found

The ATSB found that the intake valve in the engine’s number-four cylinder sustained thermal damage, which led to reduced engine performance and required the pilot to conduct a forced landing. During the forced landing, for reasons that could not be determined, the tail rotor drive shaft fractured as a result of torsional overstress.

In addition, an unapproved modification was present on the leading edge of the tail rotor blades, although this probably had no influence on the occurrence.

Safety message

This occurrence illustrates the potential seriousness of unexpected yaw and reduced engine performance during flight. These characteristics in piston engine helicopters may be symptomatic of developing engine intake valve damage. If this condition remains unattended it can lead to an increased risk of induction backfire events and significant loss of engine power.

When a loss of engine power or abnormal operation is encountered, an appropriately‑licensed maintenance engineer should complete an engine cylinder inspection in accordance with the helicopter and engine manufacturer’s requirements before further flight.

There is an opportunity to improve the understanding of engine intake valve issues with better quality data. Maintenance organisations are therefore encouraged to follow the Civil Aviation Safety Authority (CASA) recommendation for borescope inspections of valves and report any defects or nil-defect findings.

The investigation

The occurrence

On 11 January 2022, a Robinson Helicopter Company (RHC) R22 Beta helicopter, registered VH‑VHE and operated by Diamond T Helicopters, was participating in a mustering operation south‑west of Mitchell, Queensland. The pilot was the only person on board.

Prior to the first flight of the day, the pilot prepared the helicopter for operation and completed pre‑flight checks, with no anomalies observed. The pilot then performed a stabilised hover check[1] and conducted a 1.1 hour positioning flight to the mustering location, departing at 0530 Eastern Standard Time.[2] After shutting down, the pilot then refuelled the helicopter and completed the first mustering flight, which took 2.5 hours. The pilot shut down the helicopter and again refuelled. Both flights were reported to be free from incident or abnormalities.

At about 0950, the pilot commenced a second mustering flight. About 25 minutes into the flight, the pilot heard a ‘pop’ and a ‘crack’, then experienced a significant uncommanded yaw[3] and abnormal vibration. The pilot conducted a forced landing, with a degree of pedal control, from height of 10 to 15 ft above ground level. As the skids touched the ground, the pilot heard another ‘crack’ and felt the helicopter go ‘into a spin’. 

The pilot remained in the helicopter while completing shut down, including waiting for the cylinder head temperature to reduce. Inspection following shut down revealed that the tail rotor drive shaft was fractured. 

Subsequent technical inspection by the maintenance organisation revealed that the number-four cylinder could not hold the required pressure and the intake valve had significant thermal damage and guttering[4] (see the section titled Context - Examination by maintenance organisation).

Context

Examination by maintenance organisation

The helicopter was examined by a maintenance organisation who initially identified the following:

• the tail rotor drive shaft had fractured approximately 5 mm forward of the rear flange weld (Figure 1 and Figure 2)
• rotational scoring was present on the internal surface of the tailcone adjacent to the fracture location
• the tail rotor blade surfaces were free from evidence of foreign object damage, and the skid tube (shown schematically in Figure 1) was free from contact indications
• there was erosion on the leading-edge of the main rotor blades and the tail rotor blades
• epoxy additions were present on the tip of the leading-edge of the tail rotor blades (Figure 3), which were reportedly applied to increase blade wear resistance.

Figure 1: Tail rotor drive shaft fracture location

The fracture location is approximately 5mm forward of the rear flange weld. 

Source: Robinson Helicopter Company, annotated by the ATSB

Figure 2: Fractured tail rotor drive shaft 

The fracture shows significant plastic deformation. The corrosion product was not present at time of the occurrence.

Source: ATSB

Figure 3: Tail rotor blade leading edge erosion and blade modification

Both tail rotor blades exhibited erosion and a leading-edge tip modification. There was no indication consistent with a strike to either tail rotor blade.

Source: Maintenance organisation, annotated by the ATSB

During the post-occurrence inspection of the engine, maintenance engineers identified that the number-four cylinder was unable to hold pressure when subjected to a hot-compression check. The maintenance organisation removed the cylinder and, on further inspection, identified thermal damage and guttering to the intake valve (Figure 4). The bottom of the valve exhibited an asymmetric burn pattern, which was evidence of severe and uneven heating. The guttering present on the valve face provided a gas leakage path between the cylinder and the intake manifold.

Figure 4: Damaged engine intake valve from number-four cylinder 

The asymmetric burn pattern on bottom on valve is evidence of severe and uneven heating. The valve face guttering is a leakage path between the cylinder and the intake manifold. The valve was ground by the maintenance organisation, following removal, to ‘clean’ the face to enhance the appearance of the thermal damage. The appearance of the face immediately following the occurrence was not recorded.

Source: ATSB

The maintenance organisation inspected the drive train components, and observed that:

• the intermediate flex plate was undamaged
• the tail rotor gearbox chip detector was free from debris and the gearbox oil was clear
• the clutch assembly operated without defect or roughness, and the lubricating fluid was free from contaminants or fragments.

ATSB technical examination

Additional examination of the damaged number-four cylinder intake valve, tail rotor gearbox, and the fractured tail rotor drive shaft was completed at the ATSB’s technical facilities. 

The ATSB examination found:

• confirmation of the thermal damage and guttering on the intake valve[5] 
• correct assembly of the tail rotor drive shaft to the gear box yoke connection 
• no anomalies with the gearbox bevel gears and bearings that could have restricted movement of the drive shaft were identified
• the turning resistance of the tail rotor gear box was consistent with manufacturer's advice. 

The direction of the torsional deformation of the driveshaft was indicative of either drag on the tail rotor assembly or over-torque from the driving side of the failure location. 

Examination at high magnification, using both an optical and scanning electron microscope, of the mating fracture surfaces of the drive shaft revealed only gross ductile overstress features. A polished and etched cross section, and hardness measurements taken through the weld connecting the shaft to the flange, found the weld to be heat-relieved, which was consistent with the manufacture’s specifications. Energy dispersive spectroscopic analysis found the composition of the alloy to also be consistent with the manufacturer’s specifications. 

There were no pre-existing defects identified from manufacture, or operational damage such as surface scoring, fatigue cracking, denting, or corrosion that might have otherwise weakened the shaft and predisposed it to fracture.

Operational and maintenance history

VH-VHE was manufactured by the Robinson Helicopter Company (RHC) in 2010 and first registered in Australia in the same year. The helicopter was powered by a Lycoming Engines (Lycoming) four-cylinder O-360-J2A engine. The helicopter had accumulated 2,135.5 hours total time in service. 

The helicopter was overhauled in August 2021 and new Lycoming engine cylinder kits[6] were fitted at that time. The pilot stated that the engine seemed underpowered, when compared with engine power prior to the overhaul. The tail rotor drive shaft was also replaced with a new item at the overhaul.[7] 

The next 100-hourly scheduled inspection (75.7 hours following overhaul),[8] included compression checks[9], service limitations for valves checks,[10] and valve guide condition checks.[11] The compression checks found inadequate compression in the number-two and number-three cylinders, and the cylinders were reconditioned. At this inspection, the number‑four cylinder recorded adequate compression (78/80 pounds per square inch). The pilot stated that the power was improved following this inspection and cylinder reconditioning; however, was still underpowered compared with prior to the overhaul. The maintenance organisation did not perform, nor was there a requirement for them to perform, borescope inspection of the valve faces.

The helicopter accrued 93.3 hours in service from overhaul to the occurrence flight. The pilot reported that they did not observe any adverse indications or unusual behaviour of the engine, such as rough running, yawing, or backfiring during the numerous flights leading up to the occurrence. 

The maintenance organisation had a practise of applying epoxy to the leading edge, adjacent to the tips, of the tail rotor blades to extend the interval between re-painting. The epoxy was applied to the blades at the first instance re-painting was required. The maintenance organisation described the modification as 

• common practise in rural operations 
• low mass
• inspected during scheduled maintenance
• checked for balance, as part of blade balancing, following their addition and/or repainting. 

The maintenance organisation did not observe a change in the condition of the epoxy after the occurrence, except that the paint had eroded. 

Manufacturers’ advice in response to the occurrence

The ATSB sought comment from Lycoming regarding the condition of the damaged intake valve and the acceptable compression check result for the number-four cylinder, which occurred 17.6 hours prior to the occurrence. However, no advice was received at the time of writing. 

With regards to the subject occurrence, RHC advised that the VH-VHE tail rotor drive shaft fracture was consistent with a tail rotor strike, described as anything that creates a sudden resistance or drag to the tail rotor blade(s). They went on to state that a strike does not always result in damage to the blades, nor always leave an identifiable residue. The location of the tail rotor drive shaft fracture, in the event of a strike, can vary, but the fracture location in this occurrence would not be unusual for a strike. 

ATSB investigation AO-2021-008 (see the section titled Other occurrences - Tail rotor drive shaft fracture) had similarities to this occurrence. Consequently, the ATSB asked RHC whether there could be causal link between the engine response to intake valve failure, the instinctive pilot control inputs, the landing, and fracture of the tail rotor drive shaft. RHC stated that in the event of a sudden power loss followed by an instant increase in power, the input spike could distort the tail rotor drive shaft, possibly to the extent of twisting the shaft, provided that the contributing circumstances were extreme.

RHC also advised that the epoxy blade modifications were unapproved and would prevent inspection of the blade surface for cracking or corrosion. In addition, RHC stated that a helicopter could experience a more catastrophic consequence of the blade modification if an out-of-balance condition developed, possibly resulting in the loss of the entire empennage. 

Other occurrences

Cylinder head inlet valve distress

In 2019, prompted by increasing industry reports of engine issues involving RHC R22 and R44 helicopters, the Australian Helicopter Industry Association (AHIA) released an independent industry-supported investigation into durability issues affecting Lycoming O-320, O-360, and 
O-540 engines, fitted to RHC R22 and R44 models.[12] The report identified that poor operational durability of intake valves was principally due to accelerated valve, valve guide, and/or valve seat wear. The investigation further stated that these durability issues lead to a loss of cylinder compression and the potential for partial power loss events. The valve damage could result in an induction backfire, engine power loss, and airframe yaw. No one factor was identified as the reason for the accelerated intake valve wear. The report recommended helicopter operators remain vigilant of the issues outlined within the AHIA report and ensure they continue to follow best-practice operation and maintenance protocols.

In October 2020, RHC issued a Safety Alert in response to reports of burned engine intake valves in their R44 model helicopters.[13]^, [14] The Safety Alert listed three precautions for pilots to follow:

• Perform a complete run up and stabilized hover check prior to every flight. Do not initiate flight if there is any indication of engine roughness or sudden yaw.
• If engine roughness or sudden yaw occurs in flight, land as soon as practical and be prepared to land immediately.
• Following any engine roughness or sudden yaw, have a mechanic check valve condition before further flight. The mechanic should listen for sound of leakage at each intake valve while performing a compression check. Any intake vale with audible leakage requires repair prior to further flight. Check may be done with engine hot or cold.

Subsequently, Lycoming released a service instruction to collect early-life intake valve in-service data for the same RHC R44 models covered in the RHC Safety Alert to further understand the engine durability issues.[15] The ATSB asked Lycoming whether the collected data had revealed any safety relevant information data. At the time of writing, no response had been received, although it was also noted that R22 models were not included in this safety instruction. 

In July 2021, Lycoming published a service instruction allowing for the incorporation of rotator-type intake valves into O-360-J2A model engines[16] (the engine model fitted to VH-VHE) to improve valve seating and minimise compression loss at the intake valve seat. Valve rotation is known to improve face and seat wear distribution and cleaning. This optional modification had not been installed on VH-VHE.

The most recent issue of Civil Aviation Safety Authority (CASA) Airworthiness Bulletin 85-025 RHC R22/R44 Engine Intake Valve and Valve Seat Distress, published in August 2021, detailed many of the factors that may contribute to intake valve damage.[17] CASA made numerous recommendations to operators regarding operating procedures and limitations, which included: 

• expanding the three recommendations made in the RHC Safety Alert to include all R22 and R44 models
• guidance for operating temperature limitations 
• guidance for cool down and shut down procedures.

In addition, CASA strongly encouraged the implementation of regular borescope inspections and reporting of both defect and nil-defect findings to CASA. At the time of ATSB consultation, CASA had not received any borescope inspection results.

Tail rotor drive shaft fracture

A search of the Australian and United States aviation service difficulty reporting databases[18] identified two similar reported instances of a RHC R22 tail rotor drive shaft fracture events, both were Australian‑registered helicopters: 

• A defect report, which noted that the failure of the tail rotor drive shaft occurred during take-off and that the helicopter may have experienced an overspeed event or have operated within a resonant speed range.[19] 
• The ATSB investigated occurrence AO-2021-008, involving an R22 with registration VH-HCX. This investigation detailed notable similarities with this occurrence. In both occurrences the investigation found a thermally damaged intake valve, a fractured tail rotor drive shaft, no strike indications on the tail rotor blades or stinger, and both pilots stated they did not strike any object. A distinct difference between the two occurrences, however, was that the pilot of VH‑HCX had experienced reduced engine performance and associated momentary uncommanded yawing in the flight just prior to the occurrence, whereas the pilot of VH-VHE experienced no abnormal indications prior to the occurrence. The VH-HCX investigation also found that ‘the [intake] valve from the number-four cylinder of the engine sustained thermal damage, which led to reduced engine performance during the occurrence flight and the requirement for the pilot to perform a forced landing’. The ATSB concluded that the fracture of the tail rotor drive shaft occurred during the running landing; however, the investigation was unable to conclusively determine the reasons for the driveshaft fracture. 

Safety analysis

Engine power loss

Approximately 25 minutes into a mustering flight VH-VHE experienced an engine power reduction, induction backfire(s), and yawing event resulting in a forced landing. The unusual engine behaviour was likely due to thermal damage to the intake valve from the number-four engine cylinder, which reduced the effectiveness of the seal between the valve face and the seat. 

Neither the pilot nor the maintenance organisation observed any indications of the progressing damage to the intake valve. This included a successful compression test 17.6 flight hours prior to the occurrence. While not required, a borescope inspection may increase the likelihood of detecting valve damage that can lead to degraded engine performance. In this instance, the degradation in engine performance fortunately occurred where conditions were favourable for the forced landing.

Tail rotor drive shaft

The ATSB examination identified that the tail rotor drive shaft fractured in overstress from torsional loading. The reported pedal control, up until the skids touched the ground, suggested that the drive shaft fracture was concurrent with the helicopter touching the ground.

Noting the absence of identified mechanical or material issues with the tail rotor drive train, the ATSB considered two hypotheses to explain the fractured tail rotor drive shaft. These were:

• the tail rotor blades encountered a foreign object
• the altered engine performance, coupled with the pilot control inputs, and the ground contact led to fracturing of the tail rotor drive shaft. 

However, there was insufficient evidence to support either of these hypotheses, with the reasons for the driveshaft fracture remaining unknown.

Blade modifications

The tail rotor blade modifications found on VH-VHE were unapproved and had the potential for significant consequences. Using approved and documented engineering processes to make changes to an aircraft system is important, as it ensures the design complies with applicable airworthiness standards or an equivalent level of safety.

Findings

From the evidence available, the following findings are made with respect to the engine failure and forced landing of a RHC R22 Beta helicopter that occurred 130 km south-west of Mitchell, Queensland, on 11 January 2022.

Contributing factors

• Damage to the number-four engine cylinder intake valve allowed the fuel and air mixture, and exhaust gases into the air induction system. This resulted in the degraded engine performance and the forced landing of the helicopter.

Other factors that increased risk

• For reasons that could not be determined, the tail rotor driveshaft failed through torsional overstress, likely during the forced landing.
• An epoxy addition present on the leading edge of the tail rotor blades was an unapproved modification.

Sources and submissions

Sources of information

The sources of information during the investigation included:

• the pilot of VH-VHE
• helicopter maintenance organisations
• Robinson Helicopter Company
• Helicopter Rebuilds Pty Ltd
• Lycoming Engines

Submissions

Under section 26 of the Transport Safety Investigation Act 2003, the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. That section allows a person receiving a draft report to make submissions to the ATSB about the draft report. 

A draft of this report was provided to the following directly involved parties:

• the pilot of VH-VHE
• Helismart Pty Ltd
• Approved Aircraft Maintenance Pty Ltd
• Lycoming Engines
• Robinson Helicopter Company
• Civil Aviation Safety Authority
• United States National Transportation Safety Board

A submission was received from:

• Robinson Helicopter Company

The submission was reviewed, however it did not result in any change to the report text.

Purpose of safety investigations The objective of a safety investigation is to enhance transport safety. This is done through: identifying safety issues and facilitating safety action to address those issues providing information about occurrences and their associated safety factors to facilitate learning within the transport industry. It is not a function of the ATSB to apportion blame or provide a means for determining liability. At the same time, an investigation report must include factual material of sufficient weight to support the analysis and findings. At all times the ATSB endeavours to balance the use of material that could imply adverse comment with the need to properly explain what happened, and why, in a fair and unbiased manner. The ATSB does not investigate for the purpose of taking administrative, regulatory or criminal action. Terminology An explanation of terminology used in ATSB investigation reports is available here. This includes terms such as occurrence, contributing factor, other factor that increased risk, and safety issue. Publishing information  Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2022 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this publication is licensed under a Creative Commons Attribution 3.0 Australia licence. Creative Commons Attribution 3.0 Australia Licence is a standard form licence agreement that allows you to copy, distribute, transmit and adapt this publication provided that you attribute the work. The ATSB’s preference is that you attribute this publication (and any material sourced from it) using the following wording: Source: Australian Transport Safety Bureau Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

Investigation summary

What happened

On 3 March 2022, the pilot of a Cessna U206G, registered VH-JVR and operated by MAGSPEC Aviation Pty Ltd, was conducting a low-level geophysical survey, about 120 km west of Norseman, Western Australia. Recorded data showed the first survey line commenced at 1252 local time, and the aircraft’s last recorded position occurred at 1343, in the survey area.

At about 1430, the ground operator observed that the satellite tracking system was no longer reporting the aircraft’s position, and they were subsequently unable to contact the pilot. When the aircraft failed to return to Kalgoorlie by its estimated arrival time of 1630, a search was initiated. At about 1852, the wreckage was located 3.2 km west of the aircraft’s last recorded position. The injured pilot had extricated themselves from the wreckage but shortly after succumbed to their injuries. The aircraft was destroyed. 

What the ATSB found

The ATSB found it was likely that, during a manoeuvre to intercept the next survey line, for undetermined reasons, control of the aircraft was lost at a height from which recovery was not possible. While an aerodynamic stall situation was a plausible explanation for the loss of control, this remained only a possibility due to the lack of recorded data beyond the last known position of the aircraft and no witness observations. 

Although the aircraft’s satellite tracking system had stopped at 1343, an emergency response was not initiated until 1700. This was in accordance with the operator’s emergency response plan, in which a search and rescue response was to commence 30 minutes after the estimated time for arrival. However, an earlier response was very unlikely to have altered the outcome due to the extent of the pilot’s injuries. Minimising the time for a search and rescue is essential to increasing the chances of a successful outcome in the event of an accident. 

In accordance with the operator’s training, its pilots routinely used high angle of bank (45‍–‍60°) turns at low level to manoeuvre between survey lines. Steep turns at low level increases the risk of an aerodynamic stall from which a recovery may not be possible. ATSB analysis of the available satellite tracking data identified that, although the pilot was conducting steep turns, they had flown the previous 24 turns without incident.

The pilot was not wearing any protective clothing or a helmet nor were they required to do so by the operator. Wearing of such items has been recommended by industry bodies, as they may offer some protection in the event of an accident, particularly from fire but also as environmental protection following an accident.

The ATSB identified that the operator’s risk management processes did not include a pre‑operational risk assessment that considered the generic risks and hazards common across their low‑level survey operations. Further, a risk register was not maintained, which limited the operator’s ability to track, monitor, and mitigate all known hazards, and assess the effectiveness of the existing risk controls.

Also, the operator trained its pilots to routinely fly survey patterns utilising steep turns at low level. However, the procedures or limitations specific to these manoeuvres were not included in the operations manual. It was also noted that, the operator’s aircraft were fitted with a satellite tracking system, but there was no requirement nor supporting procedures to confirm the set‑up and functionality of the system prior to flight.

Although not contributory, the ATSB identified that the regulatory oversight of the operator had not specifically examined the primary activity of low‑level geophysical survey flights or the processes and procedures designed to mitigate any associated risks.

What has been done as a result

MAGSPEC Aviation has consolidated its manuals, with its health, safety and environmental management system manual incorporated into its operations manual. In addition, it has implemented a range of measures regarding its low‑level survey operations including:

• an updated job safety analysis, which provides for the consideration of likelihood, consequence and details of any risk mitigations
• procedures for the use of satellite tracking including a requirement for a pre-flight check
• conduct of procedure turns including how the turns should be flown, with minimum speeds indicated and what to do if those speeds could not be achieved
• the requirement for fixed emergency locater transmitters on all aircraft, and limitations for flight if the unit is unserviceable or not present
• updating its low-level training syllabus to include specific parameters (which mimic its previous practical training), so that competency can be formally assessed against those parameters.

MAGSPEC Aviation has also advised it no longer operates at survey heights below 30 m above ground level and it provides its pilots with an individually registered portable locator beacon, which they are required to wear on their person. It is also progressing operational amendments to enable Flight Safety Foundation’s Basic Aviation Risk Standard accreditation. Just prior to final publication of this report, the operator advised that it had been awarded the Basic Aviation Risk Standard accreditation.

Safety message

Geophysical survey operations are generally conducted at low level, necessitated by the requirement for high quality, accurate data acquisition. This creates a high‑risk operating environment that requires effective risk management.

Risk management should include a pre‑operational risk assessment to consider hazards and risks common to an operation. This can then be used to inform the management of risk for specific taskings and assist in developing appropriate mitigations. Tools such as a risk register can assist an organisation to effectively monitor its risk profile and continually improve its risk mitigation strategies.

Policy and procedures form part of effective risk mitigation strategies and will establish safety and the operating standards to be met and maintained. Documented policies and procedures can ensure the correct set‑up and functionality of operating equipment and systems. It can also minimise opportunities for deviation from an operator’s expectations and the erosion of safety margins.

This accident further highlighted that regulatory oversight activities should ensure that an operator’s primary activity is examined in sufficient detail. Not doing so potentially limits the opportunity to assess an operator’s ability to manage the risks associated with its proposed operations.

The occurrence

On 3 March 2022, a Cessna Aircraft Company U206G aircraft, registered VH-JVR, was being operated by MAGSPEC Aviation Pty Ltd for low-level, geophysical survey flights of an area about 120 km west of Norseman, Western Australia (Figure 1).

The aircraft was based at Kalgoorlie for this survey task along with 2 pilots and a ground operator. One pilot would operate the aircraft in the morning and the other in the afternoon. The ground operator was responsible for the technical and logistical aspects of the survey. 

At about 1125 local time, at the completion of the morning survey flight, the aircraft was returned to Kalgoorlie. The second pilot commenced their pre‑flight preparations at about 1130 and discussed the morning survey flight with the returning pilot. That pilot advised of a minor concern about a fuel imbalance that developed during the approximate 4‑hour flight, however, they did not report any impact on aircraft handling or engine operation. 

The ground operator prepared the survey equipment and assisted the pilot to fully fuel the tanks. 

Figure 1: VH-JVR's operating area in proximity to Kalgoorlie and Norseman

Source: Operator’s satellite tracking data, overlaid on Google Earth, annotated by the ATSB

At about 1200, the aircraft departed for the survey area with the pilot as the sole occupant. Recorded GPS data showed that the first survey line was commenced at about 1252, picking up where the morning survey flight had been completed. The last position uploaded to the tracking system was at 1343, which showed the aircraft was on a westerly heading at a ground speed of 116 kt and a GPS height of 1,398 ft above mean sea level in the target survey area.

At about 1430, the ground operator checked the satellite tracking system for VH‑JVR and noted that the aircraft’s position was no longer being reported on the system. The satellite tracking system had the ability to automatically alert the operator 15 minutes after tracking data was no longer being uploaded to the system’s servers. The operator reported they did not receive an automatic alert. 

The ground operator then attempted unsuccessfully to call and text the pilot’s mobile phone. Although the aircraft carried a satellite phone as part of its survival kit, it was not routinely switched on during operations. The ground operator then advised the operations manager, who directed them to continue the attempts at making contact and prepare the ground vehicle for a potential response. Further attempts at contact were unsuccessful and the operations manager directed that no further action could be taken other than to monitor the situation and wait until the aircraft’s estimated time of arrival at Kalgoorlie.

The aircraft did not return to Kalgoorlie by the estimated time of arrival of 1630. At 1700, in accordance with its emergency response plan, the operator contacted the Australian Maritime Safety Authority’s Joint Rescue Coordination Centre (JRCC). Another company aircraft and pilot that was at Norseman was dispatched by the operator to VH-JVR’s last known position, however, the pilot was not able to locate the aircraft.

The JRCC initiated a search and rescue operation at 1739. The aircraft wreckage was located at 1852, approximately 3.2 km west of its last recorded position (Figure 2). The search aircraft’s crew were unable to establish communications with the pilot of VH‑JVR. The JRCC also deployed a rescue helicopter to the site, and at 0042, they found the pilot, fatally injured a short distance from the wreckage.

Figure 2: VH-JVR's accident site location and last recorded position

Source: Operator’s satellite tracking data, overlaid on Google Earth, annotated by the ATSB

Context

Pilot information

Qualifications and experience

The pilot held a commercial pilot licence (aeroplane) issued in 2014 and a valid class 1 aviation medical certificate. They held a multi‑engine aeroplane instrument rating and a grade 3 instructor rating, although neither were current, nor were they required to be. 

At the time of the accident, the pilot had about 1,822 hours total aeronautical experience, of which about 570 hours were with the operator, primarily in Cessna 210 aircraft. The pilot had accrued over 350 hours on Cessna 206 aircraft prior to joining the operator and had about 12 hours on VH‑JVR.  

The pilot commenced and completed a low‑level (aeroplane) rating in June 2021, which comprised 6.7 hours of dual training in a Cessna 152 aircraft, including a flight test. In July 2021, the pilot then commenced low‑level survey training with the operator in a Cessna 210 aircraft. The pilot’s logbook detailed 5 initial training survey flights totalling 18.6 hours. These were followed by about 21 hours of solo low‑level survey, culminating in a check flight of 5.6 hours. 

The chief pilot (CP) conducted the pilot’s low‑level survey training and their geophysical survey operations check flight. The CP reported that the pilot ‘was one of those pilots who picked it up very quickly’ and was ‘very switched on’. In total, the pilot had conducted about 500 hours of low‑level survey operations. 

Although the pilot had previously flown high‑level surveys, MAGSPEC Aviation was the first operator that the pilot had flown low‑level surveys for. The operator also reported that the pilot had been recently offered and had accepted the role of deputy chief pilot/deputy head of operations.

Recent history

The pilot had been on leave since 23 February 2022. The pilot’s partner reported that, on 1 March 2022, the pilot woke at about 0700, went to bed at about 1930‍–‍2000 and did not fly that day. On 2 March 2022, the day prior to the accident, the pilot woke at 0400 and arrived at Perth Airport at 0500 to take a scheduled passenger flight to Geraldton, where VH‑JVR had been undergoing scheduled maintenance. The pilot then ferried the aircraft about 700 km to Kalgoorlie, arriving around midday. Later that day, the pilot accompanied the second company pilot assigned to the survey area on a 30‍–‍40 minute local flight to familiarise the second pilot with VH‑JVR, as they had not previously flown that aircraft. 

The pilot’s partner received a text message from the pilot at 1922, advising that they were cooking dinner in their accommodation and had no plans to go out that night. There was no further evidence of the pilot’s activities prior to the accident flight, but the pilot usually woke around 0600‍–‍0630, exercised and studied in the mornings before conducting the afternoon survey flight. 

The day of accident was the first day of that survey task. Neither the other pilot nor the ground operator expressed any concern for the pilot. Based on the available recent history, there was no evidence the pilot was likely experiencing a level of fatigue at the time of the accident. 

Aircraft information

General

The Cessna Aircraft Company U206G Stationair was a high‑wing, fixed tricycle undercarriage aircraft powered by a single Continental IO‑520‑F piston engine, with a 3‑bladed constant speed propeller. VH‑JVR was manufactured in 1978 in the United States and was first registered in Australia in 1998. The aircraft was acquired by the operator in 2021. 

Factory fitted standard equipment included:

• a vane‑type aerodynamic stall[1] warning system in the leading edge of the left wing designed to activate the audible warning horn 5‍–‍10 kt above the stall speed in all configurations
• 2 vented fuel wing tanks, which were an integral part of the metal wing structure[2] and supplied fuel via gravity feed to 2 reservoir tanks, and a fuel selector valve with selections for LEFT, RIGHT and OFF
• 3-point safety harness restraints.

Modifications

In October 2021, VH‑JVR had been modified and equipped to conduct geophysical survey operations in accordance with engineering orders approved by a Civil Aviation Safety Authority (CASA) authorised aeronautical engineer and supplemental type certificates.[3] These modifications included:

• A magnetometer boom installed at the rear of the aircraft and associated survey equipment, with its own power supply, mounted in the rear cabin.
• A fuel selector valve, which enabled the selection of LEFT/BOTH/RIGHT with a pull‑out fuel shut off valve installed to cut off fuel flow.
• A survey data acquisition and navigation system, which included flight path guidance via a digital display mounted on top of the instrument panel, allowing the pilot to monitor aircraft position in relation to the pre-programmed survey lines.[4]
• A 4-point inertia safety restraint harness.

Maintenance

The aircraft was being maintained by an approved maintenance organisation in accordance with the CASA maintenance schedule 5 and regulatory requirements. The last periodic inspection was completed on 2 March 2022 at 7,982.4 hours total time‑in‑service. The current maintenance release was not recovered and likely destroyed in the post‑impact fire. A review of previous maintenance releases and maintenance records did not identify any major repairs or recurring airworthiness issues with the aircraft.

Emergency locator transmitter

Civil Aviation Safety Regulations (CASR) 1998 Part 91 General operating and flight rules Manual of Standards (MOS) required that VH‑JVR carry an emergency locator transmitter (ELT) or a survival ELT for its intended operation. At the time of the accident, VH‑JVR was not fitted with an ELT but carried a survival ELT (refer to section titled Emergency beacons).

Weight and balance

The CP provided a recreated weight and balance sheet of the accident flight to the ATSB. The morning pilot witnessed the aircraft depart with full fuel in the survey configuration. The weight and balance sheet identified that VH‑JVR weighed about 1,488 kg on departure, about 150 kg below the maximum take‑off weight of 1,636 kg. The centre of gravity on take‑off was near the centre of the allowable range. Therefore, it was very likely that VH‑JVR was within the weight and balance limits at the time of take‑off. 

Meteorological information

The pilot who flew the morning survey flight reported that the weather at that time was fine with good visibility, except for some light turbulence.

The Bureau of Meteorology forecast for the area, valid from 1300, was for visibility to be greater than 10 km and no significant weather for the time of the accident. Winds were forecast to be southerly at about 15 kt. Satellite imagery indicated no cloud cover over the survey area. The nearest weather stations to the accident site were at Norseman (124 km east) and Hyden (158 km west‑south‑west). There was no significant weather reported at either location. Recorded winds at 1330 were south‑westerly at about 14 kt at Norseman and south‑easterly at 10 kt gusting to 17 kt at Hyden.

According to Geoscience Australia’s geodetic calculator, the sun azimuth was north‑west at about 56° elevation about the time of the accident. This was relatively high in the sky and sun glare affecting the pilot was considered not likely.

Recorded data 

Spidertracks data

A Spidertracks Spider X tracking system was installed on the aircraft, which provided near real‑time tracking via satellite and/or cellular networks, recording position, altitude, track heading and groundspeed at 15‑second intervals, increasing during aircraft manoeuvring. The data was transmitted to Spidertracks servers once every minute. The Spidertracks system also had an automatic watch function whereby an alert would be sent via text and email to a nominated person(s) in the event that the data transmissions from the device were not received for a period of 15 minutes.

Spidertracks data was able to be recovered for analysis due to its cloud‑based operation. The physical unit, and other possible data sources of recorded data identified in the wreckage, including personal electronic devices, the engine data monitoring device and the geophysical survey data equipment, were all damaged in the post‑impact fire, preventing data recovery. 

The last known position transmitted by Spidertracks was about 3.2 km east, and approximately 1 minute away (at the last recorded speed) from the accident site (Figure 3). Position data was recorded by Spidertracks once every 15 seconds, increasing to about once every 3 seconds during a turn. However, as the data was only transmitted once every minute, it was likely that the accident occurred before Spidertracks was able to transmit the last data packet to the cloud storage.

The recovered data showed that the pilot had conducted procedure turns (refer to section titled Survey pattern) at the end of each of the completed 24 survey lines, over a period of about 50 minutes. Although all turns were observed to be conducted in a similar manner, one particular turn commenced at a greater distance away from the survey area. During interview, the CP suggested that the pilot may have done so in order to have a drink or attend to a flight‑related task. The ATSB noted no evidence to suggest any concern with this particular turn. The aircraft was on the 25th survey line when the data stopped, and the accident site was in the vicinity of the expected 25th procedure turn (Figure 3).

The recorded data showed the survey lines were being conducted in an east‑west orientation, with left turns conducted to the west and right turns to the east. The average survey line speed was 114 kt at an average height of about 78 ft above ground level (AGL).

Figure 3: Recorded flight path (excluding transit from Kalgoorlie)

Source: Operator’s satellite tracking data, overlaid on Google Earth, annotated by the ATSB

Procedure turn analysis

The available Spidertracks data was analysed to assess the aircraft handling during the accident flight. The recorded ground speed data and forecast wind and direction were used to estimate the true airspeed during the survey flight. Based on the available atmospheric conditions, true airspeed was assumed equal to indicated airspeed and is used throughout the following analysis.[5]

Using the available recordings, the average bank angle, rate of turn and G load for each turn for the accident flight were calculated. These calculations assumed steady coordinated turns, at constant altitude and airspeed, with a constant wind speed and direction. 

The ATSB’s analysis of the Spidertracks data from the aircraft identified that during the procedure turns:

• The angle of bank ranged from 43° to 60° and was typically between 50° to 60°.
• The rate of turn ranged from 10° to 18° per second and was typically between 14° to 18° per second.
• The G load[6] ranged from 1.3 G to 2 G and was typically about 1.7 G to 1.8 G.
• The indicated airspeed ranged from 89 kt to 109 kt and was typically between 94 kt to 104 kt.
• The altitude during turns were between 150 ft to 300 ft AGL, with the average being 200 ft.

Comparison flights

Spidertracks data from 2 previous flights for the accident pilot, which were in VH‑JVR, as well as the morning survey flight conducted in VH‑JVR by another pilot were made available to the ATSB. These were analysed for comparison to the accident flight.

Morning flight 

The morning flight conducted by the other pilot consisted of 50 survey lines and 49 procedure turns. These survey lines were typically flown at 85 ft AGL and 120 kts. This set of survey lines were immediately adjacent to the accident flight survey, with the procedure turns occurring in a similar area. Analysis of these procedural turns identified:

• The angle of bank ranged from 24° to 56° and was typically between 40° and 50°.
• The rate of turn ranged from 4° to 16° per second and was typically between 10° and 13° per second.
• The G load ranged from 1.1 G to 1.8 G and was typically between 1.3 G and 1.5 G.
• The airspeed ranged from 92 kt to 109 kt and was typically between 95 kt and 104 kt.
• The altitude during the turns varied between 210 ft and 550 ft, with the average being 320 ft AGL.

These turns, while generally comparable with the accident pilot’s turns, and considered steep turns,[7] were typically flown at lower angle of banks, rates of turn and G load, and at higher heights above ground level. The ATSB noted that the morning pilot had recently completed their survey training with the accident pilot on the Cessna 210. This was their first low‑level survey flight in the Cessna 206.

Previous flights (accident pilot)

Spidertracks data from 2 prior survey flights in VH‑JVR for the accident pilot were available. These flights were conducted in a different location, over undulating terrain with dense vegetation, with a higher average survey height of about 140 ft AGL at about 113 kt. Each flight consisted of just over 50 procedure turns and survey lines. A summary of the analysis of these procedural turns is contained in Table 1 below (refer flights 2 and 3), with comparison to the accident flight, and the morning pilot (flight 1).

Table 1: Comparative turn analysis results

The ATSB’s analysis of the Spidertracks data from the pilot’s previous survey flights identified that those turns were flown at slightly lower angles of bank, rates of turn and G load when compared with the accident flight. While the pilot was operating at a different location, which may have influenced the way they conducted their turns, the reason for the differences was not able to be determined from the evidence available.

Wreckage and impact information

Wreckage distribution

The aircraft was located in moderately dense scrubland with small to medium trees. The terrain was relatively flat, with some low ridges in the surrounding area. 

The distribution of the wreckage indicated that the aircraft initially struck trees in an upright orientation, with an approximate 20° left angle of bank, and a nose‑down attitude at about a 30° angle of impact. The initial tree strike resulted in the left wingtip and aileron separating from the aircraft. The aircraft then impacted the ground on its left side and continued through the bush in a southerly direction, coming to rest about 45 m from the initial point of impact, where it was consumed by a post‑impact fire.

The wreckage trail consisted of a number of felled trees and aircraft components, including the nose gear assembly, left main gear and fairing, left door, section of the left wing flap, windscreen and sections of the lower engine cowling and lower engine components. There was no indication of fire in the wreckage trail or detached aircraft components (Figure 4).

Figure 4: Wreckage trail looking north towards the impact area

Source: ATSB

The propeller had separated from the engine and was located towards the rear of the wreckage and the engine was upside down and detached from its mounts. Although the left wing was significantly affected by fire, the wing spar was still distinguishable. The right wing was relatively intact as was the magnetometer boom, albeit damaged by fire (Figure 5).

Figure 5: Main wreckage

Source: ATSB

Wreckage examination

Although post‑impact fire damage precluded examination of a significant proportion of the aircraft, inspection of the site and wreckage found:

• no evidence of any pre‑existing structural, mechanical or flight control defects that would have prevented normal operation
• the wing flaps were in the fully up (retracted) position
• a small, yet intense fire zone indicative of a significant amount of fuel, with ignition occurring from the left‑wing integral fuel tank rupturing during the accident sequence
• the fuel selector was in the ‘BOTH’ orientation
• damage to the propeller indicated that the engine was producing power at the time of the impact.

Extensive fire damage to all instruments and avionics resulted in no useful switch position information. The windscreen, located part way along the debris trail, did not exhibit signs of birdstrike, nor were feather or bird remains identified in the area. In addition, the morning pilot reported that, while they had observed bird activity on survey flights, none had been sighted that day. 

The reason for the fuel imbalance noted by the morning pilot could not be determined from the wreckage examination.

Medical and pathological information

Post-mortem examination

A post-mortem examination of the pilot was undertaken by a qualified pathologist on behalf of the Western Australia Coroner. The pathologist determined that the pilot’s cause of death was a result of a combination of traumatic injuries (both soft tissue injuries and multiple fractures) and the effects of fire from significant thermal injury and smoke inhalation. There were multiple fractures to the nasal bones but none to the skull or pelvis.

The pathologist assessed that the traumatic injuries sustained were potentially survivable with immediate medical assistance, but those injuries were compounded by the thermal injury and smoke inhalation. The ATSB’s aviation medical specialist also advised that the impact injuries were likely not fatal, however, they would have been severely incapacitating. The extensive thermal injury and, in particular, the smoke inhalation was likely to have rendered the pilot unconscious within minutes. They also stated that immediate intervention would have been required but the sustained thermal injuries were likely not survivable.

The post-mortem report indicated that the pilot did not have any significant natural disease. Further, toxicological analysis did not detect the presence of alcohol or common drugs and carbon monoxide[8] levels were not significantly raised (at less than 5% saturation). 

The pilot was reported by their partner to be fit and healthy with no known illnesses.

Survival aspects 

Impact protection

Due to extensive fire damage to the fuselage, there was limited evidence available about the survivable space/intrusions, or seat and seatbelt condition. Therefore, the ATSB was not able to determine survivability with regard to the cabin area. The left (pilot) door indicated an intrusion/compression and the left main landing gear leg was detached, consistent with high impact forces and the injuries sustained to the pilot.

Post-impact fire

Metal fuel tanks are prone to rupturing during an accident impact, allowing fuel to escape and increasing the risk of a post‑impact fire. To improve crashworthiness, the addition of fuel bladders and fuel cells that have been constructed of flexible materials have proven less prone to rupturing during an impact. They are able to withstand greater deformation and puncture less readily and are less likely to expand or tear to form a larger opening from which fuel can escape. Such systems may provide occupants with more time to egress the aircraft and/or reduce the risk of any fire‑related injury.

ATSB investigation report AO-2021-052, extensively discussed post‑impact fire safety and referenced studies by the United States National Transportation Safety Board and Transport Safety Board of Canada. Those studies concluded that post‑impact fire had been shown to contribute significantly to injuries and fatalities in accidents that were otherwise potentially survivable. A potentially survivable accident is one in which the impact forces are within the limits of occupant tolerance, the aircraft structure preserves the required survival space, and the occupant restraint is adequate. 

As a result of investigation AO‑2021‑052, the ATSB identified that the aircraft (an Air Tractor AT‑400) was not required to be fitted with a crash‑resistant fuel system under United States Federal Aviation Regulations. A safety issue was raised and the ATSB recommended that the United States Federal Aviation Administration take action to address certification requirements for crash‑resistant fuel systems for fixed‑wing aircraft, in an effort to reduce the risk of post‑impact fire. At the time of writing this report, the ATSB recommendation remained open and the Federal Aviation Administration had advised that the results of a study into post‑crash fire accidents was being reviewed to determine their next action (AO‑2021‑052‑SI‑01).

Protective clothing and helmets

For the accident flight, the pilot was reported to be wearing a t‑shirt, shorts and trainer type shoes and was not using a helmet. The operator did not require its pilots to wear protective clothing or helmets, nor were they required to do so by regulations. The CP explained that this decision took into account temperature, fatigue and pilot comfort balanced against mitigating the potential risks. In addition to comfort and fatigue factors, the bulk of a helmet may not be suitable to the smaller cockpit of the aircraft. However, the CP stated that no formal risk assessment had been completed to support this decision. The operator reported that it issued each pilot (including the accident pilot) with company polo shirts made of 100% cotton as a measure of fire protection and the use of other protective clothing and helmets was left to individual pilots’ discretion. The CP indicated that some of their pilots did wear such items. It could not be determined if the accident pilot was wearing the company polo shirt. The ATSB noted that the operator’s job safety analysis (refer to section titled Job safety analysis) included consideration of protective equipment and clothing as methods of reducing risk factors.

The ATSB’s aviation medical specialist advised that if protective clothing and an appropriate helmet was worn, in most general circumstances, this would have reduced the severity of injury in an accident. However, they were unable to comment on the effectiveness of these items for this accident and noted that protective clothing and a helmet would not have prevented any smoke inhalation injury.

The International Airborne Geophysics Safety Association (IAGSA – refer to section titled International Airborne Geophysics Safety Association) recommended that appropriate clothing should be worn by all flight crew involved in geophysical surveys to minimise the immediate risk of fire in the event of an accident and for protection from exposure in a survival situation. These include:

• cotton undergarments covered by long trousers and long‑sleeved shirt or an appropriate flying suit
• closed shoes
• have gloves available at all times
• layers of clothing appropriate for the conditions
• cold weather clothing should include felt lined boots, down parka with attached hood and large mittens.

IAGSA also recommended that for fixed‑wing operations, each individual operator should determine the appropriateness of the use of an industry approved helmet. A case by case, risk assessed approach should be adopted, taking into account the relevant variables for each specific survey task.

The Flight Safety Foundation’s Basic Aviation Risk Standard[9] is a set of risk‑based aviation industry standards. The standard covers a wide variety of aviation applications of which airborne geophysical survey operations were included. The standard implementation guidelines for survey operations also recommended appropriate clothing for crew such as non‑synthetic long trousers and pants or flying suit. It also recommended that helmets should be worn when operating below 500 ft AGL unless a risk assessment stated otherwise. 

Flight following

Satellite tracking

Operator requirements

The operator had implemented flight following through use of the Spidertracks satellite tracking system installed on each of its aircraft. The company operations manual stated:

In addition to the required safety equipment the Company equips all aircraft with a real-time satellite monitoring system with a refresh rate of at least every 5 minutes and automatic alerting (to company mobile phone and email) in the event of an emergency.

Should the satellite monitoring system alert be inadvertently activated by the pilot an “ops normal” call should be made to the company as soon as practicable.

The ground operator assigned to each survey job was the primary person responsible for flight following. They were to monitor the aircraft’s location via the tracking system and initiate an emergency response, if required.

The morning pilot could not recall any specific pre‑flight requirements for the Spidertracks device and another company pilot reported that there were not any checks required, the device turned on once the aircraft’s electrical system was on.

The operations manual did not include flight following as a specific duty for the ground operator or any other staff member. Further, the manual did not detail procedures for the conduct of flight following, nor were there procedures or guidance to confirm that the tracking system was correctly configured and operating as expected prior to flight.

Automatic watch function

The operator was surprised that a Spidertracks automatic alert was not received during the accident and advised that, on a number of occasions, their satellite tracking had experienced dropouts. On some of those occasions, contact was made with the pilot and a system reset restored normal function. On other occasions, when contact with the pilot was not possible, the aircraft returned by the nominated estimated time of arrival (ETA). The operator had not contacted Spidertracks about the dropouts or conducted any other troubleshooting.

Spidertracks advised the ATSB that the automatic watch function on the aircraft’s device had not been activated on the accident flight, nor was it active for earlier flights on 2 and 3 March 2022. They further advised that there was no indication of any service‑related issues, confirming that up to the loss of data, the aircraft’s device was operating as expected. Diagnostic logs for the device were not available due to this data only being transmitted via mobile phone networks. Spidertracks confirmed that the length of time with no transmissions received, or a data loss or delay was not typical and could be indicative of a power or device failure, transmission interference or installation issue. Spidertracks found no recorded issues with the satellite service or their cloud platform at the time of, or leading up to, the accident.

Emergency response plan

The operator had a phased emergency response plan, predicated on an elapsed time since the aircraft’s ETA. Each phase was commensurate with an escalating level of concern. Satellite tracking was referred to in the plan, within the section Phase 1 - Uncertainty. Phase 1 commenced 0‍–‍15 minutes after the aircraft’s ETA had expired. The plan directed a check of the satellite tracking and if there was an abnormal or no indication in the system, the next step was to attempt contact with the crew. If contact with the crew was not possible and overall operations were assessed as not normal, the plan directed that the operations manager, as primary contact, to be notified, then the chief executive officer and CP as alternates.

The plan did not elaborate any further on required actions for an abnormal or no indication in the system prior to advancing to the next step, which was Phase 2 - Alert. Phase 2 commenced 15‍–‍30 minutes after the ETA had expired and directed the primary or alternate contacts to establish the final status of the aircraft via the tracking system. It included a note that, if there was no contact with company operations then the ground operator was to contact the Australian Maritime Safety Authority’s Joint Rescue Coordination Centre or local search and rescue services direct. Commencement of Phase 3 - Distress was at 30 minutes after ETA had expired or whenever the aircraft was confirmed as missing. 

The operator commenced phase 3 at 1700, 30 minutes after the ETA for VH‑JVR had expired and then contacted the Joint Rescue Coordination Centre.

Emergency beacons

Emergency locator transmitter 

The company operations manual stated that all company aircraft were to be fitted with an approved ELT or a portable ELT if the fixed device was inoperative or otherwise not serviceable. The operator was not able to determine why VH‑JVR was not fitted with an ELT.

The ATSB research report (AR‑2012‑128) discussed the potential safety benefits of an approved, fitted ELT, which were designed to automatically activate following an impact normally associated with a collision. While the research noted some limitations with the effectiveness of ELTs, the fitment of a crash‑activated ELT greatly increases the early notification for search and rescue efforts and arrival of potentially life‑saving medical treatment especially when occupants or crew are incapacitated.

Personal locator beacon 

The CP and morning pilot stated that a personal locator beacon (PLB) was carried in the aircraft as part of a survival kit, which was secured to the passenger seat. The PLB was routinely carried in the aircraft, and not as an alternative to an ELT but as an additional item. In accordance with regulations, the PLB was classed as a survival ELT, and an alternative to a fixed ELT. However, the PLB was not identified in the wreckage and was likely consumed by fire. 

The ATSB research report (AR‑2012‑128) suggested that carrying a PLB will most likely only be beneficial to safety if it is carried on the person, rather than being fixed or stowed elsewhere in the aircraft. The CASR Part 91 MOS stated that a survival ELT must be carried either on the person of a crew member, in or adjacent to a life raft, or adjacent to an emergency exit. 

Operational information

Airborne geophysical survey flights

Airborne geophysical survey flights are conducted by a variety of rotary and fixed‑wing aircraft which have been specifically modified and equipped with geophysical sensors. Survey flights were normally flown below 500 ft AGL over the desired area via a pre‑determined pattern and at heights designed to maximise the quality of the data captured. The data provides a detailed below ground composition of the surveyed area, primarily to inform mining and resource industry activities.

Requirement for CASA low‑level rating 

Operations requiring flight below 500 ft AGL, such as geophysical surveys, required a pilot to hold a CASA Part 61 low‑level rating. A low‑level rating is specific to various types of flying operations (such as aerial survey, firefighting or agricultural), however, the training and testing is not specific to any one type of operation. To obtain a low‑level rating a pilot must demonstrate competency in certain operational techniques, which included, but were not limited to, steep, maximum rate and minimum radius turns, procedure turns, recovery from approach to stalls (level and turning). In addition to holding a low‑level rating, MAGSPEC Aviation required prospective pilots to have a minimum of 500 hours as pilot in command. MAGSPEC Aviation then provides training specific to its operational requirements.

International Airborne Geophysics Safety Association (IAGSA)

The IAGSA is an international industry association comprised of airborne geophysical survey organisations with an overall objective to promote and enhance safety in the airborne geophysics survey industry. IAGSA publishes a safety manual for its member organisations, which details its standards and recommended safety practices. 

IAGSA is a non-regulatory body and holds no authority to compel its members to follow its standards and recommended practices, which are not a replacement for the regulatory requirements that each individual organisation may operate under. However, members have agreed under the terms of membership to follow those standards and practices where they are more stringent or not covered by regulations, except where they have filed a notification of difference. Members are also required to complete an annual self‑audit. At the time of the accident, MAGSPEC Aviation’s most recent self‑audit outlined a number of differences to IAGSA standards and practices. Although IAGSA had requested it, it had not received a formal notification of differences from the operator.

Survey height

The operator was issued a CASA instrument in 2017, which allowed it to conduct operations at a height lower than that permitted by Civil Aviation Regulation (CAR) 157.[10] The instrument was last renewed in 2021 and was valid until 2024. 

The instrument did not specify the lowest height that could be flown. The CP stated that survey flights would often be flown at 30 m (100 ft) AGL, although a standard or minimum height was not documented in the operations manual. Rather, the survey height would be requested by the client. 

The CP explained that the requested survey height was assessed during the planning stage, through a review of maps of the survey area and the conduct of a reconnaissance flight. A detailed guide on how to conduct a reconnaissance flight was included as an appendix to the operations manual. This process would confirm if the survey could be flown at the requested height. The operator stated that, on numerous occasions this process resulted in the survey being flown at heights higher than requested.

The client for the accident survey specified a height of 25 m (82 ft) AGL.

The IAGSA safety manual acknowledged that there were increased risks associated with low flying and that operating at such heights can ‘aggravate the consequences of mechanical malfunctions or human error’. When discussing minimum safe survey heights, and while recognising that lower heights may improve the quality of survey data, they noted many differences of opinion among its members. 

Having a predetermined height had been debated among the members, however, they concluded that ‘no single universal “minimum safe survey height” can be designated given the wide variety of survey conditions and aircraft characteristics’. As such, IAGSA indicated that the safety issue was not necessarily the survey height, but more importantly, could the survey be safely flown at the requested survey height. Consequently, IAGSA recommended that:

Clients specify the maximum clearance height possible, consistent with the objectives of the survey to be flown and that operators, prior to commencing a survey, conduct a detailed risk analysis in accordance with an internationally recognized procedure considering, but not limited to, the following factors and Appendix IV of this manual:

•  terrain relief, elevation & vegetation canopy thickness

•  aircraft type

•  aircrew flight and duty times

•  prevailing weather conditions

•  anticipated density altitude

•  pilot experience and recency

•  planned flight speed.

Survey pattern

From the recorded data, the accident survey flight was flown in a back‑to‑back pattern, which was a series of consecutive parallel lines followed by a procedure turn used to establish the aircraft onto the next line in the reciprocal direction (Figure 6). This was the routine pattern used by the operator in its geophysical surveys. The client had specified 25 m spacing on east‑west survey lines.

The CP explained that procedure turns consisted of an initial climbing turn to establish the aircraft at about 300 ft AGL and about 400‍–‍500 m lateral offset from the next line. After this, a level turn would be commenced (into wind) at a 45‍–‍60° angle of bank (referred to as a steep turn) to intercept the next line. Descent to the survey height commenced once the aircraft wings were established straight-and-level.

Figure 6: Back-to-back pattern and procedure turns

Source: Aerial Application Association of Australia, annotated by the ATSB

IAGSA highlighted the risks associated with turns at low level:

Turns at low level present a considerable hazard, particularly if the terrain presents visual illusions; the aircraft descends in the turn, airspeed is low, or the angle of bank is steep. An excessive angle of bank, often resulting from close line spacing or drifting in strong crosswind conditions, is insidious as the stall speed of the aircraft increases with the angle of bank (assuming a level turn) whilst at the same time the aircraft’s speed is reduced from increased drag.

During straight and level flight there may be a significant margin above the stall speed, however in a steep turn the stall speed may be reached quickly with little warning and a stall in the turn at low level will likely result in a fatal accident.

For manoeuvring at low level, IAGSA recommended:

All turns at low level should be limited to a maximum angle of bank of 30 degrees and be done at a constant altitude. No climbs or descents should be carried out during the turn. If the terrain dictates that a climb is necessary the aircraft should be climbed to the required height prior to commencing the turn and any descent back to survey height should only be done after established in a wings level attitude.

The CP explained that the back‑to‑back pattern with procedure turns was the most efficient method and enabled the capture of higher quality survey data. They reported that flying consecutive lines was less workload intensive for their pilots, especially regarding obstacle hazard avoidance. This allowed a pilot to deal with a particular hazard for a short period as they moved away from it. 

Operations manual – Special operations 

Volume 2 of the operations manual, valid at the time of the accident, discussed aircraft operations and included a section titled Part 2D Special Operations specific to survey operations. It included sub sections on low flying, survey tolerances and safety considerations during surveys. However, there was limited detail with regard to the process or procedures for the conduct of low‑level survey flights. 

As previously discussed, a standard or minimum survey height was not included in the operations manual. 

In Part 2D1.1 Low Flying, height was discussed in terms of a minimum height when overflying occupied structures, vehicles or livestock but it did not include reference to other obstacles such as terrain, vegetation canopy or masts/antennas.

Part 2D1.4 Survey Tolerances stated:

Track, height and groundspeed tolerances for the survey will be established by the client and should be adhered to as closely as possible. Significant deviations will require the line to be re-flown.

Client established survey tolerances, although important to data accuracy, remain secondary to safety and pilots should disregard them as necessary to ensure the safety of aircraft, personnel, equipment, and environment.

Part 2D1.5 Safety considerations during survey, required a minimum survey speed to be established for each aircraft type operated by MAGSPEC Aviation. This was to be the greater of the 130% of clean stall speed (wing flaps up and landing gear up if retractable), 110% of the best single‑engine climb speed or 110% of the take‑off safety speed.

Although procedure turns and a back‑to‑back survey pattern were taught to, and routinely flown by all company pilots, the special operations section did not refer to these manoeuvres. The CP stated that there was no other reference document that outlined how the operator expected its pilots to conduct the procedure turns, nor were there any documented limitations such as a maximum angle of bank or minimum height AGL prior to commencing the procedure turn.

Aerodynamic stall 

A wing generates lift as a result of the pressure differential created by airflow over the wing’s surface. The angle between the incoming or relative air flow and wing chord is known as the angle of attack (AoA). As the AoA increases, lift increases up to a certain angle, known as the critical AoA. At this point, the airflow over the upper surface of the wing becomes separated. This condition is referred to as an aerodynamic stall (or simply a stall) and results in a significant loss of lift and an increase in drag. Due to the sudden reduction in lift from the wing and rearward movement of the centre of lift, typically an uncommanded aircraft nose‑down pitch results.

A loss of altitude also occurs during the recovery from a stall and it is possible to stall with insufficient height above the ground to recover. The pilot’s operating handbook (POH) for the U206G stated that the maximum altitude loss during a stall recovery may be as much as 240 ft in power off conditions and straight and level flight. The U206G has a stall warning vane[11] and warning horn to alert the pilot of an impending stall.

Most general aviation aircraft typically have a critical AoA of around 16°. This critical AoA can be exceeded at any airspeed, any (pitch) attitude and any power setting. However, as most small aircraft are not fitted with an AoA indicator, the AoA at which the stall occurs may be referenced to an airspeed. 

When banking or turning an aircraft, it is necessary to increase the amount of lift generated to ensure that the aircraft does not descend. This increases the AoA resulting in lift and drag greater than normal straight and level flight. This increases the load factor on the aircraft above 1 G. As the angle of bank increases, the lift required to maintain a constant altitude also increases, requiring the pilot to apply back pressure on the control column. The effect is, as the angle of bank and load factor increases, the stall speed increases. At 45° angle of bank, the load factor is 1.41. This results in an almost 19% increase in the wings level stall speed. At 60° angle of bank, the load factor is 2, resulting in an increase in stall speed of 41%. 

The U206G POH provided the stall speeds at maximum weight with power off, flaps up, various angles of bank (up to the POH limit of 60°) and centre of gravity (CoG) positions (Table 2Table 2).

Table 2: U206G stall speeds (extracted from the POH)

Recovery from a stall requires reducing the AoA by moving the control column forward, which normally means lowering the aircraft nose (pitching down).

Organisational and management information

MAGSPEC Aviation

MAGSPEC Aviation Pty Ltd commenced operations in 2017 to provide airborne geophysical survey services across Australia. It operated a fleet of 2 Cessna 206 and 2 Cessna 210 aircraft. At the time of the accident, they operated under a CASR Part 138 (aerial work) air operator’s certificate. Part 138 came into effect on 2 December 2021. Since commencing low‑level survey operations, the operator had experienced the following occurrences:

• In 2018, the engine of a Cessna 210 failed necessitating a forced landing. The pilot was uninjured, however, the aircraft was substantially damaged. It was identified that sufficient fuel had not been transferred from the aircraft’s tip tanks to the main fuel tanks. The operator undertook action to review training and to reinforce fuel management procedures.
• In 2019, a Cessna 210 struck a powerline and the aircraft sustained minor damage. The aircraft was safely flown back to its departure point. The operator reported that the aircraft had been flown on north‑south lines due to the sun’s position before transitioning to east‑west lines after the sun was no longer a factor. Following this, the pilot had flown 50 m from the powerline before making contact on the reciprocal heading. Subsequently, the operator incorporated a national database of powerlines into its pre‑survey assessment process.
• In 2020, the engine of a Cessna U206G lost partial power and the pilot conducted a forced landing, resulting in substantial damage to the aircraft. The pilot was uninjured. Fuel starvation was determined as the reason for the engine power loss. It was identified that the aircraft had been operated with only one fuel tank selected instead of both. The operator updated checklists to incorporate the requirement for fuel tank selection to BOTH and amended the survey data acquisition system to provide periodic fuel check messages as a reminder to pilots.

Following the 2018 occurrence, the ATSB completed an occurrence brief (AB-2018-058). This was a short summary report and not an investigation, to allow for greater industry awareness of potential safety issues and possible safety actions. The ATSB did not investigate the other occurrences. CASA conducted 2 surveillance events following the 2018 occurrence. In response to the draft report, on 28 January 2025, the operator reported that they had requested assistance from CASA after the other occurrences but reported no assistance was provided. The ATSB reviewed CASA records, which indicated no action had been taken following the 2019 or 2020 occurrences (refer to section titled Regulatory oversight activity. 

Safety management system

Managing safety

According to the International Civil Aviation Organization (ICAO, 2018), a safety management system is a systematic approach to managing safety that seeks to proactively mitigate risks before they result in an accident or incident. This includes defining the necessary organisational structures, accountabilities, responsibilities, policies and procedures.

At the time of the accident, there was no CASA regulatory requirement for the operator to have a safety management system. However, the operator had implemented a Health, Safety and Environmental Management System (HSEMS), for the purpose of describing:

…the process by which MAGSPEC aviation manages risks has been developed to meet the Civil Aviation Safety Authority SMS requirements in addition to providing more generalized guidance on the management of risk within the organisation.

Specifically, one of the operator’s policy commitments was to:

…minimize the risks associated with operational activity to a point that is as low as reasonably practicable/achievable…

While the HSEMS was submitted to CASA, a review of CASA records found that it had not been assessed, nor was there a requirement to do so.

Safety risk management

Risk management is a key component of safety management and includes hazard identification, safety risk assessment, safety risk mitigation and risk acceptance. It is an ongoing process as the aviation system is constantly changing, with new hazards introduced, and some hazards and associated risks changing over time (ICAO, 2018).

CASR Part 138 required an operator conducting aerial work to undertake risk assessments of its operations. This was a new requirement introduced with Part 138. The Part 138 MOS detailed a layered approach to risk assessment and that before conducting an operation:

(a) the operator’s operations manual must contain: 

(i) pre-operational risk procedures [refer below] for risk assessments and mitigation processes applicable to the operation; and 

(ii) procedures for post-flight risk review; and 

(b) the operator must have a flight risk management plan based on a pre-operational risk assessment in accordance with the procedures; and 

(c) the operator must ensure that the operator and each crew member is satisfied, in a pre-flight risk review, that the flight risk management plan will eliminate, reduce or mitigate risks and hazards to the extent that it is safe to conduct, and continue, the operation without unacceptable risk to the crew members, any aerial work passengers, the aircraft or any other person or property.

The MOS further specified what was to be included in an operator’s pre‑operational risk procedures:

a. processes for identifying, reporting and recording hazards; 

b. processes for analysing identified hazards and assessing the risks they may pose, including for pre-flight, in-flight and post-flight stages of operations; 

c. processes to mitigate the risks or control the risks, including processes for the incorporation of risk controls into standard operating procedures; 

d. the creation and management of: 

i. a risk register; and 

ii. records of dedicated risk assessments performed to address each type aerial work operation that is to be conducted, including details of the risk assessors; 

e. procedures to ensure that the pilot in command and the other crew members are familiar with the pre-operational risk assessment and the associated standard operating procedures (SOP); 

f. in-flight procedures for the pilot in command and the other crew members to consider and manage the risks associated with aerial work operations.

The operator’s HSEMS stated that safety risk management begins with hazard identification and then assessing the risks associated with the hazard in terms of likelihood and severity. The manual further stated that, once the level of risk was identified, appropriate remedial or mitigation measures could be implemented to reduce the risk to as low as reasonably practicable. The risk management process detailed in the HSEMS followed a 5‑stage process:

• Stage 1 - Identify the hazard and associated risks
• Stage 2 - Assess the risk in regard to severity and likelihood
• Stage 3 - Evaluate risk tolerability
• Stage 4 - Treat/mitigate the risk
• Stage 5 - Monitoring

Stage 1 of the process listed sources for hazard identification and stated that the safety manager was to use the Donesafe[12] system to manage and record these hazards. It also noted that:

Due to the varying nature of MAGSPEC Aviation’s operating environment a separate Operational Job Safety Analysis (JSA) (see appendix 3) was undertaken by the Chief Pilot or his designee to assess site-specific risks prior to each job provide an overall risk rating for the job. 

Where non-site-specific items are identified as part the JSA or field crew safety meeting these will be reported to the HSEMS system via the DONESAFE “Hazards” report tab. 

Each survey task, including the risks associated with that task, were assessed via the job safety analysis (JSA). Any risks identified in the JSA that were not specific to the survey tasking location were entered into the Donesafe system. The risks specific to the location were not captured in the system.  

Pre-operational risk assessment

One of the key requirements for managing risk was that an operator should undertake an overarching assessment (pre‑operational risk assessment) to consider and evaluate the risks associated with its proposed operations, in this case, low‑level geophysical survey. This assessment recognised the underlying principles of CASR Part 138, where the risks and hazards associated with a type of aerial work operation are common to that type of operation. The matters to be considered in the assessment included, but were not limited to the (CASA, 2023): 

− nature of the intended operation and its particular characteristics 

− location (if known) of the intended operation and its particular characteristics 

− aircraft to be used in the intended operation and their performance profile and impacts of serviceability status 

− qualifications and experience of the FCMs [flight crew members] and support personnel to be used in the intended operation 

− generic or known hazards particular to the type of aerial work operation, external to the aircraft, that may be met during the operation.

CASA advisory circular 138‑05 v2.1 Aerial work risk management, stated that an operator should use data from the risk register and dedicated risk assessments to inform the pre‑operational risk assessment. Once populated, the assessment should then be updated over time and from operational experience, to incorporate lessons learnt from previous operations. Further, to ensure it is readily available to all crew members, it should form part of the company’s operations manual.

The ATSB’s review of the HSEMS and operations manual did not identify any requirement for a pre‑operational risk assessment to be completed.The CP also confirmed that, at the time of the accident, such an assessment had not been conducted.

Risk register

Safety risk management activities should be documented, including any assumptions underlying a risk assessment, decisions made, and risk controls implemented. A risk register could be used to ensure identified hazards and risks that emerged during planning or day‑to‑day operations were tracked and mitigated as part of formal risk management processes. An operator’s risk register can also be incorporated into the pre‑operational risk assessment. The register could include the hazard, potential likelihood and consequences, assessment of the associated risks, when or where it applied, and any controls put in place to mitigate the risk. Notably, (ICAO, 2018):

Maintaining a register of identified hazards minimizes the likelihood that the organization will lose sight of its known hazards. When hazards are identified, they can be compared with the known hazards in the register to see if the hazard has already been registered, and what action(s) were taken to mitigate it. 

The CP reported that a risk register was not maintained for the company’s operations. Although it was noted that the operator did retain a fatigue risk register. 

Flight risk management plan

The results of the pre‑operational risk assessment were to be considered when preparing a flight risk management plan, which was specific to an individual flight or task within the type of operation. The plan should outline the specific mitigators or risk controls that were to be used during the flights. The flight crew should also have sufficient time to review and confirm the plan prior to the commencement of the operation. 

Job safety analysis

Components 

As required by the company operation’s manual, the JSA was the documented risk management process designed to address the safety concerns with each project the operator conducted, that is, for each specific survey task. The ATSB’s interpretation of the Part 138 risk assessment requirements was that the JSA was equivalent to the flight risk management plan, as discussed above. The JSA consisted of 5 parts:

• Part A - Pre-survey risk assessment: This assessment was to be completed by the operations manager at the time a tasking was quoted and included details on the activity, hazards, hazard effects, initial risk score, risk mitigators, residual risk, and a final risk score. This used a pre‑populated risk matrix with 14 hazard areas, each of which were assigned a descriptor and risk score of 1 (negligible) to 5 (unacceptable). The total risk score determined if any further action was required, such as a need for additional risk controls or stopping the tasking until the risk was reduced.
• Part B - Operational job safety analysis. This was to be completed by the CP or other suitable person prior to commencing the survey task. This considered any operational limitations relating to aircraft performance, obstacles and human performance, whether any hazards affected the safety or technical performance of the survey, and if any changes were required. The risk level for the task was assessed using a pre-populated matrix with 27 hazards, but with instructions to add more as appropriate. The final risk level determined if the survey could proceed as planned (low risk), or if the survey could proceed with approval from the CP and amendments to the plan or additional risk mitigators (medium risk), or if the survey was not to proceed as currently planned (high risk).
• Part C - Field crew safety meeting: The meeting was to be completed by operational personnel at the survey site, prior to commencing survey operations and every crew change. This section was a yes/no answer sheet covering a range of operational areas designed to assess any additional hazards and risks not identified in Parts A and B. At the direction of the CP, a reconnaissance flight could also be performed to assess the survey area for any additional risks or hazards not already identified in the original JSA.
• Part D - Post-survey field crew meeting: This meeting allowed the operator to better understand any issues faced on the job and if anything needed to be accounted for, either at that specific location or for an ongoing basis.
• Part E - Emergency response plan: This plan was to be reviewed during the field crew safety meeting and crew members were to ensure that the contact and procedure details were correct.

Neither Part A nor Part B referred to consideration of previous JSAs for any applicable risk information that may be relevant to the current JSA.

Survey task assessment

Parts A-C of the JSA completed for the accident flight survey task are discussed in the following paragraphs.

Part A was completed by the operations manager and listed hazards including the 25 m survey height, which was assessed with the highest risk score of 5. It did include mitigating factors of carrying a portable personal ELB and portable GPS, conducting operations with satellite flight following and a comprehensive pilot briefing including maps. 

Several elevated risk areas were identified on the matrix, such as operations below 100 ft AGL and operating in hot conditions between 35‍–‍40 °C. Overall, the initial risk rating for the survey task was determined to be low, based on a score of 36 (Figure 7)Figure 8. 

Figure 7: Part A – Initial pre-survey risk assessment for the accident task

Source: MAGSPEC Aviation, annotated by the ATSB

The CP and operations manager approved Part B of the JSA, identifying that the survey height and antenna/masts, and the survey location with regard to other aircraft activity were concerns. When considering if there were any hazards that would affect safety or the technical performance of the survey, the survey height of 25 m was noted, and the possibility of trees, powerlines and masts in the area were low still but still a risk. As such, it was determined that a detailed reconnaissance flight was to be conducted. 

The final risk level was assessed as low, with 4 hazards identified (Figure 8Figure 9). The ATSB noted that the hazard of ‘terrain clearance less than 30 metres’ had not been ticked. However, if it had been selected, the final risk level would have remained at low.

Figure 8: Part B – Hazards risk matrix for the accident tasking

Source: MAGSPEC Aviation

Part C had been originally completed by another pilot and the ground operator. That pilot had conducted a reconnaissance flight of the survey area and signed part C noting that no additional risks had been identified. That pilot was subsequently assigned to another task. 

The ATSB noted that the survey height was referred to in the question, Can the job be flown at the suggested survey height? and this was answered as yes with no amplifying comments.

The day prior to the accident, the ground operator met with the accident and morning pilot to conduct another field crew safety meeting. They discussed the JSA, and the ground operator reported that they advised the pilots about some taller trees in the area, which had been identified in the reconnaissance flight (but not noted in Part C).

Text messages between the accident pilot and CP showed that conducting another reconnaissance flight was discussed. The CP suggested that another could be done if the pilot felt it was required but there was no direction from the CP to do so. The morning pilot completed another reconnaissance of the survey area prior to commencing their survey. 

In reference to the utility and sufficiency of the JSA, the operator advised on 28 January 2025 in response to the draft report that they considered the JSA to be their risk assessment process and was a combined risk register, pre‑operational risk assessment, flight risk management plan and record of the crew meeting. The operator further advised that the JSA was reviewed by CASA during the transition to Part 138 and: 

This risk assessment was approved by CASA during the 2nd December 2021, Part 91 / 138 AWK [aerial work] changes. It has been accepted and approved by multiple third-parties, including those that represent BARS [Flight Safety Foundation’s Basic Aviation Risk Standard].

The ATSB sought clarification from CASA to determine if the JSA met the requirement of Part 138 and whether it had been approved by CASA. On 1 May 2025, CASA advised: 

The JSA as described in the report does not meet the requirement of a pre-operational risk assessment.

The reasons for this advice are:

• The ATSB report outlined that Part A of the JSA had a pre-populated risk matrix with 14 hazard areas and Part B of the JAS had a pre‑populated matrix with 27 hazards and instructions to add more as necessary.

• CASA’s AC 138-05 identifies how a risk register is a critical component to the creation of a pre‑operational risk assessment and CASA notes that the ATSB report mentions that the operator’s Chief Pilot (CP) “reported that a risk register was not maintained for the company’s operations” which supports that a pre-operational risk assessment was not produced.

• CASA agrees with the ATSB that this activity is not specifically considering or evaluating the risks associated with the type of aerial work operation to be conducted, i.e. the JSA process is basically done on a per task basis, which is not the same as the pre-operational risk assessment as the pre‑operational risk assessment is intended to be an enduring document that is regularly updated from risk register updates and post-flight risk reviews (see the first sentence of CASA AC 138‑05 paragraph 4.2.5).

• Effectively, the JSA Part A is potentially covering elements of risk assessments that would support the updating of the pre-operational risk assessment but is not creating the pre-operational risk assessment itself.

• CASA further advised:

• The accident occurred 3 months after the commencement of the new flight operations regulations, of which Part 138 of CASR and its supporting Manual of Standards was one element.

• Under the transitional rules in Subpart 202.EAA of CASR, holders of AOCs authorising aerial work under the pre-2 December 2021 paragraph 206(1)(a) of CAR, where the AOC was in force immediately before 2 December 2021, had these AOCs recognised as legally being an aerial work certificate and such operators were required to ensure their operations manuals complied with Part 138 of CASR and contained all necessary content to enable that compliance.

• As the operator was the holder of an AOC authorising aerial work under the pre-2 December 2021 rules, the content of their operations manual would have been approved by CASA as part of them holding that AOC. Compliance with the new flight operations regulations for all such operators would be reviewed at the next appropriate CASA oversight event.

Regulatory oversight activity

Regulatory framework

CASA was responsible, under the provisions of Section 9 of the Civil Aviation Act 1988, for the safety regulation of civil aviation in Australia and of Australian aircraft outside of Australia. Section 9(1) stated the means of conducting the regulation included:

(c) developing and promulgating appropriate, clear and concise aviation safety standards;

(d) developing effective enforcement strategies to secure compliance with aviation safety standards…

(e) issuing certificates, licences, registrations and permits;

(f) conducting comprehensive aviation industry surveillance, including assessment of safety‑related decisions taken by industry management at all levels for their impact on aviation safety…

The 2 primary means of oversighting a specific operator’s aviation activities were:

• assessing applications for the issue of, or variations to its air operator’s certificate (AOC) and associated approvals (including approvals of key personnel)
• conducting surveillance of its activities. 

CASA was required by Section 28 of the Civil Aviation Act 1988 to satisfy itself about various matters when processing an application for the issue of, or variation to, an AOC. The matters included whether the organisation was suitable and whether it had suitable procedures and practices to ensure that AOC operations were conducted safely.

CASA provided records related to their assessment of MAGSPEC Aviation’s initial AOC and low flying applications. 

Initial issue of air operator’s certificate 

The CASA entry control process involved assessing an application for the issue of a new AOC or a variation to an existing AOC. The worksheet used by CASA for an AOC assessment was intended to be used in conjunction with the AOC Process Manual, AOC Handbook, other relevant technical assessor handbooks and applicable legislation. This worksheet contained the criteria required for an assessor to undertake a technical assessment. It focused on generic regulatory requirements applicable to most operators and there was no specific criteria that referred to assessing an operator’s primary activity, in this case, low‑level survey operations. Although the assessment process confirmed that the operator had processes and procedures to support its operations, there was no evidence that these were examined in any detail for their suitability for the proposed operations. 

However, the CASA officer processing the AOC application acknowledged that, while it was a new operation, the organisation included personnel from a previous operator, and that these personnel had experience and exposure to low‑level survey operations. The officer further stated that this experience was evident during the assessment, interview, and inspection phases of the assessment. 

As part of the AOC application, CASA was to also approve the appointment of the CP. The CP’s records included an assessment paper, interview record, and notes from an assessment flight. The assessment focused on the CP’s ability to manage the regulatory requirements of an AOC holder, yet did not indicate how the operator would conduct its low‑level survey operations. The assessment flight did not include any low‑level flying as CASA did not permit its officers to undertake low flying. 

The AOC was issued to the operator on 3 October 2017.

In December 2021, CASA amended its AOC entry control procedures to include more emphasis on assessing the proposed primary activity. A specific worksheet was introduced for assessing a Part 138 application and included reviewing the processes that allow an operator to safely conduct and manage its aerial work operation in compliance with the regulations. 

Low flying approval

The AOC Handbook acknowledged that low flying was an operational requirement and that an application for low flying under CAR 157(4)(b) was required. This assessment was conducted at the same time as the initial AOC application.

The worksheet for the AOC application did not record any assessment undertaken by CASA to approve the operator to undertake low flying below the levels permitted in CAR 157. However, the assessing officer indicated that a key component of issuing the low flying approval was that the operator had a legitimate requirement and that its CP and line pilots held the required low‑level rating. There was no record of any in depth assessment of how the operator would address the risks associated with low flying. Further, the AOC Handbook did not provide any guidance or instruction on how such an application should be assessed.

The low flying instrument was issued to the operator on 22 September 2017. 

Surveillance post‑AOC issue (pre‑accident)

Post‑authorisation review

As at 2017, following the issue of an initial AOC, CASA was to conduct a post‑authorisation review (PAR) of the operator to ensure that all the entry control requirements were being met. This surveillance activity was to be conducted within 6‍–‍15 months following the initial issue.[13] As described in the CASA Surveillance Manual, a PAR was a type of level 1 surveillance, which was a structured, forward planned larger surveillance event. 

The CASA records showed that a PAR, as defined in its surveillance manual, had not been undertaken on the operator. 

In response to the draft report on 3 February 2025, CASA advised the ATSB that:  

At the time of the accident CASA conducted its surveillance planning under the National Surveillance Selection Process (NSSP), which was a risk-based methodology for the selection and prioritisation of surveillance events. Under the NSSP an operator such as this did not require a post authorisation review. CASA has since implemented a multi-year surveillance approach whereby all Aerial Work Operators undergo surveillance on a regular basis, irrespective of the degree of risk that CASA has assessed. This multi-year surveillance approach is one of many core elements of CASA’s National Oversight Plan.

On 3 April 2025, the ATSB and CASA had a follow‑up briefing to seek clarification on its responses to the draft report. CASA advised that the obligation for a PAR could also be achieved through an alternative activity, and in this case had been accomplished through the conduct of a level 2 surveillance event. CASA indicated that the level 2 surveillance, conducted in July 2018 (discussed below), was noted in its surveillance system as ‘post authorisation’ and would likely have been similarly scoped to a PAR. Therefore, this was considered an equivalent activity at the time. CASA had not provided the ATSB any supporting documentation indicating that an equivalent activity was permitted, or what should have been considered if this was to be undertaken. 

Surveillance in 2018

A level 2 surveillance event was a less formal interaction with an operator and could be in the form of checklist-based compliance and product checks of a specific section of its systems. A level 2 surveillance event took place in July 2018, after the operator’s first occurrence. The planned scope included airworthiness assurance, fuel load control, operational standards, and safety assurance. Nil findings were issued, and the CASA surveillance team noted that the operator was still in the process of reviewing its operating procedures with changes to be reviewed at the next surveillance event scheduled for later in 2018. It was not evident to what extent that low‑level survey operations were examined.

A second level 2 surveillance took place in October–November 2018. The planned scope included the same areas as the previous surveillance with the additional items of:

• airworthiness control
• implementation of the drug and alcohol management plan
• crew scheduling
• flight systems
• safety risk management
• assessments
• training infrastructure
• training management. 

Three safety findings and 7 safety observations were issued as a result of that surveillance. The surveillance team noted that the operator was actively trying to mitigate some of the operational risks, but the operations manual was lacking some of the procedures followed by the operator. The findings related to non‑conformance with operations manual procedures, uncontrolled documents, and aircraft defect management. 

Observations were not required to be actioned by the operator, although CASA did encourage them to do so. One observation related to fuel load control and procedures for addressing discrepancies in fuel quantity. Another observation related to limited procedures for the completion of the JSA. The surveillance report did not make any findings or observations on overall processes or procedures for low‑level survey operations. 

Periodic assessment tool

The authorisation holder performance indicator (AHPI) was a questionnaire‑based tool used by CASA to assess ‘the apparent risk to safety presented by an authorisation holder [operator]’. The AHPI tool consisted of a number of factors and sub‑factors associated with organisational characteristics and performance, commonly thought to affect or relate to safety performance behaviour. This was used by CASA to assist with determining whether any risk‑based surveillance of an organisation was required, and to scope the areas for that assessment.

A number of AHPI assessments had been completed by CASA and 5 out of the 6 AHPIs did not trigger any higher priority for surveillance. The last AHPI prior to the accident was completed in February 2022. CASA noted that there was no record in its system of a formal PAR being conducted, and there was no record of action following the 2019 and 2020 occurrences. The CASA officer recommended surveillance take place at the earliest possible convenience. 

Transition to Part 138 regulations

In response to the draft report, on 1 May 2025, CASA advised the ATSB that the accident occurred about 3 months after the commencement of the new flight operations regulations, which included Part 138. Under the transitional arrangements to the new regulations, current holders of an AOC authorising aerial work before 2 December 2021 had this certificate legally recognised as being an ‘aerial work certificate’. These operators were required to ensure that the operations manual complied with Part 138 and contained all the necessary content to enable that compliance. 

In this case, as the operator held an AOC prior to 2 December 2021, the content of its operations manual would have been approved by CASA. The operator’s compliance with Part 138 was to be reviewed at the next appropriate CASA oversight event.

While operators were not required to submit their entire operations manual for assessment prior to 2 December 2021, as part of the transition, CASA required operators (no later than 60 days prior) to submit extracts from their operations manual covering 2 key measures. These included change management and procedures for the carriage of aerial work passengers required under Part 138. CASA only required operators to submit their entire updated operations manual immediately before the commencement date of the new regulations. 

On 27 September 2021, MAGSPEC provided CASA with its complete operations manual, the HSEMS, including the change management and carriage of aerial work passenger procedures. CASA notified MAGSPEC via email in April 2022 that its initial submissions related to these 2 aspects were not yet compliant. Following a further submission by the operator, on 18 August 2022 CASA advised via email that these areas were ‘compliant’. CASA did not provide any advice related to an assessment of any other parts of the operations manual including those related to operational risk management. 

Surveillance post-accident

Shortly after the accident, CASA conducted an initial review to determine if further surveillance of the operator was required. The CASA officer noted that the operator had never undergone a level 1 surveillance event or had a PAR conducted. The officer recommended that a response surveillance event should be scheduled.

On 31 August 2023, the ATSB briefed CASA on the draft investigation findings, which included discussion on the operator’s risk management processes in place at the time of the accident. A similar briefing had also been provided to the operator on 19 July 2023. CASA completed a level 1 systems audit of the operator in September 2023. This level of surveillance is a structured, larger-type event that considers the specific activities conducted by the operator. The audit included a review and follow-up of this accident. The scope included: 

• airworthiness and maintenance aspects
• aircraft and passenger loading control
• crew scheduling and fatigue management
• operational standards, and data and documentation
• authorised activities
• flight systems and operational support systems
• safety assurance and safety risk management. 

The audit identified 6 safety findings and one observation for the operator related to maintenance documentation, fatigue management, and operations manual compliance. Specifically, one finding regarding the operations manual noted that it did not include all the content as required by the regulations. This included limited detail regarding:

• processes and procedures relating to low‑level operations and manoeuvring and role specific equipment usage (data acquisition system)
• processes and procedures relating to the training of operational and handling procedures with regards to low‑level operations and manoeuvring and role specific equipment usage (data acquisition system).

There were no other findings or observations made regarding the operational risk management processes and procedures.

At the time of the audit, the operator advised CASA that it was waiting for the completion of the ATSB’s investigation prior to initiating any changes to its processes or procedures. However, it was noted that, while the audit was not an investigation, a number of potential latent issues existed, including:

• No active or consistent fatigue monitoring of flight crew during operations other than the required recording of flight and duty records.

• Limited detail in operational risk assessments pertaining to operations in general and the additional fatigue obligations as required by legislation.

• Limited documented operational procedures and associated training relating to low level flight techniques and procedures including the use of operational equipment utilised during survey operations [as detailed above].

• Limited documented process and procedure, and subsequent detail established when conducting internal investigation following operational incidents.

Similar occurrences

Regulatory oversight

Regulatory oversight of air operations has been discussed in detail in previous ATSB investigation reports. These included a fatal Cessna 172 accident at Agnes Waters, Queensland in 2017 (AO‑2017‑005), a fatal Eurocopter EC120B helicopter accident at Hardy Reef, Queensland in 2018 (AO‑2018‑026), and a fatal Eurocopter EC135 helicopter accident at Port Hedland, Western Australia in 2018 (AO‑2018‑022).

These investigations identified that CASA’s regulatory oversight activities had not specifically examined the nature of the operator’s primary activities. The findings were as follows:

The Civil Aviation Safety Authority’s procedures and guidance for scoping a surveillance event included several important aspects, but it did not formally include the nature of the operator’s activities, the inherent threats or hazards associated with those activities, and the risk controls that were important for managing those threats or hazards. (safety issue AO-2017-005-SI-08) 

Although the operator’s primary helicopter activity was conducting charter flights to pontoons at Hardy Reef, regulatory oversight activity by the Civil Aviation Safety Authority had not specifically examined the operator’s procedures and practices for conducting operations to these helicopter landing sites. 

Although the operator’s primary helicopter activity was conducting marine pilot transfers, regulatory oversight activity by the Civil Aviation Safety Authority had not specifically examined the operator’s procedures and practices for conducting approaches and landings to ships at night in degraded visual cueing environments. 

In response to safety issue AO-2017-005-SI-08, CASA amended its surveillance and scoping form to require consideration of current activities. Further, it proposed the addition of an operator profile report to provide current, contextual information on an operator with a view to provide a more effective audit scoping process in which consideration and documentation of an operator’s activities was mandatory.

In response to the draft report, on 3 February 2025, CASA advised the ATSB that:

CASA’s Surveillance activities do not include specific checks of the suitability or effectiveness of the processes and procedures in these specific areas. However, under CASR [Civil Aviation Safety Regulation] Part 138.370, the operator is required to conduct risk assessments and mitigation processes before conducting any particular aerial work operation (that is, they must consider the risk of the specific operation and introduce appropriate mitigants). Surveillance is carried out under set scope of 138 operators under this regulation to ensure the operations are suitably risk assessed. 

Based on the actions in response to AO-2017-005-SI-08, the ATSB was of the understanding that an operator’s primary activity was considered in entry control processes and surveillance scope. As such, the above comment was also discussed at the ATSB‑CASA meeting on 3 April 2025. CASA advised that an operator’s primary activity is considered, and when assessing a Part 138 operator it uses a standardised worksheet to ensure there is a consistent approach to that assessment and to maintain a record of the decision‑making process. 

Low-level accidents

There has been a number of ATSB investigations into fatal accidents that resulted from a loss of control at low altitude, from which a recovery was not possible.

ATSB investigation AO-2012-059

On 29 April 2012, the owner-pilot of a Cessna 150 aircraft, registered VH‑UWR, was aerial stock mustering on a cattle station about 55 km north‑east of Bourke, New South Wales. The aircraft was observed circling over an area (where cattle were not moving), then entered a steep descent followed by the sound of an impact. The aircraft was substantially damaged, and the pilot sustained fatal injuries.

The ATSB found that, while manoeuvring at low level, the pilot inadvertently allowed the aircraft to aerodynamically stall, resulting in a high rate of descent and collision with terrain. There was insufficient information about pilot control inputs to establish the factors that precipitated the stall.

ATSB investigation AO-2014-192 

On 29 December 2014, a Cessna 172S aircraft, registered VH‑PFT, departed Cambridge Airport, Tasmania to photograph yachts participating in the 2014 Sydney Hobart race. On board the aircraft were the pilot and a photographer.

At about 1815 local time, the aircraft commenced low-level photographic runs on yachts to the east of Cape Raoul. Shortly after completing a run on one yacht at a height of about 50 ft, the aircraft entered a steep climbing turn. The aircraft had almost completed a 180° turn when the upper (right) wing dropped sharply while the aircraft’s nose pitched down to almost vertical. The aircraft impacted the water’s surface in an almost vertical nose‑down attitude with wings about level. Both aircraft occupants were fatally injured, and the aircraft was substantially damaged.

As a result of the steep climbing turn, the aircraft’s upper wing aerodynamically stalled, resulting in a rapid rotation out of the turn. The steep pitch attitude indicated that, due to the stalled upper wing, the aircraft entered a spin. There was insufficient height for the pilot to recover the aircraft.

ATSB investigation AO-2021-016 

On 13 April 2021, a Cessna R172K aircraft, registered VH-DLA, departed Canberra Airport, Australian Capital Territory, with a pilot and an observer on board to conduct powerline survey work to the north of Sutton township, New South Wales.

About 3 hours into the flight, while conducting a powerline inspection in the vicinity of Tallagandra Lane, nearby witnesses observed the aircraft flying low above the trees before commencing a left turn that continued into a steep descent and collision with terrain. The pilot and observer were fatally injured, and the aircraft was destroyed.

The ATSB found that, while manoeuvring to align the aircraft to inspect a powerline, the aircraft aerodynamically stalled and entered a spin at a height that was insufficient for recovery prior to the collision with terrain.

ATSB investigation AO-2021-052 

On 4 December 2021, the pilot of an Air Tractor AT‑400 aircraft, registered VH‑ACQ, was conducting aerial spraying operations on a property 75 km west‑south‑west of Moree, New South Wales.

At 0632 local time, the aircraft took off from the property’s airstrip with the first spray load. The pilot then completed 10 spray loads, each time returning to the airstrip to replenish its load. 

Prior to departing with the 11th load, the aircraft was refuelled to full and its spray load refilled. The aircraft then returned to the western side of the target block, but after descending to recommence spraying towards the south, the aircraft climbed and turned away to track north and overfly a flood-affected area. The pilot radioed the company operations manager expressing concern about the weather conditions and the potential for chemical to drift onto a neighbouring property. About 5 minutes later, the aircraft returned to the target block, this time on the eastern boundary.

The pilot then conducted 2 ‘smoker’ runs to assess the drift, followed by 5 back‑to‑back (parallel) spray runs. At the end of the 5th spray run, the aircraft was observed to climb then enter a right procedure turn. During the turn, the aircraft descended rapidly, collided with terrain, and was subsequently destroyed by fire. The pilot sustained fatal injuries.

The ATSB found that the aircraft was too close to the start of the spray run during the turn, which probably resulted in the pilot tightening the turn. This almost certainly resulted in an aerodynamic stall at a height too low to recover before colliding with the ground.

Safety analysis

Introduction

In the afternoon of 3 March 2022, a Cessna U206G, registered VH‑JVR, was being operated on a low‑level geophysical survey flight. When on the 25th survey line, the aircraft collided with terrain and was destroyed in the post‑impact fire. The pilot was fatally injured. 

This analysis will discuss the potential reasons for the loss of control and the initiation of the emergency response. The angle of bank regularly used for procedure turns, and the benefits of protective clothing and helmets are also examined. It will also consider the operator’s risk management processes and supporting procedures for low‑level geophysical survey flights, and regulatory oversight of these activities. 

Loss of control

The last recorded position of the aircraft was on a survey line consistent with the planned survey parameters. Likewise, a comparison of the available flight data indicated that the aircraft’s location was in a similar position to that of previous procedure turns. Therefore, it was likely that the aircraft was being manoeuvred onto the next survey line at the time of the accident.

The wreckage examination determined that the aircraft impacted the trees in a left angle of bank, with a steep angle of impact and a nose‑down attitude, indicative of a loss of control. The ATSB considered several reasons for the loss of control. There was no evidence to suggest any airborne impact with a bird nor that the weather conditions affected the pilot’s ability to maintain control of the aircraft. To the extent possible, the ATSB determined that the aircraft was structurally intact, there was no flight control malfunction, and the engine was producing power at the time of impact. Further, the pilot did not have any reported health issues, and the post‑mortem and toxicology examinations did not identify the presence of any natural disease or substances. While some causes of incapacitation may not always be identified post‑mortem, there was no evidence to suggest that the pilot had become incapacitated during flight.

The pilot had previously conducted 24 survey lines with no apparent issues identified from the recorded data. Based on the ATSB’s analysis, the aircraft’s airspeed had remained above the aerodynamic stall speed during the procedure turns. Therefore, there was no indication of a near or actual stall on the previous turns. Although, the data identified that the procedure turns were consistently conducted at heights below which a recovery from a stall and loss of control may not be possible. A stall situation was a plausible explanation, as found in previous ATSB investigations into low‑level accidents. However, in this case, this remained only a possibility due to the lack of recorded data beyond the last known position and no witness observations, making it difficult to determine the precise circumstances that led to the loss of control and collision with terrain.

Delayed emergency response

The operator had a phased emergency response plan, predicated on an elapsed time since the aircraft’s estimated time of arrival (ETA). Each phase was commensurate with an escalating level of concern. Unless there was a notification of an accident by other means, the distress phase would commence at 30 minutes past the ETA, at which point the Joint Rescue Coordination Centre (JRCC) would be contacted to initiate search and rescue activity. Each phase of the plan required accessing the satellite tracking system to ascertain the location of the aircraft. 

When the satellite tracking data stopped at 1343, the automatic watch function did not send an alert to the operator after 15 minutes had elapsed, as this function had not been activated. This was not noticed until 1430 when the ground operator conducted a periodic check. While the operator reported previously experiencing dropouts of the satellite tracking system, Spidertracks confirmed that up to the loss of data, the device was functioning as expected and that there were no recorded system outages. 

The ground operator was not able to contact the pilot via mobile phone and there was no emergency beacon activation. The ATSB noted that, while the aircraft was not fitted with an emergency locator transmitter, a personal locator beacon was carried in the aircraft. However, as this was not worn by the pilot, this was not readily accessible following the accident. 

As such, the status of the pilot and aircraft could not be established. The operations manager then advised the ground operator to follow the emergency response plan and wait until the ETA. When the aircraft did not arrive at the ETA of 1630, about 3 hours after the accident, in accordance with the plan the operator contacted the JRCC at 1701. 

The JRCC promptly initiated search and rescue procedures, and an aircraft departed at 1739 and located the aircraft wreckage at 1852. A search and rescue helicopter crew physically located the pilot at 0042. With the time taken to access the pilot, whose injuries required immediate medical care, it was very likely that even if the search and rescue activities had commenced when the satellite tracking data was lost, the pilot would have succumbed to the injuries received.

When an accident occurs and any injuries that result are potentially survivable, a timely response is essential. Minimising the time for search and rescue and enabling emergency services to respond as quickly as possible may increase the chances of a successful outcome. 

Low-level manoeuvring

The operator’s pilots routinely used high angle of bank procedure turns to manoeuvre between consecutive survey lines. This back‑to‑back pattern was described by the chief pilot (CP) as the most efficient method and it also reduced pilot workload, especially for obstacle and hazard avoidance.

The CP stated that procedure turns were trained to be at about 300 ft above ground level and a 45‍–‍60° angle of bank (known as a steep turn). Analysis of the accident flight indicated that the pilot was flying high angle of bank procedure turns consistent with the operator’s training. However, they were turning at an average of about 200 ft, lower than what was explained by the CP. 

The ATSB noted that, during the preceding 24 turns, the pilot had maintained sufficient margin above the stall speeds listed in the pilot’s operating handbook. However, as aircraft data was not available up to the loss of control it was not possible to determine if the aircraft stalled nor the exact circumstances that existed.

The similar occurrences discussed involved a loss of control that was preceded by a stall at very low heights. The U206G pilot’s operating handbook stated that up to 240 ft may be required to recover from a stall, but this height was based on flight testing in controlled conditions and that significantly more height may be required.

High angle of bank turns at low level present a significant risk and International Airborne Geophysics Safety Association (IAGSA) recommended that the angle of bank should be limited due to the stall speed increasing with increasing angles of bank, thus reducing the safety margins available. These margins can be quickly eroded if a pilot tightens the turn to ensure they intercept the survey line, which can increase the load factor further, resulting in reaching the stall speed quicker.

Protective clothing and helmets

At the time of the accident, the pilot was not wearing any protective clothing or a helmet. The use of such equipment was not required by the regulator, or the operator having considered environmental, comfort, and cockpit space aspects. Instead, it was left to the individual pilot’s discretion. 

The ATSB’s aviation medical specialist indicated that protective clothing and helmets may reduce the magnitude of injuries in an accident. While they were unable to comment on the effectiveness of these items for this accident, it was acknowledged that these items would not have protected the pilot from smoke inhalation. 

In some cases, occupants survive an accident only to succumb to hazards such as fire, drowning or environmental elements such as heat and cold (Shanahan, 2004). IAGSA recommended that survey pilots/crew wear protective clothing, not just as fire protection in the event of an accident but to also provide coverage from the elements while waiting for rescue or in a survival situation. Similarly, the Basic Aviation Risk Standard also recommended that all flight and aircrew wear protective clothing during operations. It is therefore important that these hazards have been considered to enable the best opportunity for survival in the event of an accident.

Risk management

In August 2019, MAGSPEC Aviation implemented a health, safety and environmental management system (HSEMS), which was intended to meet the Civil Aviation Safety Authority (CASA) safety management system (SMS) requirements. It also provided more generalised guidance on the management of risk within the organisation. However, as an SMS was not required by regulation, it was not assessed by CASA. The ATSB acknowledges that an operator’s SMS, in this case HSEMS, will evolve and mature with time. Significant events like accidents provide an opportunity to assess if the system is operating in a way that assures the highest level of safety given the nature of their operations. 

Acknowledging the CASR Part 138 requirements for aerial work operators to undertake risk assessments, which came into effect about 3 months prior to the accident, the ATSB reviewed the safety risk management component of the HSEMS. The HSEMS manual had detailed the 5‑stage process for risk management. However, there was no requirement in the operator’s manuals, as required by Part 138, to conduct a pre‑operational risk assessment nor had one been completed, as confirmed by the CP. 

The absence of a pre‑operational risk assessment did not allow formal mitigation strategies, nor provide assurance that the risk level associated with low‑level survey operations was as low as reasonably practical. For example, the operator reported that the survey flights were often flown at about 30 m (100 ft) above ground level. While low operating heights were specified in the job safety analysis (JSA), there was no formalised risk controls referred to in the JSA. The accident survey flight had been flown at 25 m (85 ft) and this had been accepted on the basis of the reconnaissance flight, without any formal identification and implementation of supporting risk controls. While IAGSA acknowledged that stipulating a fixed minimum safe survey height was not practical given differences in survey conditions and aircraft characteristics, a pre‑operational risk assessment for operations at a reasonably anticipated operating height may have provided a foundation from which to adequately assess any variations to this height.

Further, the operator routinely flew consecutive survey lines, which used increased (steep) angle of bank procedure turns (45‍–‍60°) to manoeuvre between the lines. There was no assessment to identify any risks associated with conducting higher angle of bank turns at low-level to ensure that reasonable mitigations were implemented, and appropriate safety margins were applied. 

The HSEMS further outlined that non‑site‑specific hazards were reported in their online SMS program, but due to the varying nature of operations, a separate site‑specific assessment of each survey job would be completed using the JSA. With the JSA, each survey task was assessed in isolation, with no reference to previous JSAs to ensure that applicable risk controls continued to be applied and/or were appropriate. Also, the JSA did not benefit from being informed by an overarching pre‑operational risk assessment. Therefore, as the predominant method for assessing operational risk, the JSA did not provide assurance that all hazards would be identified, and the associated risks would be assessed and mitigated. 

As noted by the International Civil Aviation Organization (2018), safety risk management activities should be documented and the Part 138 Manual of Standards stipulated that the operator’s pre‑operational risk procedures were to include the use of a risk register. The CP reported that an operational risk register was not being maintained as part of their HSEMS at the time of the accident. While non‑site‑specific hazards would be recorded and assessed in the online SMS program, there was a missed opportunity to record the site‑specific hazards identified from the individual JSAs. Therefore, without a risk register, the operator’s ability to track, monitor, and mitigate all known hazards, and assess the effectiveness of existing risk controls was limited. 

In interactions with CASA during the transition phase to Part 138 and the level 1 surveillance event post‑accident, there was no commentary related to the adequacy of the operator’s operational risk management processes. On that basis, the operator was of the understanding that the JSA met the risk assessment requirements of Part 138 inclusive of a pre‑operational risk assessment. However, CASA has since advised the ATSB that the JSA did not meet the requirements of a pre‑operational risk assessment.

Survey manoeuvres

As emphasised by IAGSA, conducting steep turns at low level can present challenges for fixed‑wing aircraft as any margin above the stall speed can quickly diminish and there may be limited height within which to respond to an unexpected situation. The ATSB’s analysis of the flight data showed there was some variation in the angle of bank used during the procedure turns on the accident flight, ranging from 43° up to the Cessna U206G angle of bank limit of 60°. At this limit, the stall speed increases by about 40%. The data also indicated that the procedure turns were conducted at an average of about 200 ft above ground level, which was lower than what was explained by the CP.  

Similarly, a review of the morning pilot’s flight and discussions with other company pilots identified that there were differences in the individual turn techniques, demonstrating variations of the procedure turns.

While 45‍–‍60° angle of bank procedure turns were being taught, the operations manual did not include any policy or procedure for this manoeuvre. Further, there were no specific limits identified, such as a minimum turn height or maximum angle of bank, to establish appropriate safety margins such as that recommended by IAGSA. 

ATSB research investigation report B2004/0337 discussed the importance of operational procedures:

The absence, deficiency or inappropriateness of operating procedures for operators may increase the risk to aviation safety.

The absence of standardised procedures means there may be considerable differences in the techniques used by different operators and contracting organisation staff to conduct tasks. Processes that are used to accomplish a particular task will evolve through a process of experience and passing on this information, often by word of mouth. There will be inconsistencies in how the task is accomplished, as different staff and operators will have differing levels of competence and experience, and different solutions to the same problem will have naturally evolved. The organisation that is managing the operation in such an uncontrolled environment will not be in full control or fully aware of how its tasks are being accomplished and therefore will have less control over the safety of the operation.

While the ATSB was unable to determine the circumstances that led to the loss of control, an operator’s expectations and desired safety margins should be documented to minimise variation and ensure operations are performed safely. Otherwise, without formal procedures, pilots are required to exercise judgement based on their experience, skills and knowledge.

Flight following

Survey flights are often conducted over remote, inhospitable terrain where regular communication services may not be available. Therefore, the use of satellite‑based flight following services are essential for providing real‑time monitoring of an aircraft’s location and for an efficient search and rescue response. 

The operator had installed Spidertracks to all its aircraft and there was an expectation that an automatic alert would be received from the system in the event of emergency. However, there was no requirement and supporting procedure to check the functionality of the system prior to each flight. In this case, the alert function had not been activated for multiple flights, including the accident flight, which potentially influenced the delayed emergency response. Also, while the ground operator assigned to each tasking was responsible for providing flight following services, there was no expected schedule for checking the satellite tracking nor any procedure detailing the expectations of this role. Despite this, given the severity of the pilot’s injuries, it was very unlikely that a prompt emergency response would have changed the outcome. 

It is important that an emergency response plan clearly identifies the notification and escalation triggers to avoid delays. Satellite tracking systems are useful in their ability to provide early notification of an emergency, especially in cases where the occupants have been incapacitated or otherwise unable to raise an alarm. However, their usefulness can only be realised if, when installed, they are correctly configured and operating as expected, otherwise increasing the risk of a delayed response.

Regulatory oversight activity

Since 2019, 3 ATSB investigations have been published identifying that regulatory oversight did not formally include the nature of the operator’s primary activities, the inherent threats or hazards associated with those activities, and the risk controls for managing those threats or hazards.

The ATSB acknowledges that CASA’s regulatory oversight activities were subject to normal constraints of time and resources, which may limit an ability to identify issues. Therefore, regulatory surveillance cannot examine every aspect of an operator’s activities, nor identify all the limitations associated with these activities.

The initial air operator’s certificate assessment of the operator and CP focused on the generic regulatory requirements and there were no criteria to evaluate their primary activity of low‑level survey operations. Although CASA had considered the operator and its key personnel as being suitable to conduct the proposed operations, there was no evidence that the processes and procedures for the primary activity had been specifically examined. Likewise, while the operator’s pilots held the appropriate low-level rating, there were no records to indicate that a detailed assessment of how the operator would address the risks associated with low flying had been conducted as part of the low flying approval. 

Prior to this accident, the operator had undergone surveillance twice in 2018, following the first occurrence. Operational standards were included in the scope for both surveillance events but there were no related findings made by CASA nor was there any indication to what extent the operator’s low-level survey operations were examined.

The last AHPI review in early 2022 also noted that no post‑authorisation review (PAR) had been conducted following the initial issue of the air operator’s certificate (AOC) and there were no CASA records of action following the 2019 and 2020 occurrences. Consequently, it was recommended that a surveillance activity take place. The same recommendation was also made following this accident. The subsequent surveillance event in 2023 identified that the operator had limited documented operational procedures and training related to low‑level flight techniques. 

CASA advised that a PAR (a type of level 1 surveillance) was likely to have been covered by an alternative level 2 surveillance event. The ATSB noted that a level 1 event was more comprehensive than a level 2, and for a PAR, was intended to ensure that the entry control requirements were being met following the initial issue of an AOC. As the CASA officers had made comments about a PAR having never been conducted, it was unclear whether the level 2 was sufficient to have been considered as having met the requirement of a PAR. 

While none of CASA’s activities specifically focused on topics related to low‑level survey operations, for example survey patterns and heights, it was difficult to determine whether additional focus, through the conduct of a level 1 PAR for example, would have identified the specific aspects as found in the post‑accident surveillance event. However, CASA has since strengthened its AOC entry control procedures and surveillance planning and scoping to include more emphasis on assessing the primary activity including the use of a specific worksheet that highlights areas specific to Part 138 operators. Consideration of the primary activity provides a level of assurance that operators continue to meet the established requirements and function at the level of competency and safety required to undertake the activity for which they have been approved to perform. 

In addition to the above, and as previously discussed, the ATSB identified deficiencies with the operator’s risk management processes. As the requirement for risk assessments only came into effect about 3 months prior to the accident, with the introduction of Part 138, there was limited opportunity for CASA to review these processes within that period. It was also noted that CASA had intended to look at the operator’s Part 138 compliance at the next scheduled surveillance event. 

The 2023 surveillance event, which was also a review and follow-up to the accident, was a level 1 surveillance and included risk management within the scope of that activity. However, there were nil findings or observations identifying that there was no pre‑operational risk assessment and risk register, although required under Part 138. As an unintended consequence of this and the transition process to Part 138, the operator was of the understanding that the JSA satisfied this requirement. However, CASA has since indicated to the ATSB that the JSA did not meet the requirement of a pre‑operational risk assessment. While post‑accident, the 2023 surveillance event was a missed opportunity for CASA to identify the deficiencies in processes and inform the operator’s understanding of their risk assessment obligations under Part 138.

Findings

From the evidence available, the following findings are made with respect to the collision with terrain, involving a Cessna U206G registered VH-JVR, 124 km west of Norseman, Western Australia, on 3 March 2022. 

Contributing factors

• It was likely that, during a manoeuvre to intercept the next survey line, for undetermined reasons, control of the aircraft was lost at a height from which recovery was not possible, resulting in a collision with terrain. 

Other factors that increased risk

• An emergency response was not initiated until 30 minutes after the aircraft's estimated time of arrival, which was 3 hours after satellite tracking had stopped and attempts to contact the pilot had been unsuccessful. Although an earlier response was very unlikely to have altered the outcome in this case, minimising the time for search and rescue and enabling emergency services to respond as quickly as possible may increase the chances of a successful outcome.
• In accordance with the operator’s training, pilots routinely used increased angle of bank (45‍–‍60°) turns at low altitude to position the aircraft onto survey lines. This increased the risk of an aerodynamic stall at altitudes from which recovery may not be possible.
• The operator did not require its pilots to wear protective clothing or helmets during low‑level survey operations, nor were they required to do so by regulations. However, the use of such has been recommended by industry to improve survivability in the event of an accident.
• MAGSPEC Aviation's safety risk management processes did not include a pre‑operational risk assessment that recognised the generic risks and hazards common across that type of operation nor was a risk register maintained. Consequently, there was limited assurance that all the risks had been identified and that all reasonable mitigations had been applied. (Safety issue)
• The operator’s pilots were trained to, and routinely flew survey patterns utilising steep turns at low level. However, procedures or limitations specific to these manoeuvres were not included in the operations manual, which increased the risk of inconsistencies in the application of those manoeuvres and reducing the safety margins available.
• While the operator’s aircraft were fitted with a satellite-based flight following system, there was no requirement nor were there supporting procedures to confirm the set‑up and functionality of the system prior to flight or to monitor the system during flight. This increased the risk the system not operating as expected and not providing early notification of an emergency.
• The Civil Aviation Safety Authority regulatory oversight of the operator had not specifically included the primary activity of low-level survey flights, or the processes and procedures designed to reduce the risks associated with that activity.

Safety issues and actions

Central to the ATSB’s investigation of transport safety matters is the early identification of safety issues. The ATSB expects relevant organisations will address all safety issues an investigation identifies.  Depending on the level of risk of a safety issue, the extent of corrective action taken by the relevant organisation(s), or the desirability of directing a broad safety message to the aviation industry, the ATSB may issue a formal safety recommendation or safety advisory notice as part of the final report. All of the directly involved parties are invited to provide submissions to this draft report. As part of that process, each organisation is asked to communicate what safety actions, if any, they have carried out or are planning to carry out in relation to each safety issue relevant to their organisation.  Descriptions of each safety issue, and any associated safety recommendations, are detailed below. Click the link to read the full safety issue description, including the issue status and any safety action/s taken. Safety issues and actions are updated on this website when safety issue owners provide further information concerning the implementation of safety action.

Risk management framework

Safety issue number: AO-2022-011-SI-01

Safety issue description: MAGSPEC Aviation's safety risk management processes did not include a pre‑operational risk assessment that recognised the generic risks and hazards common across that type of operation nor was a risk register maintained. Consequently, there was limited assurance that all the risks had been identified and that all reasonable mitigations had been applied.

Safety recommendation to MAGSPEC Aviation Pty Ltd

The ATSB makes a formal safety recommendation, either during or at the end of an investigation, based on the level of risk associated with a safety issue and the extent of corrective action already undertaken. Rather than being prescriptive about the form of corrective action to be taken, the recommendation focuses on the safety issue of concern. It is a matter for the responsible organisation to assess the costs and benefits of any particular method of addressing a safety issue.

Safety recommendation number: AO-2022-011-SR-01

Safety recommendation description: The Australian Transport Safety Bureau recommends that MAGSPEC Aviation Pty Ltd develops and maintains a pre-operational risk assessment and risk register that is separate to its existing job safety analysis process. This should encompass the generic risks and hazards common across its operations and allow it to fully consider operational risks beyond individual survey tasks.

Safety action not associated with an identified safety issue

Whether or not the ATSB identifies safety issues in the course of an investigation, relevant organisations may proactively initiate safety action in order to reduce their safety risk. The ATSB has been advised of the following proactive safety action in response to this occurrence.

Additional safety action by MAGSPEC Aviation Pty Ltd

In response to this accident, MAGSPEC Aviation has taken the following safety action:

• The JSA has been revised to include the selection of consequence and likelihood to determine a risk for each identified hazard. It also recorded mitigations that would be applied. The operator advised the overall and highest risk scores determined whether the survey could proceed and/or if it was likely to increase the fatigue and safety of the operation to unacceptable levels.
• The health, safety and environmental management system has been incorporated into the operations manual as an appendix.
• The emergency response plan was revised to clarify initiation triggers and accounted for a satellite tracking system failure. The operator has also equipped its operations room with 2 dedicated monitors for the sole purpose of tracking aircraft.
• The operations manual now includes a minimum speed versus angle of bank section and pilot actions if an aircraft cannot achieve or maintain the required speed.
• Guidance on procedural turns has been formalised in the operations manual. Although there is a description of how to conduct the turn, the manual also explains that this was the desired turn method and may not always be possible (due to terrain, obstacles, block shape et cetera).
• The operations manual has been amended to clearly state that an aircraft was required to have a fixed emergency locator transmitter. If this becomes unserviceable or has to be removed, the aircraft can only be flown for the purpose of having the issue rectified.
• Each pilot has been issued with a personal locator beacon, individually registered with the Australian Maritime Safety Authority. Pilots are required to keep the device on their person while operating company aircraft. The operations manual also states that the personal locator beacon cannot be carried/used if not tested.
• Follow-up with Spidertracks is to be made on each occurrence of dropout, service interruption or delay in tracking updates and numerous improvements made to the interface. The operator advised that the SOS automatic watch function and alert has been investigated and rectified. A checklist item has been added to ensure Spidertracks is correctly functioning prior to departure.
• They no longer operate at survey heights below 30 m.
• The operator identified that its low‑level training syllabus was lacking parameters to mark a pilot as competent, especially in critical phases of flight. This has been formalised to match what had been done practically.
• Its operations manual is currently under review by CASA. This includes items to enable Flight Safety Foundation’s Basic Aviation Risk Standard accreditation. Just prior to final publication of this report, the operator advised that it had been awarded the Basic Aviation Risk Standard accreditation.

Glossary

AGL	Above ground level
AHPI	Authorisation holder performance indicator
AoA	Angle of attack
AOC	Air operator’s certificate
CAR	Civil Aviation Act
CASA	Civil Aviation Safety Authority
CASR	Civil Aviation Safety Regulations
CoG	Centre of gravity
CP	Chief pilot
ETA	Estimated time of arrival
ELT	Emergency locator transmitter
G	Gravity
GPS	Global positioning system
HSEMS	Health, safety and environmental management system
IAGSA	International Airborne Geophysics Safety Association
JRCC	Joint Rescue Coordination Centre
JSA	Job safety analysis
MOS	Manual of Standards
PLB	Personal locator beacon
POH	Pilot’s operating handbook
SMS	Safety management system

Sources and submissions

Sources of information

The sources of information during the investigation included:

• MAGSPEC Aviation Pty Ltd
• other pilots who conducted flights for the operator
• recorded data from the satellite tracking device
• the maintenance organisation
• aviation medical specialist
• Pathwest Laboratory Medicine WA
• Civil Aviation Safety Authority
• Bureau of Meteorology
• Western Australia Police Force
• Australian Maritime Safety Authority Joint Rescue Coordination Centre
• Spidertracks Ltd.

References

Australian Transport Safety Bureau. (2005). Aviation Research Investigation Report B2004/0337 Risks associated with aerial campaign management: Lessons from a case study of aerial locust control. Retrieved from /publications/2005/aerial_locust_control/

Australian Transport Safety Bureau. (2013). ATSB Research Investigation AR-2012-128 A review of effectiveness of emergency locator transmitters in aviation accidents. Retrieved from /publications/2012/ar-2012-128

Civil Aviation Safety Authority, (2021). Civil Aviation Safety Regulations 1998 Part 138 - Aerial work Operations. Retrieved from https://www.legislation.gov.au/F1998B00220/2021-12-02/text

Civil Aviation Safety Authority, (2021). Part 138 (Aerial Work Operations) Manual of Standards 2020 Retrieved from https://www.legislation.gov.au/F2020L01402/2021-12-02/text

Civil Aviation Safety Authority, (2022). Aerial work risk management (advisory circular AC138-05 v2.0) Retrieved from https://www.casa.gov.au/aerial-work-risk-management 

Federal Aviation Administration, (2021). Airplane Flying Handbook, FAA-H-8083-3C. US: FAA. Retrieved from Airplane Flying Handbook | Federal Aviation Administration (faa.gov)

Flight Safety Foundation, (2022). Basic Aviation Risk Standard Implementation Guidelines. (Version 9 2022). Retrieved from https://flightsafety.org/bars/the-bar-standards-and-manuals/

International Airborne Geophysics Safety Association. (2017). Safety Policy Manual (v2017-1201). Retrieved from https://iagsa.ca/

International Civil Aviation Organization. (2018). Safety Management Manual, fourth edition, Montréal: International Civil Aviation Organization.

Shanahan, D.F. (2004). Human tolerance and crash survivability. RTO HFM Lecture Series on ‘Pathological Aspects and Associated Biodynamics in Aircraft Accident Investigation’. Madrid, Spain.

Submissions

Under section 26 of the Transport Safety Investigation Act 2003, the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. That section allows a person receiving a draft report to make submissions to the ATSB about the draft report. 

A draft of this report was provided to the following directly involved parties:

• MAGSPEC Aviation Pty Ltd
• other pilots who conducted flights for the operator
• Civil Aviation Safety Authority.

Submissions were received from:

• MAGSPEC Aviation Pty Ltd
• Civil Aviation Safety Authority.

The submissions were reviewed and, where considered appropriate, the text of the report was amended accordingly.

Purpose of safety investigations The objective of a safety investigation is to enhance transport safety. This is done through:  identifying safety issues and facilitating safety action to address those issues providing information about occurrences and their associated safety factors to facilitate learning within the transport industry. It is not a function of the ATSB to apportion blame or provide a means for determining liability. At the same time, an investigation report must include factual material of sufficient weight to support the analysis and findings. At all times the ATSB endeavours to balance the use of material that could imply adverse comment with the need to properly explain what happened, and why, in a fair and unbiased manner. The ATSB does not investigate for the purpose of taking administrative, regulatory or criminal action. Terminology An explanation of terminology used in ATSB investigation reports is available here. This includes terms such as occurrence, contributing factor, other factor that increased risk, and safety issue. Publishing information Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2025 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Commonwealth Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this report is licensed under a Creative Commons Attribution 4.0 International licence. The CC BY 4.0 licence enables you to distribute, remix, adapt, and build upon our material in any medium or format, so long as attribution is given to the Australian Transport Safety Bureau.  Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

[1]      Aerodynamic stall: occurs when airflow separates from the wing’s upper surface and becomes turbulent. A stall occurs at high angles of attack, typically 16˚ to 18˚, and results in reduced lift.

[2]      Integral fuel tanks are part of the aircraft structure. They are manufactured by assembling parts of the aircraft structure with sealant to form a fuel-tight compartment, most commonly in the wings.

[3]      A supplemental type certificate is a type certificate issued when an applicant has received regulatory approval to modify an aeronautical product from its original design. The supplemental type certificate, which incorporates by reference the related type certificate, approves not only the modification but also how that modification affects the original design.

[4]      The system was pre-programmed by the ground operator prior to departure, minimising any manipulation required by the pilot during flight.

[5]      From the pilot’s operating handbook, the calibrated airspeed and indicated airspeed in the range of interest were within 1 kt, so for the purposes of the analysis were considered equivalent. The term indicated airspeed is used throughout the analysis.

[6]      G load: the nominal value for acceleration. In flight, G load represents the combined effects of flight manoeuvring loads and turbulence and can have a positive or negative value.

[7]      A steep turn is one greater than a 45° bank angle.

[8]      Carbon monoxide is a colourless, odourless, tasteless, and poisonous gas that is produced as a by-product of burnt fuel. Exposure to a leak from the exhaust of an aircraft engine into the cabin can lead to elevated levels of carbon monoxide, which can impair cognitive function.

[9]      https://flightsafety.org/bars/the-bar-standards-and-manuals/

[10]    CAR 157 (2) requires that an aircraft shall not fly over a populous area at a lower height than 1,500 ft or any other area at a lower height than 500 ft. CAR 157 (4)(b) states that the provisions of CAR 157 (2) shall not apply if the aircraft is engaged in aerial work operations and the operator has a permit from the authority (CASA) to do so.

[11]    As the wing to which the stall warning vane is mounted approaches the critical AoA, the relative air flow changes direction and will push the vane up, closing a switch in the device. This will activate a warning horn.

[12]     Donesafe is a web and application‑based safety management system tool.

[13]    This timeframe has since changed with the most recent being 12-18 monthsfollowing the initial issue, depending on the type of authorisation issued (CASA Surveillance Manual version 5.2 ‑ May 2024).