Runway excursion

Investigation summary

What happened

On 29 May 2025, a Cessna 310R, registered VH-NXA and operated by Marthakal Yolngu Airline, was conducting a non-scheduled passenger air transport flight from Darwin to Lake Evella, Northern Territory. On board were the pilot and 4 passengers. 

During the approach at Lake Evella Aerodrome, recorded data indicated that the aircraft touched down just prior to halfway along the runway. The pilot subsequently applied braking, but the aircraft did not decelerate as expected. This resulted in a runway excursion and the aircraft subsequently collided with a perimeter fence which substantially damaged the left wing. The pilot and 4 passengers were uninjured.

What the ATSB found

The ATSB found that the pilot conducted the approach above the standard profile and crossed the threshold above the normal approach speed. This resulted in the aircraft floating during the landing flare for a prolonged period in ground effect, and a landing beyond the planned touchdown point. After the aircraft touched down, the pilot subsequently commenced braking about halfway along the runway. The long landing reduced the available distance to decelerate on the runway.  

At the aircraft’s landing weight, the remaining runway length should have provided sufficient stopping distance, but degraded braking capacity meant the aircraft could not be stopped before the runway end and it subsequently departed the runway.

During a scheduled maintenance event prior to the occurrence, a licensed aircraft maintenance engineer believed an apprentice had replaced the main-wheel brake pads. An inspection after the occurrence found that the right brake reservoir was empty and that the right pads were worn beyond limits, which reduced braking capacity on that side.

The ATSB also identified that the operator’s procedures allowed the use of self-reported passenger weights without additional allowances, and that the electronic weight and balance system had been configured with higher maximum weights applicable to a modification not fitted to this aircraft. In combination with calculation errors on the day, this resulted in the aircraft being operated above the maximum permitted ramp and take‑off weights.

What has been done as result

Marthakal Yolngu Airline advised that all references to the use of self-reported passenger weights for the purposes of weight and balance calculations will be removed in the next amendment of the operations manual. 

The operator also advised that the electronic weight and balance system will be amended to reflect the correct maximum weights for VH-NXA prior to its return to service. 

The safety manager briefed company personnel about the proposed changes and advised existing pilots that actual weights for passengers must be used for all flights. 

The ATSB will monitor these safety actions until the proposed changes to the operations manual and weight and balance system have been formally implemented. 

D & T Aircraft Engineering advised that, following the occurrence, a debrief with maintenance personnel highlighted the importance of increased vigilance during inspections conducted after maintenance tasks, and that the organisation also identified opportunities to improve internal processes following similar occurrences.

Safety message

Factors such as additional airspeed over the threshold can result in a landing beyond the intended touchdown point, increasing the risk of a runway overrun excursion. While adherence to a pre-determined stabilised approach criteria can effectively mitigate such risks, pilots should always exercise vigilance and ensure the aircraft is flown within the assumed conditions used to calculate landing performance.     

Pilots are therefore encouraged to continue to actively monitor the flight path using cockpit instrumentation and external visual cues until a safe landing is assured. This should include identifying and nominating an appropriate touchdown point on the runway to ensure a go‑around can be executed if a touchdown beyond this point is likely to occur. 

Additionally, maintenance organisations should ensure that effective systems are in place to disseminate important information to all maintenance personnel, so that emerging defects are identified and rectified before they affect flight operations.

The occurrence

Pre-flight preparation 

On the morning of 29 May 2025, a Cessna C310R, registered VH-NXA and operated by Marthakal Yolngu Airline, was being prepared for a non-scheduled passenger air transport flight from Darwin to Lake Evella, Northern Territory.

When the pilot arrived at the airport, they observed a licensed aircraft maintenance engineer (LAME) and their apprentice performing maintenance on the aircraft braking system. During this period the pilot prepared their flight plan and when they returned to the aircraft, the LAME and the apprentice had completed the maintenance.    

The pilot, who was operating their first flight as pilot in command of a multi-engine aircraft, commenced their pre-flight checks. Due to the recent work on the braking system, they taxied to an aircraft bay to conduct a static engine run-up.[1]

With both engines at 1,700 RPM, they recalled that the aircraft moved forward slightly with the brakes applied. They physically increased their braking pressure, after which, the aircraft remained stationary. After completing the run-ups, the pilot taxied back to the terminal where the passengers were waiting. At this time, they also discussed the brakes with another C310 pilot who advised them that quite a lot of brake pressure was required during run-ups. 

The pilot reported feeling rushed, and elected to use the self-reported passenger and baggage weights prior to boarding for weight and balance calculations, which were recorded on the manifest. These weights were entered into an electronic weight and balance system, which indicated that the planned load complied with the aircraft weight and balance limitations. The passengers were then taken to the aircraft where the pilot conducted a safety briefing before they boarded. 

Occurrence flight 

At 0857 local time, VH-NXA departed from Darwin Airport with the pilot and 4 passengers on board for Lake Evella (Figure 1). When approaching the Lake Evella Aerodrome, the pilot reported becoming visual with runway 08 at about 15 NM (28 km) and tracked for a straight in approach. At 1051, the aircraft was established on final approach for runway 08.

Figure 1: VH-NXA flight planned track

Source: Google Earth, annotated by the ATSB

The pilot recalled that there was a south-easterly wind between 8–10 kt for the approach, with a right crosswind component. Another pilot, who arrived at Lake Evella about 3 minutes after VH-NXA, recalled the wind was from an easterly direction at about 10 kt. 

The pilot of VH-NXA recalled that the approach ‘seemed stable,’ (see Stabilised approach criteria) and stated that they generally used the runway threshold as their aiming point. They estimated crossing the threshold at their calculated approach speed of 90 kt or ‘just above.’ 

A navigational application (OzRunways) [2] was installed on a tablet computer and an Android phone on board the aircraft and broadcast flight data (see Recorded information). The OzRunways data taken from the Android phone, overlaid on a Google Earth image (Figure 2), showed the aircraft crossing the runway threshold at a height of 55 ft with a ground speed of 94 kt (Figure 2, A). The runway in Lake Evella was not equipped with visual slope guidance and the pilot relied on their visual assessment of ‘how the runway should look at certain height.’

Figure 2: VH-NXA ground speed at key points in landing sequence

Source: Google Earth, annotated by the ATSB

The pilot and the passenger seated directly behind them estimated that the aircraft touched down approximately 200 m past the threshold. ATSB analysis of recorded data indicated touchdown at 1051:29 (the corrected altitude of the aircraft matched the terrain elevation of runway 08), which was 402 m past the threshold (Figure 2, B). The passenger seated in the front row beside the pilot recalled passing the taxiway immediately after touchdown (Figure 2, C).

Another pilot on the ground standing at the apron who witnessed the landing reported observing VH-NXA a few feet above the ground in the ‘flaring attitude’ about a third of the distance along the runway. They also recalled that the aircraft was travelling faster than what they thought was normal and landed just beyond the taxiway (Figure 2, C). 

The pilot reported that after touchdown, they applied the brakes passing the apron area about halfway along the runway (Figure 2, C). At that point, they reported that the aircraft did not appear to be slowing as expected and the passenger in the last row recalled the aircraft passing the apron ‘very fast’. 

The pilot recalled increasing their braking pressure and when they saw the end of the runway approaching, they shut both engines down by selecting the mixture controls to idle cut-off. The pilot then elected to steer the aircraft to the left of the runway centreline to increase the runway distance for the deceleration required.

The aircraft departed the left side of the runway, 118 m from the runway end (Figure 2, D) (Figure 3, inset left), while the passenger seated beside the pilot verbally prompted the passengers to ‘brace.’ Recorded data indicated the aircraft was travelling at 45 kt at this point. The left wing subsequently collided with a fencepost (Figure 2, E)  (Figure 3, inset right) located 193 m from the point the aircraft departed the runway. 

Figure 3: VH-NXA ground roll following runway excursion 

Source: Marthakal Yolngu Airline, annotated by the ATSB

Following the collision, the aircraft came to a stop and the pilot and passengers disembarked through the right cabin door. There were no injuries to the pilot or passengers, however the aircraft sustained substantial damage (see Post-accident inspection).

Context

Pilot information

The pilot held a Commercial Pilot Licence (Aeroplane) issued in 2021 and a class 1 aviation medical certificate. They also held a multi-engine aircraft class rating, which was issued in 2022 and renewed with a flight training organisation on 24 May 2025.

The pilot had accumulated 1,066 hours of total aeronautical experience, which included 71 hours of multi-engine time accumulated under the supervision of an instructor. 

They reported they had flown about 70 hours in the last 90 days, including a total of 11.7 hours on the Cessna C310R, which was conducted during the course of their training. 

The pilot’s training was conducted by a flight training organisation (FTO) in Darwin, on behalf of Marthakal Yolngu Airline. FTO training records detailed that the pilot commenced line training for the C310R on 15 May 2025. 

This line training took place over 9.8 flight hours, after which they were assessed as proficient by a flight examiner during a combined line check and operator proficiency check for the C310R on 24 May.

The pilot had not operated the C310R to Lake Evella Aerodrome during the course of their training and had not operated there in any aircraft type prior to the occurrence. They reported sleeping about 8 hours the night before the occurrence and had been awake for about 7 hours at the time of the occurrence and feeling ‘fully alert.’

Aircraft information 

The Cessna 310R is a twin-engine, low-wing (with a wingspan of 11.3 m), 6-seat, unpressurised aircraft equipped with retractable landing gear and powered by 2 Continental IO-520 piston engines. VH-NXA was manufactured in the United States in 1978 and first registered in Australia in 1989. A maintenance organisation located in Darwin became the registration holder on 4 March 2020.

Braking system

Section 7 of the Cessna 310R Pilots operating handbook (POH) contained the following description of the braking system: 

The airplane is provided with an independent hydraulically actuated brake system for each main wheel. A hydraulic master cylinder is attached to each pilot’s rudder pedal. Hydraulic lines and hoses are routed from each master cylinder to the wheel cylinder on each brake assembly. No manual adjustment is necessary on these brakes. The brakes can be operated from either pilot’s or co-pilot’s pedals. 

Meteorological information

The graphical area forecast and the applicable grid point wind and temperature forecast for the flight indicated:

• prevailing visibility greater than 10 km
• scattered cloud[3] with bases 1,500 ft above mean sea level (AMSL)
• isolated areas of smoke reducing visibility to 5,000 m
• isolated rain showers and thunderstorms reducing visibility to 2,000 m and 1,000 m respectively, and broken cloud with bases 800 ft above AMSL
• moderate turbulence below 4,000 ft in thermals and dust/sand whirls (dust devils)
• wind 130° at 21 kt and temperature of 24°C at 1,000 ft above AMSL.

Aerodrome information

Lake Evella Aerodrome (YLEV) is situated at an elevation of 278 ft AMSL and comprised of a single sealed runway, 08/26, measuring 1,065 m in length and 18 m in width and was sloped 0.5% up toward the east. The aerodrome is uncontrolled and operated on a dedicated CTAF,[4] and is subject to animal hazards.

Maintenance information

Aircraft maintenance manual 

The Cessna 310R Aircraft maintenance manual (AMM) contained a troubleshooting guide to assist maintenance personnel to rectify defects relating to systems fitted to the aircraft. The section that covered the wheels and brakes included the following information (Table 1):

Table 1: Cessna 310R maintenance manual troubleshooting

The AMM also described the brake wear limits on the C310R, which included:

Check back plate and pressure plate linings for wear. If worn to a thickness of 0.125 to 0.100 inch, the linings should be replaced.

Scheduled maintenance 

The aircraft was flown to Darwin on 15 February 2025, where the authorising licensed aircraft maintenance engineer (LAME) planned to conduct a corrosion inspection at their maintenance facility. The LAME also performed a ‘check 1’ inspection, which they stated was the equivalent of a 100-hour inspection.

During the inspections, additional maintenance was conducted due to leaking brake callipers, which was common to the brakes on the C310 according to the LAME. This involved the removal, bleeding, resealing and refitting of both callipers and was performed by an apprentice. It was also the LAME’s expectation that the brake pads would be replaced during this maintenance task because this was routine practice, although not in the procedure. 

As part of the 100-hour inspection, the LAME performed an engine run-up and observed the aircraft did not hold under brakes at this time. Believing that they had been replaced, they believed that the new brake pads needed to be bedded or burnt in. The AMM stated ‘brake burn in is required to minimize glazing of the friction surfaces’ when new brakes are installed. They subsequently completed the engine run up on one engine at a time, which allowed the aircraft to remain stationary. 

Following the completion of the inspections and associated maintenance tasks, including the additional work carried out on the brakes, the LAME certified the aircraft maintenance logbook on 26 May 2025.  

The authorising LAME later stated that new brake pads should have been installed before the callipers were refitted to the landing gear, however they did not verify that this had occurred. They reported that the brake pads were last changed on 15 December 2023 and had 494 landings prior to the occurrence. 

Pre-departure maintenance

The aircraft underwent a post‑maintenance verification flight the day prior to the accident flight, with a flight instructor and the occurrence pilot as an observer. After the flight, the instructor advised the LAME by text message that the brakes felt ‘spongy.’

On the morning of 29 May, prior to the accident flight, the LAME checked the aircraft brakes, reporting that they were acceptable, even though the brake pedal travel felt more than usual. The decision was made to bleed the brakes to remove any air or water in the brake lines and top up the brake fluid. With assistance from an apprentice during this process, the LAME reported that hydraulic fluid spilled onto the right tyre and was subsequently wiped down. The aircraft was then released back to service. 

Post-accident inspection

Following the occurrence, the LAME inspected the aircraft at Lake Evella Aerodrome on 18 June 2025 and documented the aircraft damage. The aircraft had sustained significant damage to the left wing (Figure 4), which separated from the fuselage outboard of the left engine nacelle. The pitot tube, right tip tank, propeller and nose gear door were also damaged following the runway excursion. 

Figure 4: VH-NXA damaged left wing

Source: Aircraft maintainer, annotated by the ATSB

The LAME found that the right brake reservoir was empty, with evidence of hydraulic fluid leakage on the right tyre, however stated that the right brake disc was serviceable. 

They identified that the right brake pads were ‘heavily worn.’ Images supplied by the LAME also indicated the presence of hydraulic brake fluid, originating from the brake piston adjacent to the brake line (Figure 5). They also indicated that the hydraulic fluid on the tyre may have been from fluid spilling when the brakes were topped up on the morning of the occurrence flight. 

Figure 5: VH-NXA right hand brake components and hydraulic brake fluid

Source: Aircraft maintainer, annotated by the ATSB

No defects were identified on the left brake system and the right brake calliper was removed and tested in Darwin by the LAME. The right brake calliper was bolted onto a brake disc with sufficient pressure applied to prevent calliper movement. After 13 days, sufficient hydraulic fluid had leaked, which allowed the calliper to be moved in relation to the brake disc. The LAME subsequently disassembled the right brake calliper and identified a ‘very small’ hydraulic fluid leak, which they did not consider was the cause of the fluid loss during the occurrence. 

Operational information

Weight and balance

The operator’s standard operating procedures (SOP) stated that during the conduct of air transport operations, prior to each sector, the pilot in command must complete an aircraft load and trim sheet. 

An operator‑approved electronic load sheet was available to pilots for the purpose of completing weight and balance calculations in accordance with the POH weight and balance limitations.

The operator’s Cessna 310R Flight crew operating manual (FCOM) included the following statement regarding the possible modification of company operated aircraft:

The Company operates C310R aircraft in several possible modification states, which may affect limiting weight.

The only modification listed in the FCOM that affected the weight limitations for VH-NXA was the fitment of a vortex generator (VG)[6] kit. The FCOM also contained information relating to the fitment of the VG kit including increased weight limitations, changes to various airspeeds and stated:

If less than 84 vortex generators are in place or undamaged, the aircraft must be operated in accordance with the original AFM performance data (ie nil VGs).

Electronic weight and balance system

The electronic weight and balance system was developed by a third party to calculate the weight and balance for each flight. In the system, each aircraft was configured with a weight and moment arm[7] when empty. 

The pilot would enter the pilot and passengers’ weights, their seating positions and fuel to calculate both the weight and centre of gravity of the aircraft at take-off and landing. The system was designed to alert the user if any weight and balance limitations were exceeded. 

For VH-NXA, the electronic weight and balance system incorporated an increase in weight in accordance with a supplemental type certificate number for a C310R VG modification. However, the maintenance organisation that owned and maintained the aircraft stated that VH-NXA had not been fitted with the VG modification, and they were not aware of any modifications that increased the standard maximum permitted weights as prescribed in the POH. 

 As a result, the weight and balance system contained the following increases to the standard POH weight limitations which were not applicable to the aircraft (Table 2):

Table 2: Cessna 310R maximum weights

Passenger-declared weights 

The operator’s SOP stated that for the purposes of calculating the aircraft’s weight and balance, ‘passenger weights must be actual, or self-reported.’ Following the occurrence, a passenger reported that their body and baggage weights were requested without the use of a calibrated scale. The pilot did not indicate that any adjustments of additional amounts were applied to the passenger reported weights. 

The CASA multi-part AC 121-05, AC 133-04 and AC 135-08 – Passenger crew and baggage weights, described acceptable weight calculation methods that could be defined in operating procedures. The circular stated that:

The use of actual weights is the most accurate method of maximising payload capacity. Appropriately calibrated weighing scales should be used. Actual weighing is more commonly used by Part 133 [helicopter passenger transport] and 135 [smaller aeroplane passenger transport] operators. This is, in part, due to the smaller number of passengers being carried, which makes this option less disruptive than it is for Part 121 [larger aeroplane passenger operations] operators. 

Operators should have procedures to identify when passenger-declared weights are not appropriate, such as when operating close to limitations. Under these circumstances, the use of actual weights may be required to ensure limitations are not exceeded. 

Passenger-declared weights have inherent inaccuracies as passengers may not know their actual weight, especially when fully dressed. An adjustment allowance should be added to any passenger-declared weight, as a factor or a fixed additional amount.

Weight and balance calculations

Following a review of documentation provided by the operator and pilot, the ATSB identified several discrepancies contained in the operational documentation from the day of the occurrence. 

The passenger and baggage weights recorded in the manifest by the pilot indicated a combined weight of 387 kg. However, the corresponding load sheet indicated a combined passenger and baggage weight of 337 kg (excluding the pilot). 

Additionally, the fuel plan prepared by the pilot indicated a total fuel figure of 441 kg. By comparison, the fuel figure on the load sheet was recorded as 432 kg.    

Due to the identification of the combined discrepancy of 59 kg, the ATSB recalculated the aircraft’s weight and balance for the flight. This identified the following updated weights and exceedances (Table 3) prescribed in the POH for aircraft not fitted with a VG kit.

Table 3: VH-NXA calculated weights and exceedances

Landing performance calculations

The operator’s SOPs stated that company aircraft are subject to the requirements of Civil Aviation Safety Regulations Part 135 Manual of Standards (MOS) with respect to take-off and landing performance requirements. Chapter 10 of the Part 135 MOS stipulated ‘that the aeroplane crosses the runway threshold at a height of 50 ft’ unless an approved short landing operation was being conducted. Additionally for landing, the FCOM stated that the reference landing approach speed (V_ref)[8] should be achieved at 50 ft above the landing surface.

For aeroplanes, take-off and landing distance calculations to determine maximum take‑off weight or the maximum landing weight are achieved through a manual calculation using the limitations given in the POH for the specific aircraft type, taking into account: 

• environmental conditions
• runway length.

The FCOM also required pilots to apply landing distance factoring of 1.20 for all calculations. The pilot stated they had calculated their landing performance based on the aircraft’s maximum landing weight and calculated a factored landing distance of 680 m on the flight plan with a V_ref of 90 kt for their landing at Lake Evella. 

Calculations using the ATSB recalculated landing weight and the estimated ambient conditions at the time of the occurrence determined that the required landing distance (with the 1.20 factoring) with a 50 ft threshold crossing height was 659 m. This figure included a landing ground roll distance of 195 m and a corresponding V_ref of 91 kt.

Stabilised approach criteria 

The SOPs stated that, ‘unless the aircraft meets stabilised approach criteria at the specified altitude, a missed approach must be executed.’

A stabilised approach was described in the SOPs as an approach to land that met a number of criteria by 300 ft above the runway during a visual approach. These included the following:

• the aircraft is on the correct flight path

• only small changes in heading & pitch are required to maintain the correct flight path

• the aircraft speed is V_ref to V_ref +20 kt

• sink rate is not greater than 1,000 fpm or pre-briefed limits. 

Recorded information

The pilot used a flight planning application (OzRunways) on an iPad and an Android phone for en route flight planning and navigation. The flight planning software provider was an approved source of electronic aeronautical charts, however the application could not be used as a primary means of GPS-based navigation as the iPad and Android phone GPS did not meet certification for aviation use. The pilot reported that the iPad was placed on the floor for the approach while the Android phone was in their chest pocket. By examining the combination of groundspeed and derived deceleration data, in addition to the best direct line of sight to satellites, it was assessed that the Android data had the highest positional accuracy.

The recorded data had limitations due to an altitude resolution of 100 ft, while filtering and adjustments were also applied to smooth the data and are known to affect the accuracy of small sections. Additionally, the altitude data of VH-NXA was corrected to match the terrain elevation during the landing ground roll (Figure 6).

Based upon the operator’s stabilised approach criteria, the decision to continue the approach, or conduct a go-around, became applicable by the time the aircraft reached 300 ft above the runway.

The following recorded parameters were observed from below 300 ft to the threshold:

• the aircraft crossed the threshold of runway 08 at a height of 55 ft
• aircraft speed remained within V_ref to V_ref +20 kt
• sink rate (vertical speed) less than 1,000 fpm. 

The recorded data indicated (purple line in Figure 6) that at the 300 ft (578 ft corrected altitude) stabilised approach gate, the aircraft was about 42 ft above the normal 3 degree slope (orange line), and remained above it until reaching 104 ft above the aerodrome elevation. During this period, the sink rate exceeded 800 fpm for 8 consecutive seconds between 1051:06 and 1051:14 local time.

At 1051:20, the aircraft crossed the threshold with a groundspeed of 94 kt. The pilot reported the wind component during the approach was a south-easterly wind between 8‍–‍10 kt, which would have resulted in a 3‍–‍4 kt headwind component. Accordingly, the aircraft’s airspeed was likely around 97‍–‍98 kt as it crossed the runway threshold, which was 6‍–‍7 kt above the V_ref of 91 kt.

At 1051:29, the corrected altitude of the aircraft matched the terrain elevation, which indicated that the aircraft landed 402 m along the runway with a groundspeed of 81 kt. The pilot reported applying brakes as the aircraft passed the apron area, which occurred about 2 seconds after touchdown. Following a ground roll distance of about 540 m, the aircraft decelerated to a groundspeed of 45 kt when it vacated the left side of the runway at 1051:48.

Figure 6: VH-NXA approach and landing

All times are coordinated universal time (UTC). Local time was Central Standard Time (CST), which was UTC +9 hours and 30 minutes. The aerodrome elevation is 278 ft. Source: ATSB, data provided by OzRunways and Google Earth

Related occurrences 

ATSB investigation AO-2024-056

On 2 November 2024, a GippsAero GA8-TC Airvan, was being used to conduct a scenic flight from Whitsunday Airport (Shute Harbour), Queensland. During the landing the aircraft departed the upwind end of the runway before entering marshy ground and coming to a stop in a ditch. 

The ATSB investigation identified that the aircraft's approach was above profile with a high airspeed and the pilot had an incorrect understanding of the required approach speed. Subsequently, the pilot did not initiate a go-around, resulting in a landing beyond the planned touchdown point. The ATSB also identified that the operator’s weight and balance system used an incorrect empty weight moment arm to calculate the aircraft's centre of gravity, and passengers were not weighed in accordance with its procedures.

Safety analysis

On the morning of 29 May 2025, a Cessna 310R, registered VH-NXA, was being operated by Marthakal Yolngu Airline on a non-scheduled air transport flight from Darwin to Lake Evella, Northern Territory, with a pilot and 4 passengers on board. 

During a straight-in visual approach, without visual slope guidance to runway 08 at Lake Evella, the pilot assessed that the approach was stable and continued with the landing.  After the aircraft crossed the runway threshold, it floated for a prolonged period and subsequently landed before reaching a taxiway located about halfway along the runway. 

When the pilot applied braking passing the airport’s apron area, the aircraft did not decelerate as expected. The aircraft subsequently overran the runway and collided with a fence. The pilot and passengers were uninjured, however, the aircraft sustained substantial damage.

This analysis examines how the condition of the aircraft braking system, and the conduct of the approach and landing, contributed to the runway excursion. It also explores the operator’s self-reported passenger weight procedures and electronic weight and balance system, and how the latter, in combination with incorrect pre-flight weight calculations, led to the aircraft being operated above the weight limits specified in the pilot’s operating handbook.

Pre-flight maintenance 

Maintenance which was completed on the aircraft 3 days before the occurrence involved numerous concurrent tasks. These were conducted by a licensed aircraft maintenance engineer (LAME) with the assistance of an apprentice. One of the tasks involved the apprentice conducting maintenance on the braking system due to leaking brake callipers. It was the LAME’s expectation that the apprentice had replaced the main wheel brake pads during this maintenance task.

This expectation influenced their assessment that the aircraft rolled forward during post‑maintenance engine run-ups, due to the new brakes requiring ‘burning in.’ In this case, a physical verification of the brake pads was not conducted as a result. 

On the morning of the occurrence, the LAME carried out corrective maintenance in response to the flight instructor text message report of ‘spongy brakes’ the day prior. While the occurrence pilot was aware of this report, they were not aware of the LAME’s experience with the aircraft rolling forward during the engine run-up. 

When the pilot commenced the pre-flight engine run-up for the occurrence flight, the brakes failed to keep the aircraft stationary. The pilot physically increased the brake pressure and successfully kept the aircraft stationary, but did not advise maintenance personnel. The pilot’s limited experience on multi-engine aircraft led them to consider that this might be normal, which was reinforced during a brief discussion with another Cessna 310 pilot. 

Gaps in communication and incorrect assumptions allowed a latent defect to persist into operation, contributing to the runway overrun in this occurrence.

Approach

Lake Evella Aerodrome was not equipped with visual slope guidance, and as a result, the pilot relied on their assessment of visual cues of the runway itself to assess whether they were on the correct approach path while they typically used the runway threshold as their aiming point.  

Recorded data from the pilot’s Android phone indicated that the aircraft was higher than the usual 3 degree ‘correct flight path’. The rate of descent exceeded 800 fpm for a period of 8 consecutive seconds until the aircraft descended below 140 ft relative to the runway. At the time, the pilot recalled the approach ‘seemed stable,’ while the operator’s stable approach criteria permitted rates of descent up to 1,000 fpm. 

The aircraft subsequently crossed the threshold of runway 08 at a height of 55 ft with a ground speed of 94 kt. ATSB analysis concluded that the aircraft’s airspeed was likely 6‍–‍7 kt above the V_ref for the recalculated landing weight 91 kt.

When the aircraft neared the point of touchdown, it was subjected to ground effect, which meant that excess airspeed at the point of flare would result in a considerable float distance due to the reduction in drag and lack of power-off deceleration in ground effect (Federal Aviation Administration, 2023). 

Additionally, landing distances provided in the aircraft flight manual are based on the aircraft achieving V_ref (plus wind and gust additives) at 50 ft above the runway surface. As a result, any additional airspeed will result in a later touchdown and reduce the remaining landing distance available (Federal Aviation Administration, 2023). 

In this case, the additional airspeed crossing the threshold likely resulted in a prolonged float in ground effect. This resulted in the aircraft touching down 402 m beyond the runway threshold which was the pilot’s usual aiming point. Subsequently, the pilot applied braking about 2 seconds after the touchdown, at which point, there was about 585 m of remaining distance available to decelerate on the runway.

Excursion

The pilot first became aware of an issue with the braking system when they applied brake pressure during the landing roll with about 585 m of runway remaining. Witness accounts recalled the aircraft was travelling at high speed as it passed the taxiway and apron area without any significant deceleration. Additionally, recorded data showed the aircraft only slowed from 81 kt at touchdown to 45 kt when it vacated the left side of the runway following a ground roll distance of about 540 m. 

At the aircraft’s landing weight, the ATSB calculated ground roll distance required was 195 m, which was sufficient to bring the aircraft to a stop within the remaining length of the runway had the brakes been functioning correctly. However, the loss of hydraulic brake fluid and the worn brake pads on the right-hand brake reduced the available braking capacity. As a result, the braking capacity was insufficient to arrest the aircraft’s forward momentum before the end of the runway. The pilot attempted to increase the available stopping distance by steering left and departing the runway, however it was insufficient, and the aircraft subsequently collided with the perimeter fence.

Passenger weights

The operator’s exposition permitted the use of self-reported passenger weights for weight and balance calculations, without requiring the application of additional allowances or validation. This practice introduced errors into the weight and balance data used for pre-flight planning.

Research has found that people tend to underestimate the weights of themselves and others. Further, people are less accurate at estimating the weight of others than they are of themselves.[9] To cater for the variation in weight, it is recommended that operators weigh passengers or apply adjustment factors to self-reported values (Civil Aviation Safety Authority, 2025). In contrast, the operator’s reliance on unadjusted self-reported passenger and carry-on baggage weights provided no systematic mitigation for potential inaccuracies, which increased the likelihood that the aircraft would be operated overweight or at centre of gravity limits outside the manufacturer’s requirements.

Electronic weight and balance

The operator used an electronic weight and balance system to calculate aircraft loading data for each aircraft in operation. In that electronic system, VH-NXA had been configured with the higher maximum weight limits applicable to aircraft fitted with a vortex generator (VG) modification. However, the aircraft did not have the specified modification installed. Consequently, the programmed maximum zero-fuel, ramp and take-off and weights in the system exceeded those authorised in the aircraft’s POH.

This configuration error meant the electronic weight and balance system allowed VH‑NXA to be loaded in excess of the certified weight limitations, while still indicating that the loading complied with those limitations. This created an ongoing risk that the aircraft could be operated above the approved maximum weights.

Weight exceedances  

During the occurrence flight, the aircraft was operated above the certified maximum ramp and take-off weights due to cumulative errors in the pilot’s weight and balance calculations. As a result of the configuration errors in the electronic weight and balance system, no alert to the overweight condition was made. 

The pilot, who was conducting their first multi-engine command flight, reported feeling rushed during pre-flight preparation, which likely reduced the opportunity for careful verification of passenger weights, totals and data entry. Review of the weight and balance documentation from the occurrence identified multiple inaccuracies, indicating that the overweight condition arose from a breakdown in the usual cross-checking processes rather than a single isolated error.

Although the overweight condition did not result in the aircraft exceeding its maximum landing weight, operating above certified weight limits is known to increase take-off and landing distances and degrade braking performance. Additionally, excessive weight reduces the available safety margin if an in-flight emergency condition should arise (Federal Aviation Administration, 2016).

Findings

From the evidence available, the following findings are made with respect to the runway excursion involving Cessna 310, VH-NXA, at Lake Evella Aerodrome, Northern Territory, on 29 May 2025. 

Contributing factors

• The certifying licensed aircraft maintenance engineer did not verify that the brake pads had been replaced by an apprentice during scheduled maintenance, which resulted in the aircraft being returned to service with worn brake pads on the right brake system.
• The pilot conducted the approach above the standard profile and crossed the threshold above the normal approach speed. This resulted in a landing beyond the planned touchdown point, and the pilot applied braking about halfway along the runway, which reduced the available distance to decelerate on the runway.
• Due to the worn right brake pad and the lack of hydraulic fluid in the right brake system, there was insufficient braking capacity available to prevent a runway overrun following the landing and the application of brakes about halfway along the runway.

Other factors that increased risk

• Marthakal Yolngu Airline’s procedures did not require that additional allowances were applied when using self-reported passenger weights for weight and balance calculations. (Safey issue)
• Marthakal Yolngu Airline’s electronic weight and balance system used incorrect maximum weights for the aircraft, which increased the risk of flight crew operating the aircraft above the certified weight limitations. (Safey issue)
• The aircraft was operated overweight due to incorrect weight and balance calculations, as well as errors in the electronic weight and balance system.

Safety issues and actions

Self-reported passenger weights

Safety issue number: AO-2025-024-SI-01

Safety issue description: Marthakal Yolngu Airline’s procedures did not require that additional allowances were applied when using self-reported passenger weights for weight and balance calculations. 

Weight and balance system

Safety issue number: AO-2025-024-SI-02

Safety issue description: Marthakal Yolngu Airline’s electronic weight and balance system used incorrect maximum weights for the aircraft, which increased the risk of flight crew operating the aircraft above the certified weight limitations. 

Safety action not associated with an identified safety issue

Additional safety action by D & T Aircraft Engineering 

D & T Aircraft Engineering advised that, following this occurrence, a debriefing was conducted with maintenance personnel to discuss key learnings. It was acknowledged that increased vigilance would be exercised in the future to ensure that aircraft components are carefully inspected and confirmed to be in a serviceable condition following the completion of maintenance tasks. Additionally, the organisation stated that it would prioritise accessing the aircraft at the earliest opportunity in the future to assist in identifying potential causes of component failure, particularly in cases where perishable evidence plays a critical role in determining the cause.

Glossary

Sources and submissions

Sources of information

The sources of information during the investigation included:

• pilot of the accident flight
• the operator
• the flight training organisation
• maintenance organisation
• Civil Aviation Safety Authority
• Bureau of Meteorology
• accident witnesses
• OzRunways. 

References

Civil Aviation Safety Authority. (2025). Multi-Part AC 121-05, AC 133-04 and AC 135-08 - Version 1.2. Retrieved from https://www.casa.gov.au/sites/default/files/2021-08/multi-part-advisory-circular-121-05-ac-133-04-ac-135-08-passenger-crew-baggage-weights.pdf

Civil Aviation Safety Authority Australia. (2021). Safety behaviours human factors workbook for engineers. Retrieved from https://www.casa.gov.au/sites/default/files/2021-06/safety-behaviours-human-factors-engineers-workbook.pdf

Federal Aviation Administration. (2016). Aircraft Weight and Balance Handbook. Retrieved from https://www.faa.gov/sites/faa.gov/files/2023-09/Weight_Balance_Handbook.pdf

Federal Aviation Administration. (2023). Advisory Circular Subject: Aircraft Landing Performance and Runway Excursion Mitigation. Retrieved from https://www.faa.gov/documentLibrary/media/Advisory_Circular/AC_91-79B_FAA.pdf

Federal Aviation Administration. (2023, March). Pilot’s Handbook of Aeronautical Knowledge. Retrieved from https://www.faa.gov/regulations_policies/handbooks_manuals/aviation/faa-h-8083-25c.pdf

Federal Aviation Administration. (n.d.). Spatial Disorientation Visual Illusions. rev. 2/11. Washington DC: U.S. Department of Transportation Federal Aviation Administration. Retrieved from https://www.faa.gov/sites/faa.gov/files/about/initiatives/got_weather/archives/spatiald_visillus.pdf

Ramos , E., Lopes, C., & Barros , H. (2009). Unawareness of weight and height – the effect on self-reported prevalence of overweight in a population-based study. The Journal of Nutrition, vol. 13, pp.310–314.

Reed, D., & Price , R. (1998). Estimates of the heights and weights of family members: accuracy of informant reports. International Journal of Obesity, vol. 22, pp.827–835.

Shapiro , J. R., & Anderson, D. A. (2003). The effects of restraint, gender, and body mass index on the accuracy of self-reported weight. International Journal of Eating Disorders, vol. 34, pp.177–180.

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:

• pilot of the accident flight
• the operator
• maintenance organisation
• Civil Aviation Safety Authority
• Bureau of Meteorology.

Submissions were received from: 

• pilot of the accident flight
• the operator maintenance organisation
• Bureau of Meteorology.

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. About ATSB reports ATSB investigation reports are organised with regard to international standards or instruments, as applicable, and with ATSB procedures and guidelines. Reports 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. 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.

Investigation summary

What happened

On 11 January 2025, a Cessna 510, registered VH-SQY and operated by AirMed Australia (AirMed), was being used to conduct a non-emergency medical air transport flight from Narrandera Airport to Bankstown Airport, New South Wales. On board were a pilot, a flight nurse and a patient.

At 1103 local time, the aircraft departed Narrandera and approximately 30 minutes later commenced descent into Bankstown. Weather information at Bankstown reported variable wind conditions and that the runway was wet. During the landing, the pilot experienced reduced braking performance and the aircraft overran the end of the runway into muddy ground. None of the occupants were injured and the aircraft was undamaged.

What the ATSB found

The ATSB determined that during the approach, an undetected tailwind was encountered and the aircraft landed with a groundspeed higher than the minimum aquaplaning speed. In addition, there was likely standing water on the runway and the aircraft’s main landing gear tyres were worn to limits resulting in reduced braking performance. Subsequently, the pilot cycled the anti‑skid system, likely further decreasing braking performance. In combination, these factors resulted in the aircraft departing the end of the runway.

The ATSB also identified that AirMed required pilots to apply an incorrect landing distance factor, which reduced the safety margin when determining the required landing distance at a destination aerodrome. In addition, the type rating training provided by Air Link, a company associated with the operator, taught pilots to apply the same incorrect landing distance factor. Furthermore, AirMed’s procedures were unclear on how the factor should be applied, when the assessment should be conducted and how runway surface condition should be considered.

Finally, when determining the required landing distance at Bankstown, the pilot applied the incorrect landing distance factor prescribed by the operator. Subsequently, prior to descent and after obtaining the actual conditions at the aerodrome, the pilot did not identify that the landing distance available was insufficient for the landing.

What has been done as a result

AirMed updated its operations manual to require the use of a 60% landing distance factor and additional factoring for wet runway operations. Additionally, it introduced the requirement to conduct a landing distance calculation both before take-off and prior to landing, and included guidance on the applicability of options when using either tabulated data or flight planning software.

AirMed also provided training to all crew addressing the effects of tailwind, correct anti‑skid use, tyre limits and landing technique. In addition, it updated defect reporting procedures to encourage earlier reporting of anticipated maintenance requirements, and implemented a policy of tyre replacement when tread reaches 2 mm, representing 80% tyre wear.

Air Link amended training material for the C510 type rating to ensure that the correct landing distance factoring was applied and taught. In addition, past students were contacted to ensure that they understand performance requirements relating to the C510. Furthermore, it is in the process of including a new section in the endorsement training around wet weather operations, and has also implemented a policy of tyre replacement when tread reaches 2 mm.

Bankstown Airport amended the runway 11C/29C longitudinal slope information in the aerodrome manual to align with the En Route Supplement Australia slope values. Finally, while not in response to this occurrence, the Civil Aviation Safety Authority subsequently amended the performance section of the Part 121 guidance material as part of its continuous improvement process. These changes included:

• the addition of a section specifying that landing performance must be checked both pre‑flight and in-flight
• advice that actual landing distance data cannot be used to satisfy in-flight replanning operations
• provision of a list of known aircraft types, including the C510, that must not use actual landing distance data for in-flight landing distance calculations.

Safety message

The use of safety margins on top of calculated take-off and landing distances provides mitigation for a wide range of issues that impact aircraft performance, including unexpected environmental conditions. Operators must be familiar with any factoring applicable to their operation and should adjust procedures appropriately when regulations change. For private pilots, while not mandated, the use of safety margins is highly recommended. These recommended safety margins can be found in the Civil Aviation Safety Authority (CASA) Advisory Circular (AC) 91-02 Guidelines for aeroplanes with MTOW not exceeding 5 700 kg - suitable places to take off and land, and should be applied in conjunction with any guidance given in the aircraft flight manual.

This occurrence also highlights the limitations on accurate and timely reporting of runway surface condition, particularly following a period of intense rainfall. Only 3 mm of standing water is required for a runway to be considered contaminated, and this depth of water can accumulate rapidly before the runway surface condition is able to be assessed. Furthermore, when landing on an un‑grooved runway, braking may be degraded when the runway is very wet but not yet classified as contaminated, with significant additional stopping distance required.

The occurrence

On 11 January 2025, a Cessna 510, registered VH‑SQY and operated by AirMed, was being used to conduct a non‑emergency medical air transport flight from Narrandera Airport to Bankstown Airport, New South Wales (Figure 1). On board were a pilot, a flight nurse and a patient.

Figure 1: Incident flight 

Source: Google Earth, annotated by the ATSB

The aircraft had been repositioned earlier that morning from Wagga Wagga Airport to Narrandera Airport, where the flight nurse and patient boarded for the flight to Bankstown. Prior to departure from Wagga Wagga, the pilot obtained a weather forecast for both Narrandera and Bankstown (see the section titled Meteorology) and added sufficient fuel for both flights. The forecast conditions for Bankstown indicated rain and possible thunderstorms (see the section titled Forecast conditions) and the pilot determined that the aircraft would be below the maximum landing weight permitted in these conditions by the landing distance available charts (see the section titled Pilot assessment of landing distance available).

The aircraft departed Narrandera at 1103 local time and, approximately 30 minutes later, the pilot obtained the automatic terminal information service (ATIS)[1] information ‘Echo’ for Bankstown, prior to commencing descent. The pilot conducted the instrument approach procedure for runway 11C[2] and reported becoming visual with the runway at around 800‍–‍900 ft. The pilot recalled that the runway appeared wet, however they did not see any indications of standing or pooled water. Additionally, while they did not recall seeing the windsock, they reported that they did not feel as though there was a tailwind.

The aircraft commenced the round out for landing at 1156:32, and the pilot reported touching down firmly within the touchdown zone, deploying the speed brake and lowering the nose wheel prior to applying the brakes. The pilot stated that, despite braking, no deceleration or braking action was felt. They continued to hold brake pressure, expecting the anti‑skid system to activate (see the section titled Anti‑skid system), however the lack of discernible braking continued.

By this point, the pilot considered that the aircraft had passed the point on the runway where a go‑around could be safely conducted. Observing no system failure or warning indications in the cockpit, the pilot elected to cycle the anti‑skid system. They turned the anti‑skid system off, released and re‑applied the brakes. With no change in braking action the pilot again released the brakes, turned the anti‑skid back on, then re‑applied and held the brakes.

Realising that the aircraft would not stop prior to the end of the runway, the pilot advised air traffic control (ATC) that they could not stop. They then steered the aircraft slightly to the right to avoid the lights at the end of the runway. Departing the end of the runway at a groundspeed of 28 kt, the aircraft entered muddy ground and travelled another 30 m in a right turn before coming to a stop (Figure 2). The aircraft was undamaged and the pilot, flight nurse and patient were uninjured. After advising ATC of the situation and shutting down the aircraft, the patient was transferred to ground transport and the aircraft was recovered to a hanger.

Figure 2: VH-SQY after runway excursion

Source: Supplied

Context

Pilot

The pilot held an air transport pilot licence (aeroplane) issued in 2017 and a class 1 aviation medical certificate. They had accumulated 6,954 flight hours, of which 133 hours were operating the Cessna 510. In the previous 90 days, the pilot had accumulated 128 hours, all in the Cessna 510. The pilot had completed a type rating for the aircraft and an instrument proficiency check in October 2024 with Air Link, a company associated with the operator. The pilot had also completed line training with Air Link and had conducted operations for them, prior to commencing operations for AirMed.

Aircraft

General information

VH-SQY was a Cessna 510 Citation Mustang equipped with 2 Pratt & Whitney Canada PW615F‑A turbofan engines. The aircraft was manufactured and first registered in 2010 and was registered with the operator in 2019. It was in medical configuration, in which 2 seats and a cabinet in the main cabin were replaced with a stretcher and medical equipment.

At the time of the incident, the aircraft had accumulated 5,721 hours total time in service and was being maintained in accordance with the Cessna 510 maintenance manual. The last periodic inspection was conducted in December 2024, and the maintenance release showed no outstanding items. 

Main landing gear tyres

VH-SQY was equipped with 2 Michelin main landing gear (MLG) tyres. Maintenance records showed that both MLG tyres were last replaced in October 2024, after conducting 255 landings. Since that time, the aircraft had conducted 239 landings including the incident flight.

The manufacturer required the MLG tyres to be inflated to a loaded pressure of 88 PSI. Both the maintainer and the pilot reported that, while the tyre pressure was not recorded or checked, the pressure appeared to be normal with no signs of over or under inflation.

The tyre manufacturer provided guidance on tyre removal criteria (Figure 3) which stated:

Removal criteria for normal wear is based on remaining tread rubber as determined by groove depth or exposure of textile/steel ply material…
NORMAL REMOVAL WEAR LIMIT: Remove the tire when the wear level reaches the bottom of any groove at one point up to a maximum 1/8 of the circumference.
NOTE: When the NORMAL REMOVAL limit is reached, the tire should be replaced. If it is necessary to continue the tire in service beyond the normal wear limit, the tire should be removed either at the next maintenance base or upon reaching the EXPOSED CORD LIMIT, whichever occurs first. At the EXPOSED CORD LIMIT the tire should be removed and replaced.

Figure 3: Michelin tyre wear guidance

Source: Michelin, annotated by the ATSB

During the post-incident inspection, the maintainer determined that both main landing gear tyres were worn to limits (Figure 4) and identified evidence of flat spotting, potentially due to the wheels previously locking up under braking. The maintainer further advised that the tread remaining on the left and right MLG tyres was measured to be 0.013 inches (0.3 mm) and 0.019 inches (0.5 mm) respectively. Comparatively, the tread of a new tyre was reported to be 0.26‍–‍0.29 inches (about 7 mm), indicating that 4‍–‍7% of the original tread was remaining. Both tyres were subsequently replaced before the aircraft was released back to service. 

Figure 4: VH-SQY main landing gear tyres post‑incident

Source: Supplied, annotated by the ATSB

The pilot reported that they had inspected the condition of the tyres as part of the daily inspection of the aircraft prior to commencing operations for the day. They further reported that, while they identified that the centre tread on the tyres was low, they considered that as there was sufficient depth on the outside tread with no exposed cord the tyres were serviceable.

Anti‑skid system

The aircraft was equipped with an anti‑skid system to provide maximum braking efficiency across all runway surfaces and conditions. The anti‑skid system detects if the speed of any main landing gear wheel was too slow for the aircraft’s speed and released the brake momentarily to allow the wheel rotation to increase, preventing it from skidding. Anti‑skid systems are designed to reduce landing distance and minimise the potential tyre damage which can occur when a wheel is locked.

The aircraft was also equipped with a cockpit warning message to indicate when the anti‑skid was inoperative. While landing with anti‑skid inoperative was permitted, pilots were advised that doing so required the landing distance to be increased by between 39‍–‍45%.

In a normal landing with the anti‑skid system operative, the aircraft flight manual (AFM) advised pilots to apply brakes after nose wheel touchdown. The AFM further advised that:

to make sure of proper braking on water, snow, and ice-covered, hard-surfaced runways, and all unimproved surfaces, it is necessary for the pilot to apply maximum effort to the brake pedals throughout the braking run. When the system detects a skid and releases the applied brake pressure, any attempt by the pilot to modulate braking can result in an interruption of the applied brake signal and may increase stopping distance significantly.

The manufacturer advised that the anti‑skid system was not certified for turning off then on during a landing. They further advised that turning the system off during a landing roll may result in flat spotting the tyres if the anti‑skid was actively controlling a skid, especially on a wet runway. Additionally, when the system was powered on, it took several seconds to self‑test, during which time it would not function. 

Meteorology

Forecast conditions

The Bureau of Meteorology (BoM) issued both a Graphical Area Forecast (GAF) for the area including Bankstown Airport and a terminal area forecast (TAF) for the aerodrome that covered the pilot’s expected landing time. The GAF was initially issued at 0308 that morning and described the weather to expect around Bankstown as including:

• broken cloud from 2,000‍–‍10,000 ft[3]
• periods of scattered rain showers with towering cumulus clouds and broken cloud at 500 ft
• isolated thunderstorms with heavy rain, becoming occasional from 1300.

The GAF was re‑issued as part of the regular cycle at 0926 with no changes, other than that the thunderstorms were no longer expected to produce heavy rain and were expected to remain isolated. 

The TAF for Bankstown Airport was issued and then amended at 0352. It described the weather at the airport for the expected arrival time as:

• wind from 050°M at 14 kt, light showers of rain and broken cloud at 2,000 ft
• periods of up to 60 minutes of showers of rain with wind gusting 15‍–‍25 kt, broken cloud at 800 ft and visibility reduced to 2,000 m
• possibility of thunderstorms for periods up to 60 minutes with variable strong winds, broken cloud at 500 ft and visibility reduced to 1,000 m.

A new TAF was issued at 1026, with the only change being that the wind was now expected to be from 060°M at 10 kt.

Reported conditions

Prior to descent, the pilot reported that they listened to the automatic terminal information service (ATIS)[4] information ‘Echo’ for Bankstown Airport to obtain the reported conditions at the airport. The information included that:

• runway 11 left, centre and right were in use
• runway surface condition code was 5,5,5. Whole runway was wet
  (see the section titled Assessment of runway surface condition)
• wind was variable at 8 kt
• visibility reduced to 3,000 m in rain
• cloud was scattered at 800 ft, scattered at 1,200 ft and broken at 2,000 ft
• temperature was 23 degrees and QNH[5] was 1014.

Recorded conditions

Wind

The BoM weather station at Bankstown Airport recorded the mean direction of the wind at 1‑minute intervals in addition to the minimum, mean and maximum wind strength. Additionally, the standard deviation of the wind direction was recorded as a measurement of the variability of the wind direction within each minute.

The mean wind direction varied significantly between 1130‍–‍1215 (Figure 5). The recorded wind changed from a headwind to a tailwind on runway 11C, 6 minutes prior to VH-SQY landing. However, at around this time, the variability of the wind direction increased. At the time of the landing, the mean wind was recorded as a 6 kt tailwind. Subsequently, the tailwind further increased slightly before decreasing again, becoming a headwind again 14 minutes after the landing.

Figure 5: Runway 11C mean tailwind and wind direction variability 1130‍–‍1215

Source: ATSB

Rainfall

BoM observations recorded that Bankstown Airport received 2.8 mm of rain from 0900‍–‍1130 that morning. At 1133, heavy rain began to fall and continued to fall until 1155, 1 minute prior to the aircraft landing (Figure 6). During this 23‑minute period, 15 mm of rain was recorded. This represented a rainfall rate of 39 mm/hr, significantly higher than the threshold of 10 mm/hr required to be classified as heavy rain. Satellite imagery and CCTV footage confirmed that significant rain was observed at and around the aerodrome prior to the aircraft landing.

Figure 6: Bankstown Airport recorded rainfall 1130‍–‍1215

Source: ATSB

Runway surface condition

While the runway surface condition was reported as wet on the ATIS, the ATSB identified evidence of standing water on the runway at the time VH‑SQY landed. CCTV footage recorded the latter portion of the landing, during which spray was observed being ejected from beneath the aircraft during its ground roll (Figure 7).

Figure 7: CCTV images of water spray during landing

Source: Google Earth and Bankstown Airport, annotated by the ATSB

Additionally, a photograph of runway 11C taken at 1202, 6 minutes after the runway excursion, showed water on the runway (Figure 8). A subsequent photograph taken 9 minutes later showed that the amount of water on the runway had visibly reduced over this time.

Figure 8: Runway 11C surface after incident

Source: Supplied, annotated by the ATSB

Air traffic control observations

Air traffic control did not advise the pilot of any tailwind when giving an initial landing clearance at 1154 or when a subsequent landing clearance was given at 1155. After the incident, the next aircraft to arrive at Bankstown was advised of an occasional tailwind of 5 kt prior to receiving a landing clearance for runway 11L at 1159.

While the ATSB did not interview the air traffic controllers who were on duty at the time, Airservices Australia advised that an internal occurrence review was conducted into the incident which reported that a tailwind was not observed by the controller at the time that a landing clearance was given to VH‑SQY. They further advised that it was likely that controllers were using instantaneous wind readings in conjunction with visual observations to inform their assessment of wind conditions.

Recorded data

The ATSB analysed flight data recorded by the aircraft’s Garmin G1000 avionics. This data recorded specific flight and system parameters every 1‍–‍2 seconds throughout the flight, including during the landing and runway excursion. The aircraft was not fitted with a flight data recorder or a cockpit voice recorder, nor was it required to have them installed.

The data from the G1000 showed that on final approach, the aircraft was at 50 ft above ground level (AGL) approximately 240 m before the displaced threshold before arriving at an aim point short of the first touchdown marker (Figure 9).

Figure 9: VH-SQY flight path and landing

Source: Google Earth, annotated by the ATSB

At 1156:32 the aircraft commenced the round out and flare approximately 135 m past the displaced threshold at which point the aircraft’s groundspeed was 102 kt. The recorded data did not contain a specific data point to indicate when the aircraft was on the ground, however a positive G indication, consistent with a touchdown, was recorded at 1156:38, 430 m past the displaced threshold. At this time, the aircraft’s groundspeed had decreased to 93 kt. The aircraft’s groundspeed continued to decrease with no discernible change to the deceleration rate until 1156:58, at which time 140 m of the pavement remained. Subsequently, the aircraft departed the end of the runway at 1157:05 at a groundspeed of 28 kt.

The data recorded both the indicated airspeed and groundspeed of the aircraft, enabling the headwind and tailwind component of the wind to be determined (Figure 10). At approximately 550 ft, the aircraft was experiencing a headwind of 9 kt. As the aircraft descended on the approach, the headwind decreased prior to switching to a tailwind at approximately 250 ft. As the aircraft continued its descent, the tailwind increased up to a maximum of 6.3 kt when the aircraft was 50 ft AGL, consistent with the wind recorded at the airport. After peaking at 50 ft, the tailwind decreased throughout the remainder of the landing.

Figure 10: VH-SQY calculated wind on final approach and landing

Source: ATSB

Bankstown Airport

Runway environment

Bankstown Airport’s runway environment consisted of 3 parallel runways. The centre runway, 11C/29C, was the longest runway. It was typically used for arrivals and departures under the instrument flight rules (IFR) due to associated instrument approach and standard instrument departure procedures. The runway was un-grooved, and the runway surface was asphalt.

The En Route Supplement Australia (ERSA) provided information on Bankstown Airport, including runway dimensions, take-off and landing distances, runway slope and local procedures for operating at the aerodrome. The ERSA advised that runway 11C had a displaced threshold of 97 m and a landing distance available (LDA) of 1,259 m. The pavement surface continued for a further 60 m, however this section was not permitted to be used for either take-off or landing. The reciprocal runway 29C had no displaced threshold and a LDA of 1,356 m.

Runway longitudinal slope

The ERSA described the longitudinal slope of the runway as:

Slope W end 0.5% down to E. Centre 0.2% up to E. E end 0.2% down to W

The ERSA did not contain information on how this description should be used to determine a runway slope value for landing and take-off performance calculations. Bankstown Airport survey information detailed how the slope described in the ERSA was constructed (Figure 11). The survey information identified that the threshold‑to‑threshold slope was 0.15% up to the east. This would have been the runway slope applicable to performance calculations for an aircraft landing on runway 11C.

Figure 11: Runway 11C longitudinal slope

Source: Google Earth, annotated by the ATSB

Bankstown Airport maintained an aerodrome manual, as required under Part 139 (Aerodromes) of the Civil Aviation Safety Regulations (CASRs). The manual included technical information regarding the longitudinal slope of runway 11C/29C and reported:

• The runway slope was 1.12% slope to the south-west.
• In a different section of the manual, the runway was a code 3 non‑precision approach runway with an overall longitudinal slope of 1.34%, whereby the current standard was 1%. This was identified as non‑compliant with the current standards, however, was being used under grandfathering provision of the regulations.

An independent aerodrome technical inspection (ATI) was conducted for the aerodrome in December 2023. The inspection included a visual inspection of all movement area pavements to appraise compliance with CASR Part 139. The inspection reported that the overall longitudinal slope of runway 11C/29C was 1.34%, consistent with one of the statements in the aerodrome manual, but not consistent with survey information.

The ATSB sought clarification from Bankstown Airport regarding the variation between the runway 11C/29C longitudinal slopes reported in the ERSA, aerodrome manual, ATI and survey documentation. Airport personnel advised that the survey information was correct and that the aerodrome manual and ERSA would be amended.

Runway transverse slope and drainage

The aerodrome manual stated that the transverse slope values required by the CASR Part 139 Manual of Standards (MOS) had not been exceeded for runway 11C/29C.

The ATI did not contain a statement regarding assessment of the transverse slope of the runway. However, it did state that the drainage infrastructure of the runway:

appeared to be operating effectively at the time of inspection within minimal areas of ponding observed.

The ATSB was advised by Bankstown Airport that the ATI offered only a visual inspection of the transverse slope and that the slope was typically determined during the runway design process. Both the pilot and the operator reported that discussions had occurred between representatives of the operator and the airport regarding what they considered to be poor runway drainage. Bankstown Airport management advised that they were not aware of any conversations where this was discussed.

Assessment of runway surface condition

Bankstown Airport had procedures for assessing and reporting runway surface conditions and associated braking action under the global reporting format (GRF). The procedures defined the steps required to determine the runway condition report (RCR), a standardised report relating to runway surface conditions, and their effect on an aircraft’s landing and take-off performance.

The RCR was provided in 2 parts, a runway condition code (RWYCC) and a surface description. Additionally, each runway was divided into approximate thirds, with an RCR being determined separately for each of these thirds. The RWYCC was initially assigned based on a runway surface description (Table 1).

Table 1: Runway surface description to assign initial RWYCC

After initial assignment of an RWYCC from the runway surface description, receipt of 2 or more pilot reports of braking action less than that expected, could result in the description being downgraded. 

During tower hours, ATC was trained to determine if the runway was completely dry or wet and could create the RCR when conditions were dry or wet for the whole runway. Further to this, ATC was able to request the aerodrome reporting officer (ARO) to conduct a runway inspection to assess the runway surface. AROs were also required to carry out inspections after severe weather events, so long as the weather conditions would not pose a safety hazard.

The assessed RCR was advised on the ATIS. Additionally, if the runway surface condition was determined to be slippery wet, or had standing water, a NOTAM[6] was required to be submitted containing details of the adverse RCR.

The ATIS ‘Echo’ current at the time of the incident contained an RCR of 5,5,5, whole runway wet (see the section titled Reported conditions). ATC did not report having received any reports of adverse braking from other pilots prior to the incident that would have required the RCR to be amended. After the incident, while attending to VH‑SQY, the ARO advised the tower controller that there was no observed standing water on runway 11C. Subsequently, when the ATIS was updated to ‘Foxtrot’ at 1213 to advise of a disabled aircraft, the RCR remained the same.

Air traffic control requirements

Airservices Australia’s Manual of Air Traffic Services (MATS) included procedures for ATC relating to the issuing of an ATIS, and the conditions under which it should be revised. These procedures required that an ATIS include:

Surface wind direction and speed, including significant variations

Wind direction was further required to be reported as one of:

SINGLE MEAN DIRECTION;

TWO VALUES representing variation in wind direction, whenever:
i) the extremes in wind direction vary by 60 degrees or more; or
ii) the variation is operationally significant (e.g. the variation is less than 60 degrees, but the variation from the mean results in either a tailwind, and/or significant crosswind component on a nominated runway) (e.g. WIND VARYING BETWEEN [DIRECTION] AND [DIRECTION]);

VARIABLE, where it is not possible to report a mean wind direction, such as:
i) in light wind conditions (3 kt or less); or
ii) the wind is veering or backing by 180 degrees or more
(e.g. passage of thunderstorm, or localised wind effect).

The wind component of the ATIS was also required to:

Quote significant crosswind and any tailwind as:
a) MAXIMUM CROSSWIND (speed) KNOTS [RUNWAY (number), if
applicable]; and
b) MAXIMUM TAILWIND (speed) KNOTS [RUNWAY (number), if applicable].

MATS also advised on wind limitations when nominating a runway. When the runway was dry, a runway could not be nominated for use when the tailwind exceeded 5 kt. When the runway was not completely dry, a runway could not be nominated for use when there was any tailwind component.

ATC was required to revise an ATIS and assign a new code letter when certain items changed and were expected to remain that way for at least 15 minutes. This included changes to an RCR or when wind direction varied by 10°.The CASR Part 179 (Air Traffic Services) MOS also required that:

Changes to ATIS wind information must be provided to pilots with a take-off or landing clearance if it is considered that it would be of significance to the aircraft operation.

Landing performance

Aircraft flight manual

Landing performance data

The Cessna 510 AFM contained performance data for use in calculating the landing distance required (LDR) at a destination aerodrome. This data was contained in 2 sections. The first section provided data to calculate the landing distance when the runway was forecast to be dry. This section of the AFM was marked as approved by the United States Federal Aviation Administration (FAA), the organisation that originally issued the aircraft’s type certificate. The performance data was based on several assumptions including:

• the landing was on a paved, dry runway
• landing preceded by a steady 3° angle approach down to the 50 ft height point with airspeed at V_REF[7] in the landing configuration
• maximum wheel braking was initiated immediately on nose wheel contact and continued throughout the landing roll
• winds were to be taken as the tower winds 32.8 ft (10 m) above runway surface
• factors of 50% of the headwind, and 150% of the tailwind had been applied to winds.

The second section of the AFM provided performance data for landing on wet, slush, snow‑ and ice‑covered runways. This section was marked as advisory information and stated that:

The following information is considered the most accurate and practical guidance material available for wet and contaminated runway operations. This advisory information is not FAA approved.

The section also stated that:

The published limiting maximum tailwind component for this airplane is 10 knots, however, landings on precipitation covered runways with any tailwind component are not recommended.

Advisory and unapproved information

The Civil Aviation Safety Authority (CASA) Advisory Circular (AC) 21‑34 Aircraft flight manuals contained information relating to approved and unapproved sections of an AFM and advised that:

 - Approved parts of the AFM are approved by the applicable national aviation authority (NAA), based on the type certification requirements effective at the time of certification. 

 - Unapproved parts of the AFM are provided by the manufacturer additionally, as deemed necessary for the safe operation of the aircraft, and cannot conflict with approved parts of the AFM. Each approved part of the AFM is clearly distinguished from any unapproved part of that AFM.

CASA also published Civil Aviation Safety Regulation (CASR) Part 135 Acceptable means of compliance and guidance material (AMC/GM) - Australian air transport operations—smaller aeroplanes which provided guidance on complying with required performance data calculations. It stated that:

Some performance information presented in AFM or AFM supplements may be advisory information only and should not be used to determine performance in compliance with the provisions of regulations 135.345 and 135.350 [Take-off performance and Landing performance]. Caution should be exercised when using advisory material or when using third-party performance calculations as the results may not be based on the required AFM provided certification data.

Similar advice was contained in CASR Part 121 Acceptable means of compliance and guidance material (AMC/GM) - Australian air transport operations—larger aeroplanes.

The ATSB sought clarification from CASA as to the applicability of data from unapproved sections of an AFM for use in flight planning. They advised that this data could not be used for performance planning, in part because the advisory information was not developed in accordance with standardised conditions.

Landing distance requirements

Safety Margins

Take-off and landing performance data contained in an AFM was obtained through formal testing using specific criteria. It was therefore unlikely that a pilot could replicate the testing performance during normal flying conditions. For this reason, additional distance was added to the calculated distance to provide a safety margin. While only recommended for private operations, the use of safety margins was required when conducting air transport operations.

CASA AC 91-02 provided guidance on the purpose of safety margins including that:

These additional safety margins mitigate risks associated with a range of issues that impact on aircraft performance, including but not limited to:

 - pilot inaccuracies compared to performance flight testing (excess landing speed, excess height over threshold, increased float before touchdown, delayed use of braking and deceleration devices, inaccurate application of maximum braking techniques)

 - runway characteristics

 - aerodrome density altitude

 - changed external drag configuration of the aeroplane

 - underperforming engine compared to that used for performance testing.

Landing distance factor

CASR Part 135 - Australian air transport operations—smaller aeroplanes, under which this flight was conducted, prescribed certain landing performance requirements. As part of these requirements, if the aircraft was a jet‑driven, multi‑engine aircraft with a maximum take‑off weight (MTOW) of greater than 2,722 kg, such as the Cessna 510, then the performance calculations were required to be conducted in accordance with CASR Part 121, the regulations for larger aeroplanes.

CASR Part 121 required that a pilot determine that the runway at the planned destination aerodrome had sufficient landing distance available (LDA) to bring the aircraft to a stop. For a jet‑engine aeroplane, such as the Cessna 510, the aircraft was required to be shown to able to stop within 60% of the full LDA (Table 2). This calculation was required to be conducted both prior to departure using forecast conditions and in‑flight when actual aerodrome conditions were obtained.

If the runway surface condition was expected to be wet or contaminated, an additional 115% factor was required to be applied as a further safety margin. If the AFM contained landing performance data specific to wet or contaminated runways an alternative calculation was available. However, if this data was contained in an advisory or unapproved section of the AFM, as it was for the Cessna 510, it was not eligible to be used (see the above section titled Advisory and unapproved information).

Table 2: Summary of landing distance required calculations

The regulations stated conditions to be considered when calculating LDR, which included:

• runway surface condition
• forecast wind speed and direction. Unless otherwise accounted for in the performance data in the AFM, 50% of the headwind and 150% of the tailwind
• expected runway to be used
• expected landing weight
• aerodrome elevation
• runway slope if greater than 1%.

Actual landing distance data

CASR Part 121 allowed a single, less restrictive landing distance factor of 15% to be used when determining LDR in‑flight for aircraft where the AFM contained actual landing distance (ALD) data. CASR Part 121 AMC/GM listed key points surrounding the use of ALD including:

 - Actual landing distance information is intended to show landing performance that can realistically be achieved by flight crews in commercial operations.

 - This is distinct from landing performance demonstrated by test pilots during flight tests for aircraft type certification.

The AMC/GM also stated that to be classified as ALD data, performance data was required to be accordance with International Civil Aviation Organisation (ICAO) Annex 8 standards and that:

the applicability is also limited to aeroplanes intended for the carriage of passengers or cargo or mail in international air navigation. These are known in some States as transport category aeroplanes. This has resulted in some aeroplanes that are seemingly captured by the requirements of ICAO Annex 8 Part IIIB not having performance data that is required to be in accordance with those requirements.

CASA confirmed that performance data in the Cessna 510 AFM did not constitute ALD data.

Landing in very wet conditions

CASA guidance material contained advice for operators and flight crew on landing in very wet conditions which stated:

Operators and flight crews should be aware that the landing distance factors mentioned above – whether based on type certification testing or actual landing distance data provided by OEMs [original equipment manufacturer] separately – may not provide adequate stopping distance in very wet but not yet contaminated runway surface conditions.

Issues that contribute to such incidents include runway conditions such as texture (polished or rubber contaminated surfaces), drainage, puddling in wheel tracks and active precipitation. For un-grooved runways, wheel braking may be degraded when the runway is very wet. Research conducted by the FAA has indicated that 30 to 40 percent of additional stopping distance may be required in certain cases where the runway is very wet, but not yet classified as contaminated.

In order to manage some of the risks associated in operating to very wet runways, it is recommended that operators consider the landing safety factor of 1.15 (which is the difference between 1.67 and 1.92 for type certification data and the value mentioned in the actual landing distance data) to be a minimum value.

Flight planning software

The operator utilised third party flight planning software from Aircraft Performance Group (APG) for performance calculations. Access to this software and underlying performance data was available to pilots through the following means:

• the APG iPreFlight App available on the pilot’s tablet
• the APG Atlas website, available via a computer with an internet connection
• tabulated data for individual aerodromes, available as PDF documents via the electronic flight bag (EFB) on the pilot’s tablet.

The APG iPreFlight App and APG Atlas website provided an interface into which pilots entered the destination airport and runway, forecast or actual conditions and estimated landing weight. Additionally, options were available to configure how the landing distance was calculated which included:

• landing distance factor: 60%, 80% or unfactored
• wet runway calculation method: 115% or use of AFM advisory data
• other runway surface conditions: including 0.125 inches (3 mm) of water, snow and slush.

Of these options, 80% landing distance factor was selected by default.

The tabulated data provided both take-off and landing performance data for an individual airport. The landing performance data (Figure 12) was presented as a maximum landing weight (MLW) permitted and the LDR required at this weight, across a discrete set of temperature and wind conditions. Tables for both dry and wet (115%) runway conditions were provided, with 60%, 80% landing distance factors in addition to unfactored data. If the landing distance required for the selected runway was not available, then ‘NA’ was listed as the MLW and the actual distance required was specified.

Figure 12: Extract of tabulated landing performance data for runway 11C

Source: Supplied, annotated by the ATSB

Type rating training

The pilot completed a type rating for the Cessna 510 with Air Link, a company related to the operator. The ground school component of the type rating contained a section on aircraft performance, which included the calculation of landing distance required. The trainer reported that this training included:

• a review of relevant CASR Part 135 and CASR Part 121 requirements for landing distance calculation
• use of the advisory section of the AFM to calculate landing distance required under CASR Part 135 on a wet and contaminated runway
• use of the APG flight planning software on a company EFB to conduct the same landing distance calculation.

The pilot recalled that, during ground training and during the type rating test, an 80% landing distance factor was used. The trainer confirmed that use of an 80% landing distance factor was taught during the type rating. Air Link advised that this factor was adopted as a safety margin in consultation with CASA when both Air Link and AirMed were initially approved under previous regulations as there was no factoring required under the previous legislation.

Operator procedures

Procedures for determining landing distance available

The operator advised the ATSB that, consistent with the training provided by Air Link, pilots were required to apply an 80% landing distance factor when determining LDA. The pilot also reported that they consistently used an 80% factor.

The operator’s procedures contained requirements for pilots when determining landing distance and advised that:

The means to determine maximum allowable take-off and landing weights are based upon:

 - airport characteristics consisting of airport elevation, runway gradient and length, runway contaminants, obstructions within the take-off path,

 - airport/environmental conditions consisting of temperature, wind and pressure altitude

 - aircraft configurations consisting of power settings, flap settings, bleed configurations and MEL [minimum equipment list] inoperative components.

 - specified factoring (set by company, in accordance with CAO 20.7.1 [historical regulation])

The procedures further stated that:

The calculation of aircraft performance must be considered prior to dispatch, as part of the pre-flight planning process. 

Additional procedures for landing on a wet runway

The operator provided additional guidance for operating on wet or contaminated runways which stated:

Due to the large number of variables involved no exact formula has so far been found, but an empirical result of a 15% increase in the overall distance required has been accepted in the US [United States] and in other countries as providing an acceptable correction for landing.

It is desirable to apply some correction for take-off when the runway is considered to be significantly wet, so as to provide a distance margin to offset the reduced braking likely to arise in the accelerate-stop manoeuvre, whenever the take-off is likely to be distance limited. For take-off and landing, the 15% increase is considered to be appropriate.

It also provided the following guidance for adjusting landing distance available when expecting to land on a wet runway:

Obtain the LDA, and using only 0.85 X LDA as the effective distance available, derive a ‘wet’ length limited landing weight from the general chart in the manual.

Derive the approach climb limit in the usual way, and observe this and the structural limit.

If the ‘wet’ length limit is the most restrictive consider using another runway to provide better protection.

Note: That in both the take-off and landing cases, the use of 1.67% of the overall distance will provide an acceptable margin in that portion of the distance used. In most cases, the distance available for decelerating will be approximately doubled by this means.

Stabilised approach criteria

The operator’s procedures contained requirements for a stabilised approach that stated:

An approach to land must be stabilised by the FAF [final approach fix] in IMC [instrument meteorological conditions] and by 300 ft above the airport elevation in VMC [visual meteorological conditions]. An approach is stabilised when all of the following criteria are met:

 - The aircraft is on the correct flight path.

 - Only small changes in heading and pitch are required to maintain the correct flight path.

 - The aircraft speed is not more than V_REF +15 IAS [indicated airspeed] and not less than V_REF +5.

 - The aircraft is in the correct landing configuration.

 - The sink rate is no greater than 1,000 ft/min.

 - Power setting is appropriate for the configuration.

 - All briefings have been conducted.

The operator further advised that the expectation was for pilots to fly the aircraft within the specified range as they approach the field, reducing airspeed so that the aircraft was at V_REF at touchdown.

Pilot assessment of landing distance available

Pre-departure

The pilot reported that prior to departure, as part of their pre-flight planning process they assessed the landing distance available at Bankstown using the APG tabulated landing distance data. Given the forecast conditions (see the section titled Forecast conditions), they determined that the runway would likely be wet on arrival, therefore they consulted the section of the data for a wet runway. Rather than using the wind forecast from the TAF, they used for a more conservative 5 kt tailwind on arrival. Using the 80% landing distance factor in accordance with operator procedures, a more conservative temperature of 29° and interpolating between the values for nil wind and a 10 kt tailwind, they determined that the maximum landing weight was approximately 7,700 lb for runway 11C at Bankstown (Figure 13).

Figure 13: Pre-departure landing distance calculation

Source: Supplied, annotated by the ATSB

The pilot had also determined that the landing weight of VH‑SQY was expected to be 7,089 lb based on calculated take‑off weight and anticipated fuel consumption. Therefore, as the aircraft was expected to be below the maximum landing weight, the aircraft could depart.

In-flight

The pilot reported that upon receiving the ATIS they again conducted an assessment of the landing distance available at Bankstown using APG tabulated data and the reported conditions (see the section titled Reported conditions). Considering that the variable 8 kt wind reported could be all tailwind, the pilot used a more conservative 10 kt tailwind. Again, using the 80% landing distance factor in accordance with operator procedures, the pilot determined that the maximum landing weight was 7,401 lb. The pilot recalculated the estimated landing weight with the current fuel load as approximately the same as that estimated prior to take‑off.

As the aircraft was expected to be below the weight assessed in the chart, they determined that the landing could be conducted. In addition, the pilot advised that they had decided that if they observed a tailwind greater than 5 kt during the approach, they would request to land on the reciprocal runway 29C. At this point, the pilot also determined 89 kt as the required V_REF for the aircraft, for the expected landing weight.

ATSB assessment of landing distance required

The ATSB calculated the landing distance required using the 60% landing distance factor required for operating the Cessna 510. A pre‑departure assessment was conducted using conditions from the TAF available prior to departure from Narrandera. An in‑flight assessment using the conditions reported on the ATIS prior to descent was also performed (Table 3). The calculations were conducted using APG flight planning software, software from the manufacturer and using performance data from the AFM with the results consistent across the 3 methods. The pre‑departure calculation determined that a landing distance of approximately 1,230 m was required on runway 11C, less than the 1,259 m available. However, the in‑flight calculation determined that approximately 1,530 m was required, greater than that available.

Table 3: ATSB assessment of landing distance required

Wet or contaminated runway landing performance

Dynamic aquaplaning

Dynamic aquaplaning (also known as hydroplaning) can occur when an aircraft lands on a runway contaminated with standing water, slush or wet snow. Above a certain groundspeed and with sufficient contaminant, the tyre is lifted off the runway surface. This can have serious adverse effects on ground controllability and braking efficiency.

The depth of standing water required for dynamic aquaplaning to occur is generally accepted to be 3 mm (approximately 0.125 inches). This is also the depth required for a runway surface condition to be considered contaminated rather than wet.

The minimum speed above which dynamic aquaplaning can occur is a function of tyre pressure. Two different speeds are often quoted depending on whether the tyre is rotating. A lower speed is considered for a non‑rotating tyre as is the case immediately on touchdown (Table 4).

Table 4: Generally accepted Cessna 510 main landing gear aquaplaning speeds

The operator’s procedures contained guidance for pilots regarding aquaplaning and stated:

The formula used to determine the speed at which a tyre is likely to hydroplane [aquaplane] after touchdown on a wet runway is: Hydroplane speed = 7.7√𝑇𝑦𝑟𝑒𝑃𝑟𝑒𝑠𝑠𝑢𝑟𝑒𝑃𝑆𝐼

From the above formula, the Citation Mustang’s nose gear hydroplane speed is about 86 knots and the main gear about 72 knots. Above these speeds hydroplaning may occur.

The manufacturer advised that the aircraft’s minimum aquaplaning speed for performance calculations was 84 kt. The operator’s speed was consistent with that for a non‑rotating tyre, while the manufacturer’s speed was consistent with that for a rotating tyre.

Recorded data showed that the groundspeed of the aircraft was above both the rotating and non‑rotating tyre minimum aquaplaning speed until approximately 360 m of LDA remained (Figure 14).

Figure 14: Recorded groundspeed of VH‑SQY during landing

Source: Google Earth, annotated by the ATSB

Effect of reduced tyre tread

The ATSB reviewed research to assess the effect of reduced tyre tread on braking effectiveness when landing on a wet or contaminated runway. Research conducted by Leland & Taylor (1965) An investigation of the influence of aircraft tire-tread wear on wet‑runway braking concluded that:

On the wet runway, a gradual degradation in braking effectiveness was experienced up to about the 80 percent worn tire tread condition, where the wet-runway friction coefficients dropped markedly.

The completed worn tire was observed to develop, at higher speeds, only about one‑half the braking effectiveness of a new tire.

The research from O’Callaghan (2023) Wet‑runway overruns: still a slippery problem included consideration of the effect of tyre wear on braking performance and stated that:

For the aircraft operator, tire wear is a most important factor … the available 𝜇𝐵 [coefficient of braking] in wet conditions decreases as a tire wears. For a typical aircraft-type, rib-tread tire, when groove depths have been reduced to about 20% or less of the unworn value, the remaining tread may be ‘flattened out’ under load and the tire may then behave as if smooth

Additionally, the research paper Aircraft tyre hydroplaning and how to analyse it in runway excursion events, van Es (2018) discussed the effect of tyre tread on the depth of water required for aquaplaning to occur and stated:

The tyre tread grooves act similar to the pavement macrotexture in draining the bulk water. When there is sufficient macrotexture on the surface and/or the tyre has a sufficient number of deep circumferential grooves, full dynamic hydroplaning will normally not occur, unless the water depth is at a level that both tyre grooves and runway macro texture cannot drain the water sufficiently quick enough…

Smooth tread tyres operating on smooth pavements surfaces require the smallest fluid depth for dynamic hydroplaning, whereas rib treads tyres operating on an open textured or grooved-pavement surface require the largest fluid depths.

Related occurrences

Australia

The ATSB occurrence database contained 263 instances of runway excursions on landing in Australia between 2020‍–‍2024. The majority of these involved the aircraft veering off the runway rather than overrunning.

Of these occurrences, 3 included mention of standing water leading to aquaplaning, one of which involved a Cessna 525 operated by Air Link landing on runway 11C at Bankstown in 2022. The ATSB did not investigate these occurrences.

In 2020 the ATSB investigated the runway excursion of a Fokker 100 landing at Newman, Western Australia (AO‑2020‑002). It was found that poor braking effectiveness in wet conditions resulted in the aircraft overrunning the runway.

In 2008, the ATSB published a two‑part research report (AR‑2008‑018) titled Runway Excursions with the objective of analysing international and Australian trends in runway excursions. Part 1 of the report explored the contributing factors associated with runway excursions between 1998 and 2007. Water‑affected and contaminated runways was one of the contributing factors identified.

International

The ATSB identified the following occurrences of runway excursions associated with a wet or contaminated runway:

Runway excursion of a Beech 95-C55 on 12 January 2023 (NTSB WPR23LA089)

During the landing roll, the pilot applied the brakes but discovered that there was more standing water on the runway than expected, resulting in the airplane aquaplaning. It was determined that the wet runway contributed to the aircraft overrunning the runway.

Runway excursion of a Learjet 36 on 9 September 2022 (NTSB WPR22LA344)

During a landing following recent rainfall, the pilot reported that the aircraft did not decelerate normally and subsequently overran the runway. It was determined that a fast landing on a wet runway resulted in the airplane aquaplaning during the landing roll.

Safety analysis

Introduction

On the morning of 11 January 2025, a Cessna 510, registered VH‑SQY, was being used to conduct a medical air transport flight from Narrandera Airport to Bankstown Airport, New South Wales. During the landing at Bankstown, the pilot experienced reduced braking performance, and the aircraft overran the end of the runway.

This analysis will discuss the operator’s procedures for determining landing distance and the pilot’s use of these procedures. The environmental conditions at the time and the actions performed during the landing are also examined. Additionally, the analysis will consider the reporting of conditions at Bankstown Airport and the airport’s runway environment.

Pilot training

The pilot completed a type rating for the Cessna 510 with Air Link, a company related to the operator. The performance component of the ground school incorporated the use of Part 135 regulations. 

These regulations required the application of a 60% landing distance factor for a twin‑engine jet aircraft over 2,722 kg, such as the Cessna 510, when determining the landing distance required at a destination aerodrome. However, pilots were taught to use an 80% landing distance factor for this calculation. Use of this factor reduced the safety margin applied to mitigate issues that impacted the aircraft’s ability to achieve published landing performance. Furthermore, pilots were taught to use the aircraft flight manual (AFM) wet runway performance data for wet runway landing distances. However, this data was not permitted to be used for flight planning as it was from an advisory section of the AFM.

Operator procedures

Both the operator and the pilot reported that, consistent with the type rating training, a landing distance factor of 80% was used when determining the required landing distance. However, a 60% factor was required for these types of operations. 

The operator’s procedures contained guidance for calculating the required landing distance during the pre‑flight planning. However, while they referred to the application of specified factoring when determining landing distance available, they did not define what this factoring was. Additionally, the procedures did not advise that a landing distance assessment was required in‑flight in addition to prior to departure. 

Furthermore, the operator’s flight planning software provided 2 options for determining the landing distance required: 60% and 80% landing distance factors. However, the 80% option was incorrectly selected by default when using the APG iPreFlight App. In addition, an option to use the AFM advisory wet runway performance data was also available on the App. However, the operator’s guidance did not specify that this was not permitted to be used for flight planning.

Flight planning

Pre departure planning

Prior to departure, the pilot assessed the landing distance available at Bankstown Airport using tabulated data extracted from the flight planning software. Using the Terminal Area Forecast (TAF) at Bankstown Airport, they determined that the runway would likely be wet. While the TAF forecast a headwind on runway 11C, the pilot used a more conservative assumption of a 5 kt tailwind, despite the aircraft flight manual (AFM) stating that landing with a tailwind was not recommended on a wet runway.

The pilot used an 80% landing distance factor, as they had been taught, and determined that sufficient landing distance was available under these conditions. However, the use of a 60% factor was required for this flight.

The ATSB calculated the landing distance required using the conditions forecast in the TAF and the required 60% landing distance factor. It was determined that sufficient landing distance did exist at Bankstown due to the forecast headwind on runway 11C. Therefore, while incorrect data was used to conduct the assessment, the pilot correctly determined that a departure was possible.

In-flight planning

Prior to descent, the pilot obtained aerodrome conditions from the Automatic Terminal Information Service (ATIS). This information included that the wind was variable at 8 kt, the runway 11C was in use and the whole runway was wet. Using this information, the pilot again consulted the 80% landing distance factor tabulated data and determined that there was sufficient landing distance available to attempt a landing with up to a 10 kt tailwind. 

However, this assessment also required that a 60% factor was used. Use of this factor would have identified that insufficient landing distance was available to plan for a landing on runway 11C.

If the pilot had determined that insufficient landing distance was available to attempt the landing, options were available to hold for the weather to improve or request more up‑to‑date weather information from air traffic control (ATC).

Landing and runway excursion

Tyre condition

During the daily inspection of the aircraft, the pilot inspected the aircraft’s main landing gear (MLG) tyres and assessed them as serviceable. However, an inspection after the incident identified that the tyres were below the tyre manufacturer’s guidance for normal removal wear, and less than 10% of the tread was remaining.

Research showed that the braking performance of an aircraft on a wet runway degraded as its tyres wore, with a marked increase in effect when reaching 20% of the original tread depth. Additionally, the depth of tread on a tyre influenced the amount of standing water required to support aquaplaning. Consequently, while permitted for flight, the low tread on the aircraft’s MLG tyres likely had an adverse effect on the aircraft’s braking performance when landing on a wet or contaminated runway.

Meteorology

Rainfall and standing water

At Bankstown Airport there was a period of very heavy rain which continued until 1 minute prior to the landing. Given the intense nature of this rainfall, it is likely that parts of the runway were contaminated with standing water. Furthermore, video and photographic evidence showed that the aircraft encountered standing water during the landing roll. The pilot was not aware of the recent rainfall and had assessed the runway surface as wet but not contaminated, observing no standing water. This was consistent with the runway surface condition reported by the Automatic Terminal Information System (ATIS). Standing water on the runway provided an environment for aquaplaning. In addition, guidance material advised that when landing in very wet conditions degraded wheel braking may require an additional 30‍–‍40% of stopping distance.

Tailwind

The ATIS reported that the wind was variable at 8 kt. The pilot recalled that they considered this and had planned to use the reciprocal runway if they observed a tailwind greater than 5 kt. The pilot further reported that they did not observe, nor were they advised by air traffic control (ATC), of a tailwind. However, while the wind was initially a headwind during the approach, this headwind decreased and became a tailwind as the aircraft approached the runway and commenced the landing.

Landing with a tailwind increased the landing distance required. Additionally, while the recorded tailwind was below the maximum permitted by the aircraft flight manual (AFM), landing with any tailwind component was not recommended by the manufacturer when landing on a wet or contaminated runway.

Landing sequence

The aircraft conducted the final portion of the approach to an aim point short of the first touchdown marker. At 50 ft AGL, the indicated airspeed was within 5 kt of the V_REF and within the operator’s stabilised approach criteria. However, due to the tailwind encountered, the aircraft’s groundspeed was higher than the airspeed and above both the rotating and non‑rotating dynamic aquaplaning speeds during the touchdown. Consequently, the aircraft likely experienced dynamic aquaplaning when encountering standing water on the runway. Throughout the landing roll, the aircraft’s groundspeed did not decrease below the non‑rotating minimum aquaplaning speed of 72 kt until approximately 900 m into the landing roll.

During the landing roll, due to the lack of braking performance, the pilot elected to release and reapply the brakes and cycle the anti‑skid system. While well intentioned, this action likely further decreased braking performance as brake pressure was released for portions of the landing, and the anti‑skid system was momentarily not operational while a self‑test was conducted. Furthermore, this was not in accordance with the AFM which required that maximum braking was maintained throughout the landing roll. Damage observed on the main landing gear tyres was consistent with them having locked up under braking.

In summary, while the individual contributions of standing water, tailwind, tyre condition and pilot braking action could not be ascertained, the combination of these factors resulted in the aircraft’s reduced braking performance and subsequent runway excursion.

Bankstown Airport

Reported conditions

Wind and nominated runway

The ATIS reported the wind as variable at 8 kt with runway 11C nominated for use and no maximum tailwind advised. The Manual of Air Traffic Services (MATS) did not permit the nomination of a runway when there was a tailwind component with a wet runway. Recorded data showed that a mean tailwind had been recorded during the 6 minutes prior to landing. However, the wind direction was highly variable over this time. MATS required that ATIS information be updated when changes to meteorological conditions were expected to remain for 15 minutes, supporting the decision to report the wind as variable, and maintain runway 11C as the nominated runway. It was also required that wind significant to aircraft operation was provided to pilots with a landing clearance. However, at the time the landing clearance was given, the instantaneous wind observed by air traffic control (ATC) may not have indicated a tailwind.

Runway condition code

The ATIS reported the runway condition code as ‘whole runway wet’, however it was likely that sections of the runway were contaminated with standing water. Had the runway condition included sections reported as 2/standing water, the pilot would have been alerted to the potential for reduced braking performance.

However, ATC observations could only be used to declare a runway as fully dry or wet. Downgrading the runway surface condition required either 2 less-than-good braking reports from pilots or physical inspection of the runway surface by the aerodrome reporting officer (ARO). On this occasion, no braking reports had been received, and the ARO had not inspected the runway. After the incident, the ARO reported that there was no standing water observed. However, some water had likely drained away by this time.

Runway environment

Runway slope and drainage

Survey information showed that the longitudinal slope was approximately 0.15% up when landing on runway 11C. While the aerodrome manual and the aerodrome technical inspection (ATI) report contained conflicting information, Bankstown Airport advised that the surveyed slope was correct. 

The En Route Supplement Australia (ERSA) entry provided runway slope information expressed by dividing the runway into 3 sections and reporting the slope for each section individually. As a result, it was unclear how to use runway slope values for performance planning. However, as the reported slopes, and the actual surveyed slope were not greater than 0.5%, use of one or a combination of these values would have had little, if any, effect on landing distance calculations.

The ATSB was advised of previous discussions regarding poor runway drainage at Bankstown Airport. However, an aerodrome technical inspection had assessed that the drainage of the runway was operating effectively. In addition, the aerodrome manual stated that the runway slope was designed in accordance with regulatory requirements. Furthermore, while standing water was likely present during the landing, no standing water was observed by the ARO after the incident. Therefore, it was likely that the runway drainage was operating effectively.

Findings

From the evidence available, the following findings are made with respect to the runway excursion involving Cessna 510, VH-SQY, at Bankstown Airport, New South Wales, on 11 January 2025. 

Contributing factors

• During the approach, an undetected tailwind was encountered and the aircraft landed with a groundspeed higher than the minimum aquaplaning speed. In addition, there was likely standing water on the runway and the aircraft’s main landing gear tyres were worn to limits resulting in reduced braking performance. Subsequently, the pilot cycled the anti‑skid system, likely further decreasing braking performance. In combination, these factors resulted in the aircraft departing the end of the runway.
• The type rating training provided by Air Link taught pilots to apply an incorrect landing distance factor, which reduced the safety margin when determining the required landing distance at a destination aerodrome. (Safety Issue)
• AirMed required pilots to apply an incorrect landing distance factor, which reduced the safety margin when determining the required landing distance at a destination aerodrome. Furthermore, its procedures were unclear on how the factor should be applied, when the assessment should be conducted and how runway surface condition should be considered. (Safety Issue)
• When determining the required landing distance at Bankstown, the pilot applied the incorrect landing distance factor prescribed by the operator. Subsequently, prior to descent and after obtaining the actual conditions at the aerodrome, the pilot did not identify that the landing distance available was insufficient to attempt the landing.

Safety issues and actions

Type training

Safety issue number: AO-2025-002-SI-01

Safety issue description: The type rating training provided by Air Link taught pilots to apply an incorrect landing distance factor, which reduced the safety margin when determining the required landing distance at a destination aerodrome. 

Landing distance assessment procedure

Safety issue number: AO-2025-002-SI-02

Safety issue description: AirMed required pilots to apply an incorrect landing distance factor, which reduced the safety margin when determining the required landing distance at a destination aerodrome. Furthermore, its procedures were unclear on how the factor should be applied, when the assessment should be conducted and how runway surface condition should be considered.

Safety action not associated with an identified safety issue

Additional safety action by Air Link Pty Ltd

Air Link is in the process of including a new section in the endorsement training around wet weather operations, including the associated limitations and the effect of worn tyres on aircraft performance. In addition, it has implemented a policy of tyre replacement when tread reaches 2 mm, representing 80% wear.

Additional safety action by AirMed

The operator conducted the following additional proactive safety action:

• Provided training to all crew addressing the effects of tailwind, correct anti‑skid use, tyre limits and landing technique.
• Updated defect reporting procedures to encourage earlier reporting of anticipated maintenance requirements.
• Implemented a policy of tyre replacement when tread reaches 2 mm.

Safety action by Bankstown Airport

Bankstown Airport amended the runway 11C/29C longitudinal slope information in the aerodrome manual to align with the En Route Supplement Australia depiction of runway slope.

Safety action by the Civil Aviation Safety Authority

While not in response to this occurrence, the Civil Aviation Safety Authority subsequently amended the performance section of the Part 121 guidance material as part of its continuous improvement process. These changes included:

• the addition of a section specifying that landing performance must be checked both pre‑flight and in‑flight
• advice that actual landing distance data cannot be used to satisfy in-flight replanning operations
• provision of a list of known aircraft types, including the C510, that must not use actual landing distance data for in-flight landing distance calculations.

Glossary

Sources and submissions

Sources of information

The sources of information during the investigation included:

• the pilot and operator
• the manufacturer
• Air Link
• Bureau of Meteorology
• Civil Aviation Safety Authority
• Bankstown Airport
• Airservices Australia
• Aircraft Performance Group

References

Civil Aviation Safety Authority (2022). Aircraft flight manuals (advisory circular AC 21-34 v1.1), https://www.casa.gov.au/sites/default/files/2021-08/advisory-circular-21-34-aircraft-flight-manuals.pdf, CASA, accessed 1 April 2025.

Civil Aviation Safety Authority (2022). Guidelines for aeroplanes with MTOW not exceeding 5 700 kg - suitable places to take off and land (advisory circular AC 91-02 v1.2), https://www.casa.gov.au/guidelines-aeroplanes-mtow-not-exceeding-5-700-kg-suitable-places-take-and-land, CASA, accessed 8 May 2025.

O’Callaghan, J. (2023, August). Wet-runway overruns: still a slippery problem.  Presentation to the International Society of Air Safety Investigators (ISASI), Nashville, Tennessee, USA.

Van Es, G. (2018). Aircraft tyre hydroplaning and how to analyse it in runway excursion events. Presentation to the International Society of Air Safety Investigators (ISASI), Dubai, United Arab Emirates.

Leland, T. J., & Taylor, G. R. (1965). An investigation of the influence of aircraft tire-tread wear on wet-runway braking (Vol. 2770). National Aeronautics and Space Administration, Washington, D.C., USA

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 and operator
• the manufacturer
• Air Link
• Civil Aviation Safety Authority
• Bankstown Airport
• Airservices Australia
• Aircraft Performance Group
• National Transportation Safety Board.

Submissions were received from:

• the pilot
• the operator
• Air Link
• Bankstown Airport
• 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.

Investigation summary

What happened

On 2 November 2024, a GippsAero GA8-TC Airvan, registered VH-IDM and operated by Wave Air, was being used to conduct a scenic flight from Whitsunday Airport (Shute Harbour), Queensland. On board were a pilot and 7 passengers. During the landing the aircraft departed the upwind end of the runway before entering marshy ground and coming to a stop in a ditch. Neither the pilot nor any of the passengers were injured and the aircraft was substantially damaged.

What the ATSB found

The ATSB identified that the aircraft's approach was above profile with a high airspeed and the pilot had an incorrect understanding of the required approach speed. Subsequently, the pilot did not initiate a go‑around, resulting in a landing beyond the planned touchdown point. Additionally, despite having sufficient landing distance remaining, the pilot did not apply sufficient braking to prevent the aircraft departing the runway

It was also determined that the training, supervision and checking flights conducted by Wave Air did not identify that an excessive approach speed was routinely being used by the pilot. Additionally, the pilot’s initial training was not fully completed, and they were not assessed on several abnormal and emergency procedures prior to operating unsupervised.

The ATSB also identified that Wave Air’s weight and balance system used an incorrect empty weight moment arm to calculate the aircraft's centre of gravity, and passengers were not weighed in accordance with its procedures. Finally, it was determined that the decision height for assessing whether an aircraft met Wave Air’s stabilised approach criteria was too low.

What has been done as a result

Wave Air has taken the following proactive safety action:

• The operations manual was modified to require that the stabilised approach criteria be met by 300 ft above airport elevation in visual meteorological conditions and 500 ft in when operating under instrument flight rules.
• The empty weight moment arm of the aircraft was corrected in the weight and balance system and the data of other aircraft was reviewed.
• Passenger scales have been serviced and made accessible for routine passenger weighing in accordance with the operator’s procedure.
• The training and checking manual has been updated to more precisely detail training criteria.
• A new head of training and checking has been appointed.
• Pilots are required to complete examinations prior to commencing in command under supervision (ICUS) training and operating unsupervised.
• Updates have been made to the remedial training processes.
• 6-month flight reviews are now required for all pilots.

Safety message

Pilots should always be prepared to promptly execute a go-around if an approach for landing does not proceed as expected. Accurate knowledge of the aircraft’s reference speeds, in addition to having pre‑determined stabilised approach criteria, assist the assessment of whether an approach should be discontinued. Furthermore, routine practice of this manoeuvre will ensure that it can be performed safely when needed.

The occurrence

On 2 November 2024, a GippsAero GA8-TC Airvan, registered VH-IDM and operated by Wave Air, was being used to conduct a scenic flight with a pilot and 7 passengers on board. At 1120 local time, the aircraft departed from Whitsunday Airport (Shute Harbour), Queensland in‑company[1] with another of the operator’s aircraft as part of the same tour (Figure 1). Approximately one hour after departure, the 2 aircraft returned to the airport, joining the base leg of the circuit for a landing on runway 32.[2] VH‑IDM was leading the company and the pilot made positional broadcasts on behalf of both aircraft on the common traffic advisory frequency (CTAF).

Figure 1: Scenic flight route

Source: Google Earth, annotated by the ATSB

The airport did not have a dedicated Bureau of Meteorology (BoM) weather station, however the pilot recalled a cloud base at about 2,500 ft above ground level (AGL) and a 3‍–‍5 kt wind. Footage of the windsock at the time showed a light headwind on runway 32.

The pilot advised that, as the aircraft proceeded on the final approach to landing, they intended to maintain an airspeed of 80 kt and a flight path to arrive at the runway past the displaced threshold due to trees in the runway undershoot (see the section titled Whitsunday Airport (Shute Harbour)). However, the pilot reported that the aircraft did not descend as expected, resulting in it being above the intended approach path. In response, the pilot lowered the nose to increase the descent rate and regain the approach path, but as a result the airspeed increased from 80 to 90 kt.

At about this time, the trailing company aircraft contacted the pilot on the company frequency to request that they roll through to the end of the runway to exit after landing rather than backtracking. This was to avoid obstructing their landing. Because VH‑IDM could only broadcast on one of its 2 radios, the pilot selected the standby frequency (that was selected to the company frequency) on that radio, replied that they would roll through, and then reselected the CTAF frequency.

The aircraft continued the approach (Figure 2), remaining above the desired approach path while the airspeed varied between 85‍–‍95 kt. The aircraft passed over the displaced threshold of the runway at approximately 100 ft AGL. The pilot commenced the flare about 300 m beyond the displaced threshold, at an airspeed of approximately 90 kt. The aircraft then floated for about 640 m before touching down at a groundspeed of 65 kt with 370 m of runway remaining. The pilot recalled that throughout the approach and landing they did not consider conducting a go‑around[3] and were focused on landing the aircraft.

Figure 2: VH-IDM flight path and landing roll

Source: Google Earth, annotated by the ATSB

After touching down, the pilot retracted the flaps and recalled attempting to apply full braking pressure. They further recalled that the brakes did not perform as expected and they were unable to bring the aircraft to a stop. Subsequently, veering slightly right, the aircraft departed the end of the runway at a groundspeed of 24 kt. The aircraft travelled briefly across grass before entering marshy ground and coming to rest in a ditch, as the propellor struck the ground. Neither the pilot nor any of the passengers were injured and the aircraft received damage to the propellor and firewall (Figure 3)

Figure 3: VH-IDM damage

Source: Wave Air

After verbally confirming with the passengers that they were uninjured, the pilot advised the pilot of the trailing company aircraft via radio of the accident, before exiting and assisting the passengers to evacuate the aircraft. The trailing aircraft landed at approximately the same time as VH‑IDM came to a stop and taxied to the terminal after confirming the safety of the pilot and passengers of VH‑IDM.

Context

Pilot information

The pilot held a commercial pilot licence (aeroplane) issued in 2020 and a class 1 aviation medical certificate. They had accumulated 1,103 hours, of which 15 hours were operating the GA8‑TC Airvan and 225 hours were operating the non‑turbocharged GA8 Airvan. In the previous 90 days, the pilot had accumulated 170 hours, all in the GA8 and GA8‑TC. The pilot had been flying with the operator since June 2024 and had flown almost exclusively from Whitsunday Airport (Shute Harbour). The pilot had last conducted a flight review as part of an instrument proficiency check in December 2022.

While the pilot reported having limited sleep in the 24 and 48 hours prior to the accident, the ATSB examined the possible effect of fatigue and determined that they were not experiencing a level of fatigue known to affect performance.

Aircraft information

General information

VH-IDM was a GippsAero GA8-TC Airvan fitted with a Lycoming TIO‑540‑AH1A turbocharged piston engine. The aircraft was manufactured and first registered in 2009 and at the time of the accident had accumulated 1,240 hours total time in service.

A service bulletin that allowed an increased maximum take‑off weight of 1,905 kg and maximum landing weight of 1,860 kg had been completed on the aircraft. The aircraft was being maintained in accordance with the GA8‑TC‑320 maintenance schedule. A periodic inspection had been completed the morning of the accident and the maintenance release showed no outstanding items. The accident flight was the first flight following the inspection. The pilot advised that a different aircraft had originally been scheduled to be used for the occurrence flight, however the operator substituted VH‑IDM at the ‘last minute’.

Brake system

The aircraft’s brake system included toe brakes incorporated into the rudder pedals. Each rudder pedal was connected hydraulically to a brake unit on the corresponding main landing gear wheel and was engaged by applying pressure to the top of the pedal (Figure 4). During flight, a pilot’s feet rested on the floor in contact with the lower part of the rudder pedals to control the rudder. On the ground, a pilot would move their feet up on the pedal so that the top of the foot could be used to apply brake pressure. The heel was then used to maintain rudder control and nosewheel ground steering.

Figure 4: GA8-TC rudder pedals and brakes

Source: ATSB

The pilot did not report any issues with the braking performance of the aircraft prior to take‑off, and the ATSB was advised that no fault was found with the brakes during the post‑accident inspection. Additionally, there were no marks observed on the runway to indicate that the wheels had locked up and the aircraft had skidded. The pilot also advised that their seat position had been adjusted appropriately and that all controls, including the rudder pedals, could be used effectively.

Whitsunday Airport (Shute Harbour)

Whitsunday Airport (Shute Harbour) was a privately‑owned, uncertified airport (aircraft landing area) located in hilly terrain onshore from the Whitsunday Islands. The airport had a single sealed runway 14/32, which was 1,410 m long and 15 m wide (Figure 5). Runway 32 had a displaced threshold due to trees in the undershoot, which reduced the runway available for landing to 1,310 m. The displaced threshold and departure end of the runway were at elevations of approximately 80 ft and 20 ft respectively, giving it a downslope of approximately 1.4%.

Figure 5: Whitsunday Airport runway environment

Source: Google Earth and passenger video footage, annotated by the ATSB

The airport’s management published a Visiting pilot’s guide that provided information regarding operating at the airport and procedures for approaching and departing each runway. The arrival procedure for runway 32 specified that:

A straight in approach requires a slight right hand dog leg on final to maintain terrain clearance. After following the centre of Shute Harbour water in towards the valley, a right hand dogleg should be made prior to crossing Shute Harbour Road. When necessary to join base for runway 32, keep south of the Shute Harbour Jetty. Land after the displace threshold - this applies to both ends.

The airport management also had a website which contained the above information but also included detail that the:

Touchdown aiming point (Displaced Landing Threshold) is no shorter than the windsock on the lefthand side of the runway.

The pilot advised that landing on runway 32 required modifications to a standard approach. Firstly, as described in the pilot’s guide, high terrain to the south‑west required an oblique approach before aligning with the runway centreline to maintain terrain clearance. Secondly, trees near the arrival end of the runway necessitated a higher approach and could result in visual contact with the displaced threshold being lost. Lastly, due to the downslope of the runway it was necessary to touch down as early as possible to avoid an extended float. The pilot advised that due to the combination of the trees and the downslope they were required to get over the trees and then ‘chop the power’. The pilot of the trailing company aircraft provided similar advice regarding these considerations when landing on runway 32.

As an uncertified airport, Whitsunday Airport was not required to comply with any obstacle or terrain clearance standards. The Civil Aviation Safety Authority (CASA) Advisory Circular (AC) 91‑02 Guidelines for aeroplanes with MTOW not exceeding 5,700 kg - suitable places to take off and land provided guidance for pilots when determining the effect of obstacles on, and in the vicinity of, an uncertified aerodrome:

Pilots should be aware that uncertified aerodromes may declare an available runway length that begins and ends directly at an obstacle. Common examples might be small trees at the beginning or the end of the runway surface.

During landing, high ground or obstructions in the approach area can cause a pilot to adopt a higher than normal approach path to avoid the obstacle, but still achieve a touchdown early in the available runway length…In all cases, the likely outcome is a long landing and the subsequent psychological effect of continuing a landing from an unusual situation outside the experience of the pilot.

Recorded data

VH-IDM was equipped with ADS‑B out capability however flight data was not available for the approach and landing portion of the flight. Two passengers took video recordings of the landing and runway excursion on their mobile phones. Both passengers were seated on the right of the aircraft, one in the first row behind the pilot, and the second one row further back. The recordings showed the view through the front of the aircraft, as well as sections of the instrument panel and certain actions conducted by the pilot (Figure 6).

Figure 6: Passenger video footage

Note: Altitude is above mean sea level (AMSL). Adjustment of 30 ft applied to aircraft altimeter. Source: Passenger video annotated by the ATSB

From the passengers’ videos, the ATSB was able to determine the flight path and speed of the aircraft throughout the approach and landing. While the pilot advised their intended aim point was just past the displaced threshold, the aircraft maintained an average descent path of approximately 7° towards a landing spot 300 m beyond the displaced threshold. Overhead the displaced threshold, the aircraft was approximately 100 ft above ground level. Throughout the approach, the navigation unit displayed the groundspeed as between 90‍–‍95 kt. The airspeed indicator was also periodically visible and showed readings up to 95 kt.

After the commencement of the flare, the aircraft floated for approximately 640 m. During the float, the aircraft decelerated and with about 560 m of runway remaining was at the landing approach speed of 71 kt (see the section titled Landing Performance). Subsequently, the main wheels contacted the ground with approximately 370 m of runway remaining at a groundspeed of 65 kt. The aircraft took 18 seconds to reach the end of the runway, during which the groundspeed slowed from 65 kt to 24 kt.

The pilot was observed to reduce the throttle at several points during the final approach with the last reduction observed just prior to the flare. This indicated the approach was conducted with some power. The pilot was also observed to interact with the radio unit during the final approach, which was likely when responding to the trailing aircraft. The video confirmed that the flaps were set to 38° for landing and were retracted immediately upon touchdown. Also on touchdown, the pilot’s feet moved upwards on the rudder pedals and pressure was applied against the pedals during the landing roll. However, it could not be ascertained whether the pressure was maintained throughout the landing roll, or whether the pressure was being applied to the top of the pedals to apply brakes or to the bottom of the pedals for rudder and steering.

Weight and balance

Software

The operator used a third‑party system to calculate the weight and balance for each flight. In the system, each aircraft was configured with the weight and moment arm[4] when empty. Subsequently, for each flight, the fuel onboard, the pilot and passengers’ weights and their seating positions were recorded to calculate both the weight and centre of gravity of the aircraft at take‑off and landing. The system would provide an alert if the limits prescribed in the aircraft flight manual (AFM) were exceeded. A paper copy of the weight and balance calculation was provided to the pilot before each flight.

Aircraft empty weight

VH-IDM was last weighed on 12 August 2024 and the empty weight was determined to be 1,058 kg, with an empty weight moment arm of 1,202 mm. The operator advised that the aircraft was weighed in the freight configuration, therefore they added the weight of the operational equipment required for passenger carrying from the AFM, establishing the weight as 1,110.6 kg. The operator further advised that to provide a safety margin, a higher weight of 1,134 kg was configured in the system. 

The ATSB identified that the empty weight moment arm was not adjusted to account for the added passenger operational items, with the freight configuration arm of 1,202 mm being used. Additionally, the operator had inadvertently added the weight of one passenger seat to each row, rather than 2. The ATSB calculated that the actual empty weight of VH-IDM in passenger configuration was 1,131.3 kg, with an empty weight moment arm of 1,275 mm. This was 2.7 kg less than the empty weight and 73 mm aft of the empty weight moment arm compared to that configured in the operator’s weight and balance system (Table 1).

Table 1: Operator vs ATSB empty weight and arm

Passenger weights

The operator’s standard operating procedures (SOPs) required that in determining the weight and balance of an aircraft:

Actual weights will be determined by weighing all occupants, equipment and other baggage.

The operator’s website also required that:

Full names and exact weights per passenger must be advised when booking due to flight weight availability.

The passengers reported that their weight was requested at the time of booking, but they were not weighed prior to the flight.

The CASA Multi-part AC 121‑05, AC 133‑04 and AC 135‑08 – Passenger crew and baggage weights described acceptable weight calculation methods that could be defined in operating procedures. The circular stated that:

The use of actual weights is the most accurate method of maximising payload capacity. Appropriately calibrated weighing scales should be used. Actual weighing is more commonly used by Part 133 and 135 operators. This is, in part, due to the smaller number of passengers being carried, which makes this option less disruptive than it is for Part 121 operators.

Operators should have procedures to identify when passenger-declared weights are not appropriate, such as when operating close to limitations. Under these circumstances, the use of actual weights may be required to ensure limitations are not exceeded.

Passenger-declared weights have inherent inaccuracies as passengers may not know their actual weight, especially when fully dressed. An adjustment allowance should be added to any passenger‑declared weight, as a factor or a fixed additional amount.

Take-off and landing weight

The ATSB requested actual weights from the pilot and passengers including the baggage they took with them on the flight. Using this information, in conjunction with the revised empty weight and moment arm, the take‑off and landing weight and balance of the aircraft was calculated and compared to that calculated by the operator (Table 2).

Table 2: Operator‑ vs ATSB‑calculated take‑off and landing weight – accident flight

While the take‑off and landing weight differed by 1 kg, the actual moment arm was 42 mm aft of that calculated by the operator. This was still less than the maximum of 1,626 mm specified in the AFM. However, the ATSB identified loading scenarios where the operator’s configuration would present the weight and balance as acceptable, when the actual moment arm was aft of this limit (Table 3).

Table 3: Operator‑ vs ATSB‑calculated take‑off weight – aft centre of gravity limit scenario

Pilot training

Operator proficiency check

Prior to conducting scenic flights unsupervised, the operator’s SOPs required the pilot to successfully complete an operator proficiency check (OPC). The flight component of the OPC was conducted without passengers and provided an assessment of the pilot’s competency in normal, abnormal and emergency procedures when operating the aircraft.

Upon starting with the operator, the pilot conducted a one‑hour supervised flight in a non‑turbocharged GA8 with the head of flying operations (HOFO),[5] where initial handling training was conducted. This was the pilot’s first flight operating a GA8. The pilot recalled that the flight included conducting steep turns, stall recovery and several circuits on runway 14 at Whitsunday Airport. They reported being uncertain whether a go‑around or a short field landing was conducted during that flight. The pilot also did not recall that they had practiced applying maximum braking, nor that they had done so subsequently.

At the conclusion of the flight, the HOFO completed an OPC assessment, which recorded that several items had been assessed as ‘competent’ including a go‑around and a short field landing. However, several items were marked as ‘not yet competent’ including low‑level flying, flapless landing, basic instrument flying, engine failure and forced landing and aircraft system malfunctions as these items were not conducted during the flight. While the OPC was not completed, no subsequent OPC was conducted prior to the pilot operating unsupervised.

The operator advised the ATSB that due to the nature of some of the flight sequences, a flight training organisation (FTO) had been engaged to conduct OPCs. The FTO had last conducted training and assessment for a group of the operator’s pilots in March 2024, prior to the accident pilot’s commencement with the operator. The operator further advised that due to the timing of their commencement, the pilot had not conducted an OPC with the FTO and that this was an oversight.

Line training and line check

The operator’s line training consisted of a series of flights with a supervising pilot, with passengers on board. Following line training, a line check was conducted, after which a pilot could operate unsupervised if an OPC had also been completed. Training records and the pilot’s logbook showed that 9 supervised flights totalling 9.9 hours were conducted in June 2024 prior to a line check flight. The flights were supervised by 3 different pilots including the HOFO. Following a line check conducted by the HOFO, the pilot commenced operating unsupervised.

General emergency training

The operator required the pilot to successfully complete a general emergency procedures competency check for the aircraft type being flown. This consisted of ground‑based topics and an in‑water practical component. While training records were not available, both the operator and the pilot recalled that the ground‑based training had been completed. However, the in‑water practical component was not conducted. The operator advised that the most recent in‑water training session had occurred in May 2024, prior to the pilot commencing, and that this was also an oversight.

The pilot reported that they had completed in‑water practical training with 2 previous operators, initially in August/September 2022 and subsequently in September/October 2023. They also completed the training with the current operator after the accident and advised that the training provided by all operators involved donning and inflating a lifejacket while in water. They also reported that the while the training conducted by the previous operators was conducted in a swimming pool, the training with the current operator was conducted in open water and included carrying an injured passenger and discussion of survival skills.

Differences training

The SOPs required that differences training was conducted prior to operating an aircraft of the same type with performance differences. Additional training was therefore required prior to operating a turbocharged GA8-TC, such as VH‑IDM, when initial training had been conducted in a non‑turbocharged GA8.

The operator had provided documentation to the pilot on the differences in operating the GA8‑TC and a supervised flight was conducted with the HOFO in October 2024, prior to operating the aircraft type unsupervised. 

Recognition of prior learning

The operator’s procedures allowed flight crew members who had completed training with other operators to be eligible for recognition of prior learning (RPL). The procedures further advised that the training needed to have been completed within the previous 6 months, and could be applied to the following training events:

• general emergency training
• differences training
• line training.

Approach speed

The operator reported that pilots were taught to conduct the final approach to land at an airspeed of 70 kt with 500 ft/min descent rate. They also advised that there was no difference between the approach and landing speeds when operating the GA8 compared to the GA8‑TC. Additionally, the operator’s SOPs stated that:

During the approach phase the pilot-in-command shall ensure that the aircraft is flown at the approach speeds (V_REF) provided in the Aircraft Flight Manual for the aircraft being flown.

The FTO advised that pilots were taught and assessed in the non‑turbocharged GA8 on establishing a reference airspeed (V_REF) [6] of 71 kt on final as per the AFM. They also advised that no training or assessment had been conducted in the GA8‑TC.

The pilot reported that they considered 80 kt as the appropriate final approach speed. However, they also stated that, following discussions with other pilots after the accident, they now understood that 70‍–‍75 kt was an appropriate final approach speed. The ATSB was also advised that the pilot had been observed landing long on previous occasions, however this had not been communicated to the operator or discussed with the pilot.

The pilot of the trailing aircraft reported they typically aimed for 80 kt on final and were ‘happy to get it down to about 75 [kt] on runway 32’. They also reported that for a short field landing they used 70‍–‍75 kt on final. 

Landing Performance

The AFM provided performance charts to calculate the expected landing distance and ground roll. At a landing weight of 1,860 kg with a runway slope of −1.4% and the atmospheric conditions present at the time of the accident, the expected landing distance required was calculated to be 480 m, including a ground roll of 210 m. The AFM further described the reference speed and technique for achieving this performance:

• airspeed at 50 ft of 71 kt
• power off, 7° approach profile
• 38° (full) flap
• aircraft approaches with idle power at the given airspeed appropriate to weight
• after touch down maximum wheel braking is used to bring the aircraft to a stop
• for maximum braking effectiveness the wing flaps should be retracted and back pressure applied to the control column.

The AFM also stated that:

Care must be taken to ensure airspeed is accurately maintained during the final landing approach. Timely and appropriate use of power should be exercised to maintain the desired flight path and airspeed. Excessively high approach speeds will result in prolonged floating and increased landing distance.

The AFM also provided performance charts for a 3° approach angle with power. In this circumstance the landing distance was expected to be higher due to the lower approach angle, with the ground roll remaining approximately the same.

Stabilised approach criteria 

The operator’s procedures specified that aircraft should be on a stabilised approach as early as practical on the final approach path and that the following criteria were required for an approach to be stable:

• the aircraft is on (or close to) the correct flight path, only small changes in heading and pitch being required to maintain that path

• the aircraft speed is not more than Vref + 20 kt and not less than Vref

• the aircraft is in the proper landing configuration (except that full flap should not be selected until committed to land)

• sink rate is maximum 1,000 ft/min

• power setting appropriate to the configuration but not below any minimum power for approach specified in the Aircraft Flight Manual

• all briefings and checklist items have been performed.

In visual conditions, if these criteria were exceeded below 100 ft above airport elevation, the pilot was required to execute a go‑around.

The ATSB calculated that if an aircraft was 20 kt above the landing reference speed at the 100 ft decision height, in a power off 7° approach descent, the pilot had 2.7 seconds to reduce their speed to the landing reference speed. By comparison, a decision height of 300 ft would increase this time to 13.3 seconds while a decision height of 500 ft would increase the time to 23.9 seconds.

CASA provided guidance in AC 91‑02 on initiating go‑arounds in response to an unstable approach, stating that:

Pilot training emphasises that a safe landing is the result of a stabilised approach. If pre-determined stabilised approach criteria are exceeded, then a safe landing is not assured. The decision to execute a go-around should be made as early as possible to maximise the safety outcome. At the conclusion of an effective go-around, the pilot will then have an opportunity to consider what options are available to conclude the flight.

Additionally, the Flight Safety Australia article Quantifying the go-around (CASA, 2021) highlighted the importance of practicing go‑arounds:

It’s not enough to pass the test and fly a go-around only every couple of years when tasked by an instructor. Consciously ask yourself if you’re in the slot, judging your aeroplane’s state and trend all the way down final. By quantifying your performance, you can make the go-around decision before you are at the highest risk of loss of control.

Going around is as natural a part of flying as landing itself – or it will be, if you evaluate landing criteria every time and occasionally practice the go-around task.

The pilot advised that, in addition to not considering a go‑around during this approach, they could not recall having previously conducted a go‑around outside of training.

Related occurrences

The ATSB occurrence database contained 200 other reported occurrences of runway excursions during landing in Australia between January 2021 and December 2024. Of these, 12 resulted in injuries to the pilot and/or passengers, including 2 where the injuries were serious.

Included in these occurrences were 2 other runway excursions involving a GA8 Airvan, both of which were investigated by the ATSB:

Runway overrun involving GippsAero GA8, VH-WSB on 26 December 2021 (AO-2022-001)

During the landing, the aircraft floated significantly beyond the intended landing point. The pilot did not recognise the risk of a runway overrun and did not conduct a go‑around or apply sufficient braking to stop the aircraft on the remaining runway.

Runway excursion involving GippsAero GA8, VH-TBU on 6 April 2023 (AO-2023-016)

During the landing, the aircraft floated for a significant time and touched down approximately halfway down the runway, with insufficient remaining runway to stop. While the pilot recognised opportunities to conduct a go-around when they determined they were not on the correct approach profile, this was not conducted.

Safety analysis

Pilot actions

Approach and landing

During the approach to land, the aircraft’s flight path was significantly above that intended, with an aim point approximately one third down the runway. While the deviation was likely influenced by the associated terrain and obstacles, the pilot had conducted this approach regularly and was familiar with the required approach path to land safely. The deviation was also possibly influenced by distraction when interacting with the radio to respond to the pilot of the trailing aircraft and by implied pressure to minimise the time spent on the runway.

Attempting to regain the intended flight path, the pilot lowered the nose, but did not reduce the power to idle. Subsequently, the aircraft’s airspeed increased to approximately 95 kt. While the pilot planned 80 kt as the airspeed on final, the approach airspeed required by the operator was 70‍–‍71 kt. This approach speed was also the reference landing approach speed (Vref) in the aircraft flight manual (AFM). Therefore, the aircraft’s airspeed deviation was about 24 kt.

While recognising that the aircraft was above the intended flight path and faster than the intended airspeed, the pilot continued the approach. The operator’s procedures required a go‑around to be conducted when the aircraft was not on the correct flight path or was more than 20 kt faster than the landing approach speed below 100 ft. While the airspeed was within this limit based on the pilot’s incorrect understanding that the approach speed was 80 kt, the aircraft was not on the correct flight path and therefore a go‑around was required. However, the pilot did not consider a go‑around and commenced the flare well beyond the planned touchdown point at a high airspeed. Due to the high airspeed and the downslope of the runway, the aircraft floated significantly before touching down with less than a third of the runway remaining.

Landing roll and braking

Despite the reduced runway available, performance calculations determined that sufficient runway remained for the aircraft to stop if maximum braking was applied. However, although the pilot was observed retracting the flap and applying pressure to the rudder pedals, the aircraft did not decelerate as expected. Given the braking system was functional prior to take‑off and after the accident and there was no indication that the aircraft skidded, maximum braking application was likely not conducted effectively. Subsequently, due to the insufficient braking, the aircraft departed the end of the runway.

Training and assessment

Initial training

The ATSB considered the effect of the training the pilot received from the operator prior to the accident. Given the elapsed time since their last flight review, additional training from a flight training organisation would have given them opportunity to practice procedures such as go‑arounds and short field landings. In addition, the operator did not complete the operator proficiency check and as a result the pilot was not assessed on several abnormal and emergency procedures in the GA8. However, the pilot was assessed as competent in both go‑arounds and short field landings during the initial handling training and had completed a number of line training flights that would have given them time to practice basic handling skills. 

The general emergency training required an in‑water practical exercise that was not conducted. While recognition of prior learning was able to be applied to this training event, training with another operator was completed more than 6 months prior and so was not applicable. However, given the pilot had received this training twice in the preceding 26 months from other operators and the training was similar, it was likely that the pilot was competent in this area.

Approach speed

The pilot had received initial handling training and conducted line training flights supervised by 3 different pilots. They had also passed a line training check and had recently received differences training in the GA8‑TC. Notwithstanding this, the pilot’s pattern of using the incorrect approach speed was not identified or corrected. 

Weight and balance

The ATSB determined that the empty weight for the aircraft was calculated incorrectly, however as the operator had increased the weight in the system as a safety buffer, this did not have an effect. Additionally, the empty weight moment arm used to calculate if the aircraft’s centre of gravity was within the allowable limits, was not adjusted for the additional operational items. As a result, the system calculated the aircraft’s centre of gravity forward of the actual position. While the actual weight and centre of gravity of the accident flight was within limits for both take‑off and landing, the incorrect empty weight moment arm permitted the aircraft to be loaded in a way that the centre of gravity was aft of the limit, while presenting to the pilot as within.

Furthermore, passengers were not weighed prior to flight and instead passenger‑declared weights were used. This was not in accordance with the operator’s procedures and was not recommended when operating close to the weight limitations of the aircraft.

Stabilised approach criteria

It was determined that the decision height for assessing if the aircraft met the operator’s stabilised approach criteria was too low. As in this case, where an aircraft’s airspeed was 20 kt faster than Vref at 100 ft (the decision height), it was very unlikely that the aircraft could be slowed to the reference landing approach speed in 2.7 seconds, most likely leading to a go‑around.

While go‑arounds are a normal aspect of flying, as stated in AC 91‑02, ‘the decision to execute a go‑around should be made as early as possible to maximise the safety outcome’. However, in this case the pilot did not consider the operator’s stabilised approach criteria in their decision‑making.

Findings

From the evidence available, the following findings are made with respect to the runway excursion involving GippsAero GA8-TC, VH-IDM, at Whitsunday Airport (Shute Harbour), Queensland on 2 November 2024. 

Contributing factors

• The aircraft's approach was above profile with a high airspeed and the pilot had an incorrect understanding of the required approach speed. Subsequently, the pilot did not initiate a go-around, resulting in a landing beyond the planned touchdown point.
• Despite having adequate landing distance remaining, the pilot did not apply sufficient braking to prevent the aircraft departing the runway.
• The training, supervision and checking flights conducted by Wave Air did not identify that an excessive approach speed was routinely being used by the pilot during the final approach to land. (Safety issue)

Other factors that increased risk

• The pilot’s initial training was not fully completed, and they were not assessed on several abnormal and emergency procedures prior to operating unsupervised.
• Wave Air's weight and balance system used an incorrect empty weight moment arm to calculate the aircraft's centre of gravity, and passengers were not weighed in accordance with their procedures. (Safety issue)
• The decision height for assessing whether an aircraft met Wave Air’s stabilised approach criteria was too low. (Safety issue)

Safety issues and actions

Training and assessment of approach speed

Safety issue number: AO-2024-056-SI-01

Safety issue description: The training, supervision and checking flights conducted by Wave Air did not identify that an excessive approach speed was routinely being used by the pilot during the final approach to land.

Weight and balance calculation

Safety issue number: AO-2024-056-SI-02

Safety issue description: Wave Air's weight and balance system used an incorrect empty weight moment arm to calculate the aircraft's centre of gravity, and passengers were not weighed in accordance with their procedures.

Stabilised approach decision height

Safety issue number: AO-2024-056-SI-03

Safety issue description: The decision height for assessing whether an aircraft met Wave Air’s stabilised approach criteria was too low.

Glossary

Sources and submissions

Sources of information

The sources of information during the investigation included:

• the pilot and passengers of the accident flight and another company pilot
• the operator
• Whitsunday Airport (Shute Harbour)
• Civil Aviation Safety Authority
• Bureau of Meteorology
• Airservices Australia
• the engaged flight training organisation
• aircraft insurance company
• video footage of the accident flight and other photographs and videos taken on the day of the accident

References

Civil Aviation Safety Authority (2022). Guidelines for aeroplanes with MTOW not exceeding 5 700 kg - suitable places to take off and land (advisory circular AC 91-02 v1.2), https://www.casa.gov.au/guidelines-aeroplanes-mtow-not-exceeding-5-700-…, CASA, accessed 11 December 2024.

Civil Aviation Safety Authority (2022). Passenger crew and baggage weights (multi-part advisory circular AC 121-05, AC 133-04 and AC 135-08- Version 1.1), https://www.casa.gov.au/sites/default/files/2021-08/multi-part-advisory…, CASA, accessed 7 January 2024.

Civil Aviation Safety Authority (2021). Flight Safety Australia: Quantifying the go-around, https://www.flightsafetyaustralia.com/2021/04/quantifying-the-go-around, CASA, accessed 16 December 2024.

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 and operator
• Whitsunday Airport (Shute Harbour)
• Civil Aviation Safety Authority

Submissions were received from: 

• the pilot
• the operator 

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. About ATSB reports ATSB investigation reports are organised with regard to international standards or instruments, as applicable, and with ATSB procedures and guidelines. Reports 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. 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.