On 24 March 2010, a Boeing 727-200 being operated by Canadian company Cargojet AW on a scheduled cargo flight from Hamilton Ontario to Moncton New Brunswick failed to stop after a night landing on 1875 metre long runway 06 at destination in normal ground visibility and eventually stopped in deep mud approximately 100 metres beyond the runway end and approximately 40 metres past the end of the paved runway end strip. The three operating flight crew, who were the only occupants, were uninjured and the aircraft received only minor damage.
An Investigation was carried out by the Canadian TSB. It was noted that after visually confirming that there was no post-crash fire, the flight crew had pulled the CBs for the Flight Data Recorder (FDR) and Cockpit Voice Recorder (CVR) and useful downloads from both were thus enabled and subsequently obtained. It was established that the crew members were all experienced on the aircraft type and familiar with Moncton, although had little experience of landing on runway 06.
It was noted that the aircraft commander had been PF for the sector and had elected to land on runway 06 at destination. Following the issue of a landing clearance and the aircraft becoming fully established on the Instrument Landing System (ILS), the AP was disconnected and the landing checklist was commenced. Shortly after this, the aircraft drifted above the ILS GS slightly exceeding half scale deflection before corrective action was taken. As the aircraft crossed the FAF, its altitude was noted to be approximately 50 feet higher than profile after which the aircraft regained and followed the ILS GS until it crossed the runway threshold. At a range of approximately 2nm from the threshold, the runway became visible and at 375 feet aal the aircraft encountered heavy rain. As the decision height (DH) was reached, the wipers were selected on. The aircraft crossed the runway threshold at 165 KIAS and touched down approximately 9 seconds later between 610 and 760 metres from the threshold at approximately 157 KIAS. The automatic speed brakes deployed on touchdown and approximately 3 seconds later, reverse thrust was selected. The aircraft heading then began to veer to the right and in response, reverse thrust was reduced - although the application of manual brakes was not suspended - until the runway direction was regained. However, deceleration was insufficient to bring the aircraft to a stop before it reached the end of the runway and it passed through the 60 metre long paved runway end strip and departed the paved surface with an indicated airspeed of approximately 50 knots before coming to rest in deep mud approximately 27 seconds after touch down. Maximum manual wheel braking was applied shortly after touchdown, and maintained by the flight crew throughout the entire landing roll but did not have the desired effect because of aquaplaning. Evidence of this was subsequently found in the form of reverted rubber marks on the runway showing that the anti skid system had failed to activate even though it was subsequently confirmed to be fully functional, indicating that the minimum wheel speed threshold for system activation was never reached.
It was concluded that the overrun had been a consequence of the choice of a short runway on which late touch down - around one third of the way down the runway - was then made at a speed significantly above the applicable Reference Speed (Vref) onto a runway with standing water present and so conducive to the aquaplaning. It was considered that the reduction of reverse thrust following touchdown so as to realign with the runway centreline, although in accordance with the manufacturer's recommended practice, had increased the required stopping distance. It was considered that the decision to carry out a pilot-flown-approach, rather than a pilot-monitored-approach at night and in heavy rain, probably contributed to the excess airspeed that was maintained until touchdown. No indication that an aircraft system malfunction had contributed to the accident outcome was found and the focus of the Investigation therefore moved to an analysis of the factual aspects of the landing made and the prevailing environmental context.
The aircraft in its final stopping position (reproduced from the Official Report)
Comments on aspects of the overrun
It was noted that although runway 06/24 (1867 metres long) was shorter than the other runway at Moncton (11/29 – 2440 metres long) and in the direction used had a downslope of 0.47%, it had an ILS approach. The alternative runway 11 only had a non precision approach with higher minima, although neither cloudbase nor visibility mitigated against use of these minima at the time and the surface wind mean direction was between the two runway alignments It was noted that the aircraft commander as PF had elected to carry out a manual pilot-flown-approach and not a pilot-monitored-approach (PMA). He stated that his decision was based on the fact that the reported weather was above the mandatory weather limits for a mandatory PMA under Cargojet SOPs. The Investigation considered that when the aircraft encountered heavy rain just moments before reaching the ILS decision height, both pilots' attention was probably focused primarily on the runway which may have led to the excessive airspeed, which was sustained until touchdown, going undetected – although no observation was made about the role of the third – but considerably less experienced – crew member in this regard. It was found that the AFM factored wet runway landing distance for the prevailing circumstances was 1826 metres – more than 97% of the LDA. AFM LDR figures for the 727 were achieved through actual flight testing by Boeing plus a 15% safety margin for certification purposes. By comparison, it was noted that the standard method of simply applying a 1.92 factor to the unfactored dry runway landing distance produced an LDR of 1750 metres and thus an apparently greater ‘safety margin’. The latter had, though, been insufficient to allow for the combination of a delayed touchdown point, the higher than required touchdown speed, the slightly downsloping runway and a runway contaminated with standing water. The Investigation observed that “these differences represent several of the variables which, to a certain extent, may typically exist in many landings”. However, there is no requirement for pilots to ensure that a safety margin exists for a planned landing once en route, it was considered that “there is increased risk that pilots will select runways for landing that provide an unacceptable safety margin against runway overrun”. It was noted that aircraft manufacturer’s continue to incorporate reverse thrust credit into their landing distance calculations and although this has been criticised, at the time of the Investigation there was no regulatory prohibition of this practice in North American regulations. In the accident under investigation, the flight crew had factored in the use of reverse thrusters but reverse thrust then had to be reduced in order to regain runway centreline alignment. As a result, the credit that would have been gained by the reverse thrusters was not fully realised, and the landing distance was extended unexpectedly. The use of a reverse thrust credit, when determining landing distance required, increases the risk of overrun if the reverse thrust is not applied and maintained in accordance with the AFM.
The question of whether the onset of aquaplaning was material to the overrun was considered by the Investigation. It was noted that the crew had relied on the use of reverse thrust in accordance with recommended practice after experiencing aquaplaning and weathercocking on touchdown but although the reduction of reverse thrust in order to realign with the runway centreline was in line with recommended practice, it had the effect of increasing the LDR. In addition, the recommendation to release the brakes whilst attempting to regain runway alignment was not followed and it was therefore considered probable that the wheels did not have time to spin up to a sufficient speed to activate the anti-skid braking system for the remainder of the landing roll. Whether or not the recommended release would have provided enough time for anti-skid activation to occur could not be ascertained but it was considered that “it is likely that the higher than required touchdown speed (alone) prevented the flight crew of (the accident aircraft) from being able to stop (their) aircraft within the remaining runway available”. In respect of the selection of runway 06 for landing, it was noted that application of three out of the five available wet runway factors, the projected LDR for runway 06 would have exceeded the LDA while the for the other two, the projected LDR would have provided a safety margin of less than 50 metres whereas the LDR using all of the factors would have yielded an LDR within the LDA on runway 11. In respect of runway surface condition (RSC) reporting it was noted that the RSC report had not been updated to reflect the effect of the onset of heavy rain shortly before the accident landing, nor was this required by regulation. It was considered by the Investigation that “while it is understood that measuring the effects of water on runways presents certain challenges, the lack of clearly defined RSC reporting standards related to water on runways increases the risk of (aquaplaning) and continued overruns”.
In respect of the failure of ATC to specifically communicate for information the onset of heavy rain to the approaching aircraft, the Investigation considered that although such action is neither required nor specifically recommended, “if significant changes in weather which could adversely affect aircraft landing performance are not communicated to pilots in a timely manner, there is (an) increased risk of overrun due to unanticipated reductions in landing performance". Related to this aspect of the circumstances, it was also noted that “at the time of the occurrence, there was no requirement to issue a SPECI if light rain changes to heavy rain. However, in November 2010 Amendment 75 to ICAO Annex 3…..took effect. That amendment now requires that a SPECI shall be issued at the onset, cessation, or change in intensity of freezing precipitation, moderate or heavy precipitation, and thunderstorms (with precipitation)”. In respect of the fact that the accident runway was not grooved, the Investigation noted that while it is widely recognised that grooved runways improve both drainage and skid resistance as well as reducing the risk of aquaplaning and that studies have shown that wet, grooved runways often provide a level of braking only marginally less than dry runways, only three Canadian airports had grooved runways at the time of the Investigation. It was further noted that although the FAA specifically promotes the benefits of grooving, Transport Canada (TC) had yet to issue any similar formal recommendations and there was no related regulatory requirement. It was also noted that “despite claims about the difficulties associated with grooved runway maintenance during the winter, grooved runways are being used without problem in other geographical areas which experience similar weather” and noted that “at least two TC studies have previously identified that there are no major challenges associated with grooved runways in winter conditions”. The Investigation considered that “Canada's high rate of runway overruns versus the U.S. is likely due to the widespread use of grooved runways in the U.S. compared to Canada”. Finally, in considering the range of issues highlighted by the facts disclosed by the Investigation, the Board noted that “Runway overruns continue to pose one of the greatest risks to travelling Canadians” and had already identified the absence of any Canadian regulatory requirement for RESAs (or other engineered systems or structures with equivalent effect) as a key risk reduction measure. The Investigation noted that while TC has indicated its intent to meet the ICAO 90 metre RESA standard set in 1999, this had not yet occurred. It was accepted that the investigated event was an example of one where the overrun on a prepared surface would have resulted in a final stopping position beyond the ICAO 90 metre RESA Standard but still within the limits of the ICAO 240 metre RESA recommendation.
Investigation Findings in respect of Risk included the following:
- There is no requirement in Canada to demonstrate dispatch safety margins for contaminated runway surfaces, using actual contaminated landing distance data or an approved equivalent. As a result, there is increased risk that dispatch safety margins will be reduced due to contaminated runway surfaces.
- The current wet runway dispatch factor for Canadian Aviation Regulations Subpart 705 operators does not provide an adequate safety margin to account for operational variables which may exist upon landing. As a result, there is increased risk of runway overrun due to the reduced safety margin when landing on a wet runway.
- Current regulations do not require flight crew to ensure that a safety margin exists, based on a calculated landing distance versus available landing distance, using the actual operating and environmental conditions. As a result, there is increased risk that pilots will select runways for landing that provide an unacceptable safety margin against runway overrun.
- The use of a reverse thrust credit, when determining landing distance required, increases the risk of overrun if the reverse thrust is not applied and maintained in accordance with the airplane flight manual.
- Unfactored landing distances are not representative of routine operations and do not include any type of safety margin. As a result, there is increased risk of runway overrun if landing decisions are based on the unfactored landing distances.
- The application of Canadian Runway Friction Index 0.5 to unfactored dry runway landing distances may result in an inadequate safety margin for landing on a wet runway. As a result, there is increased risk of runway overrun based on overly optimistic wet runway landing performance calculations.
- Inadequate drainage at airports increases the likelihood of standing water, which increases the risk of runway excursion due to hydroplaning.
- The use of non-grooved runways increases the risk of runway overrun as a result of hydroplaning (aquaplaning).
- The lack of formal guidance for assessing the impact of rain on landing performance increases the risk that flight crews will underestimate the distance required to safely stop following landing on a wet runway.
- The lack of timely and accurate runway surface condition reports during periods of extended or heavy rainfall increases the risk of overrun due to miscalculations in landing performance.
- The lack of clearly defined runway surface condition reporting standards related to water on runways increases the risk of hydroplaning (aquaplaning) and continued overruns.
- If significant changes in weather which could adversely affect aircraft landing performance are not communicated to pilots in a timely manner, there is increased risk of overrun due to unanticipated reductions in landing performance.
- The lack of regulation requiring runway end safety areas (RESA) increases the risk of aircraft damage and passenger injuries at airports in Canada which do not have a RESA or other engineered systems or structures designed to safely stop planes that overrun.
The release of the Final Report: A10A0032 was authorised by the TSB on 18 October 2011. No Safety Recommendations were made.