On 9 July 2018, an ATR 72-600 (CN-COH) being operated by RAM Express, a wholly owned operating subsidiary of Royal Air Maroc (RAM), on a scheduled domestic passenger flight for the parent company from Tangier to Al Hoceima as RAM439 hit the sea surface twice in quick succession whilst on approach to destination in day IMC before successfully climbing away and, after reporting a bird strike, diverting to Nador where substantial structural and landing gear damage which rendered the aircraft unfit for flight pending major repairs was found. None of the 58 occupants were injured by the impact.
The position of sea surface impact on the approach runway extended centreline. [Reproduced from the Official Report]
An Investigation into the Accident was carried out by the Moroccan Bureau d’Enquêtes et Analyses (BEA) in accordance with ICAO Annex 13 principles. The FDR and 2 hour CVR were both removed undamaged from the aircraft and their relevant data were subsequently downloaded successfully.
It was noted that the 61 year-old Training Captain in command had a total of 13,487 flying hours which included 193 hours on type. He had obtained his ATR72 type rating four months prior to the accident just over a month after beginning employment with the operator following a long period of employment as a Boeing 737 Captain and instructor at the parent company RAM. The 25 year-old First Officer had a total of 1,063 flying hours which included 815 hours on type. He had obtained his ATR72 type rating just over a year prior to the accident after beginning employment with the operator two months earlier. An observer pilot, a new company recruit not qualified on the aircraft type, was also present on the flight deck and occupying the supernumerary crew seat during the flight sequence which included the accident.
The flight sequence operated by the aircraft and crew on the day of the accident was planned as Casablanca - Al Hoceima - Tangier - Al Hoceima - Casablanca. Before commencing the first of these flights, the crew agreed that the Captain would act as PF for the first leg and then revert to acting as PM for the remainder.
During the first flight, in the cruise at FL160, an EGPWS ‘TERRAIN’ annunciation was displayed and the amber “TERRAIN FAULT” light was illuminated, a system activation which indicated temporary loss of the system’s ‘enhanced’ modes and which was assessed as “probably linked to a degradation of the GPS signal”. This annunciation lasted a little over 6 minutes. The crew subsequently planned for and commenced an RNAV approach to runway 17 at Al Hoceima having noted that the reported ceiling was 800 feet aal. On reaching the procedure MDA of 1,030 feet, the Captain was “unable to see the runway” but decided to continue the descent anyway with a rate of descent of around 1000 fpm. One minute later, the EGPWS ‘TERRAIN AHEAD, PULL UP’ warning was activated with the aircraft 1760 metres from the intended landing runway threshold and just 60 feet above sea level. The Captain responded by pitching up and adding power and after reaching a recorded minimum height of 45 feet above the sea surface, the aircraft climbed to a recorded height of 108 feet and maintained this until the runway was acquired visually and landed. The second flight was then competed without further event.
Because of the relatively short flight time for the third, Tangier to Al Hoceima, sector, the approach briefing for Al Hoceima was carried out on the ground in Tangier based on a VOR/DME approach to runway 17 (see the illustration below) for which the MDA was 760 feet. The Captain was recorded explaining to the First Officer that “if the runway is not in sight at the MDA, we will descend until a height of 400 feet which will then be maintained until the runway is in view and that if the runway is still not in sight at 2 nm from the VOR, it will be necessary stop the approach and make a go around”. In addition, the Captain was recorded as accepting, after consulting the DDM (Dispatch Deviation Manual), the First Officer’s suggestion that the GPWS should be disabled so as to avoid alarms being activated during the descent and approach “because they think that those emitted by this system, during the first flight were untimely”.
The flight proceeded normally and, after a six minute cruise at FL 130 during which the (PM) Captain had further discussed the plan for the approach - which would involve establishing directly onto the FAT (final approach track) - with the First Officer and had given corresponding instructions in respect of it. Descent as cleared to FL 060 was commenced and Al Hoceima TWR contacted and the intention to make a VOR/DME approach for runway 17 agreed. The Captain then reminded the First Officer of the manoeuvres to be executed in the event of a go-around and stated that during the approach, he would “take care of the monitoring of speed and water” whilst the First Officer controlled the flight path.
Re-clearance to 3000 feet on QNH was given and the selected altitude set accordingly and the rate of descent to 1,500 fpm then to 1,300 fpm. The approach checklist was run and passing 4500 feet QNH, with the aircraft 11.5 nm from runway 17 threshold at 230 KCAS, the Captain de-activated the EGPWS and reminded the First Officer to select more flap to increase the rate of descent.
The VOR/DME 17 Approach Plate for Al Hoceima. [Reproduced from the Official Report]
As the heading was set to establish on final approach, the Captain told the First Officer to reduce speed from 220 knots and the First Officer set 170 knots. TWR then gave them the latest weather as “wind calm, visibility four thousand in mist, overcast at 600, temperature 23, dew point 23, QNH 1016" which the Captain acknowledged. With the aircraft established on the procedure radial and approaching the procedure vertical profile as the speed reduced through 188 KCAS, a selected altitude of 400 feet was set, and V/S mode engaged with a 1200 fpm rate of descent.
Passing 1800 feet QNH, the flaps were selected to 15° and the selected speed was 140 knots as the Captain confirmed “fully established” to TWR and the flight was cleared to land. On request from the First Officer, the Captain selected the landing gear down and was recorded telling the First Officer to “go, go to the limit". The selected rate of descent was progressively increased to 1800 fpm and flaps 30° selected. With 3.3 nm to go, the altitude capture mode engaged 400ft before the selected (400 feet) altitude was reached because of the commanded 1800 fpm descent. As 400 feet was approached with the indicated speed now reduced to 125 knots but the runway still not in sight, the Captain was recorded saying “we continue to 300” and the selected vertical mode was changed from the just-engaged ‘pitch hold’ back to V/S with a 1000 fpm rate of descent. The Captain said “keep on going” but passing 310 feet at 121 KCAS, the aircraft was still 2 nm from the runway.
Eight seconds later at 135 feet above the sea surface and an indicated speed of 128 knots, the First Officer said “it's not normal” and then spoke in Arabic, his native tongue, "now take it in manual" and disconnected the AP at a height of 80 feet, began to apply back pressure to his control column and increased power. However, at the same time, the Captain began an opposite input to his control column which neutralised the attempt to climb and a couple of seconds later, the aircraft hit the sea surface twice in quick succession, the first time with an impact which registered a maximum vertical acceleration during impact of 3.2 g and the second, at a -3° pitch attitude, which registered 3.9 g. After a few seconds, the Captain stopped his pitch down and the First Officer was able to increase his pitch up sufficiently to establish the aircraft in a recovery climb.
The Captain advised Al Hoceima TWR that the aircraft had made a go around and would divert to Nador. When the controller queried the reason for the go around, the Captain responded that it followed a bird strike which was not true. On arrival at Nador, a visual inspection of the aircraft found that serious structural damage had been sustained to the lower fuselage in the vicinity of the main landing gear which itself was subsequently, along with the nose landing gear, confirmed to have exceeded the maximum certified load and therefore required replacement. The Captain’s initial report to the aircraft operator maintained the bird strike story but this was subsequently corrected to advise contact with the sea surface prior to the go around.
The first illustration below shows the approach vertical profile flown and the selected FMS vertical mode in use and, below the main diagram, the actual rate of descent and, where appropriate, the selected target vertical speed. The second illustration below shows the vertical profile and impact vertical acceleration around sea impact.
The vertical profile flown (mauve) against the VOR/DME procedure vertical profile extended below its MDA (760 feet) using QNH altitude (1016 hPa) as the vertical reference. [Reproduced from the Official Report]
Detail of the recorded vertical acceleration and height above the sea during the impact phase. [Reproduced from the Official Report]
Procedural Non Compliance
In addition to the flagrant breaching of the MDA on both approaches to Al Hoceima, it was noted that the RAM Express OM was unequivocal on the requirement for all approaches in IMC to be stabilised by 1000 feet aal, on what constituted ‘stabilised’ and on the automatic requirement to go around if not stabilised.
The formally-stated Conclusions of the Investigation included, but were not limited to the following:
- The approach to Al Hoceima was characterised by a lack of preparation and anticipation and the flight crew's poor level of CRM made it impossible to manage effectively. Verbal communication was limited to the Captain’s instructions followed without challenge by the First Officer and call outs were non-existent and cross-checks rare and ambiguous.
- The flight crew's work was characterised by deliberate violation of procedures and hazardous operational decisions in support of a determination to continue the approach in the absence of the required conditions and beyond the limits.
- Before commencing the accident flight, the crew had decided and briefed that they would descend to an altitude which was significantly below the applicable MDA.
- The crew de-activated the system (GPWS) during the flight so as not to receive alerts during the approach.
- The VOR / DME approach at Al Hoceima was planned when the ceiling was already 600 feet, significantly lower than the MDA for the procedure to be flown.
- The execution of the approach was characterised by an obvious delay in the configuration of the aircraft and failure to properly control indicated speed, altitude and rate of descent.
- The approach carried out was therefore not stabilised as required by the SOPs of the company.
- The crew did not initiate a go-around when it became apparent that the approach was not stabilised below the specified gate.
- The crew descended below the minimum descent altitude (MDA) without having the mandated visual reference to do so.
The Probable Cause of the Accident was formally determined as “non-compliance with operational procedures, in particular; the deliberate deactivating of the EGPWS, the continuation of an unstable approach below the applicable stabilisation gate and the continuation of the approach below the applicable minima in the absence of visual reference with only the First Officer’s reaction, even though it was late, making it possible to limit the outcome to no more than significant damage to the aircraft”.
Contributory Factors were also identified as:
- the absence of any ‘advisory callouts’ by the First Officer during the approach.
- the absence of any reference in the Despatch Deviation Manual (DDM) covering a total failure of the EGPWS, which was likely to have supported the flight crew decision to deactivate it intentionally.
- the lack of any meaningful CRM on the flight deck, especially in terms of communication, coordination and adaptation to take account of the steep [[Authority Gradients|authority gradient between the Captain and the First Officer and the high level of assertiveness of the Captain which led to the First Officer being slow to challenge the Captain’s directions when aware that they were incompatible with the requirements for stabilised approaches and respect for approach procedure minimums.
Safety Action taken by RAM Express whilst the Investigation was in progress was noted as having included:
- Installation of an EGPWS switch seal and the addition of a check that this remains intact as part of the Daily Check of the airworthiness of the aircraft.
- Introduction of a DDM procedure to cover a total failure of the EGPWS.
Four Safety Recommendations were made as a result of the Investigation as follows:
- that RAM Express strengthen the process of integrating their pilots’ performance through CRM training modules adapted in such a way as to ensure the fluidity of interactions and the consistency of communications between pilots and the appropriateness of decisions. [01/19]
- that RAM Express strengthens their pilot training for non precision approaches by:
- emphasising during awareness and training sessions the need to conduct constant angle approaches when vertical guidance is not available
- adapting the aircraft equipment and associated procedures to allow the vertical profiles of such approaches to be programmed and followed. [02/19]
- that RAM Express validates during flight crews’ training and skill checks that they:
- take into account the criteria for undertaking and continuing an approach in terms of stabilisation and minima.
- fully understand the EGPWS system, the meanings of its messages and the response to be followed in the event that alerts or warnings are annunciated. [[03/19]
- that RAM Express insists during flight crews’ training and skill checks that:
- the criteria for undertaking and continuing an approach, in terms of stabilisation and minima, are taken into account.
- The meanings of the alerts and warnings which may be delivered by the EGPWS system are understood and the required responses to each are followed. [04/19]
The Final Report of the Investigation was released online on 12 July 2020 and has been published only in French. This summary of it is based solely on an unofficial translation of the content of that Official Report.