Back to home pageEI-DYG 737-800 loss of thrust on both engines

Home > Accident News > EI-DYG

Contents

Illustrated technical information covering Vol 2 Over 800 multi-choice systems questions Study notes and technical information Close up photos of internal and external components A compilation of links to major 737 news stories with a downloadable archive Illustrated history and description of all variants of 737 Detailed tech specs of every series of 737 Databases and reports of all the major 737 accidents & incidents General flightdeck views of each generation of 737's Description & news reports of Advanced Blended Winglets Press reports of orders and deliveries A collection of my favourite photographs that I have taken of or from the 737 Details about 737 production methods A compilation of links to other sites with useful 737 content History and Development of the Boeing 737 - MAX A quick concise overview of the pages on this site

spacer

21 Dec 2018 - Final report issued EI-DYG 737-800 loss of thrust on both engines at CIA 10 Nov 2008

The accident occurred on 10th November 2008 at 06.56 (07.56 local time), at Ciampino Airport and involved a B-737-8AS type aircraft, registration marks EI-DYG. The Boeing 737-8AS aircraft, registered EI-DYG, took off from Frankfurt Hahn (EDFH) airport at 05.30 on 10th November 2008 bound for Rome Ciampino (LIRA) with 166 passengers and 6 crew members on board. As it was approaching the destination airport, it collided with a thick flock of starlings on the very short final approach. At birds visual acquisition, the crew interrupted the landing procedure, set a go-around manoeuvre but both engines did not provide the necessary thrust for the manoeuvre, the aircraft rapidly lost speed and height and impacted heavily on the runway. The main left landing gear detached from its anchoring on its route on the ground and the lower part of the left engine nacelle came into contact with the runway. The aircraft continued its run on the ground until reaching a complete halt at RWY threshold 33. The 6 crew members and 166 passengers evacuated the aircraft without further inconveniences.

The accident has been caused by an unexpected loss of both engines thrust as a consequence of a massive bird strike, during the go-around manoeuvre. The loss of thrust has prevented the aircrew from performing a successful go around and has led the aircraft to an unstabilized runway contact.

The following factors have contributed to the event:

  • the inadequate effectiveness of bird control and dispersal measures put in place by the airport operator at the time of the accident;
  • the captain decision to perform a go around, when the aircraft was at approximately 7 seconds from touchdown. The above decision was significantly influenced by:
    1. the lack of instructions to flight crew concerning the most suitable procedures to adopt in the case of single or multiple bird strikes in the landing phase;
    2. the absence of specific training in the management, by the flight crew, of the surprise and startle effects in critical phases of the flight.
See more details about the book

All of the information, photographs & schematics from this website and much more is now available in a 374 page printed book or in electronic format.

*** Updated 20 Nov 2018 ***

The 737 Tech Site on Facebook The 737 Tech Site on Twitter The 737 Tech Site on Instagram

Final report issued EI-DYG 737-800 loss of thrust on both engines at CIA 10 Nov 2008

Introduction

The accident occurred on 10th November 2008 at 06.56 (07.56 local time), at Ciampino Airport and involved a B-737-8AS type aircraft, registration marks EI-DYG. The Boeing 737-8AS aircraft, registered EI-DYG, took off from Frankfurt Hahn (EDFH) airport at 05.30 on 10th November 2008 bound for Rome Ciampino (LIRA) with 166 passengers and 6 crew members on board. As it was approaching the destination airport, it collided with a thick flock of starlings on the very short final approach. At birds visual acquisition, the crew interrupted the landing procedure, set a go-around manoeuvre but both engines did not provide the necessary thrust for the manoeuvre, the aircraft rapidly lost speed and height and impacted heavily on the runway. The main left landing gear detached from its anchoring on its route on the ground and the lower part of the left engine nacelle came into contact with the runway. The aircraft continued its run on the ground until reaching a complete halt at RWY threshold 33. The 6 crew members and 166 passengers evacuated the aircraft without further inconveniences.

ANSV was informed of the accident immediately after the event by ENAV SpA and carried out an operational inspection with its own team of investigators at the accident site on the same day of the event. ANSV sent notifications of the event, in compliance with the international regulations (Annex 13 to the Convention on international civil aviation), to the following bodies: ICAO, NTSB, BEA, AAIU. NTSB, BEA and AAIU accredited their own representatives in the investigation conducted by ANSV. The significant delay in publishing this report is due, on the one hand, to the retirement of some investigators, who, over time, have been appointed as IIC (investigator-in-charge), on the other hand, to the known shortage of personnel affecting the ANSV investigation area.

HISTORY OF THE FLIGHT

The Boeing 737-8AS aircraft, registration marks EI-DYG and radio call sign RYR41CH, took off from Frankfurt Hahn (EDFH) airport at 05.30 on 10th November 2008 inbound Rome Ciampino (LIRA) with 166 passengers and 6 crew members on board. The flight took place without any significant event until the approach phase at the destination airport. The aircraft established the first radio contact with Ciampino TWR communicating that it was 9 NM from the runway and stabilised on the ILS for RWY 15. The aircraft, authorized and configured for the approach, was proceeding for landing, when, in very short final, it collided with a thick flock of birds (later identified as starlings). The crew interrupted the landing procedure, initiating a go-around manoeuvre. With both engines not delivering the necessary thrust and a flight attitude set for climbing, the aircraft rapidly lost speed and height, hitting the runway heavily in proximity of the taxiway AC . Then, on first contact with the runway, which occurred with the main landing gear properly extended and with the lower part of the fuselage tail section, the main left landing gear detached from its anchoring during the landing run and the lower part of the left engine nacelle came into contact with the runway. The aircraft stopped near RWY threshold 33. The Fire Brigade vehicles immediately arrived at the aircraft and sprayed extinguishing foam around it, particularly in the area where the engine nacelle had come into contact with the runway. The Captain then arranged for the disembarkation of the passengers and crew using a ladder truck from the right front door, with the addition of the right rear slide, later activated and used.

Damage to Aircraft

The main left landing gear was torn from its anchoring, with its support leg sticking out from the upper surface of the left hand wing. The lower part of the left hand engine nacelle came into contact with the runway, the thrust reverse panels (doors) in the open position. The lower part of the fuselage, in the tail section, presented obvious signs of sliding and deformation of the structure caused by contact with the runway. There was also buckling of the passenger cabin floor frame. Not less than 86 bird impact points were identified on the radome and the front part of the fuselage, wing leading edges, flaps undersides, engine nacelles and landing gear. There were numerous and solid organic residues and bird feathers on stator and rotor blades of the engine fans.

PERSONNEL INFORMATION

Captains Flying Experience: the pilot had been employed, as Captain, with the aircraft operator involved in the accident for about 3 years. He had been previously employed by another operator on the same type of B737 aircraft, also acquiring familiarity with Rome Ciampino airport. The Captain held a Type Rating Instructor qualification on B737 300-900 type aircraft, although did not have any training assignments with the operator involved in the accident. At the time of the accident he had flown 9883 f/hs, of which 6045 f/hs on B737 type aircraft. The accident flight was the first flight of the day for the Captain. The Captain was in his second day of duty: he had gone into service the day before, following a rest period of 5 days.

First Officers Flying Experience: the pilot had been employed, as a first officer, with the aircraft operator involved in the accident for about 6 months (since May 2008). The pilot had obtained the commercial pilot licence on 3rd April 2008 and taken the theoretical examination for the airline pilot licence with a JAA member State. The pilot had subsequently obtained the B737 300-900 type rating on 24th April 2008, i.e. about 7 months before the accident. At the time of the accident, he had flown 600 f/hs, of which 400 on B737 type aircraft.

AERODROME INFORMATION

The Ciampino Airport AIP Italia, in the additional information part, showed only the presence of hooded crows in the airport grounds, all year round. The NOTAM in force did not report any warning regarding the presence of birds at Ciampino Airport.

Procedure for bird strikes prevention at Ciampino Airport

In 2003, a B737-400 experienced an impact with a flock of starlings a few metres before the contact with the RWY when landing at Ciampino. The occurrence was considered in the analyses of birdlife conducted by the airport operator. The naturalist environmental research and the observation activity conducted by the operator in the following years, as described in the 2008 annual report, highlighted that between 2004 and 2008 (until the EI-DYG accident) the number of bird strikes reported against the number of movements varied between 3.16 and 1.36 per 10,000 movements.

On the basis of the data for this activity in 2005, the operator had selectively cut down 11 stone pine trees which were near the airport terminals and parking areas. It can be seen from the study published by the airport operator after the accident (precisely on 19/11/2008) that the operator, when defining the risk for planning an adequate bird dispersal system, took into consideration not just the above, but also the data from a comparison of the same period of the year between 2007 and 2008, and in particular starlings at Ciampino Airport, which showed a reduction of about 56%.

Scheduled inspections on the day before the accident and on the day of the accident

The first two scheduled inspections (sunrise and 13.00 hours) on the day before the accident did not detect any presence of birds. A scheduled inspection (sunset) took place on the evening of the day before the accident, starting at 16.25 and ending at 17.00, which had shown the presence of about 1000 starlings, resting on the grass, affecting sectors C2B and C3A and 300 starlings resting on the grass in sectors C1B and C1A. The BCU dispersed the birds using a total of 35 gun salvoes. No bird carcasses were found in that circumstance.

On the morning of the accident, the scheduled inspection was carried out starting at 06.20 hours LT (05.20 UTC) and ending at 06.55 hours LT (05.55 UTC). These times have been taken from the inspection form and are consistent with the times associated with radio communications between the BCU and the TWR. During the inspection no presence of birds was reported. As from communications between TWR and BCU at 05.21.15 hours and reported above, the latter had carried out the above inspection on the manoeuvre area not interesting the runway.

Flight data recorders

From birds visual acquisition to impact with the ground

From 06.55.49 it has been recorded a reduction of the angle of descent and of the VS, which went from -752 ft/min to -391 ft/min at 06.55.51, the time when the TOGA command was recorded. In detail, at 06.55.49 (corresponding to the second when the CVR recorded the Captains exclamation of «Ahi!» repeated about 10 times, as it will be shown later, following the sighting of the birds), the aircraft was aligned with the localizer, at a radio height of 136 ft and a CAS of 149.5 kt with engines at 62% N1. At 06.55.51, as mentioned, the go-around pushbutton switch was recorded as being activated with the first push on the TOGA switch followed by a second push after 3 seconds at 06.55.54 to command the full go-around N1 limit. When the TOGA was applied, the vertical speed underwent to a further sudden reduction, and in the aircrafts descent trajectory a slight increase in the radio altimeter height has been recorded (from 108 to 173ft) up to 06.56.01. After that time, the aircraft returned to descend.

At 06.55.56, the selection of flaps from 40° to 10° position was commanded. This was reached at 06.56.12, with the aircraft on the ground. The FDR data show that the Captain took aircraft control at 06.55.58, i.e. 7 seconds after activation of the TO/GA. From the time that go around was commanded, with the consequent change of attitude, the speed recorded a progressive decrease. At 06.56.07, with a CAS of 122.75 kt, the sink rate warning activated, followed at 06.56.09 by the stick shaker, time corresponding to the maximum vertical speed of -1360 feet/min reached by the aircraft. At 06.56.10, the aircraft touched the ground (WOW switch) with an attitude of about 10° pitch and -6° roll, indicated speed of 120.75 knots, VS -1064 ft/min, vertical acceleration of 2.66 g and and lateral acceleration -0.45g.

Landing roll

After the touchdown, the aircraft decelerated at 06.56.10 until 06.56.38, using the brakes, engine no. 1 thrust reverse and the spoilers. The warning of unsafe condition of left hand landing gear activated during the ground run. With the aircraft stationary on the runway, the flaps were lowered to 40° and the spoilers set to zero.

Attitude and decelerations

The following data plot highlights how the aircraft, from the visual acquisition of the flock of birds, progressively changed the flight path angle, which was correctly set at around -3° before the birds visual acquisition, reaching up to about 4° at the go around phase then going down to the minimum values (-7°). The activation of the sink rate warning (at about -6° flight path angle) and of the stick shaker were recorded in these phases. At stick shaker activation, the angle of attack was about 21°. In the phase when the flight path angle went from -3° to about 5°, the pitch angle changed from about -1° to 9° and then remained a few degrees positive until the impact with the ground. The aircrafts magnetic heading was also characterised by left and right deviations for the runway direction (150°), together with deviations on the roll axis. The aircraft impacted the runway with a roll angle of about 6° to the left and a pitch angle of about 10°.

Engines behaviour

With regards to the behaviour of the two engines, both experienced a stall condition (state 3) when TO/GA was selected, the right hand engine at 06.55.51, the left 1 second later; the N1 RPM of both engines dropped from 66% to a value around 40%, where they remained until impact with the ground, after which the left engine increased RPM up to 81% through activation of the thrust reverse, whilst the right remained at values below 45%, until being shut down. A sudden rise of the EGT of both engines was recorded coinciding with the reduction of the N1 rpm.

CVR: From visual bird acquisition to impact with the ground

It should be noted that, at 06.55.49, whilst the FO was commenting «Nice», the Captain had begun to exclaim «Ahi!» repeated about 10 times in the following two seconds.
At 06.55.51, a loud bang was heard, with a change of in the engines noise, which had become irregular.
At 06.55.52, the Captain ordered «Go around... go around... go around».
At 06.55.53, the FO confirmed «Go around, flaps 15».
At 06.55.54, the Captain repeated «Go around... go around».
At 06.56.07 and in the next second, the synthetic sink rate warning was played, and repeated twice.
At 06.56.09, the activation of the stick shaker was heard and after a second the exclamation of the Captain and the noise due to the hard landing.

CVR: deceleration run on the ground and immediate action after landing

The aircraft hit the ground at 06.56.10; at 06.56.18, with the aircraft on the runway and decelerating, the Captain confirmed «My control». At 06.56.38, the aircraft came to a halt on the runway and simultaneously the Captain, via interphone, ordered «Remain seated» and repeating the order twice. From 06.56.44 to 06.56.47, the Captain turning to the FO ordered «We do... open the outflow valve.», starting the procedure to make the aircraft safe. Simultaneously, the FO communicated to Ciampino Tower by radio «Is maintaining on the runway, MAYDAY.». After which the recording ended when both engines were switched off.

Ground evidence and wreckage distribution

Ground evidence that can be associated with the hard contact between the aircraft and the ground are near the taxiway AC intersection with the runway, intersection positioned about half way along its total length. They are shown in the following image, in order of progress:

  • marks of impact of the lower part of the tail fuselage with the asphalt surface of taxiway AC ;
  • marks of impact of left hand landing gear on the runway asphalt surface;
  • marks of impact of the right hand landing gear on the grass adjacent to the right edge of the runway;
  • continuous marks of left hand engine and left hand landing gear scraping from the first point of contact to the final resting point of the aircraft.

An angle of about 20° between the direction of the aircraft on contact with the runway and the axis of the runway emerges from the set of these marks. In the next image, an overall view of the marks is reported.

A large number of remains of dead birds were spread in the area at about 100 metres before the runway 15 threshold and more precisely in the grass section between Center line barrette no. 2 and Center line barrette no. 3 of the CALVERT approach system. The majority of the organic remains were concentrated inside an ellipsoidal area, closer to the Center line barrette no. 2 pole, the larger axis of which was about 35 metres and the shorter one about 10 metres.

About 120 starling carcasses or remains of them were identified and recovered during the clearing of the area, which were carried out with the support of an ornithologist. No organic remains belonging to other species of birds were identified in the recovered material.

Wreckage examination

The aircraft came to a stop close to RWY 33 threshold, resting on the main right landing gear, front landing gear and left engine nacelle. The right hand side slide had been deployed during the disembarkation of the passengers and crew from the aircraft.

Fuselage

The fuselage shows multiple bird strikes (no less than 86), visible in the next images. The windshield panels show some bird strikes in their top central part. There are clear signs of damage caused by the impact and scraping along the asphalt runway on the lower tail part of the fuselage. The lower part of the passenger cabin floor structure shows extended deformation.

Wings and relative surfaces

There are multiple damage points on the left hand wing and buckling on both the top surface and underside of the wing. The flaps showed some dents.

Landing gear

The main left hand landing gear is seriously damaged, its shock absorber has gone back into the wing. The main left hand landing gear doors are heavily damaged.

Engines

Both engines show many bird strike points on their respective air intakes. The left hand engine nacelle shows damage from scraping along the asphalt; the thrust reverse doors are locked in a partially open position.

TESTS AND RESEARCH

Both engines CFM56-7B26/3, S/N 896379 (#1) and S/N 896387 (#2) were removed from the aircraft on 14th January 2009 and sent to the GE facilities in Cardiff (UK) where, starting from 5th February 2009, the tear down operations were carried out on the modules and components of both engines. Both engines had 2419 hours and 1498 operating cycles since new. The significant elements emerging from the investigation carried out are shown below. The external visual inspection of the fans showed 55 impact points with birds on the left hand engine fan, 30 points on the right hand engine fan, both fans without apparent damage and deformation of the blades aerodynamic profile, slight deformation of the fan case engine #1, due to the contact with and subsequent scraping of its nacelle along the runway.

There was a considerable amount of organic material inside the fan, the compressor booster, the high pressure compressor and the combustion chambers of both engines, a greater quantity in #2 compared to engine #1, despite lesser traces of impact in its FAN. The organic residues found produced a severe alteration of the compressor blades aerodymanic profiles and partial airflow disruption, affecting the various sensors, and inside the combustion chambers. A limited presence of metal splatter was also identified on the hot parts of both engines, not such as to change the functional characteristics of the parts concerned. The accessories of both engines underwent specific inspections and functional tests, without any anomalies being found. No faults were found on both engines prior to the ingestion of the birds.

Based on the engine parameters recorded by the FDR and following the analysis of the data taken from the ECC of both engines, the investigation carried out allowed to understand the engines behaviour following the massive ingestion of birds. Immediately after the bird strike and application of the TO/GA command (corresponding to the vertical red line in the following plotting), both engines experienced a drop in N1 RPMs, from 62% to 52% and 46% for the left hand and right hand engine respectively. The N2 remained substantially stable at about 83% (a drop of about 3%), whilst the EGT temperature underwent a sudden increase of about 300°C in the next 24 seconds. The N1 drop and rise in the EGT took place at the same time as a series of surges2 experienced by both compressors and indicative of a severe aerodynamic disturbance of the air flow inside the HPC.

These repeated surges caused a fluctuation of the pressure values transmitted from the High Pressure Compressor Discharge Pressure (CDP), with a corresponding drop in the N1 (from 62% to about 40%) and N2 (stabilised on 83%) rpm and a sudden increase in the EGT values, due to a loss of air flow crossing the engine. Considering that N1, N2 and CDP are parameters used by the FADEC to control the fuel flow; the drop in N1 and N2 rpm caused the FADEC to try to restore the rpm by increasing the fuel flow. However, the latter is limited by the engine's acceleration data, which are a function of the CDP, N2 and T25 parameters. The low CDP values caused by the compressor surges led to fuel flow values that were too low to maintain the N1 level required by the crew. As can be seen from the next plotting, the engine levers remain in full forward position until 06.56.05, i.e. 5 seconds before impact on the ground, to then be pulled back slightly (MCL position); during the deceleration run after impact with the ground, both the engine levers were commanded from MCL to IDLE and the thrust reversers (T/R) were activated, after which, as follows, different behaviour for the engines were observed:

  • the N1 on the left hand engine increased to about 80%, ensuring the proper functionality of relative thrust reverser,
  • the N1 rpm on the right hand engine did not increase, staying at values of 40% and thus not generating thrust inversion through the T/R. The same engine only seemed to restore its proper functionality in the final phases of the aircrafts decelleration on the runway, with the engine lever at idle.

The vibration levels recorded for the FANS and HPC of both engines shown in the previous plotting do not indicate anomalous vibrations in the seconds prior to the TOGA command being activated, vibrations that may possibly be an indication of bird strikes. The FAN and HPC vibrations trend in the approach and short final in the previous flights indicates engines vibrations levels similar to the ones indicated in the accident flight.

Operating procedures

The go around is a manoeuvre that is performed for safety reasons when the pilot decides or is instructed to interrupt the approach and abort the landing.

In conclusion, no considerations or guidelines were found in the operations manuals in force at the time of the events which could have supported the decision making process that determines the execution of a missed approach rather than a landing when a non-normal condition arises in the terminal phases of an approach. Whilst, for example, with regard to the procedure to abort on take-off, a highly critical situation both for the time available to make the decision and for the aircrafts performances, there are detailed guidelines in the FCOM regarding the criteria to adopt for the decision to interrupt or proceed, this does not occur for the landing phase and missed approach. As also highlighted recently by specific studies4, the go around manoeuvre can present critical issues, even though it falls under normal procedures.

 

Analysis

AIRCRAFT

Based on the witness statements, data and evidence available, the aircraft did not show any technical malfunctions until the bird strike. The damage caused to the fuselage and flight controls by the bird strike did not affect the ability to control the aircraft; however the same bird strike severely affected the engines operations. A specific section dealing with engines behaviour follows.

Fuselage

The damage on the fuselage is fully consistent with the attitude and speed with which the aircraft came into contact with the runway and the impact with the flock of birds.

Landing gear

The high vertical speed (1064 ft/min) and the aircraft attitude, not levelled (about 10° pitch up and 6° roll to the left), has caused the aircraft to initially touch the runway/taxiway AC with the left landing gear and tail and then with the right and nose landing gear. The vertical acceleration recorded at the time of contact with the runway (2.66 g) is not particularly high; the structural damage to the left hand landing gear can be explained by the asymmetry of the previously described landing, where the energy of the impact was absorbed by the left hand landing gear, causing its strengthening structures to collapse and the shock absorber to penetrate the left hand wing structures. The collapse of the left hand landing gear caused the lower part of the left hand engine nacelle to come into contact with the runway surface. During the landing run, the braking action was exercised through the brakes, the spoilers on both wings and thrust reverse on the left hand engine, which doors remained open because of the interference of the engine nacelle with the runway surface.

Engines

The operations of both engines were regular until the TO/GA command was activated. When the go around was ordered with the simultaneous application of the TO/GA command, there was a significant drop in N1 rpm in both engines (about 17% for the right hand engine and 10% for the left hand engine) and a slight reduction of N2 rpm (about 3%), whilst the EGT temperature of both engines experienced a significant rise, about 300°C in the next 24 seconds. A stall mode 3 condition was recorded on both engines within 1 second of the TO/GA command being applied, this is indicative of a deep stall on both engines and consistent with the drop in N1 and the sudden increase in the EGT. The N1 rpm sampling, carried out every 4 seconds, does not allow to exclude that the drop in N1 could have started in the 3 seconds that constitute the time interval between the datum recorded at 06.55.48 (N1 at about 62%) and the datum recorded at 06.55.52 (significant drop in N1); in particular, it does not exclude that the drop in rpm could have started in the two seconds preceding the activation of the TO/GA command, which occurred at 06.55.51. With regard to this aspect, it is useful to analyse the behaviour of one of the other parameters indicative of a stall condition, the EGT; in the time frame consisting of the 2 seconds between the last recorded N1 value (06.55.48) and the application of the TO/GA command (06.55.51), no significant changes in temperature were recorded, but its value increases significantly only after the TO/GA command is activated (by about 100°C, 2 seconds after application of TO/GA). The analysis carried out on the vibration speeds recorded for the FAN and HPC of both engines indicates the absence of anomalous vibrations in the seconds preceding the application of the TO/GA command. The investigation carried out at GE Cardiff indicates the presence of considerable quantities of organic remains in the FAN boosters and cores of both engines as the cause of the surges experienced by both engines, which significantly changed the aerodynamic flow inside them, so much so as to generate the surges as of above. The available data indicate that the effects of this change in aerodynamic flow (large changes in N1 and EGT) are clearly simultaneous to the application of maximum power via the TO/GA command; this change of the aerodynamic flow did not produce effects on the engine parameters before TO/GA application.

The considerations above allow two hypotheses to be drawn regarding the time when the ingestion of the birds by the engines could have occurred:

  1. at the same time as the application of the TOGA, given the absence of significant changes both in terms of vibrations and engine parameters before that time;
  2. before the application of the TOGA command, without this causing changes in the main engine parameters.

The first hypothesis is without any doubt the more probable one; it can be reasonably assumed that the birds were ingested at the same time the TO/GA command was applied, since the FDR had not recorded significant changes in the operation of both engines before that moment. The number and total weight of the birds ingested must be evaluated for an assessment of the engines compliance with the certification requirement.

The number of impacts on the fans (55 and 30 for left hand and right hand respectively), though not being directly indicative of the number of birds that were actually ingested by the individual engines, given the fragmentation of the bird that can occur after the first impact and the subsequent multiple impacts that the same bird can cause, is however indicative of a considerable quantity of ingested birds. This consideration is confirmed by the significant quantity of organic remains found inside both engines (greater in the right hand engine, despite the smaller number of impacts on its fan) and considered the cause of the surges experienced by both engines. The above leads to assume that the number and weight of the birds ingested by the engines was certainly high, even if they cannot be quantified precisely. The applicable certification standards (CS-E 800 and, with some differences to the previous, FAA 14 CFR 33.76) require, in the event of impact with small size birds that their ingestion by the engines does not lead to any of the following conditions:

  1. loss of more than 25% of power under TOGA thrust;
  2. the engine being shut down.

The requirements set in terms of number of birds and total weight vary according to the Engine Inlet Throat Area (the area with Dt diameter, shown in the image below). The Engine Inlet Throat Area for the CFM56-7B26/3 engine is 2570 in2 (1.658 m2). For this Engine Inlet Throat Area value, the standard shows a maximum of 16 birds with a unit weight of 85 g (corresponding to the average weight of an adult starling). In light of the discussion concerning the inability to precisely determine the number of birds actually ingested by both engines, it is nevertheless possible to assume that the quantity of birds ingested by engines was very likely greater than the max amounts required by certification standards for which the CFM56-7B26/3 engines were certified. According to the above, there were no design/certification deficiencies regarding the engines, since they had ingested a number and a weight of birds that were greater than the max amounts indicated by the certification requirements.

FLIGHT EXECUTION

From birds visual acquisition to ground contact

At 06.55.49, whilst the FO was busy observing the altitude and speed instruments to regain the correct parameters and commented «Nice» to confirm the positive correction in progress, the Captain exclaimed «Ahi!» (repeated about 10 times in the next 2 seconds), realising that the flock of birds was seen again on the approach path.

At 06.55.51, at an altitude between 136 and 112ft radalt and a distance of 100 m from the runway threshold, almost simultaneously ed «Go around», repeated three times and the TO/GA pushbutton was activated. Based on the witness statements, it cannot be excluded that the Captain intervened instinctively on the throttles, anticipating, de facto, the action ordered verbally to the FO.

The throttles were positioned fully forward and the flaps commanded from 40° to 10°, overcoming the detent positioned in the 15° position to prevent retraction beyond the position requested for the go around manoeuvre. From the statement made by the Captain, the selection was not voluntary nor was there any awareness of the flaps selection error. This action is indicative of a reaction to an unexpected situation, typical of a phenomenon defined as startle effect

On activation of the TOGA, when the throttles were full forward, the CVR CAM channel recorded a loud bang , a noise confirmed by the Captain during the interview and attributed to the impact with the flock.

The CVR evidence, the engine parameters and the position of the maximum concentration of the bird carcasses on the ground all indicate that the impact with the birds (or better the crossing of the flock which caused multiple impacts) occurred simultaneously with the activation, via the pushbutton, of the go around. The go around activation coincided with the stall of both communication « go around», when the aircraft was about 1 second from reaching the RWY 15 threshold and about 7 to 8 seconds from the normal point of contact with the runway, with the aircraft between 136 ft and 112 ft radalt, 149.5 kts and N1 62%.

On activation of the TOGA (06.55.51), the Flight Director symbols appeared indicating the necessary attitude for the go around, the FO set that attitude, requesting the flaps at 15°: both engines, instead of increasing rpm and providing the necessary thrust for the go around, experienced an N1 drop from about 62% to values around 40%; the vertical speed had a sudden reduction, causing a slight increase in the radalt (from 112 to 173 ft), until 06.56.01. The above represented a substantial absence of positive rate of climb, which therefore did not induce the crew to take the subsequent actions provided by the checklist for go around, among these the landing gear retraction. At 06.55.56, the flaps selection from 40° to 10° was commanded. This position was reached at 06.56.12, with the aircraft on the ground. The FDR data show that the Captain took the aircraft control at 06.55.58, i.e. 7 seconds after activation of the TO/GA. There were no communications between pilots regarding the handover of aircraft control until it started its deceleration run on the ground. The Captain reported that in that circumstance he promptly intervened on the controls and noted that he did not have the thrust to continue the go around, then turned all his attention to "look outside". He then realised he was inside the airport perimeter (Captains exclamation «On est dedans») and capable of bringing the aircraft, which had lost its alignment on the right likely due to asymmetric engines thrust at go around, to land on the runway.

The change of attitude and lack of engines thrust led also to a rapid airspeed drop and angle of attack increase. From 06.56.01, the aircraft began to lose height, despite the nose up command still applied by the aircrew; the sink rate activation was recorded at 06.56.07, with a CAS of 122.75 knots, the stick shaker activation, at 06.56.09, corresponding to the max vertical speed of 1360 feet/min reached by the aircraft (indicative of an aerodynamic stall condition) and a progressive speed reduction and angle of attack increase, which reached values of 120.75 kts and 21° of AoA respectively, 1 second before the runway contact, when the aircraft was at a radalt height of 21 ft. At 06.56.10, the aircraft touched ground (WOW commutation) with an attitude of about 10° pitch and -6° roll, an indicated speed of 120.75 knots, variometer of -1064 ft, vertical and lateral acceleration of 2.66 g and -0.45g respectively.

The aircraft had therefore lost height because of the aerodynamic stall, with the engine levers at maximum power position but without response from the engines and impacted the ground near taxiway "AC", about half way along the total length of the runway, with a vertical acceleration of 2.66g. The possible reasons behind the Captainsdecision to go around were investigated. The following statements from the captain provided some elements: «we are heavy, we have a bit of tailwind, we know that Ciampino has a bit short runway, it is CAVOK condition, we know we have Fiumicino close-by.»; «we know at that point that we hit something important. So maybe we have damages to the gear. We dont know. I have doubt. As we were taught at that time, in case of doubt: go around. We were trained for that! And I believe it is still applicable. In case of doubt, go around.». The Captain also confirmed that the Ciampino RWY length (2207 m), defined as a bit short compared to other airports such as the one he came from (Hahn) or the alternate one (Fiumicino), was considered a critical issue by him if associated with any problem or an instability on the parameters, such as to influence the length of the landing run, either as a consequence of an imprecise point of contact or through the reduction of the braking performance due to a failure.

In this regard, the landing distance necessary to bring the aircraft to a stop was calculated; it was extrapolated from the QRH Advisory Information in order to assess whether this perception of the critical issue was actually justified in the specific conditions of the event under analysis.

Considering the mass of the aircraft as 61,100 kg on landing, a tailwind component of 5 kt, a velocity of Vref+10 kt, a temperature of about 10° below ISA, dry runway, max manual braking, use of T/R and height of 100 ft on the threshold (50 ft above the predicted glide path), the ALD 7 is about 1341 m., therefore quite less than the available runway length. Nevertheless, although this value may not be critical, it is necessary to point out that this datum, taken from the QRH, is only applied in emergency conditions or during the tactical re-planning phase. In fact, the value that is considered during planning, as provided by the applicable regulations, is the RLD8, which is calculated multiplying the ALD by a factor of 1.67. It is also a common perception, supported by performance tables and confirmed by other pilots operating at Ciampino Airport with the B737, that with an aircraft landing mass above 60.000 kg, the presence of a tailwind component of more than 5 knots makes necessary to pay particular attention to maintain the parameters and be precise with the runway point of contact.

Moreover, it is obvious that a recalculation of the ALD in the variety of contingent cases and in light of any additional failure cannot be carried out when approaching, but only once the latter has been interrupted, with the aircraft under control and separated from obstacles. Though the data considered during the approach planning are not available, it is therefore possible to assume that the Required Landing Distance known by the Captain was such that the Captain felt the runway length available as critical, in presence of contingent penalties (destabilisation of the aircraft in the final phase, tailwind component).

Landing roll

The ground marks indicate that the impact first occurred with the lower part of the fuselage tail and the main left landing gear against the asphalt surface of taxiway AC , then with the main right landing gear on the grass adjacent to the right edge of the runway, followed by the collapse of the left hand landing gear and runway contact with the left engine nacelle. The aircraft continued its ground run decelerating through the brakes, spoilers and left engine thrust reverse only, which reached a maximum N1 value of 81% during the deceleration run. During the ground run, the Captain confirmed he had aircraft control, uttering «My control». 9 seconds after runway contact (06.56.19), the left landing gear unsafe warning activated via the L GEAR WARNING caption. The aircraft came to a stop on the runway at 06.56.38, about 50 m from the end of the runway. Both engines were cut off at 06.56.52 (ENG CUTOFF).

After landing, the wind screen showed some traces of bird strike, but not so as to compromise the flight crews outside field view, as confirmed by the Captain during the interview.

HUMAN FACTOR

Bird strike: survey of existing guidelines/instructions Regarding the go around manoeuvre, it seemed advisable to carry out a survey of the existing guidelines/instructions related to actions to be taken in the event that birds are encountered during the short final of the approach phase.

Human factor final considerations

The go around is a manoeuvre carried out for safety reasons, for example:

  • no visual acquisition or loss of landing references;
  • wind speed sudden change, such as to compromise the safely approach;
  • evidence of a runway incursion;
  • in case of a not-stabilised approach.

The failure to not carry out the go around, or to not carry out it promptly, are amongst the main factors that determine significant occurrences and runway excursions. Nevertheless, the manoeuvre in itself and the management of the subsequent flying phase following this manoeuvre are not risk-free

The go around manoeuvre ordered by the captain does not find anything against such decision in the training and operational documentation examined and in force at the time of the accident. The set of regulations, in fact, emphasized the opportunity to proceed with a go around in case of unexpected events in landing final phases; consequently and consistently with this approach, no guidelines had been implemented as a result of a preventive assessment regarding the opportunity to go on with a go-around or to complete the landing manoeuvre in circumstances such the ones occurred in the accident. The above was confirmed by the commander in his declarations, when he stated: «As we were taught at that time, in case of doubt: go around. We were trained for that! And I believe it is still applicable. In case of doubt, go around.».

The decision to go around in the very short final, following the bird strike, placed the aircraft in a very critical situation, much more critical than, with hindsight evaluation benefit, the aircrew would have faced in case of an immediate landing. We could consider, for instance, a case in which the aircraft has begun to lose height outside the airport boundaries or, following an initially positive variometer reading, the crew has decided to retract the landing gear. In such scenarios, with a possible failure of both engines to provide the thrust needed, whilst the aircraft is at an height and position no longer suitable to guarantee an emergency landing on the airport runway, the risk associated with the decision to go around is higher than to land. The theoretical preparation to execute the go around manoeuvre and its practical training cannot ignore to identify specific scenarios where the risk associated with the decision to go around changes significantly. Among these scenarios, it seems useful to consider and characterize a bird strike scenario on approach phase. An assessment of the literature analysed, leads to consider that, in circumstances such as the ones occurred, it would have been less risky to land than to go around. Such assessment is based on the fact that, from an energy viewpoint, the landing of the aircraft (at 7/8 seconds from touchdown) was assured. It has been also considered that the process leading to the decision to perform a go around took place in few seconds (likely 2) and in a flight phase where the crew, at least the FO PF, was prepared for an imminent landing. This decision-making process took place without any precursor or warning, as can be deduced from the relaxed climate in the cockpit during the final phases of the approach, the absence of exclamations/comments indicative of a warning/alarm situation. f a degradation of situational awareness, typical of the surprise effect.

Questions were raised regarding the possibility that a go around minded pilot in similar cases can proceed very likely with a go around, not properly assessing the consequences of such manoeuvre, in order to avoid marginal damages during landing, which could be later attributed This can lead to the consequence that the pilot, doubtful and with very little time available for decision making, decides to go around, without an adequate risk assessment. In the analysis of the present case, it has been considered that the go around decision could be due to one or more of the following reasons: 1. an attempt to avoid the flock of birds; 2. a loss of visual contact with the runway; 3. a verified or suspected failure, to be assessed and resolved before landing; 4. a loss of stabilisation from the approach path following an birds avoidance manoeuvre.

The interview with the Captain allowed to eliminate the first two hypothesis, the decision to go around was rather justified with a combination of the last two. The decision was certainly influenced by: the awareness to land on an airport with a limited runway length (2207 m) with an aircraft almost fully loaded with passengers and with a tailwind component, a situation in which every significant deviation from the approach parameters or destabilisation entails a go around; having an alternate airport, such as Fiumicino, available in the immediate vicinity with a runway of 3 kilometres, with optimal meteorological conditions and with no concerns about the fuel available; the fear that, following multiple impacts, the braking system could have been damaged or functionally limited; the possible conditioned reflex caused by the training received, where in case of doubt it is better to go around together with the awareness that accidents often are cause by the decision to continue a not stabilized approach rather than proceed with a go around. It is assumed as highly improbable to be able to conduct an adequate and sound risk assessment in situations like the one the aircrew faced, due to lack of available time to decide which course of action to adopt. Such risk assessment should be carried out preferably before the approach. In absence of specific norms and guidelines on the matter (as confirmed by EASA to ANSV) it is considered essential to provide flight crews with specific guidelines, in the theoretical and simulator phase of training, with reference to the different possible scenarios, by operators. In fact, differently from what happens, for example, with the take-off phase when many considerations are made in the FCOM and FCTM of most aircraft regarding when to abort or not, for landing phase there were no such instructions in the B737 FCTM at the time of the accident; they were only introduced in the same manual after the EI-DYG accident. The accident operator deemed opportune to introduce a training aimed at managing such scenarios, but only after experiencing this event.

In conclusion and in light of the previous considerations, it is considered essential that guidelines similar to the ones introduced by Boeing in the FCTM on bird encounter and bird strike in critical flight phases, such as take-off approach, and landing, are extended by the manufacturers to all types of similar aircraft and become part of the operators theoretical/practical training for flight crews.

The event is characterised (at least for the EI-DYG flight crew) by the absence of warning signs, in a phase where the aircrew was mentally prepared to land, close to transition from the approach phase to the landing phase . Such situation most probably generated a surprise effect , confirmed by the captain not standard exclamation («Ahi!» repeated about 10 times) rather than a less instinctive verbal communication (e.g. birds ahead ).

The startle effect can also explain the instinctive reaction that led to exceed (for reasons the pilot could not explain) in selecting the flaps lever to the 10° position instead of 15°, as requested by the go around procedure, passing the physical detent that prevents the 15° position from being overcome.

The surprise effect or startle effect can negatively impact situational awareness which, with little time available, generates inability to correctly carry out the information processing, decision making and problem solving: cognitive science teaches us that the reactions and decisions taken instinctively are not necessarily the most correct. Such negative impact should be eliminated/minimized through a specific training. In particular, the negative effect should be addressed in the most critical phases of flight. In such phases, possible unexpected scenarios (although plausible) should be considered, against which to develop procedures and strategies able to mitigate the negative effects. The above training should prepare flight crews to properly respond to the unexpected events. Among these training tools, also clear and positive communication techniques should be developed to indicate a threat .

In the subject case, the event of ird encountering on final" should generate, i.e., a call out by the pilot who first acquires the threat , like: irds at 12 o'clock , rather than an instinctive and too generic expression as Ahi Such standard phraseology, in a CRM perspective, could take the other crew members at the same level of situational awareness, for better decision making and actions. Ie I land or go around.

It is believed that types of training aimed at managing th urprise/startle e proposed in a way that permits to properly deal with non-routine situations. The BEA, following more recent events16, highlighted the following: Initial and recurrent training as delivered today does not always promote and test capacity to react to the unexpected. Indeed, the exercises are repetitive and well known to crews, and do not enable skills in resource management to be tested outside of this context . For such reasons BEA sent to EASA the following recommendation: «EASA review the requirements for initial, recurrent and type rating training for pilots in order to develop and maintain a capacity to manage crew resources when faced with the surprise generated by unexpected situations; [Recommendation FRAN 2012 042]»; «EASA ensure that operators reinforce CRM training to enable acquisition and maintenance of adequate behavioural automatic responses in unexpected and unusual [Recommendation FRAN 2012 043].». BEA refers to a situation taking place at high altitude and during cruising, hence different from the one investigated here; however it is believed that the principle included in the above recommendations, to have a specific training for unexpected events, it is also applicable to the present event. -off and landing, such training should be considered as essential.

CONCLUSIONS


3. GENERAL

This chapter shows the facts ascertained during the investigation and the causes of the event.

3.1. EVIDENCE

  • The flight crew held the necessary aeronautical qualifications to carry out the accident flight.
  • It was the first flight of the day for the crew.
  • The FO (CM-2) was PF and the Captain (CM-1) PM.
  • The start-up, taxiing, take-off and climb were carried out uneventfully.
  • The meteorological conditions at the destination airport were optimal.
  • The PF carried out a thorough approach briefing, including the missed approach procedure.
  • The Ciampino Airport AIP, in the additional information , warned only about the presence of hooded crows on the airport all year round.
  • The NOTAM in force did not report any warning regarding the presence of birds on Ciampino Airport.
  • The crew had not been informed by any means about birds presence/risk; consequently they did not deal with this aspect in the briefing.
  • The Airport Manual comprised the bird dispersal procedures in accordance with the ENAC APT-01A circular.
  • The ENAC APT-01A circular indicated that BCU should have intervened not only in case of bird dispersal but should also provide continual vigilance on the environment and disturbance of the fauna, in such a way as to make the latter perceive the airport as un unpleasant and unsafe place.
  • The Airport Manual (section MOV/11) dictated the execution of scheduled inspections (sunrise, 01.00 pm and sunset), on request inspections (with the intervention of the BCU limited to the time necessary to perform inspection and bird dispersal) or assessments following a reported presumed bird impact.
  • The first of the three scheduled inspections, dictated by the Airport Manual in force at the time of the accident, had been duly carried out between 05.20 and 05.55 on the accident day, without any significant birds sighting.
  • According to the communications between the TWR and the BCU, the latter had carried out the morning inspection not interesting the runway.
  • Before the accident, 16 movements (take-off or landings) were recorded for RWY 15 at Ciampino Airport (8 movements after the BCU inspection, which ended at 05.55); the last of these flights
  • It is reasonably possible that the flock positioned itself in the place from where it later took off, between the landing of a Saab 340 aircraft and the moments immediately preceding the impact with the EI-DYG.
  • None of the crews of the 8 flights that interested the runway after the end of the BCU inspection reported any presence of birds affecting the operations.
  • The approach to Ciampino was flown manually (no autopilot, no autothrust, no FD) to train the CM-2, with an ILS-Z procedure for RWY 15.
  • The ILS was flown as a stabilised approach consistent with the requirement of the operator approved procedures.
  • The manual approach showed some deviations from the localizer, corrected by the CM-2 following verbal input from the Captain.
  • The final configuration for the landing with flap selection at 40° was reached at 06.54.52, at a radio altimeter altitude of 996 ft and a distance of about 2.5 NM from the field.
  • On the MINIMUMS callout there were also almost simultaneous callouts of Captain («Continue») and the FO («Land»).
  • At a radalt height of 136 ft and a distance of about 300 m from the runway, the Captain, A in rapid sequence.
  • At a radalt height between 136 and 112 ft and a distance of about 100 m from the runway, the TO/GA pushbutton was activated.
  • The FO gave the acknowledge: «Go around, flaps 15», setting the go around attitude.
  • Both engines stalled and the CVR recorded a loud bang at the same time as the TO/GA was activated.
  • The bang corresponds to the impact with the flock of birds, which occurred when the aircraft was about 100 m from the RWY 15 threshold, with the aircraft position just overhead the area were the maximum concentration of bird corpses was found.
  • On go around, the flap lever was positioned at 10° instead of position 15°, required by the go around procedure with two engines.
  • On TOGA, both engines, instead of increasing rpm, experienced a drop, from 62% N1 to about 41%.
  • When the TOGA was applied, the vertical speed decreased rapidly; a slight increase in the radalt (from 112 ft to 173 ft radio altimeter) has been recorded until 06.56.01; the aircraft then continued to lose height, despite the nose-up command.
  • There was a progressive speed reduction and an increase of the angle of attack until the activation of the stick shaker, recorded at 21 ft radalt.
  • The aircraft hit the ground in aerodynamic stall condition halfway along the total length of the runway, with a landing mass of 61,100 kg and 3800 kg of fuel, with flaps transiting from 40° to 10° (position actually reached 12.1°) and a vertical acceleration of 2.66g.
  • 9 seconds after the contact with the ground, the unsafe left landing gear indication was activated.
  • The aircraft decelerated by means of brakes, spoilers and thrust reverse of the left engine only.
  • The aircraft came to a stop on the runway at 06.56.38, about 50 metres from THR 33.
  • orders, one of the rear slides, unlike the s orders.
  • Until the time of the accident, neither the manufacturer nor the operator had provided guidelines or procedures in the applicable manuals (FCOM, FCTM and OM) referring to the actions to take in the case of bird encounter or bird strike in the approach/landing phase.
  • After the event, the aeronautical literature underlined through articles the risks associated with a go around after a bird strike, providing reasons and considerations shown below: the full extent of the damage to the aircraft, and particularly the engines, might not be apparent until it is decided to apply power; after a missed approach manoeuvre, the pilot could be in a situation where the is not in a condition to fly; at an engine rpm typical of the approach phase, the damage can most probably be confined to engine fans and does not affect the cores; the high rpm, resulting from a missed approach manoeuvre, while penetrating the flock of birds, could cause greater damage to the engine and consequent loss of thrust.
  • The literature considered, suggests to land after a bird strike when in the approach/landing phase and not to carry out a go around.
  • After the event, Boeing with B737 FCTM introduced a recommended technique as prevention strategies and guidelines, to be followed by the crew in the event of bird impact in the take-off and landing phases. The concepts expressed herein agree with the ones in the above-mentioned aeronautical literature.
  • There is no evidence of guidelines adopted by other aircraft manufacturers in their flight and training manuals, guidelines similar to the ones introduced by Boeing in case of bird strike.
  • After the event, the operator developed specific training for birds encounter and bird strike on take-off and landing.
  • There are no guidelines at regulatory level (EASA) covering specific training for such scenarios to be followed by operators.
  • At the time of the event, there were no training guidelines aimed at managing and mitigating the surprise and startle effects, effects that influenced the actions carried out by the flight crew.

CAUSES

The accident has been caused by an unexpected loss of both engines thrust as a consequence of a massive bird strike, during the go-around manoeuvre. The loss of thrust has prevented the aircrew from performing a successful go around and has led the aircraft to an unstabilized runway contact.

The following factors have contributed to the event:

  • the inadequate effectiveness of bird control and dispersal measures put in place by the airport operator at the time of the accident;
  • the captain decision to perform a go around, when the aircraft was at approximately 7 seconds from touchdown. The above decision was significantly influenced by:
    • 1. the lack of instructions to flight crew concerning the most suitable procedures to adopt in the case of single or multiple bird strikes in the landing phase;
    • 2. the absence of specific training in the management, by the flight crew, of the surprise and startle effects in critical phases of the flight.

SAFETY RECOMMENDATIONS

4. RECOMMENDATIONS

In light of the evidence collected and the analyses carried out, the ANSV considers it necessary to issue the following safety recommendations.

RECOMMENDATION ANSV-12/1525-08/1/A/8 Type of recommendation: SRUR/SRGC. Reasoning: during the investigation, it has been considered that the training and operating instructions might not always provide sufficient elements to enable pilots to make decisions in time critical situations, based on previous theoretical analysis of the possible scenarios, taking into account risks and psychological conditions associated. The subject accident can be characterized as a flight through a flock of birds, in the very short final, with the aircraft fully configured for landing and stabilised on the descent path. During this phase of flight, the engines are selected at a relatively low thrust level, at which possibly ingested birds may very likely not affect the core of both engines (however, in the subject accident the stall happened at a relatively low rpm) therefore reducing the chances to cause significant damage to them and to permit the aircraft to land safely with the selected approach thrust. In such conditions, the decision to carry out a go around manoeuvre, where maximum thrust is applied to engines, could increase the likelihood to cause engine damages and malfunctions with consequent loss of available thrust, which could prevent the go around manoeuvre from being safely completed. The lack of clear indications to aircrew regarding the opportunity to carry out the go around manoeuvre in this kind of scenario, can lead the crew to apply, uncritically (and without the necessary awareness about the potential consequences) such manoeuvre which might present greater risks than landing the aircraft. The above considerations are consistent with the ones expressed and recommended, for instance, by Boeing, Airbus ,UK CAA, all related to the above scenario. Following the event, Boeing introduced in the FCTM of the B737 a recommended technique referred to the case of bird strike, providing both prevention strategies and guidelines concerning the actions to be implemented by the flight crew in case of bird impact in take-off or landing phases. Addressee: EASA and FAA. Text: ANSV recommends to provide flight crews with guidelines and/or operational and training procedures, based on a careful assessment of the risks associated with the conduct of the aircraft in approach in case of birds encounter or single/multiple bird strikes. These guidelines should include the following topics: to discuss bird strike risks during take-off and approach briefings when operating at airports with known or expected bird activity; in case of single/multiple bird strikes in the final approach phase, if landing is assured, it is preferable to land maintaining the lowest engine power setting possible rather than carrying out a go-around procedure (in case of birds ingestion, especially a massive one, the engines damages can be greater at high engine rpm, the latter typical of a go-around manoeuvre); to consider that engines damages or malfunctions might prevent the go-around manoeuvre from being completed safely and consequent impossibility to land on the runway.

RECOMMENDATION ANSV-13/1525-08/2/A/18 Type of recommendation: SRUR/SRGC. Reasoning: the initial and recurrent training currently provided, does not seem to optimize, promote or develop skills to manage unexpected situations able to generate surprise and startle effects. Following the event, the accident operator developed and implemented specific training on the matter, however ANSV has not found, at regulatory level, indications aimed at adopting such type of training. At training level, therefore, particularly for critical phases of flight, should be identified unexpected but plausible events, capable to generate such surprise and startle effects. Crews should be trained to deal with such events, through visualization and conditioning exercise. Training to cope with the negative consequences of these effects would be considerably valuable in helping to minimize the likelihood to adopt inadequate decisions by flight crews as, for instance, in case of bird strike / bird encounter during approach and landing phase. Addressee: EASA and FAA. Text: ANSV recommends to provide indications in order to adopt specific trainings for flight crews, to cope with the surprise and startle effect, particularly in critical phases of flight such approach and landing.

The Italian ANSV released their Final Report on 20 Dec 2018, 10 years after the event stating that:

"The significant delay in publishing this report is due, on the one hand, to the retirement of some investigators, who, over time, have been appointed as IIC (investigator-in-charge), on the other hand, to the known shortage of personnel affecting the ANSV investigation area."

The Captain of the flight has stated since publication of this report that:

"No regret nor doubt on my decision that day. It’s so easy to re-write the story afterwards. Short runway, tail wind and heavy plane mean go- around if something wrong happens. Losing both engines was a very remote outcome..."

Footer block

This site has had visitors to date.