B738, en-route, south west of Beirut Lebanon, 2010

B738, en-route, south west of Beirut Lebanon, 2010


On 25 January 2010, a Boeing 737-800 being operated by Ethiopian Airlines on a scheduled passenger flight from Beirut to Addis Ababa in night IMC disappeared from ATC radar soon after departure from Runway 21 and was subsequently found to have impacted the sea in an unintentional out of control condition some five miles south west of the airport less than five minutes after getting airborne Impact resulted in the destruction of the aircraft and the death of all 90 occupants.

Event Details
Event Type
Flight Conditions
Flight Details
Type of Flight
Public Transport (Passenger)
Flight Origin
Intended Destination
Take-off Commenced
Flight Airborne
Flight Completed
Phase of Flight
5NM South West of Beirut, Lebanon
Distraction, Fatigue, Flight Crew Incapacitation, Inappropriate crew response - skills deficiency, Ineffective Monitoring, Manual Handling, Procedural non compliance, Spatial Disorientation, Violation
AP Status Awareness, Flight Management Error, Extreme Bank, Take off Trim Setting, Aerodynamic Stall
Damage or injury
Aircraft damage
Hull loss
Non-aircraft damage
Non-occupant Casualties
Occupant Fatalities
Most or all occupants
Off Airport Landing
Causal Factor Group(s)
Aircraft Operation
Safety Recommendation(s)
Aircraft Operation
Air Traffic Management
Investigation Type


On 25 January 2010, a Boeing 737-800 being operated by Ethiopian Airlines on a scheduled passenger flight from Beirut to Addis Ababa in night Instrument Meteorological Conditions (IMC) disappeared from ATC radar soon after departure from Runway 21 and was subsequently found to have impacted the sea in an unintentional out of control condition some five miles south west of the airport less than five minutes after getting airborne. Impact resulted in the destruction of the aircraft and the death of all 90 occupants.


An Investigation was carried out by the designated Lebanese CAA Investigator in Charge. Assistance was received from the French BEA in respect of readout and analysis of the DFDR and Cockpit Voice Recorder (CVR) data. A Preliminary Report was issued on 25 February 2010 followed by Progress Reports on 10 February and 25 August 2011.

It was established that the aircraft commander, although having been a professional pilot with the Operator for over 20 years, had only acquired an ATPL and his first multi crew command in 2008 on a Fokker 50, with conversion to the Boeing 737-800 taking place the following year and being completed the month prior to the accident flight. The First Officer had joined the Operator as a cadet a year prior to the accident and had become qualified on the Boeing 737-800 as his first type five months prior to the investigated event. The Investigation noted that the Ethiopian Airlines procedure for restriction of type-inexperienced pilot pairing - a Captain with less than 300 hours and a First Officer with less than 100 hours on type should not be scheduled together - was complied with on the accident flight. The aircraft commander had only 188 hours as Captain on type but was accompanied by a First Officer with 350 hours on type.

The Investigation recognised that some form of subtle incapacitation might conceivably have affected the aircraft commander’s performance. It was also noted that the flight crew had commenced duty at Beirut to operate the accident flight after a rest period which circumstantial evidence indicated had, for both pilots, included interrupted sleep which they had attributed casually to a meal eaten after their previous duty period. However, post mortem examination of the bodies had not been possible and no other evidence to confirm such possibilities had been found.

No evidence was found that the airworthiness of the aircraft or the weather conditions prevailing at the time had affected the ability of the crew to control the aircraft and it was noted that the cloud base in the area was around 2000 feet amsl with the control difficulties occurring in IMC or its ‘dark night’ equivalent. Recovered recorded aircraft and ATC data allowed the Investigation to reconstruct the aircraft trajectory for the whole duration of the flight and one of the resultant depictions included in the Official Report is reproduced below.

It was established from the recorded data that the aircraft commander had been PF for the departure and had flown the aircraft manually, although with somewhat coarse control inputs and with the headings set correctly on the MCP following ATC clearances not properly followed. Terrain Avoidance and Warning System (TAWS) Bank Angle alerts had begun 2.5 minutes after becoming airborne and shortly before a call for AP engagement from the PF at an altitude of around 4000 feet which was neither acknowledged nor actioned. The aircraft had remained in manual control mode for the remainder of the flight and both bank and pitch control had steadily deteriorated from then on with control inputs becoming increasingly confused and the aircraft being flown out of trim and with crossed controls. After a delayed ‘recovery’ from a first stick shaker activation, a second one occurred with the aircraft at an altitude of 9000 feet and as the aircraft remained stalled, a roll to the left continued which culminated in a spiral dive in which the recorded bank angle reached nearly 120° to the left and pitch attitude reached 48° nose down. Eventually, recovery became impossible and impact followed.

The Investigation considered that the request from PF for the AP to be engaged indicated “that he felt uncomfortable with manually controlling the aircraft and that he was looking for a solution” It was noted, however, that the reason why the AP did not (or could not have been engage(d) was that, as specified in the Boeing FCTM “the airplane should be in trim and the Flight Director commands should be satisfied before autopilot engagement” whereas at the time of the call these necessary conditions were not satisfied.

It was noted that the stabiliser trim setting for departure had been wrongly set to a position one unit more nose down than that prescribed for the aircraft as loaded. This had resulted in a greater need for significantly more (manual) control column back pressure to achieve the required initial climb profile.

A 3D flight profile reproduced from the Official Report

Recorded data suggested that the performance of the First Officer as PM was normal in respect of normal operations but CVR evidence suggested little response to the failure to follow headings, and the alerts, warnings and abnormal attitudes which became progressively more prevalent during the flight.

The Investigation concluded that the Probable Causes of the accident were:

  • The flight crew’s mismanagement of the aircraft’s speed, altitude, headings and attitude through inconsistent flight control inputs resulting in a loss of control.
  • The flight crew failure to abide by Crew Resource Management principles of mutual support and calling deviations hindered any timely intervention and correction.

The following nine Contributing Factors were also identified:

  • The manipulation of the flight controls by the flight crew in an ineffective manner resulted in the aircraft undesired behaviour and increased the level of stress of the pilots.
  • The aircraft being out of trim for most of the flight directly increased the workload on the pilot and made his control of the aircraft more demanding.
  • The prevailing weather conditions at night most probably resulted in spatial disorientation to the flight crew and led to loss of situational awareness.
  • The relative inexperience of the Flight Crew on type combined with their unfamiliarity with the airport contributed, most likely, to increase the Flight Crew workload and stress.
  • The consecutive flying (188 hours in 51 days) on a new type with the absolute minimum rest could have likely resulted in a chronic fatigue affecting the Captain’s performance.
  • The heavy meal discussed by the crew prior to take-off (had) affected their quality of sleep prior to that flight.
  • The aircraft 11 bank angle aural warnings, 2 stalls and final spiral dive contributed (to) the increase of the crew workload and stress level.
  • Symptoms similar to those of a subtle incapacitation have been identified and could have resulted from and/or explain most of the causes mentioned above. However, there is no factual evidence to confirm without any doubt such a cause.
  • The F/O reluctance to intervene did not help in confirming a case of Captain’s subtle incapacitation and/or to take over control of the aircraft as stipulated in the Operator’s SOPs.

The following 11 Safety Recommendations were made as a result of the Investigation:

  • The Operator should revise its CRM program in order to stress F/O assertiveness and leadership requirements especially in periods of abnormal performance.
  • The Operator should consider its classification of airports where non-technical constraints might affect flight operations and brief their flight crew accordingly.
  • The Operator should re-examine (their) crew pairing and scheduling policies in order to ensure a less stressful cockpit environment.
  • The Operator should consider establishing write up criteria for pilots’ training files in order to avoid the adverse effects of any mis-interpretation by the trainees.
  • The Operator should consider developing his safety oversight program in order to detect such potential flight crew performance.
  • The Ethiopian CAA should ensure that the recommendations (made) to the Operator have been implemented.
  • The Ethiopian CAA should re-examine the regulations concerning crew pairing policies.
  • International Civil Aviation Organisation (ICAO) should re-examine the international requirements for the identification, training and reporting of subtle incapacitation symptoms and (causes).
  • The Lebanese Government should establish requirements to ensure that responses to such accidents are made systematically without reliance on foreign ad hoc assistance.
  • The Lebanese DGCA should re-evaluate the working conditions of the ATC personnel.
  • The Lebanese Government should consider establishing administrative and logistic support for (accident) investigations (such as this).

The Final Report – ET 409 of the Investigation was published on 17 January 2012.

Further Reading

SKYbrary Partners:

Safety knowledge contributed by: