French investigators are seeking improvements to convective weather detection capabilities after a serious turbulence incident involving an Air France Airbus A330-200 over eastern Africa.
The crew did not detect the convective zone – a task exacerbated by the rapid development of cells in the region – as the aircraft cruised at 36,000ft over Tanzania.
Dar es Salaam air traffic control twice asked whether the crew could climb to 38,000ft but the pilots declined, in order to maintain a sufficient margin from the operating ceiling.
The crew had expected convective zones while passing latitude 12.5° south. Some 10min after crossing this boundary, however, the crew set the navigation display scales to 160nm and adjusted the weather radar tilt to monitor possible isolated cells.
Flight-recorder data suggests the disturbance began a few minutes later, in the vicinity of Dar es Salaam, the airspeed increased from M0.81 and the pilots saw flashes and cloud on the right side of the A330 – although there was no return on the weather radar. According to the captain, who was flying, the radar tilt had been set to 1.5° down.
As the airspeed reached M0.83 the crew selected lower speeds – a normal precautionary response to convective weather – and briefly extended the speedbrakes, but the aircraft slowed to M0.79 before accelerating to M0.82.
The aircraft (F-GZCG) then encountered “violent” turbulence, says French investigation authority BEA, and its autopilot disengaged.
BEA says the A330 started to climb despite the captain’s pitch-down commands. Although the crew reactivated the autopilot, it disengaged again, along with the autothrust.
The A330’s altitude increased to 38,000ft before the crew stabilised the flightpath, and the aircraft started to descend 10s later. It returned to its assigned altitude and the flight continued without further incident.
BEA analysis found that the vertical speed peaked at 8,500ft/min during the 40s event, with the pitch varying from 6° nose-down to 11° nose-up. The captain’s control inputs were mainly nose-down, notably for 10s after the autopilot disconnected.
Airbus determined that the aircraft remained in its flight envelope throughout the incident which, the BEA says, comprised “highly dynamic” turbulence in all three axes. The A330 was subjected to a sudden sharp downdraft of 110kt and then updrafts of 70kt and 40kt.
While the crew’s selection of the 160nm range, rather than 80nm, on the navigation display was not “optimum” for detecting the storm cell, the BEA says the pilots’ response enabled them to maintain control of the aircraft despite the sudden onset of difficult flight conditions.
The captain told the inquiry into the 27 February 2012 event that he was startled by the ferocity of the turbulence – which made the instruments unreadable and verbal communication impossible – and that it caused sensory illusions over whether the aircraft was climbing or descending.
Given that the cell was not visible on infra-red satellite imaging just 12min before the event, the BEA acknowledges that a reasonable weather radar setting might not have made a difference.
Air France reviewed its radar procedures in the wake of the loss of flight AF447, another A330-200, in the vicinity of convective weather over the south Atlantic in June 2009. Investigators also noted the use of sub-optimal radar settings before an in-flight upset involving an Air France A340 in July 2011.