Unperceived Predicament

On the night of Oct. 16, 2012, a Bombardier CRJ700 ran off the end of a contaminated runway after touching down long with a tailwind at an airport on the coast of Brittany. None of the 57 people aboard the regional jet was hurt, but aircraft damage was assessed as severe by the Bureau d’Enquêtes et d’Analyses (BEA) of France.

“The investigation showed that the accident was due to the crew’s failure to decide to carry out a missed approach [and their lack of awareness] of the degree to which runway conditions were contaminated or of the remaining length of runway available,” said the BEA’s final report.

Moreover, the report concluded that the pilots’ situational awareness and decision making had been numbed by fatigue and routine, as well as the absence of clear communication about the condition of the runway. Investigators also found that an opportunity had been missed for the flight crew to review recent lessons learned and best practices for landing in adverse weather conditions.

The report said that the operator, Brit Air, “did not have a true picture of the safety performance of its operations.” For example, the regional airline was not aware of data showing that one-third of the landings made by its CRJ700s on Runway 25 at the Lorient Lann Bihoué Naval Air Base, where the accident occurred, were “overshoots,” or landings beyond the runway’s touchdown zone.

The report also said that the operator’s fatigue risk management system did not account for the effects of flying trip sequences that incorporated multiple legs.

Fifth and Last Leg

The flight crew had completed four flight segments before departing from Paris at 2030 local time for the last leg to Lorient. Both pilots told investigators that they felt tired before beginning the flight. The captain noted that this was typical for the fifth leg of a trip sequence, especially at night.

“In general, five-leg flights are tiring,” the report said. “This is felt by a majority of pilots, but few pilots, including [the captain], inform the airline of this.”

The captain, 42, was the pilot flying. He had 6,910 flight hours, including 4,025 hours in type. He was hired by Brit Air as a cabin attendant in 1992 and was promoted as a flight crewmember after earning his airline transport pilot (ATP) certificate in 1999. He earned a CRJ700 type rating in 2001 and had been flying as a captain since 2007. During the three months preceding the accident, he had conducted 10 landings at the Lorient Lann Bihoué airport — seven at night and three during the day.

The copilot, 45, had 5,244 flight hours, including 3,014 hours in type, and had joined the airline in 2004. He held an ATP certificate and type ratings in the CRJ700 and CRJ1000. He had conducted eight night landings and two day landings at the airport during the previous three months.

Lorient Lann Bihoué Naval Air Base is a joint-use facility on the southern coast of Brittany that is operated by the French navy.

Around the time of the accident, showers and frequent thunderstorms had been forecast throughout Brittany. Before beginning the descent to the airport, the crew reviewed the current automatic terminal information system (ATIS) broadcast, which indicated that surface winds were from 170 degrees at 18 kt, visibility was 10 km (6 mi) and there were broken ceilings at 1,000 ft and 1,500 ft. The ATIS also said that the runway was “wet with water puddles” and that the precision approach radar (PAR) procedure to Runway 07 was being used.

A few minutes later, however, the approach controller told the crew that the surface winds were from 160 degrees at 17 kt, gusting to 26 kt, and that visibility was 3,000 m (about 2 mi). The flight crew requested and received vectors for the instrument landing system (ILS) approach to Runway 25, which had a longer available landing distance than Runway 07.

‘Fatigue and Weariness’

The cockpit voice recording showed that both pilots expressed their “fatigue and weariness” several times during the flight, the report said. The effects of fatigue were manifested, in part, in the crew’s conduct of the “Descent” checklist. The copilot called out completion of items on the checklist without waiting for confirmation from the captain; he also forgot certain callouts.

At about 2106, the controller advised the crew that there was a “big squall on the field at the moment” and that visibility had decreased to 2,000 m (1 1/4 mi) in heavy rain. The controller reiterated that the runway was “wet with some puddles” and advised that the crew of the preceding aircraft had encountered “difficulties during landing due to aquaplaning.” (The crew of that aircraft, an Embraer 145, later reported that they had temporarily lost control after touching down on the slippery runway.)

“This information did not trigger any particular reaction by the crew or an additional briefing taking into account the potential threats associated with it,” the report said.

The pilots were conducting the “Approach” checklist when the controller issued a new surface wind report —150 degrees at 17 kt, gusting to 25 kt — and cleared the crew to conduct the ILS approach to Runway 25.

The crew told investigators that, due to the risk of wind shear, they decided to conduct the approach with the flaps extended 30 degrees and at 140 kt, a landing reference speed (V_REF) that was appropriate for the airplane’s gross weight and the selected flap configuration.

No Change of Plan

At about 2124, “the controller again indicated the presence of heavy rain, the condition of the runway, the aquaplaning and the difficulties of the preceding aeroplane,” the report said. “This information did not alert the crew and did not change their plan of action [i.e., to land with a flaps 30 configuration].”

The crew’s selection of flaps 30 complied with the operator’s existing instructions for an approach and landing with known or suspected wind shear. According to the report, however, the crew was not aware of another critical factor: that the runway was contaminated with standing water.

The report said that a runway is considered contaminated, in part, “if more than 25 percent of its surface area is covered by a film of water more than 3 mm [0.125 in] deep.” Although no measurements had been taken by airport personnel at the time, the BEA determined that the runway was indeed contaminated by standing water.

However, the repeated reports that the runway was “wet with water puddles” — phraseology that was not proper for pilot-controller communications — had led the crew to believe that the runway was wet, rather than contaminated with standing water, which reduces braking performance and typically results in a longer landing distance.

The report indicated that a flaps 45 configuration would have been more appropriate for the conditions because it would have provided for a shorter landing distance.

Runway 25 was 2,230 m (7,316 ft) long and surfaced with ungrooved concrete. The runway was known to have water-retention problems. Following two landing excursions by military aircraft, the navy in 2010 had approved a recommendation for runway reconditioning, including grooving the concrete to improve water drainage. However, “as of the date of the accident, no reconditioning had yet been undertaken,” the report said.

Investigators calculated that the CRJ’s landing distance on the contaminated runway was 2,117 m (6,946 ft). It is important to note that this calculation assumed that the aircraft would pass 50 ft over the approach threshold at the reference landing speed.

‘Stable, Continue Approach’

As the aircraft descended through a radar height of 1,000 ft, the captain called out “stable, continue approach.” Recorded flight data confirmed that the approach was stabilized at that point but also that the latter portion of the approach was conducted with increasing airspeed and with a tailwind component of about 4 kt.

The crew saw the runway approach lights at about 800 ft. The captain disengaged the autopilot and told the copilot to set the windshield wipers at maximum speed. Both pilots recalled that it was raining heavily.

“For about 10 seconds, the airspeed increased above 150 kt, with a maximum of 155 kt,” the report said. “The aeroplane went through 500 ft radio altimeter height at a speed of 154 kt.”

Numerous roll-control inputs were recorded before the CRJ crossed the runway threshold at 56 ft. Indicated airspeed was 153 kt (13 kt higher than V_REF), and, due to the tailwind, groundspeed was about 157 kt. The faster speed and the tailwind increased the calculated landing distance on the contaminated runway from 2,117 m to about 2,500 m (8,202 ft), according to the report.

Just before touchdown, the copilot remarked that visibility was “bad” and told the captain that the aircraft was left of the runway centerline. The captain later told investigators that he had difficulty estimating height and aligning the CRJ with the centerline due to the aircraft’s inefficient landing lights and the deficient runway marking and lighting. The runway did not have centerline lights.

Long Landing

The captain told investigators that he perceived that the flare was initiated too high and that the aircraft floated, but he did not know how far from the approach threshold the aircraft actually touched down.

The report said that groundspeed was 140 kt when the CRJ touched down about 1,130 m (3,707 ft) from the approach end of the runway at 2122. “The crew did not realise that the runway was contaminated and that the landing was long. At no time did they envisage a go-around.”

The touchdown occurred with 1,100 m (3,609 ft) of runway remaining. “This length was sufficient for a complete stop of the aeroplane on a dry or wet runway,” the report said. “It was inadequate on a contaminated runway.”

The spoilers extended after touchdown, and the crew deployed the thrust reversers and applied maximum reverse thrust and wheel braking. A few seconds after touchdown, the captain told the copilot that he could not brake the aircraft. The report said that white tire tracks found on the runway after the accident indicated that aquaplaning likely had occurred.

The aircraft turned right just before it overran the runway at 66 kt. The CRJ came to a stop in a grassy area about 200 m (656 ft) from the end of the runway after the left wing struck localizer antennas.

The captain ordered an emergency evacuation, and the 53 passengers exited through the left front door and the overwing exits. The report said that the CRJ incurred major structural damage; the landing gear required replacement and the engines had to be removed for repair.

Unrecognized Threats

The report said that the pilots had accounted for the possibility of wind shear and a wet runway in their planning, but they had not identified the threats of a contaminated runway and an overshoot, even after receiving information about the previous crew’s difficulties in landing.

The airline recently had incorporated principles of threat and error management (TEM) in recurrent ground school sessions. However, “this was not put into effect during simulator sessions,” the report said. “In addition, by the date of the accident, only the captain had been given awareness training. The crew therefore was not predisposed to apply it.”

Moreover, the report noted that the airline’s training had not incorporated the “lessons learned and best practices” for landing in adverse weather conditions that had been covered in a symposium hosted in 2010 by the Direction Générale de l’Aviation Civile (DGAC), the civil aviation authority in France. “This symposium specifically addressed the risk of runway excursions,” the report said.

Investigators also found discrepancies in the information provided by the airline and by the aircraft manufacturer regarding flap selection. Bombardier had recommended using the minimum flap setting appropriate for the available runway length when landing with suspected or confirmed wind shear. Based on this information, Brit Air’s operations manual specified a flaps 30 configuration in this situation. However, the operations manual was not revised after the manufacturer in 2010 removed the instruction to use the minimum flap setting in wind shear.

“Bombardier stated to the BEA that this instruction had been removed because [the CRJ700 was] certified only for the standard flaps 45 configuration,” the report said, noting that among several changes made by Brit Air after the accident was an instruction to use a flaps 45 configuration for all landings.

The BEA issued several recommendations based on the findings of the accident investigation. The recommendations included improvement of drainage and elimination of known areas of water retention on Lorient’s Runway 07/25; extension of civil airport certification and safety management requirements to all military airports where civil aircraft operations also take place; integration of TEM in recurrent training and checks by commercial aircraft operators; and implementation of effective and thorough fatigue risk management systems. The BEA also called on the DGAC to ensure improvement by Brit Air of its processes for checking and updating its documentation. 

This article is based on the English translation of the BEA report “Accident on 16 October 2012 at Lorient Lann Bihoué (56) Aerodrome to the Bombardier CRJ-700 Registered FGRZE, Operated by Brit Air.”