Lanzarote Overrun: I have nothing planned

2 Mar 12 5 Comments

It was an uneventful night. The crew went on duty on Thursday night at 20:40 for scheduled passenger flights on a Boeing 737-800 from Lanzarote to Glasgow and back.

Friday morning, on the return leg, their work is almost done. The First Officer is flying and the Captain has taken the opportunity to have a short rest. There are 74 passengers and six crew on board for this 4-hour and 19 minute flight to Lanzarote.

At 06:50 (UTC and local time) they are cruising at FL 390. The Lanzarote ATIS (recording giving up-to-date airfield information) looks good:

06:50 Runway in use 03. Transition level 70. Wind touch down zone variable 3 knots, variable between 200 and 270 degrees. Visibility 10km or more, clouds few at 1,800 feet, broken 9,000 feet. Temperature 18 degrees dew point 14, QNH 1014, QFE 1013. Report Lanzarote ATIS information A.

There is no mention that the runway is wet, which should be specified in the ATIS and by air traffic control, as wet runways can require a longer landing distance.

The First Officer has already programmed the Flight Management Computer for the 03 runway. He gives a detailed briefing of the approach. The briefing includes a review of the airport chart and the landing distances required. The captain mentions that the approach also includes Precision Approach Path Indicator (PAPI) visual aids to help incoming aircraft acquire the correct approach path.

They continue to descend. The Captain contacts the handling agent in Lanzarote with an estimated arrival time of 07:22.

“Just out of curiosity,” says the Captain to the FO, “ask him how the wind is for runway 21…I don’t think they’ll let us, but anyway…” He said later that runway 21 saved time when arriving from the north although it was a more difficult approach.

The FO makes the request and then comments, “The thing is, if we head for 21, I have nothing planned, so if they end up giving it to me, I’m going to laugh.”

Canaries Approach responds: “Europa 196 you may proceed to the 11 mile fix, runway 21 to land on 21, continue descent to 5,000 feet, QNH 1014.”

The First Officer now has to work out the new approach. He talks to himself as he inputs the new information into the Flight Management Computer.

From the final report (PDF):

The FO asks the captain in a troubled voice, “We’re going to here, to mile 11, right?” The captain replies, “Mile 11, yes.” The FO again asks “There?” and the captain replies, “I don’t know but descend.”

The FO states, “No, no, something’s not right here… we’re now heading for mile 11, ah … no…” The captain replies, “Set 4,000 for now,” adding shortly afterward, “Here, to this point.”

The FO once more asks, “What point are we going to? I’m not seeing this point at mile 10, mile 11, it has to be F.” The captain replies, “CD21” and the FO, in a relieved voice, says “CD21, OK, that makes sense! Of course, I thought something was off!”

The carefully set-up approach is in a mess. The Captain repeatedly requests a faster descent. The Flight Officer knows he’s not prepared for the runway change and clearly struggles to set up the Flight Management Computer fast enough. At 10,000 feet they are 21 nautical miles from the runway and going 315 knots. They are too high and too fast. The Captain requests “a bit of speedbrake” to alleviate the issue.

They continue the descent. The Captain reads the descent check list. The First Officer continues to express surprise at the indications of the Flight Management Computer.

Before the runway change, the descent had been planned at 300 feet per nautical mile. At the point of the runway change, they are suddenly 23 nautical miles closer to their runway and require an immediate increase in descent rate. The table below is from the accident report and shows the descent rate required to reach runway 21 as the flight progresses.

At 1,700 feet they are 3 nautical miles from the runway and travelling at 180 knots. The First Officer confirms landing in sight. The approaching minimums automatic call-out is sounded.

The First Officer asks for flaps 25 and then a few seconds later requests flaps 30. The Captain tells him to wait a couple of knots: they are going too fast to lower the flaps any further.

At 900 feet, they are descending at 2,000 feet per minute. The cockpit fills with a caution from the Enhanced Ground Proximity Warning System: SINK RATE, SINK RATE.

“It’s all right,” says the captain and lowers the flaps to 30°.

500 feet. They are going 177 knots and descending at 1,900 feet per minute. They have not achieved the glide slope. The warning system continues to sound: SINK RATE, SINK RATE, PULL UP, PULL UP.

Boeing’s Flight Crew Operations Manual states that the PULL UP warning requires an immediate go around, specifically because it is associated with an unstabilised approach.

The First Officer says “We’ll keep it there, OK?”

500 feet is the decision point, at which point the pilot monitoring should make a statement to continue the approach if stabilised and break off if it is not. The pilot monitoring, in this case the Captain, must demand a go-around if the approach is unstable. Reasons to interrupt the approach include confusion among the crew, checklists completed late, crew overloaded, and the approach altitude, speed, glide slope or configuration becomes unstable.

Top tip: if warning calls from the Enhanced Ground Proximity Warning System are filling the cockpit, the approach is not stabilised.

The captain did not make the 500-foot call out.

300 feet, 176 knots, descending 1,700 feet per minute and more warnings: TOO LOW TERRAIN, SINK RATE. TOO LOW TERRAIN, SINK RATE.

The Captain realises that the runway is wet and increases the thrust setting on the autobrake system to maximum.

According to Boeing 737 FCTM “Flight Crew Training Manual”, the maximum braking landing distances included in the QRH must not exceed the 60% of the real available distance, in order to provide adequate safety margins. Considering this, the minimum required runway length would be 1,712 m. Lanzarote runway is 2,400 m long, thus it exceeds by more than 700 m the minimum runway length required for this type of aircraft.

After a prolonged flare, the airplane touches down halfway down the runway at a speed of 157 knots. That’s 23 knots over the VREF – the landing reference speed at 50 feet above the threshold which is 1,300 metres behind them as they touch down.

Groundspeed 137 knots: Someone accidentally presses a brake pedal and the autobrake disengages five seconds after touchdown. The Captain calls out “Autobrake disarm!” immediately and both crew members engage manual brakes with full force.

Groundspeed 96 knots: The Captain calls for the reversers to be applied, 13 seconds after touchdown. They need time to spool up and thus have no immediate effect. The tower controller activates the alarm.

Groundspeed 51 knots: The next sound on the cockpit recorder is the sound of the aircraft colliding with the row of lights at the end of the runway.

Groundspeed 40 knots: Sounds of collisions with beacons.

The aircraft crosses the 60 metre stopway and slows to a halt just as the reversers finally kick in. It stops a metre away from the jet blast barrier for Runway 03.

The Captain transmits, “Lanzarote Europa 196…”
The Tower cuts him off, “Sending you firefighters Air Europa.”

There are no injuries and no damage to the aircraft other than the landing tyres. The Flight Officer is heard lamenting the event and blaming the reversers.


The cause of the incident was a high-energy unstabilized approach followed by a landing with excessive speed, 1,300 m past the threshold, with a wet runway. In addition, neither the autobrake nor the reverser was used efficiently. A contributing factor to the incident was a combination of deficiencies involving several aspects of CRM (Crew Resource Management).

There’s more detail in the final report, including a long list of what went wrong.

In the end, it strikes me that the greatest problem here was the reluctance to break off the approach once things began to go wrong.

“The captain undoubtedly noticed that the FO could not adapt properly to the circumstances and that the situation was worsening, and yet he decided to maintain his role and not take control of the airplane himself. The captain made every decision. At no time did he ask the FO his opinion. He decided to request the runway change, he decided to have the FO continue flying and he decided to continue the approach in spite of the presence of a multitude of indications that suggested or required that the maneuver be stopped. He eventually decided to land, increasing the autobrake setting to MAX instead of ordering a go around when he saw they were landing too fast in the middle of a wet runway.”

Even if we accept that asking for Runway 21 at that late stage was reasonable, the approach should have been broken off based on the descent profile alone. It should have been broken off when the FO was clearly overloaded and it should have been broken off when the warnings sounded and it should at the latest have been broken off at the 500 metre call-out point.

Fatigue and operating in the low-performance phase of the circadian rhythm were brought up as possible contributing factors.

This is certainly a text-book example of bad decision making.

Source: Comision de Investigacion de Accidentes e Incidents de Aviacion Civil (CIAIAC) final report IN-041/2008 (in English)

Analysis: Incident: Air Europa B738 at Lanzarote on Oct 31st 2008, departed runway on landing

Photographs by Manuel Estévez, Erik Cabrera Jensen and Keith Hatfield.

Category: Accident Reports,


  • Another striking thing is that with all the EGPWS warning callouts and Spd/ RoD exceedences for fairly long duration, the crew never once thought of discontinuing the approach—- weren’t they concerned about post flt data monitoring and questioning by the management? It only proves very poor post flt data monitoring or lax followup. Management too needs to introspect.

  • Well, if the only thing stopping them was the chance of a post-flight bollocking, that’s pretty worrisome in itself! ;)

    But I don’t think it was that well thought through. Fixation and fatigue are a bad mix.

  • Post-flight bollockung ?? what about puting all those passengers in a real danger ?
    I beleive -as said- it is always wise to ask for a go around ..

  • “If all ends well no questions asked” appears to be the prevalent practice here. A 100% Post flt monitoring and questioning works as a gud prevention with no questions asked for any go rounds/ diversions. Bollocking when required is a good thing.

Post a comment:

Your email address will not be published. Required fields are marked *


This site uses Akismet to reduce spam. Learn how your comment data is processed.