Friends are asking me about a video of a KLM Boeing 777 landing at its home base in Amsterdam. “We Asked A Pilot About The ‘Terrifying’ Landing Going Viral” is a Huffington Post article where they ask an airline pilot for his opinion:
“It looks fairly dramatic, but it was not dangerous,” he reassured us. Videos of landings tend to look a lot more dramatic than they actually are. Those judders and swerves? Just normal reactions to gusty crosswinds.
“For the passengers on board, I’m sure there was some jostling, but the plane didn’t come anywhere close to crashing,” he says.
For example, you may have noticed that the plane tilts to its right just before it touches down. It looks like it comes close to flipping over into a fireball of death. Actually, Smith says, that’s a deliberate correction to the crosswind. Pilots are trained to land at an angle, so the wheels closest to the wind land first. That helps stop the wind from pushing the plane sideways across the runway.
“That landing was well within the capabilities of the pilots and also the aircraft,” Smith stresses.
Pilots are trained to abort landings if they feel unsafe, and they don’t lose face among the crew for doing so, he says.
What do readers think? Were they close to scraping a wingtip or engine nacelle? That would not be a sign of prudent planning in the airline world (since you can always choose to land somewhere else and send the passengers onward via bus or train and/or wait a day).
The idea that they could have aborted the landing at the point shown in the video seems questionable. The engines on a 777 take a long time to spool up and then pilots seem not to handle these well (see my explanation in this posting about landing the CRJ, a much lower-inertia aircraft; also see this NTSB report of damage to a Boeing 747 after a low go-around at the very same airport)
What do pilot readers think? Was landing in the worst July storm in a century just another day in the front office or a case of get-there-itis? Did we watch awesome skill or questionable judgment and a bit of luck?
Related:
- engine nacelle strike on B747 at Heathrow in gusty crosswind
- Lufthansa A320 striking wingtip in crosswind (and then managing to go-around! my posting on the incident)
- YouTube compilation video of a bunch of windy crosswind landings in Dusseldorf (they all seem more stabilized than the KLM/Amsterdam)
- somewhat crazier video on an icy runway
I don’t know about the technical stuff but around 0:03-0:05 with the particular camera angle it really looks like a mean humongous bird of prey with outstretched wings and belly and talons trailing behind!
I have a theory that the tendency for pilots to pursue risky landings is a great manifestation of the two different modes of thought described by the Nobel (economics 2002) winner Daniel Kahneman in his book Thinking Fast and Slow. (Phil I just realized it may have been you through this blog which recommended the book to me).
Perhaps the Dutch pilot flying this turbulent approach was so focused on maintaining a stabilized glide path that he did not have enough reserve mental energy to do the critical thinking necessary to decide to go around?
For example, repetition has conditioned and habitualized a pilot’s System 1 (subconscious reaction) to perform at maximum potential during approach and landing more so than in almost any other human endeavor. The Dutch Pilot Flying was using all of his mental energy to perform subconscious routines like:
“Pitch for Glide Slope, pitch for the VASI”
“Check the Airspeed, adjust power”
“Crab into the wind, stay on the center line”
“Here comes ground-effect, get ready”
“Kick the rudder and opposite aileron, Don’t leave a big skid mark!”
Combined these seem to demand that all of the brain’s energy be diverted to System 1.
Perhaps the reason the Dutch Pilot Flying didn’t have enough mental reserve to hit the gas is because in the moment his System 1 (Subconscious) performance left no energy to either decide to switch to System 2 (Kahneman says switching requires energy) or to actually compute the virtue of the landing attempt:
“Has the Airspeed so far remained steady enough to imply a safe descent?”
“Has my Vertical Speed been too erratic?”
“Am I at a height above the ground where, if I push the throttles, I can ride out ground effect into a climb? I don’t want the wheels to touch and then either side-load the gear or get blown off the runway with power…”
Who knows what the Pilot Monitoring was thinking. Possibly terror.
Do you think this theory has any merit? If it is true I imagine the best practical solution may be to have pilots practice many more go-around and generally no-go decisions.
It was a great landing. Full deflection at the end there.
I’d call the decision to put it down more iffy than I’d like. However, I use to fly into Omaha a lot, Chicago > Omaha. There were many times when we landed in the middle of a borderline white out blizzard, at night. The pilots would say “buckle up ladies and gentlemen, it’s a little windy tonight.” We’d bounce around and seesaw roll much like the video. But more often than not when we touched down the plane was square with the runway and wheels kissed once, no bounce. Pilots on this route had a lot of practice under these conditions. And the snow probably looked worse than it was using underwing landing lights, like using high-beams. My guess, since this was a KLM plane this was not the crew’s first landing under these conditions.
Looks like the landing was well under control. They could not really move to other airports because the storm hit a large part of Europe that day.
I used to fly Amsterdam-Paris and back every weekend in 1999. Schiphol was almost invariably gusty and turbulence-ridden, making for white-knuckled landings for the nervous flyer I am.
I’m not a pilot but it doesn’t exactly strike me as the height of probity to continue a landing like that. There is clearly pretty severe turbulence near the ground – what if a microburst or other shearing wind happened?
I watched it a few times. It didn’t look that bad to me. I know when I am landing in stuff that is a little gusty I am judging two moments. The first is on the “stabilized” approach on final. Am I really in control of the airspeed, direction and attitude of the plane? If I am not, I go around. If I am still not, I go elsewhere. The second is the moment I enter ground effect, because (at least for my little DA40) everything changes. I have a reserve of lift, so I can slow way down, but that also means I have larger control surface deflections. Usually if I make it to ground effect I’m on the ground. A few times I’ve been too fast and the runway isn’t long enough, so I go around.
I’d say it was not technically a wrong decision to continue, but nobody would have faulted a go around. Definitely not the ride you want to sell at business class fares.
That video by itself is not enough to judge the full impact of such landings. I would like to see videos of the following:
1) A wide angle view of the passengers inside the airplane.
2) A wide angle view of the pilot while making this landing.
3) A wide angle view off pilot’s cockpit looking out the window into the runway.
4) A wide angle view off few passengers looking inside the airplane.
5) [bonus] A wide angle view off the belly of the airplane.
Question: How much in control is the pilot in such landing? Full or partial? If partial, by how much and in what way? I mean is there any kind of automatic adjustments a computer is doing while the pilot is maneuvering the airplane?
George: It is all hand-flying at that point. (My first time flying a turbojet I disconnected the autopilot in order to hand-fly an instrument approach from about 5000′ above the ground. The guy in the right seat said “You’re hand-flying now. Do you want to declare an emergency?”) Generally speaking the autopilot is disconnected no higher than 150 or 200′ above the ground and typically shortly after breaking out of the clouds (e.g., 500′ or 1000′ above the ground). Autopilots on lighter airplanes don’t work that well in heavy turbulence. They will often disengage if you’re flying through a cumulus cloud, for example. Maybe the B777 autopilot is truly impervious, as noted above, the last phase of the flight will be hand-flown.
The point at which the plane is technically committed to the landing is at thrust reverser deployment. Spool up time for a go-around has not been a factor in jets for a very, very long time. (This was the initial idea behind the so-called stabilized approach – balancing some drag against thrust so that the engines would be partially spooled during the approach.) All transport jet engines now have an approach idle setting – or equivalent – that keeps the engines partially spooled. (Usually triggered by the initial flap or slat extension) While it may take eight to ten seconds for a flight idle engine to spool up an engine at approach idle spools up immediately. The approach depicted didn’t look that dramatic to me btw.
In my opinion, that looks more or less like how someone who was used to flying taildraggers regularly would land in those conditions. Keep the plane lined up at all costs, stay active on all of the control surfaces, and keep flying it until you’re at the gate. I wish there was a second camera angle to get a better view of the rudder and/or ailerons, but certainly from about 0:18 to 0:23 it’s clear that he’s making continuous, timely inputs on the elevator.
Is this sort of landing appropriate for the airline world, where stabilized approaches are the law of the land? I don’t know. But at the very least, it’s clear that the pilot has stick and rudder skills in spades.