The approach illustrates several classic problems. Whilst initially stable, there is a drop associated with windshear and then the flare starts. During the flare, the pilots decide to level the wings and yaw out the drift. This creates the classic full wing lift on the upwind (starboard side) and blanking on the downwind side. Right wing produces full lift, left less lift. Aircraft rotates away from the crosswind. The left winglet actually struck the ground in this case.

Lufthansa have had an accident doing this before, which produces a potential psychological precurser to land wings level. It would have been better to dump it down on right main gear first, which pops the most of the spoilers on the left hand wing and dumps the lift so the left wing drops down too. A reasonable amount of airspeed for rudder control was available.

The pilots probably did not have enough crosswind information. This is caused by three separate factors. Firstly, the presentation of the wind as th current vector which the pilots have to resolve into head/crosswind components. Secondly, the time at which they are told of the wind by ATC, which is usually at landing clearance time, and that can be a couple of minutes prior to landing. Thirdly, the ICAO rules in Annex 3 and 11 state that Air Traffic Controllers should give the two minute average wind direction and speed. The two minute average direction can shift very slowly when compared with the actual instant wind direction (in the event of a microburst).

Spottie has a campaign to sort this out!

In this case, the vertical impact rate of descent was very low, so not much chance of significant damage. You would need to drop on at the type of vertical descent rates shown earlier in the video without flare to really bend it.

Happy landings.

Spottie