Sighting in Sundhoffen, May 19, 2020

During the day, the witness's family had heard gunfire several times near their home. Around 9 p.m., the witness went out into the fields with binoculars to look at the surroundings. They then saw a cloud literally "explode" in the sky, in a shape that recalled an atomic mushroom cloud, only smaller and translucent, visible thanks to the cloud it had struck. The witness fetched their brother, and together they again heard gunfire at regular intervals. The clouds pulled apart and split into three horizontal layers, fairly quickly and in a very geometric way, a little like lenticular clouds. The brother saw a projectile rise into the sky. The witness went on to describe clouds with unusual shapes and colours, and a storm forming. Both witnesses finally said they saw an unknown craft, first hovering, then moving among the clouds. The witness estimated the observation lasted 30 minutes.

The investigation established that gunfire was heard all day, including between 9 and 10 p.m., and that a night-time military exercise involving live firing was scheduled for that evening. The witnesses found the detonations different from the usual shots, but sounds are perceived differently in the evening or at night than by day, because the background noise is not the same. Shots to protect crops, notably against pigeons, were also fired that evening. The clouds were lenticular clouds of the "stacked" type, which spread out and vanish as night falls, with the sunset heightening the contrasts the witnesses described. Through binoculars that were not perfectly focused, the witnesses saw a black mass making various jerky movements and likened to a large fan; this strongly matches a paramotor microlight flying in the evening, since its parachute allows aerial acrobatics.

GEIPAN classed the case B: a main misidentification of an aircraft, which was in fact a multiple misidentification (a day and night firing exercise, shots to scare unwanted animals from crops, lenticular clouds forming, and paramotors). The hypothesis is considered highly probable.