10 December 2018, before dawn, approximately 07:20–07:45 local across the published accounts; about 90 seconds
Outcome: Open (Unresolved). GEIPAN retained D1 after a field interview and reconstruction. A drone remained the strongest ordinary suggestion but was not identified; agricultural lighting, projected light and drifting objects did not satisfactorily fit the reported sequence.
A light over a field before work
A driver leaving a rural property near La Forêt-Fouesnant on 10 December 2018 described a yellow luminous bar with a conspicuous fluorescent-green centre. The account included slow motion, an apparent turn and pause, then a movement away into mist. The view was interrupted as the observer drove and stopped along the access lane.
The published summary and original narrative give somewhat different clock-time ranges. This article preserves a broad before-dawn interval instead of inventing a second-by-second timestamp. The roughly 90-second duration is the witness’s estimate, not a measured recording.
How the report developed
GEIPAN initially investigated remotely and later conducted a field visit with a cognitive interview and reconstruction on 5 October 2019. The technical report is dated 28 May 2020. Setting photographs, maps and drawings supplied detail about the view, but no photograph or video of the appearance itself was available.
The distinction is important because the report contains illustrated reconstructions. A later photograph marked with a suggested position can help explain a testimony; it cannot independently show that the original light was there, its exact shape or its distance.
Testing a drone and projected-light explanations
The enquiry checked agricultural activity and found no relevant machine operating in the fields concerned. It compared the reported sharp outline, slow movement and changes of direction with reflected or projected light, drones and a wind-borne lantern. A drone was its strongest candidate but not sufficiently supported to adopt.
The report mentions restrictions on night flying as context. A restriction alone cannot rule out an unauthorised flight, and the investigators also considered that possibility. Their assessment depended on the combined reported appearance, scale and movement, rather than legality being physical proof of absence.
Why moving-observer reconstruction matters
The driver changed position along the lane and looked away while continuing the journey. The report used the setting and the observer’s account to reconstruct three phases of apparent movement. It did not obtain a continuous independent record of a source turning towards a person.
That leaves two questions distinct: what appeared to happen from the moving observation point, and what a physical object, if present, actually did. The investigation found the reported sequence difficult for the hypotheses considered, but its reconstructed geometry remains dependent on the accuracy of those observations.
The radio claim and the unresolved outcome
The witness also reported a radio problem after the observation. The reviewed material does not establish a controlled equipment test linking that problem to the light. Temporal association should not be promoted into proof of electromagnetic interference.
GEIPAN retained D1 and UKPX records Open (Unresolved). This reflects a documented enquiry that failed to identify a satisfactory cause. It does not establish an encounter with a vehicle or an observer-directed manoeuvre. Independent imagery, a specific matching drone operation or additional contemporaneous accounts would make a reassessment substantially stronger.
Location and map reference
La Forêt-Fouesnant, Finistère, France. Broad commune reference from the French government geographic API; not the witness home, precise observation position or object location. This listing does not advertise access to private property.
Sources and review
CNES–GEIPAN official case record and the original investigation report. Current official summary;23-page report pp 1–4 and 21–23, including original time range, October 2019 fieldwork and conclusion. Radio malfunction retained as an unverified association. Sources and current classification checked 13 November 2026. The incident date above is distinct from the later investigation and archive-update dates.