The Aunay-les-Bois Triple-Witness Sighting

Theoretical Alignment

This case's strongest alignment sits with the Extraterrestrial Hypothesis, though at a moderate rather than dominant level, driven mainly by the negative result of GEIPAN's own exhaustive process of elimination: three genuinely independent conventional explanations were tested against real, checkable data — a formal radar request, three meteor-observation databases, and wind data at multiple altitudes — and GEIPAN judged none of them satisfactory (scores of 0.300 to 0.475). That kind of comprehensive, well-documented failure to explain a physically-described object is the strongest evidence this file offers for something outside the conventional inventory, without itself proving an extraterrestrial origin specifically. Interdimensional framing draws modest support from the case's genuinely ambiguous physical status: GEIPAN's investigators explicitly could not determine whether the object was a single coherent craft or several smaller light sources moving in tight formation, an unresolved ontological status more amenable to speculative frameworks than to a simple identified-craft reading. Psycho-Social framing carries real but limited weight: all three witnesses behaved rationally throughout, independently sketching what they saw before comparing notes and spending hours cross-checking their account against drones, aircraft, weather balloons, and satellites rather than immediately reaching for an exotic explanation — a pattern that argues against fabrication or panic-driven misperception, even as GEIPAN's own noted concern about witnesses conferring on some answers keeps this file from being read as fully independent triple corroboration.

Background

Understanding why three friends' five to ten seconds of confusion around a garden fire turned into a two-year investigation involving a national air-defense radar request and a French Navy drone specialist requires understanding both the quiet Normandy commune where it happened and the specific institutional machinery GEIPAN has built for evaluating an account with no photograph, no video, and nothing beyond three human memories and three hand-drawn sketches.

Aunay-les-Bois: A Small Farming Commune in the Orne Countryside

Aunay-les-Bois is a small commune in the Orne department of Normandy, part of the arrondissement of Alençon, at roughly 150–200 meters above sea level, crossed by the Tanche. Its population, 138 in 2023, gives it a population density of only about 16 people per square kilometer — genuinely rural even by the standards of the wider Orne, one of Normandy's most sparsely populated departments. The commune covers roughly 853 hectares of farmland, hedgerow, and small woodland parcels, the kind of countryside where a family might keep a house visited only occasionally, as GEIPAN's report describes this one: in T2's words, the family home of her partner, a longère in a lieu-dit a few kilometers outside the village center, surrounded on all sides by fields and animal enclosures for cattle and horses, with woods nearby. This detail matters directly to the investigation: because the witnesses only visited the property intermittently, GEIPAN's field investigators were never able to conduct their site visit and angle measurements in the witnesses' actual company, and instead reconstructed the sightline from their own visit, the witnesses' photographs, and a later video the witnesses made on site.

GEIPAN's Investigative Apparatus, Tested Under Pandemic Conditions

GEIPAN, the Groupe d'Études et d'Informations sur les Phénomènes Aérospatiaux Non-identifiés, operates within the Centre National d'Études Spatiales (CNES) in Toulouse, tracing its lineage to GEPAN, founded there in 1977 by Claude Poher, reorganized as SEPRA in 1988, and re-established under its current name and public-facing mandate in 2005. This particular case arrived during an unusual moment in that history: the COVID-19 pandemic forced GEIPAN's field investigators to conduct all three of this case's formal cognitive interviews by videoconference in August 2021, rather than in person — the first time, per the investigators' own note in the case file, that this specific interviewing technique had been deployed remotely. The cognitive interview is a structured, non-leading recall technique distinct from simply re-asking the original written questionnaire, designed to help a witness surface additional accurate detail without introducing suggestion; GEIPAN's investigators judged the technique successful even conducted over video, finding "excellent agreement" in the description of the phenomenon's appearance and movement across the interviews and the original questionnaires — with one important exception, discussed at length below, that some of the witnesses' matching answers appeared to reflect after-the-fact consultation rather than three fully independent perceptions.

The investigation that followed was substantial for a case with no photographic evidence: a first investigator conducted the initial remote analysis across 2021 and 2022; two investigators then traveled to the property in the witnesses' absence on June 6, 2022, to photograph the site and take angle measurements matching the witnesses' description; GEIPAN formally requested radar data from the French air-defense radar network CAPCODA on August 10, 2020, receiving a response on September 8, 2020; and, in a final step, the witnesses were asked to return to the property themselves to photograph and film their own sightlines, footage GEIPAN's own IPACO image-processing software then used to calculate minimum object size and speed. Few single-episode, no-photo GEIPAN cases in this site's archive have received a comparably resourced multi-year investigation.

Complete Timeline

Date	Event
1977	GEIPAN's predecessor, GEPAN, is founded within CNES in Toulouse under Claude Poher.
Night of June 2–3, 2020, approx. 00:00–00:05	Sighting begins. Three friends around a bonfire in the garden of a holiday home outside Aunay-les-Bois spot two bluish-white lights crossing the sky low over the horizon to the south-southeast. T3 first exclaims "a comet"; T1 shouts that it is not a comet but a UFO; T2, arriving, repeatedly insists it must be a plane.
Same night, ~5–10 seconds total	The object passes behind a line of poplars, briefly reappears, and disappears from view to the east. None of the witnesses moves to follow it or manages to photograph it.
Same night, immediately after	The witnesses draw sketches of what they saw and compare them, then spend the rest of the night researching drones, aircraft lights, electrical arcing, weather balloons, and satellites online without finding a match.
June 4, 2020	Witness T1 completes and signs GEIPAN's technical questionnaire (the file's recorded August 4 signature date is a witness dating error; GEIPAN received the submission June 4).
August 4, 2020	Witness T2 completes GEIPAN's technical questionnaire.
August 10, 2020	GEIPAN formally requests radar data from the French air-defense radar network CAPCODA.
September 8, 2020	CAPCODA returns its radar data to GEIPAN.
October 18, 2020	Witness T3 completes GEIPAN's technical questionnaire, the last of the three.
2021–2022	A first GEIPAN investigator conducts a remote, desk-based analysis of the case.
August 2021	Cognitive interviews. All three witnesses are separately interviewed by videoconference — the first time GEIPAN conducted this technique remotely — due to pandemic restrictions. Recordings show excellent agreement with the original questionnaires, alongside GEIPAN's specific concern about consultation on certain matching answers.
June 6, 2022	Field investigation. Two GEIPAN investigators travel to the property in the witnesses' absence, take photographs, and make angle measurements reconstructing the witnesses' point of view.
After the field visit	The witnesses are asked to return to the property themselves to photograph and film their own sightlines from each of their original vantage points, for use in GEIPAN's IPACO image-processing analysis.
Following the full investigation	Final classification. GEIPAN's 34-page report (dated November 3, 2025) rates the consistency "good" and the strangeness still "high" after investigation, scores the three hypotheses 0.300, 0.450 and 0.475, finds none satisfactory, and classifies the case « D » (case ID 2020-06-51034).

Detailed Narrative

"Regardez, une comète!" — Three Independent First Impressions

GEIPAN preserved each witness's own written account, and its investigator noted that each was retranscribed "without any modification or correction" specifically to preserve how each witness naturally expressed the event. Witness T1's account begins: "In the night of Tuesday June 2 to Wednesday June 3, between 00:30 and 01:00, three of us adults saw a luminous phenomenon moving across the sky. We were in the commune of Aunay-les-Bois (61), in the garden of a country house in the middle of farm fields and animal enclosures. We observed a luminous phenomenon moving in a straight line from right to left, parallel to the horizon. It passed behind the trees for a few seconds, reappeared, and then left our field of vision. We could see two fairly widely spaced light points that seemed to belong to the same object, with several smaller light points beneath. The color of the lights was slightly bluish." T1 continues: "There was no perceptible engine noise and no wind that evening. The observation lasted around five seconds. The object's speed was constant. The first of us to spot it exclaimed: 'Look, a comet!' I then saw the object myself but it didn't look like a comet (I'd seen one before) and I immediately thought of a flying object given its speed and horizontal trajectory. The third person initially thought of a plane, but we realized it wasn't the usual lights of a plane, and above all the total absence of noise ruled out that option. We then thought of a drone, but the gap between the two lights seemed far too large for that… the object looked fairly large relative to the treetops, but it seemed to be a good way behind them. We then thought of electrical arcing on power lines, but on checking, those lines run in front of the trees, not behind them. In shock, it didn't occur to us to move to follow the object, we were so surprised. We spent a good hour checking online what this object could be, without finding an answer."

Witness T2's independent account describes the same sequence with its own specific detail about who saw what first: "The first of us to spot the phenomenon had walked a little away from us to smoke a cigarette; he was standing in front of the garden's wooden gate with a view over the fields and the treeline beyond. He thought he saw a comet and told us; his partner joined him, exclaiming that it didn't look like a comet, and I then joined them in turn, saying it was probably a plane, and I saw it myself. Then we saw the second point appear from behind the trees that had been hiding it until then. I kept repeating mechanically that it was a plane, but looking more closely I noticed smaller light points aligned underneath, and I thought that it didn't look like the planes I've often had occasion to observe around here — and above all, the place being silent, we're used to hearing them when they fly a bit low, yet the phenomenon was silent. The object finally disappeared from our field of vision; we were sort of stunned, a little puzzled by what we'd just seen, and decided to draw the observation. Comparing our drawings and discussing them, we found we had observed the phenomenon in the same way. We then spent the rest of the evening looking for a rational explanation, looking at various images of drones, atmospheric phenomena, electrical phenomena, farm equipment, planes, etc."

Witness T3, T1's partner and the first to spot the light (T2 places him at the garden gate, smoking; T3 himself says he was walking back toward the house while clearing the table), gave the most vivid account of the moment of realization: "We were at the table with friends in the garden of their country house at Aunay-les-Bois (61). Around 00:30, we'd started clearing the table. I walked toward the house and, turning around, I first saw a single white/bluish light. I first exclaimed 'Oh, a comet!' — we'd talked that evening about our friends' sighting of a comet a few months earlier. Seeing the object, [T1] exclaimed 'That's not a comet, that's a f*cking UFO!' It seems [T2] had just come out of the house and said, sounding scared, 'no, no, that's just a plane.' I initially saw only a single blue point, then a second one entered my field of vision. I then understood it was a single set. The object looked enormous to me and seemed fairly close behind the trees, maybe around 200 meters. For me the object consisted mainly of two whitish-blue lights with a luminous halo beneath. The lights disappeared and reappeared behind the trees. We didn't move, none of us managed to get out a phone to film, we were as if hypnotized. In my estimation, the sighting lasted 8 to 10 seconds. We then each drew what we had seen, to compare. And we spent the evening searching online for anything resembling it, without success."

A Rational, Methodical Response to an Irrational Few Seconds

What stands out across all three independent accounts, and what GEIPAN's own investigators repeatedly noted as a point in the witnesses' favor, is how deliberately unspectacular their behavior was after the sighting ended. None reached immediately for an exotic conclusion; instead, each first tried, in sequence, a comet, an aircraft, a drone, and overhead power-line arcing, testing and discarding each against a specific, checkable detail — the absence of a tail for the comet hypothesis, the absent-but-expected aircraft lighting and total silence for the plane hypothesis, the improbably wide gap between the two main lights for the drone hypothesis, and the physical layout of the power lines relative to the treeline for the arcing hypothesis. Only after sketching what they'd seen, comparing those sketches, and searching the internet without success did they contact GEIPAN. As witness T1 put it in her questionnaire: "We tried every possibility we know of but nothing matches what we saw. The absence of engine noise, the color, the trajectory, etc., nothing seems to correspond to anything we know, nor to an atmospheric phenomenon. We think it's a flying object. We're turning to you to try to get an answer" — a closing line that reads less like a UFO enthusiast's excited report and more like a frustrated, methodical process of elimination that had simply run out of road.

GEIPAN's investigators drew a carefully qualified conclusion in their cognitive-interview analysis: "the general impression that emerges from the study of the questionnaire and interview material is that the witnesses undoubtedly observed together the same phenomenon, physically present, which carries for them a strong sense of strangeness essentially because of its uncommon appearance, but also because of the absence of noise in a particularly silent environment. All three were marked by this observation, T1 and T2 being more emotionally affected, even fearful in T2's case." The questionnaires fill in the reactions: T1 wrote that her partner (T3) "remains fairly indifferent because for him there was nothing extraordinary about it," that T2 was "very frightened" once she realized it could not be a plane, and that she herself was "quite excited and convinced it was not common." T3 described the moment as "quite exciting and at the same time very strange," wrote that T2 "was terrified," and said simply: "I just want to know."

Comparing Sketches: A Real Strength With a Real Caveat

The witnesses' instinct to draw the object immediately after the sighting (T1 says she and T2 drew separately and then showed their sketches to T3) gave GEIPAN's investigators an unusually direct way to check consistency independent of memory or suggestion. Comparing the three sketches, the investigators found strong agreement on the object's core visual structure — two bluish main lights of matching appearance, surrounded by numerous smaller lights and/or a halo of the same color — while also finding genuine, informative small differences: T1 and T2 both described the lights as more purely "bluish," while T3, who initially glimpsed only a single light before the second one emerged from behind the trees, described a "whitened blue." Only T1's sketch depicted the object's disappearance behind the treeline, though all three witnesses mentioned this detail in their written accounts; and T3 uniquely sketched the object as a solid black shape, treating the two lights as visible markers on a single, otherwise-invisible body rather than as free-floating points. GEIPAN's investigators judged the sketches "no doubt more faithful" than the written testimonies themselves, noting that T2 and T3 completed their formal questionnaires two and four months after the sighting respectively, a gap that leaves room for what the investigators specifically termed "false memories" to creep into the written record even without any intent to mislead.

Against that real strength sits an equally real, explicitly documented caveat. GEIPAN's cognitive-interview analysis states plainly: "we observed excellent agreement in the descriptions given during these interviews regarding the appearance and movement of the observed phenomenon, agreement already verified in the description provided in the initial testimonies, in the [questionnaires], but with certain answers identical in every respect, showing that the witnesses consulted each other before writing the [questionnaires]." The report's consistency summary returns to this point directly: "we can only regret that certain information, such as the assessment of the PAN's apparent size, seems to have been determined after mutual consultation." This is a materially different, and more serious, evidentiary limitation than simply comparing sketches for fun after the fact — it means that at least one specific, quantitatively important data point (the estimated 30mm gap between the two lights held at arm's length, used directly in GEIPAN's later size calculations) may reflect group agreement rather than three fully independent measurements, a limitation this site's own analysis carries forward rather than smoothing over.

The Radar Request, the Meteor Databases, and a Genuine Institutional Effort to Disprove the Sighting

Unlike many single-witness GEIPAN files, this case received a real attempt at instrumented corroboration. GEIPAN's formal request to the French air-defense radar system CAPCODA (Centre Air de planification et de conduite des opérations et de défense aérienne, the French Air and Space Force's air-defence operations centre) went out on August 10, 2020 and returned data on September 8, 2020. The result was informative mainly by elimination: CAPCODA recorded exactly one aircraft passage anywhere near the relevant time window, traveling from southwest to northeast around 22:00–22:01 UTC — between thirty minutes and a full hour before the witnesses' own estimated sighting window — passing to the west and north of the witnesses' position rather than the southeast-to-east path they described, and flying at roughly 34,000 feet (about 10,300 meters), far too high to match a phenomenon the witnesses judged large relative to a nearby treeline. GEIPAN noted that the track's orientation matched the phenomenon's but that this was "the only concordant element," and judged the aircraft hypothesis very unlikely, also because the phenomenon showed none of the usual aircraft light colors or blinking.

A parallel effort checked three independent databases of meteor and fireball observations: the International Meteor Organization's database, which recorded one bolide seen by thirteen observers from the south of France at 22:14 on June 2; the FRIPON automated all-sky camera network, whose Belfort and Dijon stations jointly recorded a second bolide at 00:01:45 — within the relevant window — but confined to a low elevation over the southwest horizon, inconsistent with the witnesses' southeast-to-east path; and the Base des Observateurs Amateurs de Météores, which recorded no event at all matching the date and time. GEIPAN's investigators concluded that an atmospheric-entry event large enough to match the witnesses' description — and it would need to have been large, given the implied minimum dimensions discussed below — should have been visible over a wide area and recorded by at least one of these systems, and that its total absence from all three argued against the hypothesis even though the object's reported bluish color, its apparently elongated form, and its dead-straight horizontal path were all independently consistent with a fragmenting meteoroid.

A French Navy Drone Instructor Weighs In

Of the three hypotheses GEIPAN formally scored, the drone hypothesis scored highest, and the investigators consulted a military-drone specialist serving in the French Navy (Marine Nationale), described in the conclusion as a drone instructor, about blue lighting. According to the report, he said blue is not used by drones except on certain Parrot models, and that those lights blink; the conclusion summarizes this as blue lights being usable "but blinking," which does not match the witnesses' steady lights. The investigators layered a second, independently gathered fact on top of that consultation: during their June 6, 2022 site visit, they happened to hear a light touring aircraft flying roughly in the direction of the sighting, audible despite a contrary ground-level wind. GEIPAN itself cautioned that it "will weigh" its conclusions on noise, since engine power and especially the observation distance, which cannot be determined here, remain unquantifiable.

Key Witnesses

The Five Observables Assessment

The Five Observables framework, developed by the Pentagon's Advanced Aerospace Threat Identification Program to describe performance characteristics beyond known conventional technology, is applied here for comparative consistency across the site's case files. This brief sighting offers limited data on most axes; the only "Present" rating rests on the reported silence.

Assessment: The only signal in this case that bears on the observables is acoustic. All three witnesses reported total silence in a very quiet setting, where they had heard cattle and an animal cry minutes earlier. During their June 2022 site visit, GEIPAN's investigators happened to hear a light touring aircraft flying roughly in the direction of the sighting, despite a contrary wind. But GEIPAN explicitly weighted its conclusions on noise, because the object's distance (at least 200 m, possibly much more) and any engine's power cannot be known, so the silence is suggestive rather than decisive. Every other observable reads as absent: constant, moderate, horizontal motion with no acceleration or hover, and calculated speeds (16–20 km/h at 200 m, 40–133 km/h at 500 m, 120–400 km/h at 1,500 m) within conventional performance.

Evidence Assessment

Why "Open"

This case is stamped Open because it combines two things that pull in opposite directions with roughly comparable force: a genuinely well-corroborated, well-investigated account, and an explicit, GEIPAN-acknowledged limitation on just how independent that corroboration really is. On the strength side, three adults, not one, described the same core visual event — two matching bluish lights with a shared halo, moving at constant speed on a dead-level path, in total silence — and did so via a real evidentiary process: written questionnaires filed months apart (June, August, and October 2020), sketches drawn immediately after the sighting, and video cognitive interviews GEIPAN's own investigators judged to show "excellent agreement" with the original accounts. That evidentiary process was reinforced by real institutional effort rarely seen on a no-photo single-episode file: a formal national radar data request, cross-checks against three separate meteor-detection systems, and an expert consultation about military drone lighting. None of those checks produced an explanation, and GEIPAN's hypothesis scores (0.300 for wind-borne balloons, 0.450 for atmospheric entry, 0.475 for drones) left it concluding that "no hypothesis is satisfactory."

Weighing against that strength is the specific, honestly documented consultation problem: GEIPAN's own investigators state plainly that some of the witnesses' matching answers, most importantly the apparent-size estimate that fed directly into the case's later minimum-dimension calculations, appear to have been settled after group discussion rather than independently measured. That doesn't mean the witnesses invented anything — the investigators remained confident all three saw a real, shared phenomenon — but it does mean the file's numerical size and speed calculations rest on a data point that isn't fully independent triple corroboration, a genuine complication this site's own scoring reflects rather than smooths over. Combined with the sighting's very brief five-to-ten-second duration, which limited how much detail any witness could gather regardless of honesty or attentiveness, and the absence of any photographic or instrumented evidence of the object itself, this case earns a score toward the stronger end of the site's published GEIPAN files without reaching the tier reserved for cases with either hard instrumented corroboration or fully independent multi-witness triangulation.

Sensor & Instrumentation Detection Profile

No photograph, video, or purpose-built instrument recorded the phenomenon itself at the time of the sighting — a real limitation given the object's reported proximity, but one partially offset by an unusually thorough after-the-fact instrumentation effort. GEIPAN's formal CAPCODA radar data request is itself a form of instrumented detection attempt, even though its result was a negative (no matching aircraft track). The FRIPON automated all-sky meteor camera network, whose Belfort and Dijon stations did record a fireball within the relevant time window, functioned as an independent, purpose-built optical detection system whose data GEIPAN was able to check directly against the sighting's reported azimuth and time, ruling it out rather than confirming it. Finally, the witnesses' own later reconstruction photographs (T1, T3) and video (T2, shot on an iPhone 11 Pro), made at GEIPAN's request, fed the agency's IPACO image-processing software, which calculated the object's minimum plausible dimensions and speed envelope from the reconstructed geometry — a genuine, if indirect, instrumented analysis of an event no camera captured live.

Environmental & Geospatial Context

The meteorological picture for this sighting comes from the Alençon station, roughly 17 kilometers southwest of the observation site as the crow flies. Between 00:00 and 01:00 local time (22:00–23:00 UTC) on June 3, 2020, the station recorded near-calm winds of just 1 to 1.2 meters per second, oriented between north-northwest and north-northeast; no precipitation; horizontal visibility of 20 kilometers; and no clouds detected by the station's own ceilometer. These are, in short, close to ideal observing conditions, which strengthens rather than weakens the reliability of what the witnesses reported seeing, while also ruling out cloud reflection or precipitation-related optical effects as a contributing factor.

Astronomically, GEIPAN's Stellarium reconstruction for 00:30–01:00 local time placed the Moon to the south (azimuth 187°) at about 26° elevation and, potentially after 00:45, Jupiter to the southeast (azimuth 124°) at just 2°. A JPL Horizons query for the site gives the Moon (89% illuminated, waxing gibbous) at azimuth 199–207° and 28–30° elevation between 00:30 and 01:00, and Jupiter rising at about 00:47 at azimuth 122°. The witnesses also wrote that Venus was visible, but Horizons puts Venus about 15° below the horizon (it was at inferior conjunction on June 3, 2020). Neither the Moon nor Jupiter corresponds to the reported trajectory, which ran from about azimuth 150° as the witnesses gave it (about 190° after GEIPAN's site check) to the east (90°) at about 15° elevation.

Geographically, no airport sits near Aunay-les-Bois itself; the nearest, Alençon's aerodrome, lies some 20 kilometers to the southwest. The one aircraft CAPCODA's radar data did record near the relevant window flew far too early, on the wrong heading, and at an altitude of roughly 10,300 meters — consistent with high-altitude airline traffic rather than anything resembling the witnesses' account. France's official Géoportail mapping service shows the observation site sits within a drone-restricted zone prohibiting unauthorized flight above 50 meters altitude, with that restriction's southeastern and eastern boundary roughly 1.4 kilometers away; GEIPAN's investigators found no evidence of any nearby installation, farm operation, or private property with a specific, identifiable reason to be operating a drone that night.

Observer Credibility & Occupational Profile

None of the three witnesses' occupations are disclosed in GEIPAN's published file. T1 and T3 are a couple; T2 is their friend, whose partner, the fourth person present that evening, did not see the phenomenon and did not take their account seriously. GEIPAN found the three accounts closely matching, with small differences (T3 described a "whitened blue"), and noted that T1 and T2 were the more emotionally affected, T2 to the point of fear. According to T1, T3 remained "fairly indifferent" afterward; T2 said she is "systematically discredited" when she mentions it; and T3 wrote only that he wants to know what it was.

Each witness reported spending real, sustained effort trying to explain what they'd seen before ever contacting GEIPAN — independently sketching the object, comparing sketches, and then spending hours searching online for a matching drone, aircraft light configuration, satellite pass, weather balloon, or electrical phenomenon. According to T1, T3 was "fairly interested in strange aerospace phenomena," while she and T2 were "very interested in astronomy and aerospace phenomena"; T2 had been interested in ufology as a child. T2 said she expects a scientific explanation and would be reassured by one; T1 and T3 wrote that they hoped for answers, "rational or not." This combination — methodical, self-skeptical behavior immediately after the event, genuine and varied emotional responses rather than a uniform script, and an explicitly stated preference for a mundane explanation over an exotic one — supports the account's credibility, tempered specifically by the one documented instance of the three consulting each other before finalizing certain questionnaire answers.

Physical & Material Assays

This case involves no confirmed physical or trace evidence of any kind. The object never approached the ground closely enough, by any witness's account, to leave a mark, and no soil, vegetation, or ground-trace sample was ever collected or tested. All three witnesses listed a string of garden lights ("guirlande de lumière") a little further off in the garden as the only equipment switched on; the report does not discuss it further.

Historical Precedents & Archive Matches

This case's most directly relevant points of comparison on this site are two other multi-witness GEIPAN files, each illustrating a different way corroboration can be complicated even when a sighting is genuinely shared.

Villers-lès-Luxeuil (2019) — A Different Flavor of "Not Fully Independent"

The Villers-lès-Luxeuil Triangle (Open, 50/100) offers the closest structural parallel on this site: a two-witness GEIPAN case explicitly noted as "not fully independent," since one witness followed the other in a separate vehicle and unsuccessfully tried to alert her to an earlier light phenomenon before the shared triangle sighting itself. That case's independence problem is temporal and physical — one witness's attention was drawn by the other's actions in real time. This case's independence problem is different in kind: the witnesses' core visual perceptions appear to have been genuinely simultaneous and separate, but at least one specific data point (apparent size) was settled through discussion after the fact, during the process of completing GEIPAN's own questionnaire, rather than contaminated during the observation itself. Reading the two cases side by side usefully illustrates that "not fully independent" is not a single failure mode in GEIPAN's own methodology, but a spectrum of distinct ways corroboration can be complicated without being invalidated.

Bornel (2014) — Two-Witness Corroboration With Photographic Support

The Bornel Conservatory Fireball (Open, 51/100) presents a genuinely stronger two-witness case by comparison: a retired married couple whose accounts GEIPAN rated at 0.76 consistency, reinforced by actual cellphone photography, however partial, of the phenomenon itself. This case's "good" consistency rating, achieved with three witnesses but zero photographic evidence and the specific consultation caveat noted above, sits close behind Bornel's despite lacking Bornel's photographic corroboration — a comparison that highlights just how much weight GEIPAN's own methodology places on witness count and interview quality even in the total absence of physical or photographic evidence.

Fauville-Évreux (1978) — Radar Data That Cuts the Opposite Direction

The Fauville-Évreux Radar Sighting (Open, 48/100) offers an instructive contrast on the specific question of radar evidence. That case's central evidentiary asset is a real, if unconfirmed and uncorroborated, radar echo — a positive instrument reading that could never be independently verified by a second station. This case's own radar request produced the opposite kind of result: a clean, well-documented negative, with CAPCODA's data able to positively identify the one nearby aircraft in the area and just as positively rule it out as the source given its timing, heading, and extreme altitude. Both cases show GEIPAN making a genuine effort to bring instrumented data to bear on a witness account; here, unlike at Fauville-Évreux, that effort removed one candidate explanation rather than adding an unverifiable signal.

Material Analysis

As with the Physical & Material Assays section above, this case involves no confirmed physical or trace evidence of any kind at the sighting site. No landing marks, soil disturbance, or vegetation damage of any kind were reported or investigated, since the witnesses' own account places the object well above the treeline and never approaching the ground.

The Conventional Explanation Candidate: Atmospheric Entry, Wind-Borne Balloons, and Drones

GEIPAN's own file organizes its analysis around three principal hypothesis families — a fragmenting atmospheric entry, wind-borne LED balloons, and one or more drones — scoring each on a 0-to-1 reliability scale (its bands: certain 100%, strong >80%, medium 40–60%, weak 20–40%, very weak <20%). It concluded that none is satisfactory.

Atmospheric Entry (Bolide/Meteor), Scored 0.450

The object's reported bluish coloring, possible elongated form, and dead-straight horizontal trajectory are all independently consistent with a meteoroid fragmenting in the atmosphere. Weighed against this: three separate observation databases (IMO, FRIPON, and BOAM) recorded no matching event at the correct time and azimuth, despite an entry of the implied scale — large enough to be visible with the apparent size witnesses described — plausibly being expected to register on at least one system, especially given the clear, calm June evening's excellent visibility.

Wind-Borne LED Balloons, Scored 0.300

The weakest of the three candidates. Ground-level wind data showed a near-calm breeze from the north-northwest to north-northeast, essentially incompatible with the object's reported south-southeast-to-east path. Wind at higher altitudes (850–700 hPa, roughly 1,000–2,500m) blew closer to west-southwest, marginally more compatible but still badly oblique to the witnesses' line of sight, and would require an altitude and corresponding distance the witnesses never reported noticing. The scenario is further undermined by timing: festive LED balloons are typically used on weekends or during school holidays, neither of which applied to a Tuesday-to-Wednesday night in early June.

Civilian or Military Drone(s), Scored 0.475

The strongest of GEIPAN's three candidates; the report calls it the hypothesis "closest to the characteristics of the observation." In favor: the object's low elevation, calculated speed range, and light configuration are all broadly compatible with a drone flown a few hundred meters from the witnesses. Working against it: French civil aviation law prohibits civilian night drone flights outright; a French Navy drone specialist said blue drone lights, where used, blink, unlike the steady lights reported; and the witnesses' own angular data implies a minimum single-object size of roughly 9 meters, well beyond any known civilian or military drone, which would require multiple drones flying in tight nighttime formation — a scenario GEIPAN judged "still possible, but unlikely."

Related Cases

Further Reading — Recommended Literature

DISCLOSURE: Some links above are affiliate links. If you purchase through them, OverClassified receives a small commission at no additional cost to you. This helps support independent research.

Essential Viewing

No Videos Available

Sources

• GEIPAN (CNES), "AUNAY-LES-BOIS (61) 03.06.2020", Case 2020-06-51034 — official case summary
• GEIPAN (CNES), Compte Rendu d'Enquête, AUNAY-LES-BOIS (61) 03.06.2020 — full 34-page investigation report (PDF)
• GEIPAN (CNES), Témoignage T1, AUNAY-LES-BOIS (61) 03.06.2020 — witness T1 structured summary
• GEIPAN (CNES), Témoignage T2, AUNAY-LES-BOIS (61) 03.06.2020 — witness T2 structured summary
• GEIPAN (CNES), Témoignage T3, AUNAY-LES-BOIS (61) 03.06.2020 — witness T3 structured summary
• GEIPAN (CNES), official case archive & public portal — geipan.fr
• GEIPAN (CNES), Glossaire du GEIPAN — official classification definitions, A/B/C/D/D1/D2
• GEIPAN (CNES), Son histoire — institutional history: GEPAN (1977), SEPRA (1988), GEIPAN (2005)
• GEIPAN (CNES), "Missions, méthodes et résultats" — investigative structure and field-investigation practices
• International Meteor Organization, Fireball event database, 2020/2678 — cross-checked bolide event ruled out by GEIPAN
• Wikipedia (English), "Aunay-les-Bois" — commune history, geography, and population data
• Wikimedia Commons, File: Aunay-les-Bois - mairie 20180930.jpg, photo by "Pymouss" (2018), CC BY-SA 4.0 — licensing and attribution
• Wikimedia Commons, Category: Aunay-les-Bois — image archive for the commune
• Insee, Dossier complet — Commune d'Aunay-les-Bois (61013) — official French national statistics for the commune
• This site, case-files/villers-les-luxeuil-triangle.html — comparative two-witness GEIPAN case explicitly noted as "not fully independent"
• This site, case-files/bornel-conservatory-fireball.html — comparative two-witness GEIPAN case with photographic corroboration
• This site, case-files/fauville-evreux-radar-sighting.html — comparative GEPAN/GEIPAN case with an unconfirmed instrumented (radar) signal
• This site, case-files/foret-fouesnant-electric-bar.html — comparative single-witness GEIPAN case investigated via cognitive interview
• This site, key-figures.html, entry on Claude Poher — GEPAN founding and early institutional history
• This site, locations/index.html, entry on Toulouse — CNES/GEPAN/SEPRA/GEIPAN institutional headquarters
• UFO Transparency, full-text extraction of GEIPAN's Compte Rendu d'Enquête, AUNAY-LES-BOIS (61) 03.06.2020 (ref. DTN_DA_GP-2024.0012609, dated 03/11/2025, 34 pp.), used while geipan.fr was returning 404 errors (October 2026). ufotransparency.com
• NASA JPL Horizons ephemeris queries (Moon, Venus, Jupiter; Aunay-les-Bois, 22:30–23:00 UTC June 2, 2020). ssd.jpl.nasa.gov

Corrections (October 2026)

A line-by-line check against the full text of GEIPAN's 34-page investigation report and JPL Horizons led to the following changes. The witnesses' identities had been scrambled. The report's questionnaires show that T3, a man, was the first to see the light and called it a comet; T1, his partner, shouted that it was a UFO; and T2, a friend whose partner's family owned the house, kept insisting it was a plane and was the most frightened. The page had made T1 the man who saw the "comet" and the one who was "indifferent," made T2 "T1's partner" who first disputed the comet, and attributed the "UFO" exclamation to T3 (as "her"). The bracketed names in T3's quoted account have been corrected to match. The quote "We tried every possibility we know of..." comes from T1's questionnaire, not T2's cognitive interview. The witnesses' interests had also been swapped between T1 and T3. CAPCODA was misexpanded; it is the Air and Space Force's "Centre Air de planification et de conduite des opérations et de défense aérienne." The 2022 audibility of a light aircraft was an incidental observation during the site visit, not a "controlled test," and GEIPAN explicitly cautioned that it weighted its conclusions on noise because the object's distance is unknown. The claim that GEIPAN "considered and dismissed" the garden lights, the claim that GEIPAN treated the witnesses' differing emotions as evidence against a coordinated story, and a "50% threshold" for hypotheses (GEIPAN's scale puts 40–60% in a single "medium" band) were not in the report and were removed. The numeric consistency and strangeness figures (0.72/0.53) could not be re-checked because geipan.fr was offline; the report itself rates consistency "good" and strangeness "high." Venus, which the witnesses reported seeing, was below the horizon (JPL Horizons; inferior conjunction on June 3, 2020), and the Moon was 89% illuminated, not 92%. The unsourced demonym and a 175 m elevation were removed.

Following this review, the rating was changed from Open 56 to Open 54 (October 27, 2026), because the higher rating had rested partly on claims corrected or removed above.