The Sainte-Foy-de-Peyrolières Five-Phase Sighting

Theoretical Alignment

This case aligns most strongly with the Extraterrestrial Hypothesis among the frameworks this site tracks, largely on the strength of its structured-craft imagery: a coherent geometric object — first rectangular, then triangular — marked by evenly spaced lights, capable of both a slow, controlled descent and a sudden, silent, trail-less burst of acceleration which GEIPAN could not match to the aircraft, satellites, fireworks or drones it checked. It carries real, if secondary, Interdimensional alignment on the basis of the object's fluid morphological transformation — a rectangle becoming a triangle and then a rectangle again is the kind of non-fixed physicality IDH proponents cite as evidence against a simple, solid, terrestrial craft. Psycho-Social framing carries genuine, non-trivial weight specifically because GEIPAN's own investigators formulated and tested it directly as one of their seven formal hypotheses: a witness who had been looking at illuminated lights on his own house facade in the minutes before the sighting is exactly the profile a retinal-persistence explanation is built around — but GEIPAN rejected it because the lights changed configuration and the witness followed them with his eyes, discussed below.

Background

Understanding why GEIPAN treated this case with such unusual investigative intensity — two separate expert committee reviews, a full field reconstruction, and a formal cognitive interview for what remains, on paper, a single uncorroborated witness account — requires understanding both the specific richness of what was reported and the institutional machinery France has built to investigate exactly this kind of account.

Sainte-Foy-de-Peyrolières: A Fortified Refuge on the Gascon Plain

Sainte-Foy-de-Peyrolières is a commune of roughly 2,100 residents in the Haute-Garonne department, sitting on the plain southwest of Toulouse near the border with the Gers, within the Cazères canton. Its own municipal history traces its founding to around 1050, when Hélie, the lord of Samatan, donated forested land to Benedictine monks from the Abbey of Conques; the monks cleared the woods, built a church, and laid out a fortified, square-shaped settlement on raised ground for defense. A generation later, around 1070–1100, Hélie's son Odon de Samatan made a further donation that allowed the creation of a second settlement, a sauveté — a refuge zone offering peasants specific legal privileges and physical security in exchange for settling and working the land — which drove real population growth in the centuries that followed. The town's fortunes then followed the broader arc of medieval Gascon history: plague in 1348 killed more than two-thirds of the inhabitants, the Black Prince's raid of 1355 ruined the sauveté, and the Wars of Religion in the 1560s–1590s brought a further quarter-century of pillaging and terror before the community formally recognized Henri IV as its legitimate king around 1595. The present church, the third on the site, was completed in 1791 after part of its predecessor had collapsed. Population recovered to around 1,500 by 1860, and a railway line from Toulouse opened in 1900.

What makes the commune's location notable for this specific case is less its own history than its proximity to one particular institution: Sainte-Foy-de-Peyrolières sits within the same department, Haute-Garonne, that has hosted Toulouse's Centre National d'Études Spatiales (CNES) campus — and, within it, GEIPAN itself — continuously since the agency's 1977 founding. The sighting occurred, in other words, well within GEIPAN's own home department, a detail that in principle should have made rapid follow-up and thorough field investigation more logistically straightforward than for a case reported from a more distant region of France.

GEIPAN's Investigative Machinery: Volunteers, Experts, and a Two-Person Core

GEIPAN's own published account of its methodology describes an organization built around a genuinely small core: just two CNES staff positions plus limited administrative support, multiplying its actual investigative reach through a network of roughly twenty trained volunteer field investigators recruited nationally, and around fifteen scientific experts (GEIPAN lists plasma physics, meteorology, photography and psychology, among others) who form the "comité des experts" (expert committee) consulted on the agency's most complex or strangest cases. Field investigations, which are reserved for cases likely to reach a D classification, typically involve an in-person witness meeting, a cognitive interview — a structured, non-leading interview technique developed in cooperation with a CNRS cognitive psychology laboratory in Toulouse specifically to maximize accurate recall without introducing false memories — and, where the terrain and timeline allow it, a physical reconstruction of the sightline and geometry involved. This case received essentially the full version of that process: a technical questionnaire completed by the witness about two months after the sighting, a meeting on December 11, 2018 with GEIPAN's head and deputy head for a reconstruction and cognitive interview, a later review by an experienced investigator, and two sessions of the expert committee, on June 21, 2022 and May 9, 2023, reflecting how much difficulty GEIPAN's own investigators had in settling on a single leading explanation.

GEIPAN's published statistics give useful scale. The agency receives about 500 contacts a year, of which about 200 lead to an investigation published on its site. GEIPAN's methods page gives 59% of cases classified A or B (explained or probably explained), 34% C (unusable for lack of information), and about 7% D (unexplained), with about 2% D over the last ten years; its dynamic statistics for published cases give 3.1% D. About 10% of cases get a field investigation, and GEIPAN says D cases "very often" go to the expert committee before final classification. Two committee sessions and nearly six years from sighting to report put this case at the more intensive end of that range.

Complete Timeline

Date	Event
c. 1050	Hélie, lord of Samatan, donates forested land to Benedictine monks from Conques, founding the fortified settlement of Sainte-Foy-de-Peyrolières.
1977	GEIPAN's predecessor structure is founded within CNES in Toulouse — the same department in which this sighting later occurs.
July 8, 2018, about 23:29	From his garden, the witness sees what he takes to be a shooting star with a very long trail. FRIPON and BOAM meteor-camera networks later show no matching event (with the caveat that their coverage is incomplete).
July 8, 2018, approx. 23:30 (Phase 1, ~4–5 seconds)	Sighting begins. Five or six intensely luminous points appear and descend a short distance with a "cascade" effect, then dim.
July 8, 2018 (Phase 2, ~5 seconds)	Four of the points, now in an aligned formation, move slowly downward, revealing a dark, rectangular shape on which they sit in geometric alignment; the witness distinguishes dark orange hues on the shape.
July 8, 2018 (Phase 3)	The shape's motion shifts from vertical to horizontal; the rectangle progressively morphs into a triangle and begins gaining speed.
July 8, 2018 (Phase 4, ~1–2 seconds)	The shape undergoes a burst of blinding, silent, trail-less acceleration — the moment the witness later described as the strangest part of the entire event.
July 8, 2018 (Phase 5)	The shape reverts to a rectangle, slows, and fades into the night sky. Total observation duration: approximately 20 seconds.
July 8, 2018, 23:41	The witness leaves a 149-second voicemail with GEIPAN describing a 2–3-second observation toward the Great Bear: four faint points in a horizontal line, then two, a change to a triangle, a fast silent passage and a sudden disappearance; he suggests a military aircraft. About 15–20 minutes after the sighting he sees four to eight aircraft with red and green lights "turning around Sainte-Foy." On July 10 he leaves a second message saying he has no photo and does not wish to follow up.
Aug. 31 – Sept. 9, 2018	The witness sends a typed account, ten sketches, two maps and an annotated sky chart (August 31), completes GEIPAN's technical questionnaire (September 6) and adds details (September 9). GEIPAN asks the CNOA for an air-traffic radar plot on October 29, 2018.
December 11, 2018	Field investigation. GEIPAN's head and deputy head meet the witness for a reconstruction and cognitive interview. He adds further details by e-mail in 2018–2023, including video reconstructions and azimuth measurements in July 2021; he had also contacted the gendarmerie, which did not open a report.
June 21, 2022 and May 9, 2023	GEIPAN's expert committee reviews the case twice. CNES's space-surveillance center (COSE) reports no atmospheric reentries that day, no GRAVES radar detection, and about ten satellites on north-south tracks at the time but not matching the reported groups of points.
During the investigation	The DGAC at Toulouse-Blagnac supplies a radar/video replay of traffic over a 16 × 24 nautical-mile box centered on Toulouse for 21:50–22:10 UTC (23:50–00:10 local), beginning about twenty minutes after the sighting; it shows airliners over Toulouse, not aircraft circling Sainte-Foy. The CNOA plot requested in 2018 is never received; a renewed request comes too late for the data-retention period.
June 7, 2024 (published August 2024)	Final classification. GEIPAN's report classifies the case D (listed as D1 in the online database), case 2018-07-50618, strangeness 0.55, consistency 0.72.

Detailed Narrative

Twenty Seconds, Five Phases: What the Witness Described

What distinguishes this case from the great majority of single-witness UAP reports GEIPAN handles is not any one dramatic detail but the sheer structural granularity of the account as a whole. Most brief nighttime sightings, when GEIPAN's investigators reconstruct them carefully, resolve into a single continuous action — a light crossing the sky, a shape passing overhead. This witness instead described a sequence with five internally distinct phases, each with its own duration, motion characteristics, and visual details, unfolding across a total of roughly twenty seconds. Phase one, lasting some four to five seconds and beginning about a minute after a shooting star had crossed the same patch of sky, consisted of five or six intensely bright points of light appearing and then descending a short vertical distance with what the witness called a "cascade" effect — a term suggesting a staggered, waterfall-like sequential dimming rather than all points moving in lockstep — before their brightness faded. Phase two, roughly five seconds long, saw four of those points, now arranged in an aligned formation, continue a slow downward movement that gradually revealed a solid dark shape behind them: a rectangle, on which the four lights sat in clear geometric alignment, with the witness able to make out dark orange tones on the body of the shape itself.

Phase three marked the account's first major transition: the object's motion shifted from primarily vertical to primarily horizontal, and its shape began progressively changing from a rectangle into a triangle, gaining speed as the transformation completed. Phase four, compressed into just one or two seconds, was the moment the witness singled out afterward as the strangest part of the entire experience: a burst of extremely rapid, almost instantaneous acceleration, with no accompanying sound and no visible trail or exhaust of any kind. Phase five closed the sequence out: the shape reverted to a rectangular outline, slowed its motion, and simply faded away into the night sky, bringing the total observed duration to approximately twenty seconds from first light to final disappearance. GEIPAN's summary singles out the quantity of very precise information about the key phase 4, while also calling the observation "very chaotic and discontinuous." This five-phase version is the one GEIPAN analyzed; it comes from the witness's written account of August 31, 2018, not from his first report (see below).

A Second Witness Who Saw Nothing

A detail GEIPAN's file is careful to record honestly, rather than omit or downplay, is that a second person was present alongside the primary witness that evening and did not see the phenomenon at all. This is not, on its own, a mark against the primary witness's credibility — GEIPAN's summary attributes it simply to how quickly everything happened. A twenty-second event, arriving without warning in a specific patch of sky, is exactly the kind of sighting a second person can easily miss entirely if their attention was directed elsewhere even briefly. The absence of a corroborating witness therefore does not, in GEIPAN's own assessment, constitute active counter-evidence against the sighting having occurred as described; it does, however, mean the case retains the single most common structural limitation of UAP reports generally — no independent second account exists to cross-check the primary witness's specific details against.

From a Two-Second Voicemail to a Twenty-Second Account

The witness's first contact with GEIPAN, a 149-second voicemail left at 23:41, ten minutes after the sighting, is the account closest to the event, and it is much simpler than what followed. It describes an observation of 2–3 seconds toward the Great Bear: four faintly lit points in a horizontal line, then only two, a change to a triangle, a fast, silent passage and a sudden disappearance; the witness suggested a military aircraft and asked whether anyone else had seen it. Two days later he left a second message saying he had no photo and did not wish to follow up. On August 31 he sent a typed account, ten sketches, two maps and an annotated sky chart; he completed the technical questionnaire on September 6 and kept sending additions by e-mail for years afterwards, including video reconstructions and azimuth measurements in 2021. GEIPAN did not reproduce those later additions, saying they did not change its analysis. The growth from a 2–3-second glimpse to a twenty-second, five-phase sequence is the central question about this testimony. GEIPAN still rated its consistency "good," noting the large quantity of precise information about the key phase, but also called the observation "very chaotic and discontinuous" and warned that estimates of duration, distance and speed were highly uncertain.

The Field Investigation: Reconstruction and Cognitive Interview

GEIPAN's field investigation went beyond simply reviewing the completed questionnaire on paper. Investigators visited the site, conducted a physical reconstruction of the witness's original vantage point and sightline, and carried out a formal cognitive interview — the structured, non-leading recall technique GEIPAN's investigators are trained in specifically to elicit maximum accurate detail without contaminating memory with suggested content. That interview process, per the case file, confirmed the overall coherence of the witness's account across the full sequence, while surfacing exactly one minor internal contradiction worth noting honestly rather than glossing over: the witness wrote that during the key acceleration he "bounded" four or five meters to get past a large bay laurel tree and "must have lost 1 second of visual" while watching where he put his feet, yet elsewhere said he never lost sight of the object. GEIPAN's investigators treated this as a minor, non-disqualifying discrepancy rather than a serious credibility problem — the kind of small inconsistency cognitive-interview methodology is specifically designed to surface in any account, including entirely truthful ones, since human visual attention over any multi-second span is rarely perfectly continuous and rarely perfectly self-reported after the fact. GEIPAN drew a practical lesson from it: an observer who is running past a tree during a one-to-two-second phase cannot be watching continuously, and the object seen before and after the tree could in principle have been different lights, though GEIPAN judged that unlikely given how short the gap was.

Seven Hypotheses, Two Expert Committee Reviews

What sets this case apart procedurally from most single-witness GEIPAN files on this site is the sheer breadth of formal hypothesis testing applied to it. GEIPAN's investigators formulated and tested seven distinct explanatory hypotheses: a passing train of satellites; fragmentation during atmospheric reentry of space debris or a meteoroid; fireworks; a single aircraft; a squadron or formation flight of several aircraft with navigation lights illuminated; retinal persistence carried over from the witness having looked at illuminated lights on his own house facade shortly before the sighting began; and a coordinated "ballet" of multiple drones. Investigators took the case to GEIPAN's expert committee twice, on June 21, 2022 and May 9, 2023, reflecting real difficulty in settling on a single best explanation.

The Leading Candidate: A Squadron in Formation

Of the seven hypotheses tested, GEIPAN's own file identifies the formation-flight hypothesis — several civilian or military aircraft flying together at night with navigation lights illuminated — as the one most coherent with the specific elements the witness described: multiple aligned light points, a shape with hard edges, and a transition in apparent formation geometry over the course of the sighting could plausibly map onto several aircraft flying in changing relative positions, their individual lights merging visually into what looked like a single structured object from the witness's distant vantage point. Despite judging it the most coherent candidate, GEIPAN's investigators ultimately invalidated the hypothesis for two specific, independently sufficient reasons.

First, the geometry and the silence do not fit together. GEIPAN's summary puts the angular distance covered at about 90° to 110° in four seconds; its report works from the questionnaire's figure of about 135° in one to two seconds during the acceleration, from north or north-northeast to south-southeast. Taking the witness's own distance estimates of 500 m to 2 km, that implies average speeds of about 4,300 to 35,000 km/h and an object 22 to 88 meters long, which should have produced a sonic boom. Several aircraft fast and close enough to produce that apparent motion would be loud, all the more so with a light northwest wind partly carrying sound toward the witness, yet he heard nothing. GEIPAN added that the estimates are very uncertain, since the observer was moving and the key phase lasted about a second, and that a distant formation could combine apparent speed with silence early in the sighting (when it appeared at about 65° elevation) but less easily at the end (about 15°).

Second, the radar data do not support it, but they are also weak. GEIPAN asked the Centre National des Opérations Aériennes (CNOA) for an air-traffic plot on October 29, 2018; the report says that either it never arrived and was not chased, or it was mislaid, and the summary says a renewed request came too late for the CNOA's data-retention period. GEIPAN called this "truly regrettable," noting that the witness himself had written that only a radar detection could confirm his account. What GEIPAN did obtain was a radar/video replay from the DGAC at Toulouse-Blagnac covering a box centered on Toulouse from 23:50 to 00:10 local time, starting about twenty minutes after the sighting. It showed too few aircraft, all over Toulouse and at widely varying altitudes, rather than a group circling Sainte-Foy. The CNOA told GEIPAN that a military exercise on a Sunday night was improbable; possible army helicopter (ALAT) flights were considered but not investigated. GEIPAN's aeronautical expert, a former fighter pilot, judged a night formation flight improbable: "Flying at 23:30 would be of little interest and would require specific training," and the landing and refuelling lights that might look like bright points are fixed, point along the aircraft's axis, and are not used outside those phases of flight. GEIPAN also looked at, and judged unlikely, a night flight by the Patrouille de France.

The Other Six Hypotheses

Each of the remaining candidates was tested and set aside on its own grounds. CNES's space-surveillance center (COSE) found no satellite pass matching the reported groups of points (about ten objects crossed north-south or south-north in the window, at different times), no atmospheric reentry over France that day, and no detection by the GRAVES space-surveillance radar; Starlink trains did not yet exist (the first was launched in May 2019). The lights as described, stable for several seconds, without trails and "slightly golden white," also did not look like a reentry, and GEIPAN's committee judged a meteor fragmentation followed by an unrelated second phenomenon at the same place and time implausible. Fireworks were considered because it was a summer Sunday near a large city, but the initial elevation (about 65°) would have required a professional display, none was found, and the long horizontal movement after phase 3 does not fit. A single aircraft seen in perspective, turning toward the witness, could account for the apparent changes of shape after phase 3 (rectangle, chevron, rectangle), and the witness said he was familiar with aircraft; GEIPAN set it aside because the number of lights did not fit, and a Patrouille de France Alpha Jet was judged improbable.

The retinal-persistence hypothesis came from the witness's own account: just before the sighting he had been looking with his mother at the brightly lit front of the house, though GEIPAN did not know how strong that lighting was. It was rejected because the lights changed configuration (an afterimage keeps its shape), because the witness followed the lights with his eyes (an afterimage moves with the gaze, not the other way round), and because tracking with the eyes is jerky while he described smooth motion. A "ballet" of drones could explain the lights' relative movements and the changes of direction and speed, but at a drone's top speed of about 150 km/h the observed angular motion would put them only 8.5 to 17 meters away, where the witness would have recognized them and heard them; public drone shows must also be declared to the prefecture, and none was found in the Toulouse area that night.

The Verdict: Classification D

Having tested and set aside all seven hypotheses, and having consulted its expert committee twice, GEIPAN classified the Sainte-Foy-de-Peyrolières sighting as D (listed as D1 in its database): a phenomenon of unexplained nature and origin, with an official strangeness rating of 0.55 and a consistency rating of 0.72. That consistency figure, rated "good," reflects the unusually dense, internally structured nature of the single witness's account, confirmed under a formal cognitive interview and reconstruction, weighed against the case's two structural limitations shared with most single-witness UAP reports: no second corroborating account, and no photographic or video documentation of any kind.

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 case registers a genuinely strong positive match on one specific observable, with real ambiguity on two others.

Assessment: This case's clearest positive observable is Instant Acceleration: the witness singled out Phase 4's silent, trail-less burst of speed as the single strangest moment of the entire sequence, and it maps directly onto AATIP's canonical "instant acceleration" category with none of the ambiguity that hedges most of this case's other observables. Hypersonic performance is only speculative rather than confirmed because GEIPAN's own file does not establish a fixed range to the object at any point, making a literal velocity calculation impossible from the public record alone — but the angular rate reported is difficult to reconcile with any conventional aircraft close enough to be seen with this much shape detail. Low Observability is speculative for a different reason than in most comparable cases: rather than a clean negative radar check (as in the Montluel case), this file has a partial, incomplete radar picture that neither confirms nor rules out a conventional low-altitude aircraft or drone target escaping full detection.

Evidence Assessment

Why "Open"

This case is stamped Open rather than Credible, Disputed, or Debunked because GEIPAN's own investigation reached exactly that state: an unusually rigorous, multi-hypothesis, twice-expert-reviewed investigation that nonetheless failed to find any conventional explanation that survives its own scrutiny, without ever reaching the kind of multi-witness or physical-trace corroboration that would justify a stronger designation. The case earns real, substantial credit on the strength of its investigative process: a witness whose detailed written account and technical questionnaire earned a "good" (0.72) consistency rating from GEIPAN's own field investigators; a field reconstruction and formal cognitive interview that confirmed the account's overall coherence; seven independently tested hypotheses, including a genuinely strong leading candidate (aircraft formation flight) that was invalidated on two separate, specific grounds rather than dismissed casually; and two separate expert-committee consultations, reflecting real investigative difficulty rather than a rubber-stamped conclusion.

Set against that, "Credible" would still overstate the case as it stands in the public record: it remains, in the end, a single witness with no photographic or video evidence and no second corroborating observer, its first, ten-minute-old description was of a 2–3-second glimpse, not the twenty-second sequence analyzed later; and the radar check performed — unlike the clean negative result in the comparable Montluel case — was itself incomplete, leaving a genuine evidentiary gap rather than a confirmed absence of conventional air traffic. A case whose own investigating agency ran an unusually thorough process, twice consulted its full panel of outside experts, and still concluded that nothing it could test reduces the account's underlying strangeness is exactly the state "Open" is meant to describe: genuinely, carefully investigated, and genuinely unresolved.

Sensor & Instrumentation Detection Profile

This case's instrumented data are partial and inconclusive. The DGAC at Toulouse-Blagnac supplied a radar/video replay of a 16 × 24 nautical-mile box centered on Toulouse for 21:50–22:10 UTC (23:50–00:10 local), which begins about twenty minutes after the sighting; it shows airliners over Toulouse at varying altitudes and nothing matching the account or the aircraft the witness saw circling Sainte-Foy afterwards. The CNOA air-defense plot requested on October 29, 2018 was never received. CNES's space-surveillance center reported no matching satellites, no reentries and no GRAVES radar detection, and the FRIPON and BOAM meteor-camera networks showed no event matching the preceding "shooting star." No photograph, video, or audio recording exists.

Environmental & Geospatial Context

The sighting occurred on a clear night. Météo-France data for Toulouse-Blagnac, about 23.5 km east-northeast of the witness, show a light northwest wind, no cloud layer below 7,500 m and visibility of about 30 km; the witness described the sky as "magnificently starry and totally clear." The Moon (about 24% lit) was below the horizon (NASA JPL Horizons). The witness was looking north toward the Great Bear, Draco and the Little Bear, which GEIPAN confirmed with Stellarium; the phenomenon appeared at about 65° elevation and ended at about 15° toward the south-southeast. A "shooting star" with a very long trail crossed the sky about a minute before. GEIPAN found no matching event in the FRIPON or BOAM meteor-camera networks, while noting their coverage is incomplete, and raised the possibility that the "shooting star" was itself a first aircraft.

Geographically, Sainte-Foy-de-Peyrolières sits on open Gascon plain roughly 36 kilometers southwest of central Toulouse and around 18 kilometers from Muret, within reach of civilian air traffic converging on Toulouse-Blagnac Airport as well as the former Francazal military and civil aerodrome closer to Toulouse's southwestern suburbs. This places the sighting within a genuinely active regional aviation corridor — a relevant fact for weighing the aircraft-formation hypothesis GEIPAN's investigators ultimately invalidated, since ordinary civil air traffic in the area is dense enough that a coordinated multi-aircraft formation flight, had one actually occurred, should in principle have left some trace in either radar records or ordinary air-traffic logs, neither of which materialized in a form matching the witness's account.

Observer Credibility & Occupational Profile

The witness's specific occupation is not disclosed in GEIPAN's published case summary, consistent with the agency's standard anonymization practice, but the file otherwise documents a credibility profile built from behavior rather than credentials: a voicemail to GEIPAN ten minutes after the sighting (describing a much shorter event than his later account); a willingness, two months later, to complete the agency's full structured technical questionnaire in unusually granular, phase-by-phase detail rather than a vague summary; and full, sustained cooperation with a subsequent field investigation, including an in-person reconstruction and a formal cognitive interview specifically designed to surface inconsistencies. One inconsistency did surface, about whether he lost sight of the object while running past a tree. The account also kept growing: GEIPAN received additional details by e-mail through 2023 and chose not to reproduce them.

The case's structural limitation is the same one shared by the majority of single-witness UAP reports: however internally consistent and well-documented one account is judged to be, it cannot carry the same evidentiary weight as independent corroboration from a second observer. That limitation is made more visible, not less, by the fact that a second person actually was present that evening and did not see it — plausible if the event was as brief as the first voicemail described, but one that leaves the case's evidentiary foundation resting on a single account regardless.

Historical Precedents & Archive Matches

Comparing this case against the other GEPAN/GEIPAN-investigated French sightings already published on this site highlights both what makes it distinctive and where its evidentiary profile sits relative to comparable files.

Montluel (2022) — A Clean Negative Radar Check, Versus an Incomplete One Here

The Montluel cloud-distorting object (Open) offers the closest direct structural parallel on this site: another single-witness, no-photograph GEIPAN case investigated with a comparably thorough hypothesis-elimination methodology, including a formally tested and rejected retinal/optical-perception hypothesis. The two cases diverge sharply on their radar evidence specifically: Montluel's investigators obtained a clean, if incomplete-coverage, negative result from CAPCODA's (and its predecessor units') air-defense radar picture, finding no matching aircraft at all. This case's Toulouse-Blagnac civilian radar check instead returned a partial, mistimed, non-matching data set — a genuine gap in the instrumented record rather than a confirmed absence of air traffic, and one GEIPAN's own file explicitly regrets not being able to close with the fuller CNOA national air-defense restitution.

Montayral (2015) — Consistency Ratings at Opposite Ends of the Single-Witness Spectrum

The Montayral aerodrome object (Open) is instructive by contrast: another single-witness, no-photograph case whose consistency was judged only "just sufficient" to validate its D1 classification, with no upward revision at any point in its investigation. This case's own consistency rating of 0.72 ("good") sits meaningfully higher, reflecting the additional structural richness of a five-phase, internally differentiated account versus Montayral's comparatively simpler single-action sighting — a genuine difference in the density and checkability of the underlying testimony, not merely a difference in how each file happens to be written up.

Fauville-Évreux (1978) and Bornel (2014) — Positive Radar and Photographic Evidence This Case Lacks

The Fauville-Évreux radar sighting (Open) carries a genuine, if uncorroborated, positive radar contact roughly matching its visual sighting's bearing and timing — a different category of instrumented evidence than either this case's partial, non-matching Blagnac data or Montluel's clean negative. The Bornel conservatory fireball (Open), published alongside this case, carries two corroborating witnesses and an inconclusive but real photograph, illustrating the specific kind of multi-source corroboration this Sainte-Foy-de-Peyrolières file simply does not have, however unusually detailed its single account is judged to be.

Material Analysis

This case involves no physical or trace evidence of any kind — no landing marks, no recovered material, no soil or vegetation sample, and no biological or physiological effect on the witness beyond his own visual observation. Nothing in the account places the object anywhere near ground level or within physical reach of the witness at any point during the sequence; the phenomenon was airborne and, per the witness's own description, at a substantial distance throughout its roughly twenty seconds of visibility. There is accordingly nothing to sample, test, or physically examine after the fact — only the witness's own detailed verbal and questionnaire-based description of what he observed.

The Conventional Explanation Candidate: A Night Formation Flight, and Six Rejected Alternatives

GEIPAN's file treats a night formation flight of several aircraft as the most coherent conventional candidate among the seven it tested. Multiple aircraft in changing relative positions could, from a distance, merge visually into what looks like a single structured object with aligned lights, and the witness himself saw four to eight aircraft with red and green lights around Sainte-Foy fifteen to twenty minutes later, without knowing whether they were civil or military; in 2019 he spoke of "reconnaissance aircraft." GEIPAN nevertheless judged the hypothesis invalidated by two elements: the apparent speed implied by the account, which requires a nearby formation that should have been heard; and the DGAC radar replay, which showed too few aircraft, all over Toulouse, though it began about twenty minutes after the sighting. Its aeronautical expert also judged a night formation flight improbable, and the CNOA plot that might have settled the question was never received. Because the key radar data are missing and the account is uncertain on timing, this candidate is weakened rather than eliminated.

Satellite Train

A passing train of satellites was tested specifically because this category of misidentification has become increasingly common. Starlink trains did not exist yet in July 2018 (the first was launched in May 2019), and CNES's space-surveillance center found no satellite pass matching the reported groups of points. It also fails to match this case's core structural claim: a satellite train is a single, slow, uniform line of points crossing the sky in one direction, and cannot descend in a staggered "cascade," resolve into a solid rectangular body with orange coloring, transform into a triangle, or execute a sudden silent acceleration burst.

Atmospheric Reentry Fragmentation

Fragmentation during the atmospheric reentry of space debris or a meteoroid can genuinely produce multiple simultaneous points of light at varying brightness, matching at least the surface description of Phase 1. It fails, however, to explain any of the sequence's later phases: a reentry event follows a single, continuous ballistic trajectory and does not reverse direction, resolve into a solid geometric body, or transform shape across multiple subsequent phases the way this account describes.

Fireworks

Fireworks were considered because it was a summer Sunday evening near a large city. The initial elevation of about 65° would have required a professional display, none was found in the area, and the long horizontal movement after phase 3 does not fit.

Single Aircraft

A single civil or military aircraft seen in perspective, coming toward the witness and changing direction, could account for the apparent changes of shape after phase 3 (rectangle, then chevron, then rectangle again). GEIPAN set it aside because the number of lights does not fit an airliner, which the witness, familiar with aircraft, would have recognized, and judged a Patrouille de France Alpha Jet improbable.

Retinal Persistence from House-Facade Lights

This hypothesis received genuine, dedicated consideration because the witness's own account supplied a real, specific triggering circumstance: he had recently been looking at illuminated lights on his own house facade before the sighting began. GEIPAN rejected it because the lights changed configuration relative to each other, because the witness followed them with his eyes (an afterimage follows the gaze instead), and because he described smooth rather than jerky motion.

Coordinated Drone Ballet

A drone "ballet" could explain the lights' relative movements and changes of direction, but at about 150 km/h the observed angular motion puts the drones only 8.5 to 17 meters from the witness, close enough to be recognized and heard; drone shows must also be declared to the prefecture, and none was found in the Toulouse area that night.

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

Sources

• GEIPAN (CNES), "SAINTE-FOY-DE-PEYROLIERES (31) 08.07.2018", Case 2018-07-50618 — official case 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), "Missions, méthodes et résultats" — investigative structure, expert committee, and cognitive-interview methodology
• American Meteor Society, "Meteor Activity Outlook for July 14–20, 2018" — Alpha Capricornid minor shower activity context
• Mairie de Sainte-Foy-de-Peyrolières, "Histoire" — commune founding, medieval history, Wars of Religion
• Mairie de Sainte-Foy-de-Peyrolières, "Présentation & chiffres clés" — present-day commune profile
• INSEE, "Dossier complet — Commune de Sainte-Foy-de-Peyrolières (31481)" — official population and demographic statistics
• Wikipedia (English), "Sainte-Foy-de-Peyrolières" — commune geography and overview
• Opex360 / Zone Militaire, "L'armée de l'Air crée une nouvelle unité dédiée à la planification et à la conduite des opérations aériennes" — CNOA/CAPCODA institutional background
• This site, case-files/montluel-cloud-distorting-object.html — comparative GEIPAN single-witness case with a clean negative radar check
• This site, case-files/montayral-aerodrome-object.html — comparative single-witness GEIPAN D-family case
• This site, case-files/fauville-evreux-radar-sighting.html — comparative GEPAN/GEIPAN positive radar case
• This site, case-files/bornel-conservatory-fireball.html — comparative two-witness GEIPAN case published alongside this one
• Wikimedia Commons, File: Sainte-Foy-de-Peyrolières 1.JPG, photo by "Paternel 1" (January 6, 2016), CC BY-SA 4.0 / GFDL — licensing and attribution
• YouTube (france tv), "GEIPAN : dans les coulisses de l'organisme qui étudie les phénomènes aérospatiaux non identifiés" — oEmbed-verified video, general GEIPAN institutional coverage
• This site, key-figures.html, entry on Claude Poher — GEPAN founding and early institutional history
• This site, key-figures.html, entry on Jean-Jacques Velasco — SEPRA leadership and methodology evolution
• This site, locations/index.html, entry on Toulouse — CNES/GEPAN/SEPRA/GEIPAN institutional headquarters, within the same department as this sighting
• GEIPAN (CNES), case page "SAINTE-FOY-DE-PEYROLIERES (31) 08.07.2018," Internet Archive copy of 15 August 2024 (classification D1, strangeness 0.55, consistency 0.72, updated 07/08/2024). web.archive.org
• GEIPAN (CNES), "Compte rendu d'enquête, Sainte-Foy-de-Peyrolières (31) 08.07.2018," Toulouse, June 7, 2024 (67 pp.: voicemails, chronology, Météo-France data, DGAC and CNOA radar, COSE/GRAVES, seven hypotheses, expert committees, annex on the Patrouille de France). Internet Archive copy. web.archive.org
• GEIPAN (CNES), "Missions, méthodes et résultats" (staffing, cognitive interview, classification statistics: 59% A/B, 34% C, about 7% D; about 2% D over ten years). Internet Archive copy. web.archive.org
• NASA JPL Horizons, observer ephemeris for the Moon from 43.49°N, 1.15°E, July 8, 2018, 21:00–23:00 UT. ssd.jpl.nasa.gov

Corrections (October 2026)

A fact-check against GEIPAN's full 67-page investigation report (June 2024, read via the Internet Archive because geipan.fr was unavailable) found that the earlier version of this page left out the most important fact about the testimony and misstated several of GEIPAN's findings. The witness's first report, a voicemail ten minutes after the event, described a 2–3-second sighting of four faint points, a triangle and a sudden disappearance, and suggested a military aircraft; the twenty-second, five-phase account came seven weeks later and was added to for years. The page had presented the prompt call as fixing the account "before any memory decay." The radar replay GEIPAN obtained covered 23:50–00:10, starting about twenty minutes after the sighting, and the CNOA plot was requested in October 2018 but never received; the page described a slightly late window and a request made too late. The witness also saw four to eight aircraft circling the area 15–20 minutes afterwards, which GEIPAN considered in its formation-flight analysis; the page omitted this.

Following this review, the rating was changed from Open 47 to Open 42 (October 27, 2026), because the higher rating had rested partly on claims corrected or removed above.

GEIPAN's reasons for rejecting several hypotheses were misstated: a single aircraft seen in perspective could explain the shape changes (it was set aside on the number of lights); the drone hypothesis was rejected because the drones would have had to be 8.5–17 m away and no declared show was found, not because no 2018 drone could do it; fireworks were rejected on elevation, the absence of any display, and the horizontal motion. Also corrected: the expert committee met in June 2022 and May 2023 and the report is dated June 2024 (published August 2024); GEIPAN's database lists the case as D1; GEIPAN's statistics now cited are its methods-page figures (about 7% D overall, 2% over the last decade) and its published-case figure (3.1% D), and the 59% figure refers to A/B classifications; Starlink trains did not exist in 2018; the Moon was below the horizon (NASA JPL Horizons) and the night was clear, not "overcast." Removed as unsupported: a 13th-century brick church (the present church dates from 1791), 47 First World War dead, and wartime resistance activity in the commune.