0X
France — GEIPAN and COMETA
France is the only country whose government runs a permanent, publicly searchable UFO case archive — and the only one where a group of retired generals published a pro-extraterrestrial report that the world mistook for official policy.
The institution: GEPAN → SEPRA → GEIPAN
- DOCUMENTED France has had a state UFO office continuously since 1977. GEPAN (Groupe d'Études des Phénomènes Aérospatiaux Non identifiés) was created inside CNES — the French national space agency — under CNES director general Yves Sillard, with physicist Claude Poher as its first head. It sits in Toulouse, inside a civilian space agency, not a defence ministry. That placement is the whole reason the archive is public.
- DOCUMENTED Renamed twice, never abolished. In 1988 it became SEPRA (Service d'Expertise des Phénomènes de Rentrées Atmosphériques) under Jean-Jacques Velasco, with a remit broadened toward atmospheric re-entry debris. In 2005 it was reconstituted as GEIPAN, the added "I" standing for Informations — the public-disclosure mission was written into the name.
- DOCUMENTED It is small. GEIPAN is a handful of staff plus outside experts and volunteer investigators, overseen by a steering committee (comité de pilotage) chaired by Yves Sillard and drawing members from French civil and military aviation authorities, the gendarmerie, meteorological services and scientific institutions. Frédéric Courtade has directed it since January 2024.
- DOCUMENTED The gendarmerie is the intake pipe. Reports made to the National Gendarmerie are forwarded to GEIPAN, giving it something no private research group has: a national police force taking first-response statements, on standard forms, at the site.
| Period | Name | Parent | Notable head |
|---|---|---|---|
| 1977–1988 | GEPAN | CNES | Claude Poher (founding) |
| 1988–2005 | SEPRA | CNES | Jean-Jacques Velasco |
| 2005–present | GEIPAN | CNES | Frédéric Courtade (since Jan 2024) |
← swipe table →
GEIPAN / CNES, geipan.fr — "Le GEIPAN / Historique". https://www.geipan.fr/en
The public archive — what it actually is
- DOCUMENTED GEIPAN put its case files online in March 2007. Not summaries — the underlying dossiers: witness questionnaires, gendarmerie reports, investigator notes, photographs, analysis annexes, as scanned PDFs with witness identities redacted. Traffic on launch was heavy enough to take the site down.
- DOCUMENTED It is searchable and it is still being added to. The Recherche de cas interface filters by date, department, phenomenon type and classification. As of 25 June 2026 the published database held 3,368 cases. Cases are added and — importantly — reclassified as new analysis lands, so the totals move.
- DOCUMENTED No other government does this. The US AARO publishes reports about cases; the UK released historical MoD files to the National Archives as a closed set after the desk shut in 2009. GEIPAN publishes a living, per-case, classified, machine-searchable archive of an ongoing programme. It is a genuine outlier and the site's single most useful primary source on this topic.
- DOCUMENTED Volume is modest. GEIPAN has described receiving on the order of several hundred reports per year and opening formal investigations on a minority of them — the archive is a curated investigative output, not a raw sightings dump.
GEIPAN case search and statistics pages, CNES. https://www.geipan.fr/en/recherche/cas and https://www.geipan.fr/en/stats (figures read 2026). Raw dossiers are also downloadable in bulk as CSV database files; comparison programmes: UK National Archives MoD UFO files, US AARO.
The A/B/C/D classification system
- DOCUMENTED Every published case gets exactly one of four letters. The classification encodes two different things at once: whether the phenomenon was identified, and — when it wasn't — why not. That second distinction is what makes the French system worth copying.
- DOCUMENTED C is the honest category, and it is the one everyone drops. Class C means "we could not identify this because there wasn't enough usable data" — a statement about the report, not about the sky. Class D means "we had enough data to work with and still could not explain it." Conflating C and D is the most common way French statistics get misquoted.
| Class | GEIPAN definition | What it means in plain terms |
|---|---|---|
| A | Perfectly identified phenomenon | Cause established with certainty (Chinese lantern, ISS, Starlink train, aircraft, balloon) |
| B | Probably identified phenomenon | A specific cause is strongly indicated but not nailed down |
| C | Unidentified — insufficient data | Report too thin to analyse. Says nothing about strangeness |
| D | Unidentified after investigation | Data adequate, investigation done, no explanation found |
← swipe table →
GEIPAN, "Classification des cas" / case-search filters, CNES. https://www.geipan.fr/en/stats See also La méthodologie de classification au GEIPAN (strangeness/consistency scoring, 2021).
The numbers — and the number everyone gets wrong
- DOCUMENTED 3.1% of published GEIPAN cases are class D. Against a total of 3,368 cases as of 25 June 2026: A = 942 (28.0%), B = 1,306 (38.8%), C = 1,014 (30.1%), D = 106 (3.1%). Roughly two-thirds of cases get an actual identification; nearly a third fail for want of data; about one in thirty survives investigation unexplained.
- DISPUTED Is the residual "28%" or "3%"? A widely-cited figure attributed to CNES around 2007 put unidentified French cases near 28%. The current published breakdown puts class D at 3.1%. Both numbers are real; they count different things. The 28% figure tracks cases lacking an identification (dominated by data-poor class C); the 3.1% figure is the post-investigation residual. Anyone quoting "28% of French sightings are unexplained" without saying which is being measured is — knowingly or not — inflating the result by roughly tenfold.
- DOCUMENTED The statistics are dynamic. GEIPAN notes that since 2016 it computes statistics across all published classified cases combined and updates them as cases are reclassified. A percentage quoted from a 2010 press article is not the same population as today's.
| Class | Cases | Share |
|---|---|---|
| A — perfectly identified | 942 | 28.0% |
| B — probably identified | 1,306 | 38.8% |
| C — unidentified, insufficient data | 1,014 | 30.1% |
| D — unidentified after investigation | 106 | 3.1% |
| Total | 3,368 | 100% |
← swipe table →
GEIPAN statistics page, CNES, figures dated 25 June 2026. https://www.geipan.fr/en/stats Longer-run programme figures in GEIPAN, "Missions, méthodes et résultats".
Common misconception
"Class D" does not mean "alien." It is a bookkeeping label meaning this file was investigated with adequate data and no explanation was found. GEIPAN itself draws no origin conclusion from a D classification, and 106 unexplained files out of 3,368 is a statement about the limits of after-the-fact investigation, not evidence of non-human craft. Unidentified is unidentified.
Where the French model is criticised
- DISPUTED French skeptics say investigation quality is uneven. Zetetic and skeptical researchers in France — notably the Observatoire Zététique circle and investigators such as Éric Maillot — have argued over years that some GEPAN/GEIPAN dossiers accepted witness reconstructions too readily and under-pursued mundane hypotheses, and that a case can reach class D through investigative shortfall rather than genuine anomaly. GEIPAN's position is that classification is a documented, reviewable process and that reclassification of older cases is routine precisely because it revisits them.
- DOCUMENTED The archive is self-auditing in a way most are not. Because the underlying dossiers are published, the criticism above is testable by any reader — you can pull the PDF and judge the reasoning. That is a structural advantage over programmes that publish only conclusions.
- UNSUPPORTED Claims that GEIPAN suppresses or withholds "the good cases." This circulates in French and Anglophone UFO forums, sometimes attached to SEPRA's 2004 closure. No document, whistleblower deposition or withheld-file inventory has been produced to support it. The programme's budget and staffing have always been small enough that "little capacity" explains most gaps without invoking concealment.
Observatoire Zététique publications — in particular Rossoni, Maillot & Déguillaume, Les OVNI du CNES : trente ans d'études officielles (1977–2007), reviewed in AFIS, Science et pseudo-sciences n°280; GEIPAN methodology pages, CNES, and "Le GEIPAN : groupe d'études ou de recherches ?".
The COMETA report (1999) — who actually wrote it
- DOCUMENTED Title and date. Les OVNI et la Défense : à quoi doit-on se préparer ? ("UFOs and Defence: What Should We Prepare For?"), published July 1999 in France and released to the public via a special issue of the magazine VSD.
- DOCUMENTED COMETA was a private association, not a government body. Comité d'Études Approfondies registered as a French non-profit association on 24 February 1999. Its members were mostly retired — generals, engineers, scientists and officials, many of them alumni or former staff of the Institut des Hautes Études de Défense Nationale (IHEDN), the French national defence college. Association with IHEDN alumni is the origin of the persistent claim that COMETA was an IHEDN or Defence Ministry product. It was not.
- DOCUMENTED The signatories were genuinely senior. Chaired by Général Denis Letty (French Air Force, ret.). Prefaces by André Lebeau, former president of CNES, and Général Bernard Norlain, former director of IHEDN. Contributors included Jean-Jacques Velasco (then head of SEPRA), Général Bruno Domange (Air Force) and Edmond Campagnac, former technical director at Air France. Their seniority is a fact; it is not evidence for their conclusions.
- TESTIMONY Delivery to the government is attested, endorsement is not. Copies were sent to President Jacques Chirac and Prime Minister Lionel Jospin. Jospin's chief of staff is reported to have said the Prime Minister read and annotated the report "to mark his interest." No French government has ever adopted, endorsed or responded substantively to COMETA's findings.
COMETA, Les OVNI et la Défense : à quoi doit-on se préparer ?, Association COMETA, July 1999; VSD hors-série. Text later made available via GEIPAN's documentation — see GEIPAN's own notice disclaiming CNES endorsement. IHEDN.
What COMETA concluded — stated precisely
- TESTIMONY The authors asserted the phenomenon is physically real. COMETA wrote that the "almost certain physical reality" of unidentified objects displaying flight characteristics beyond known aeronautics was established by the accumulated case record. This is the report's assertion, not an independently verified finding.
- TESTIMONY They endorsed the extraterrestrial hypothesis as most probable — explicitly without proof. COMETA stated the extraterrestrial hypothesis appeared to be the most probable or credible explanation available, while acknowledging it was not scientifically proven. The report's own hedge is routinely stripped off when it is cited. Read the framing carefully: this is a group of retired officials reasoning by elimination from a case corpus, not a body presenting new physical evidence.
- TESTIMONY The recommendations were defence-and-policy, not scientific. COMETA argued France should strengthen detection, brief pilots and air-defence staff, coordinate at European level, and prepare institutionally for the consequences of the hypothesis being correct — including its assessment of US secrecy, which the report treated as a policy problem for France.
- DISPUTED Whether COMETA's method supports its conclusion. Supporters hold that a body of technically competent officers examining radar-visual and pilot cases converged rationally on the ETH. Critics — including French skeptical researchers and a number of scientists — argue COMETA relied on a pre-selected set of already-famous cases, accepted several whose prosaic explanations were live at the time, and reached the ETH by elimination without ever establishing that the elimination was exhaustive. Both readings are about the same document, which anyone can read.
COMETA report, 1999, chapters 3–4 and conclusions. (Internet Archive mirror.)
Common misconception
"The French government officially concluded UFOs are probably alien." It did not. COMETA was a privately constituted association of retired officials, publishing at its own initiative and distributing through a consumer magazine. Its members' former ranks and IHEDN connections gave it the appearance of officialdom — an appearance that has propagated for a quarter-century. The actual French state position on UAP is GEIPAN's, and GEIPAN draws no origin conclusions at all.
Trans-en-Provence, 8 January 1981 — the physical-trace case
- TESTIMONY What the witness reported. Renato Nicolaï, a retired mason in his mid-fifties, told gendarmes he saw a dull lead-grey object roughly 2.5 m across descend onto a terrace on his property at Trans-en-Provence (Var), rest briefly, then lift and depart. The observation lasted on the order of 30–40 seconds. The fact of the report is documented; its accuracy is a witness account and nothing more.
- DOCUMENTED Physical sampling happened fast, by police, under procedure. Gendarmes from the Draguignan brigade examined the terrace the following day, photographed a roughly circular ground mark and took soil and vegetation samples — including control samples at increasing distance from the trace. Rapid, procedural, chain-of-custody sampling by a police force is what distinguishes this case from nearly every other landing-trace claim.
- DOCUMENTED The report exists and is named. GEPAN published the analysis as Note Technique n°16 (1983), covering case 81/01 — the most detailed physical-trace study any government UFO programme has published. It reported: mechanical compaction of the soil consistent with a load in the region of 4–5 tonnes; evidence of heating estimated below roughly 600 °C; trace phosphate, zinc and iron oxide deposits.
- DOCUMENTED The biochemistry is the part that is hard to wave away. Michel Bounias of INRA (the French national agronomic research institute) analysed wild alfalfa from and around the trace and reported chlorophyll and pigment levels reduced by roughly 30–50% in samples nearest the mark, with the effect graded by distance from it. Bounias proposed exposure to a pulsed non-ionising energy source as a candidate mechanism. The measurements are published; the mechanism is a hypothesis.
- DOCUMENTED GEIPAN still classifies it D. Trans-en-Provence remains in the French archive as an unexplained case after investigation. That is a classification, not a conclusion about origin.
GEPAN/CNES, Note Technique n°16, 1983 (case 81/01, Trans-en-Provence); GEIPAN case file, geipan.fr.
Trans-en-Provence — the case against
- DISPUTED The trace may be vehicle damage. French skeptical investigators, principally Éric Maillot with Jacques Scornaux, have argued the ground mark is consistent with tyre slippage or scraping from a heavy vehicle on a site that had construction and vehicle traffic — and have pointed out that the original gendarmerie material itself noted a resemblance to tyre marks. On this reading the "4–5 tonne" compaction figure is unremarkable: that is roughly a loaded vehicle.
- DISPUTED Witness reliability. Skeptics cite later ambiguous or shifting statements by Nicolaï and argue the investigation treated a single uncorroborated account as the fixed point against which physical findings were interpreted, rather than the reverse. Defenders — including Velasco and French author Gildas Bourdais — argue Nicolaï had no motive, sought no publicity, and that the graded biochemical gradient cannot be produced by a tyre.
- DISPUTED The chlorophyll result itself. Critics note the samples were analysed well after collection, that the "control" plants were not matched for age, sun exposure or micro-environment, and that pigment depletion in wild alfalfa has many prosaic causes — drought stress, herbicide drift, mechanical crushing. Bounias's supporters counter that the distance-graded pattern is the finding, and that ordinary stressors would not produce a gradient centred on the mark. No independent replication was possible: the site was a one-off.
- UNSUPPORTED The "microwave weapon" and "electromagnetic propulsion" readings. Both circulate as if Note Technique n°16 established them. It did not — it recorded measurements and offered a candidate energy mechanism. Nothing in the file identifies a source, a device, or a propulsion principle.
É. Maillot / Observatoire Zététique analyses of case 81/01; G. Bourdais and J.-J. Velasco, opposing published treatments. Primary measurements in GEPAN Note Technique n°16.
Stakes and interests — proponents and skeptics both
- DOCUMENTED Jean-Jacques Velasco. Ran SEPRA, contributed to COMETA while still heading the state programme, and after leaving CNES published OVNIS, l'évidence (2004) arguing for the reality of extraterrestrial visitation. A commercial book advocating the conclusion he had reached in an official capacity is a stake, and should be weighed when his account of Trans-en-Provence is cited.
- DOCUMENTED COMETA's distribution channel. The report reached the French public through a paid special issue of VSD, a mass-market magazine. That is not disqualifying, but it is a commercial publication route, not a government gazette, and it shaped the report's framing and reach.
- DOCUMENTED The skeptical side has institutional stakes too. The Observatoire Zététique and allied French skeptical organisations exist to produce prosaic explanations and derive standing from doing so successfully. Their Trans-en-Provence critique should be read the same way as the proponents' case: on the evidence, not the affiliation. In this instance the tyre-track hypothesis has never been demonstrated to reproduce the pigment gradient, and skeptical accounts generally set that finding aside rather than explaining it.
- DOCUMENTED GEIPAN itself has an institutional interest in credibility. Its directors have been explicit that the unit must maintain scientific standing inside CNES. That incentive cuts both ways — it disciplines the analysis, and it creates pressure against classifying anything D without cause. 106 D cases in 49 years is a conservative institution's output.
Velasco, OVNIS, l'évidence, Carnot, 2004; COMETA/VSD 1999; GEIPAN public statements and "Le GEIPAN : groupe d'études ou de recherches ?".
Why this section matters more than its case count
The French contribution to this field is not Trans-en-Provence. It is the demonstration that a government can run a UFO investigation programme that publishes its raw files, distinguishes "we lacked data" from "we investigated and failed," keeps its residual honest at about 3%, and reclassifies cases when it turns out to be wrong. Every argument about secrecy elsewhere has to contend with the fact that France simply put the files on a website in 2007 and the sky did not fall.
Documents & sources — go read them yourself
- Note Technique n°16 — Enquête 81/01 : analyse d'une trace
- COMETA — Les OVNI et la Défense : à quoi doit-on se préparer ?
- À propos de la publication du rapport COMETA
- The COMETA Report (1999)
- GEIPAN — Recherche de cas (case search)
- GEIPAN — Statistiques
- GEIPAN — Classification
- La méthodologie de classification au GEIPAN
- GEIPAN — Methodology
- GEIPAN — Son histoire
- GEIPAN — Mission
- GEIPAN — Missions, méthodes et résultats
- GEIPAN — Frequently Asked Questions
- Le GEIPAN : groupe d'études ou de recherches ?
- Mise à jour des fichiers de base de données téléchargeables (CSV)
- GEIPAN — Documentation
- GEIPAN homepage
- CNES — Centre national d'études spatiales
- Les OVNI du CNES : trente ans d'études officielles (1977–2007)
- Chapter 13 — « Trans-en-Provence » (1981)
- Observatoire Zététique — Ovnis et extraterrestres archive
- Trente ans d'études du CNES, OVNI et extraterrestres
- Institut des hautes études de défense nationale (IHEDN)
- UFO reports — The National Archives
- UFO Desk: Closed — Last Tranche of UFO Files Released
- Unidentified Flying Objects (UFOs) — briefing document
- 2024 Consolidated Annual Report on Unidentified Anomalous Phenomena
- FY24 Consolidated Annual Report on UAP (PDF)
- All-domain Anomaly Resolution Office (AARO)
- Michel C. L. Bounias, "Biochemical Traumatology as a Potent Tool for Identifying Actual Stresses Elicited by Unidentified Sources: Evidence for Plant Metabolic Disorders in Correlation With a UFO Landing," Journal of Scientific Exploration vol. 4 no. 1, 1990
- GEIPAN case file for Trans-en-Provence (83), 08.01.1981, class D
- Jean-Jacques Velasco & Nicolas Montigiani, OVNIS, l'évidence, Éditions Carnot, 2004
- COMETA, VSD hors-série, July 1999
- Gildas Bourdais, published defences of the Trans-en-Provence investigation
Links verified August 2026. Unlinked items in this section are ones for which no stable public URL could be confirmed.
Plates — the documentary record
Images are public domain or freely licensed; each carries its credit and licence, and links to its source record. Served from this site rather than hotlinked.