Within Gard UFOs
How Gard UFO Mysteries Became Weaker Files
Several Gard files became less mysterious after later checks against the Moon, sky-trackers, witness contradictions and weak data.
On this page
- The A9 following light case
- Montdardier and the setting Moon
- Roquemaure and festive sky trackers
Page outline Jump by section
Introduction
Several Gard UFO files have become less mysterious not because witnesses were dismissed, but because later checks changed what the records could responsibly support. In the clearest examples, old “unexplained” files were re-read against astronomy, road geometry, witness consistency, weather, missing evidence and the changing public familiarity with lighting technology. The result is a more modest but more useful Gard UFO history: the A9 following-light case was split into one weak file and one astronomical explanation; Montdardier was reclassified from an old unexplained file to a setting-Moon misidentification; and Roquemaure moved from mystery to a very probable festive sky-tracker case. GEIPAN’s wider method explains why this matters: a French UAP case is not fixed forever unless the evidence is strong enough, and even “unexplained” can later mean “explained”, “probably explained”, or simply “too thin to analyse”.[Geipan+3Geipan+3Geipan]cnes-geipan.frGeipan[A9] DE NIMES (30) VERS MONTPELLIER (34) 05.01.1984 T1 | GEIPAN…
Why Gard verdicts changed
The key to these Gard cases is GEIPAN’s classification system. GEIPAN, the French space agency CNES unit that collects, analyses, archives and publishes reports of unidentified aerospace phenomena, says its work includes witness collection, scientific analysis, possible remote or field investigation, and public publication of documented cases. CNES gives GEIPAN’s current national proportions as 24.6% perfectly identified, 39.7% probably identified, 32.4% not identified because of missing data, and only 3.3% unidentified after investigation.[CNES]cnes.frGEIPAN | CNESGEIPAN | CNES…
That distinction is essential for Gard. A case can become less mysterious in two different ways. It can be positively explained, as with the Moon, Aldebaran or sky-trackers. Or it can be downgraded because the file is not strong enough to sustain an unexplained verdict. GEIPAN’s 2021 classification note says cases are assessed through two linked ideas: residual strangeness after investigation and “consistency”, meaning the quantity and reliability of the information. Where consistency is not strong enough, the case can fall into category C: impossible to conclude, not proof of something extraordinary.[Geipan]cnes-geipan.frGeipan La méthodologie de classification au GEIPAN | GEIPANGeipan La méthodologie de classification au GEIPAN | GEIPAN
GEIPAN itself has said older cases are not necessarily final. In its explanation of the fall in category D cases, it notes that in 2008 it refined its criteria, split unexplained cases into D1 and D2, and aimed to make classifications more objective and transparent. It also states that D1 and D2 cases can evolve if new information, cross-checking or further investigation becomes available.[Geipan]cnes-geipan.frGeipan Forte baisse des CAS D sur les 10 dernières années | GEIPANGeipan Forte baisse des CAS D sur les 10 dernières années | GEIPAN
The A9 following-light case
The A9 case is one of the best Gard examples of a dramatic experience becoming a weaker official file. On 5 January 1984, at about 2.30 am, two witnesses travelling in a medical transport vehicle on the A9 from Nîmes towards Montpellier reported a phenomenon that seemed to follow them for about 20 kilometres. A second phenomenon was reported only by the more frightened second witness. GEIPAN records that the case had previously been classified D under the name Grand-Gallargues 1984, before later re-examination.[Geipan]cnes-geipan.frGeipan[A9] DE NIMES (30) VERS MONTPELLIER (34) 05.01.1984 T1 | GEIPAN…
The reclassification did not produce a single neat debunking. Instead, GEIPAN split the event into two parts. The first phenomenon, seen by both witnesses, retained some strangeness: GEIPAN wrote that its physical characteristics and behaviour did not correspond to a known phenomenon. But the witnesses’ descriptions and drawings diverged on almost all major details of appearance and behaviour. Because the file was too weak for further verification or cross-checking, GEIPAN decided it could no longer remain a D case and classified it as C, lack of reliable information.[Geipan]cnes-geipan.frGeipan[A9] DE NIMES (30) VERS MONTPELLIER (34) 05.01.1984 T1 | GEIPAN…
The second phenomenon changed more decisively. It was seen only by the second witness, and GEIPAN classified it as an astronomical misidentification of Aldebaran. The agency also described the “following ball” effect: a distant celestial object can seem to follow a moving car because the road and foreground are changing while the star remains fixed in the sky. That is a particularly important lesson for Gard, where several striking reports are made from vehicles on roads at night.[Geipan]cnes-geipan.frGeipan[A9] de NIMES (30) vers MONTPELLIER (34) 05.01.1984 T2 | GEIPAN…
The A9 case therefore weakened in two different ways at once. One part became less evidentially useful because the witness record did not hold together well enough. The other part became explainable once astronomy and the geometry of a moving vehicle were considered. For readers, the point is not that the witnesses invented the story. It is that a powerful night-time impression can outrun the later evidential record.
Montdardier and the setting Moon
The Montdardier case shows a different kind of changed verdict: a once-unexplained file becoming a clear misidentification. In the night of 17 to 18 December 1988, at around 2.45 am, four witnesses in two vehicles on the D113 saw a luminous object that appeared to block the road. It was described as a double rectangle separated by a dark line, apparently 7 to 8 metres above the ground, silent, and receding as the cars advanced. GEIPAN notes that the case had been classed D in 1988 by SEPRA under the earlier name Le Vigan.[Geipan]cnes-geipan.frGeipanMONTDARDIER (30) 18.12.1988 | GEIPAN…
This case is especially useful because GEIPAN did not attack the witnesses’ honesty. On the contrary, it described the case as consistent, with three witnesses and a precise description, and stated that the sincerity and credibility of the witnesses had not been doubted. The verdict changed because the described object shared several features of a known astronomical body: the setting Moon. The Moon was also present in the relevant part of the sky, although the witnesses did not mention it.[Geipan]cnes-geipan.frGeipanMONTDARDIER (30) 18.12.1988 | GEIPAN…
The reclassification turns on a subtle but common problem in UFO files: people may accurately report what an object felt like in context while still misidentifying what caused it. A low Moon seen along a road, through fatigue, fear, surprise and the demands of night driving, can appear local, structured and moving with the observer. GEIPAN’s current classification is A, a Moon misidentification.[Geipan]cnes-geipan.frGeipanMONTDARDIER (30) 18.12.1988 | GEIPAN…
Montdardier matters within Gard because it preserves both sides of the case. It was not a throwaway report. It had multiple witnesses and enough detail to be taken seriously. Yet stronger reconstruction made it less anomalous, not more. That is exactly why reclassified files are valuable: they show that a credible witness and an incorrect interpretation can coexist.
Roquemaure and festive sky-trackers
Roquemaure is the most vivid example of a Gard case weakened by changing technological context. On Christmas night 1993, a driver near the counter-canal between Roquemaure and the Rhône reported two flame-red circular forms, not dazzling, first still and then moving in different directions. A second witness, from the Île de la Barthelasse at Avignon, had earlier seen two white lights. GEIPAN records that the case had previously been classed D under Roquemaure 1993 before later re-examination.[Geipan]cnes-geipan.frGeipanROQUEMAURE (30) 25.12.1993 | GEIPAN…
The later explanation was not a single recovered machine or named operator. GEIPAN instead retained a very probable explanation: beams from two sky-trackers projecting moving spots onto low cloud during a festive period. The agency pointed to several converging details: circular shapes are typical of projector impacts on cloud, the colours could vary with filters, automated or manual systems can make such lights move independently, and stratocumulus cloud at the time provided a suitable projection surface.[Geipan]cnes-geipan.frGeipanROQUEMAURE (30) 25.12.1993 | GEIPAN…
The weaknesses are also important. GEIPAN notes that the second testimony was brief, the gendarmerie enquiry was quick, angular data were absent, and a video reportedly made by the first witness was not in GEIPAN’s archives. It also says the exact origin of the lighting could not now be found because more than 25 years had passed and the search would have had to cover a radius of several tens of kilometres. Even so, it judged the overall pattern very probable: Christmas night, near the Avignon-Orange urban area, with cloud, no noise, suitable distances and a period when such devices were already in use for discotheques and entertainment.[Geipan]cnes-geipan.frGeipanROQUEMAURE (30) 25.12.1993 | GEIPAN…
Roquemaure therefore sits between “solved” and “historically reconstructed”. It is not a recovered-device identification, but it is no longer a strong mystery. The case became weaker because investigators better understood a source of visual confusion that was less familiar to the public and investigators in the early 1990s.
What the three cases show together
These changed Gard verdicts are not identical, and treating them as identical would flatten the evidence. The A9 file shows the danger of inconsistent testimony and the “following light” illusion. Montdardier shows how a structured-looking object can be the Moon when seen from a moving car at night. Roquemaure shows how entertainment lighting can create moving, silent, coloured forms when beams strike cloud.
A useful comparison is:
CaseOriginal force of the reportLater weakening factorCurrent GEIPAN outcomeA9, Nîmes to Montpellier, 1984Two witnesses, long following effect, motorway settingDivergent testimony for the first phenomenon; Aldebaran and vehicle-motion illusion for the secondPAN 1: C, lack of reliable information; PAN 2: A, astronomicalMontdardier, 1988Several witnesses, precise road-blocking luminous formSetting Moon in the relevant sky position; interpretation shaped by night driving and fearA, Moon misidentificationRoquemaure, 1993Coloured moving forms, gendarmerie record, second witnessCloud-projected sky-tracker beams during a festive period; missing video and weak angular dataA, very probable sky-tracker beams
The shared lesson is that a changed verdict does not always mean “nothing happened”. It means the strongest later explanation, or the recognised weakness in the file, changed the official evidential status. In Gard, reclassification often moves the emphasis away from extraordinary craft and towards ordinary objects seen under confusing conditions.
Why weak files still matter
A category C case can be frustrating because it does not deliver a satisfying explanation. But in Gard’s UFO history, C cases are important precisely because they prevent overclaiming. GEIPAN’s method says consistency depends on the quantity and reliability of the information, and that stronger strangeness needs stronger consistency if a case is to remain truly unexplained.[Geipan]cnes-geipan.frGeipan La méthodologie de classification au GEIPAN | GEIPANGeipan La méthodologie de classification au GEIPAN | GEIPAN
That matters for the A9 first phenomenon. GEIPAN did not identify it, but it also did not preserve it as a strong unexplained case. The problem was not merely that investigators lacked imagination; it was that the accounts and drawings diverged too much to support a firm conclusion. A reader looking only for “unexplained” labels might miss that distinction.[Geipan]cnes-geipan.frGeipan[A9] DE NIMES (30) VERS MONTPELLIER (34) 05.01.1984 T1 | GEIPAN…
The same caution applies in the other direction. Montdardier and Roquemaure are not made worthless by their explanations. They show how a serious report can be transformed by better context. In Montdardier, the witnesses were still treated as sincere; in Roquemaure, the missing video and lost trail of the exact light source were acknowledged rather than hidden. Those limits make the changed verdicts more credible, not less.[Geipan]cnes-geipan.frGeipanMONTDARDIER (30) 18.12.1988 | GEIPAN…
A continuing pattern in Gard
The same mechanisms still appear in later Gard files, which helps explain why the older reclassifications are not just historical curiosities. GEIPAN’s Gard case list includes a 2021 road observation from Remoulins towards Montfrin, classified A as a Moon misidentification, and a 2020 Uzès case, updated in 2025, classified B as a probable novelty balloon. The official list also contains other Gard cases classed as missing reliable information, aircraft, reflections, meteor, bird, balloon and other ordinary causes.[Geipan+2Geipan]cnes-geipan.frGeipan Recherche de cas | GEIPANGeipan Recherche de cas | GEIPAN
The Remoulins-to-Montfrin case is especially close to the older road cases. A driver observed a luminous patch during a roughly 30-minute journey; GEIPAN concluded that the Moon was in the witness’s line of sight at each stage, that cloud reflection helped create the perceived light, and that the impression of oscillation and pursuit came from the relative positions of the Moon, the driver, the road and the need to concentrate while driving.[Geipan]cnes-geipan.frOpen source on cnes-geipan.fr.
The Uzès case shows the modern version of the same cautious approach. Two witnesses saw a silent, matte black, oval object moving slowly in the evening sky. The photo was considered unusable, but the account was detailed. GEIPAN used local wind modelling from Météo-France’s AROME model and concluded that the object was probably a novelty balloon moving with the general westerly wind, despite the witnesses’ impression of local ground wind from another direction.[Geipan]cnes-geipan.frOpen source on cnes-geipan.fr.
Together, these later cases show why the changed verdicts in the older Gard files are not isolated debunkings. They are part of a departmental pattern: night roads, low celestial objects, cloud, weak or missing images, local wind differences, and unfamiliar lighting or floating objects repeatedly create reports that feel stranger at the time than they look after reconstruction.
How to read Gard’s changed verdicts fairly
The fairest reading is neither “all UFO reports are nonsense” nor “official explanations erase mysteries”. Gard’s reclassified files show a more careful middle ground. Some reports became explained; some became probably explained; some became too weak to carry the weight once placed on them. That is a real historical finding.
For readers comparing Gard cases, three questions are more useful than asking whether a file is simply “solved” or “unsolved”:<div class="content-enhancement content-enhancement--step-flow" markdown="1">
- Did later investigators identify a specific ordinary cause, such as the Moon, Aldebaran, a meteor, a balloon, an aircraft or a projected light?
- If not, did the record remain strong enough to support an unexplained verdict, or did it fall into missing-data territory?
- Did the later explanation account for the witness’s strongest impressions, such as pursuit, silence, colour, apparent low altitude or sudden movement?</div>
On those tests, the reclassified Gard cases mostly become weaker files. The A9 case retains an unresolved residue for the first phenomenon, but not a strong unexplained one. Montdardier is best read as a sincere multi-witness Moon misidentification. Roquemaure is best read as a historically plausible sky-tracker case, with acknowledged gaps but enough convergence to remove most of the original mystery.
Amazon book picks
Further Reading
Books and field guides related to How Gard UFO Mysteries Became Weaker Files. Use these as the next step if you want deeper reading beyond the article.
The UFO Experience
Focuses on classification, case analysis, and the evaluation of reported sightings.
UFOs
Provides context on official investigations and how governments assess unexplained aerial reports.
UFOs and Government
Examines how official bodies investigate, archive, and reassess UFO cases over time.
The Demon-haunted World
Explains the skeptical and evidence-based reasoning used when revisiting and reclassifying UFO reports.
eBay marketplace picks
Marketplace Samples
Live-tested eBay searches with available results related to this page.
Endnotes
1.
Source: cnes.fr
Title: GEIPAN | CNES
Link:https://cnes.fr/projets/geipan
2.
Source: youtube.com
Link:https://www.youtube.com/watch?v=hn2xTieploU
3.
Source: youtube.com
Title: UFOs: GEIPAN is working on the issue (Toulouse)
Link:https://www.youtube.com/watch?v=nnOX-NXZFqE
4.
Source: cnes-geipan.fr
Link:https://cnes-geipan.fr/fr/cas/1984-01-01016
5.
Source: cnes-geipan.fr
Link:https://cnes-geipan.fr/fr/cas/1984-01-50572
6.
Source: cnes-geipan.fr
Link:https://cnes-geipan.fr/fr/cas/1988-12-01157
7.
Source: cnes-geipan.fr
Link:https://www.cnes-geipan.fr/fr/cas/1993-12-01341
8.
Source: cnes-geipan.fr
Title: Geipan La méthodologie de classification au GEIPAN | GEIPAN
Link:https://www.cnes-geipan.fr/en/actualites/methodologie-classification-geipan
9.
Source: cnes-geipan.fr
Title: Geipan Forte baisse des CAS D sur les 10 dernières années | GEIPAN
Link:https://www.cnes-geipan.fr/en/actualites/baisse-cas-d
10.
Source: cnes-geipan.fr
Title: Geipan Recherche de cas | GEIPAN
Link:https://cnes-geipan.fr/en/recherche/cas/tab?field_agregation_index_value=&field_date_d_observation_value%5Bmax%5D=&field_date_d_observation_value%5Bmin%5D=&field_date_value=&field_departement_target_id=&field_document_existe_ou_pas_value=All&field_is_new_value=All&field_is_revisited_value=All&field_latitude_value%5Bmax%5D=&field_latitude_value%5Bmin%5D=&field_longitude_value%5Bmax%5D=&field_longitude_value%5Bmin%5D=&field_phenomene_target_id=&field_type_de_cas_target_id=116&order=field_departement_textuel&page=46&select-category-export=nothing&sort=asc
11.
Source: cnes-geipan.fr
Link:https://cnes-geipan.fr/fr/cas/2021-09-51253
12.
Source: cnes-geipan.fr
Link:https://cnes-geipan.fr/fr/cas/2020-08-51053
13.
Source: cnes-geipan.fr
Link:https://www.cnes-geipan.fr/fr/recherche/cas?field_agregation_index_value=&field_date_d_observation_value%5Bmax%5D=&field_date_d_observation_value%5Bmin%5D=&field_departement_target_id=Gard&field_document_existe_ou_pas_value=All&field_latitude_value%5Bmax%5D=&field_latitude_value%5Bmin%5D=&field_longitude_value%5Bmax%5D=&field_longitude_value%5Bmin%5D=&field_phenomene_target_id=&field_type_de_cas_target_id=All&page=%2C6&select-category-export=nothing
14.
Source: cnes-geipan.fr
Link:https://www.cnes-geipan.fr/en/node/58788
15.
Source: cnes-geipan.fr
Link:https://www.cnes-geipan.fr/fr/glossaire
16.
Source: cnes-geipan.fr
Title: 2015 09 01 Spatial Point Pattern Analysis of the Unidentified
Link:https://www.cnes-geipan.fr/sites/default/files/2015-09-01_Spatial_Point_Pattern_Analysis_of_the_Unidentified.pdf
17.
Source: cnes-geipan.fr
Link:https://www.cnes-geipan.fr/en/node/412
18.
Source: cnes-geipan.fr
Title: evolution classification des cas
Link:https://www.cnes-geipan.fr/en/actualites/evolution-classification-des-cas
19.
Source: academieairespace.com
Link:https://academieairespace.com/event/geipan-studies-uaps-ufos/?lang=en
Additional References
20.
Source: arxiv.org
Link:https://arxiv.org/html/2502.06794v1
21.
Source: youtube.com
Title: Why NOBODY Talks About The Real UFO Geography
Link:https://www.youtube.com/watch?v=OZn3kge28e4
22.
Source: researchgate.net
Link:https://www.researchgate.net/publication/369507030_GEIPAN_classification_with_text_mining_and_machine_learning
23.
Source: france-science.com
Link:https://france-science.com/en/caipan-ii-international-conference-on-unidentified-aerospace-phenomena-organized-by-geipan-in-toulouse/
24.
Source: utl-rouergue.fr
Link:https://www.utl-rouergue.fr/conference/phenomenes-aerospatiaux-non-identifies/
25.
Source: youtube.com
Title: Mick West: UFOs DEBUNKED! Brian Keating’s INTO THE IMPOSSIBLE Podcast (#154)
Link:https://www.youtube.com/watch?v=HJtTgIKDUZw
26.
Source: Wikipedia
Title: Groupe d’études et d’informations sur les phénomènes aérospatiaux non identifiés
Link:https://fr.wikipedia.org/wiki/Groupe_d%27%C3%A9tudes_et_d%27informations_sur_les_ph%C3%A9nom%C3%A8nes_a%C3%A9rospatiaux_non_identifi%C3%A9s
27.
Source: youtube.com
Link:https://www.youtube.com/watch?v=q4BoOzOiAvk
28.
Source: science.nasa.gov
Title: uap independent study team final report
Link:https://science.nasa.gov/wp-content/uploads/2023/09/uap-independent-study-team-final-report.pdf
29.
Source: Wikipedia
Title: Unidentified flying object
Link:https://en.wikipedia.org/wiki/Unidentified_flying_object
Topic Tree