فهرس الحالات

Abstract

This study compares the 123 case dossiers in UAPRAD’s English-language collection, using the public versions captured on 7 November 2026. It examines which reported similarities survive scrutiny of source provenance, observation conditions and the association of evidence with particular events. A structured documentary comparison of every dossier is supplemented by targeted examination of original reports, reproduced contemporary records and scholarly research on testimony, measurement and reporting. The analysis identifies a recurring difficulty in connecting observations across witnesses, instruments, times and locations. Apparent agreement can acquire greater precision in later summaries than the underlying records support. Nevertheless, the collection contains substantial positive evidence: contemporary reporting, recorded images, operational responses, physical alterations and, in some cases, measurements that constrain proposed explanations. Comparisons of radar and visual encounters show that target association requires episode-specific evaluation. Photographic and infrared cases demonstrate how geometry can restrict performance claims without resolving source identity. Physical-trace cases reveal a corresponding distinction between characterising a specimen or alteration and establishing its causal connection to an aerial observation. Witness comparisons identify selective preservation, opportunities for communication and retrospective development, while retaining differences among individual accounts. The corpus supports several bounded explanations and unresolved propositions, rather than a single physical classification. Its principal research contribution is a differentiated account of where uncertainty enters an investigation and which surviving records could reduce it. Because the collection is historically selected, these findings concern its evidential structure and cannot estimate the prevalence or origins of UAP generally.

Keywords: unidentified anomalous phenomena; comparative case analysis; testimony; sensor interpretation; provenance; causal inference.

1. The comparative problem

UAP cases are frequently compared through their most memorable features: unusual shapes, abrupt movement, close approaches, multiple witnesses or apparently associated physical effects. Such comparisons can identify subjects for investigation. Their explanatory value depends on whether the shared feature was independently observed, measured or subsequently inferred. Two accounts describing a large object may share an interpretation of uncertain distance. Two radar and visual cases may involve very different relationships between the reported light and the recorded return. A laboratory result can be precise even when the origin of the tested material is poorly documented. Comparing these cases requires attention to the route from observation to claim.

UAPRAD’s collection provides a substantial basis for that inquiry. It includes aerial sightings, photographs and films, radar encounters, close encounters, material investigations, regional waves and recent institutional proceedings. Its breadth permits comparison between records that are usually discussed within separate traditions. The school reports at Ariel and Broad Haven raise questions about the independence of testimony; the Navy encounters and earlier radar cases raise questions about the independence of instrument detections. The questions are related, because both concern whether apparent corroboration adds information about the same event. They nevertheless require different tests: interview history in one setting, clock alignment and tracking geometry in the other.

The central question is therefore: which similarities across this collection remain evidentially significant after observation conditions, source dependence and the history of investigation are taken into account? Four subsidiary questions guide the analysis. What physical properties do the records actually constrain? Which agreements arise from separate observations, and which from communication or common source material? How well do competing explanations address the same episode? What consequential uncertainty remains after a particular claim has been explained or weakened?

The resulting argument is that association is a recurrent point of vulnerability. Investigators must connect a radar return with a light, a mark with a reported landing, an illness with an exposure, or a later narrative with an earlier event. A connection may be plausible and still require evidence. Conversely, a failure to establish one connection does not erase the observations on either side of it. The analytical task is to identify the surviving proposition and the evidence needed to test it.

This approach also changes the treatment of unresolved cases. An unidentified light, an unexplained acceleration and an uncertain historical allegation are different research problems. The first concerns classification; the second requires dependable motion measurements; the third may depend primarily on document recovery. Combining them under a common residual category conceals the practical work each demands. GEIPAN’s distinction between insufficient-information cases and cases remaining unidentified after investigation offers a useful precedent for preserving such differences, although this study does not retrospectively assign its classifications to UAPRAD’s collection. GEIPAN classification guidance.

2. Corpus and method

2.1 Coverage and documentary basis

The corpus comprises all 123 entries in UAPRAD’s case catalogue at the study cut-off. Every published English page was captured, with retrieval time, page address and a digital fingerprint retained in the accompanying corpus file. Reconciliation with local records identified 111 pages carrying stored editorial revisions. The captured public versions therefore define the texts examined in this study, preserving the published assessments available at the cut-off.

Every dossier’s assessment was screened, and the associated narrative and source sections were consulted where needed to establish a comparison. Selected central claims received additional examination against original publications or identified reproductions. These checks included the official GOFAST and Aguadilla analyses, Fravor’s congressional statement, the Colorado project’s Ubatuba investigation, GEPAN’s Trans-en-Provence report and Haines’s analysis of Minot radar photographs. Scholarly supplements were selected for a specific analytical function, including witness communication, interviewing, hypnosis, observational selection and instrument design. The source-access record distinguishes these checks from claims retained on the basis of UAPRAD’s existing dossier synthesis.

This is a comparative documentary study, rather than a new investigation of every incident or a systematic review of the entire UAP literature. It neither authenticates every underlying document nor reprocesses every original recording. That boundary affects how its findings should be used: a case comparison can expose a missing connection or a misapplied result without settling the historical event. The accompanying register provides complete catalogue coverage and identifies the principal question contributed by each entry.

2.2 Units of comparison

The catalogue entry is the unit of coverage; the observation episode and individual claim are the units of analysis. These distinctions prevent several forms of double counting. Phoenix includes earlier moving formations and later stationary lights. The Roosevelt entry combines GOFAST, GIMBAL and broader squadron reporting. Nimitz includes a visual interception, accounts of radar activity and a later infrared recording. A hearing can repeat testimony about an event already represented elsewhere in the catalogue. A regional wave may contain different sources on different nights.

The comparison follows a common set of questions: when and by whom was the information recorded; what was directly observed; what measurement or transformation produced the asserted property; what connects the evidence to the event; and what competing explanation has been tested against those particulars? Source independence is considered separately from observer competence. Several competent observers may discuss a shared view before recording it. Conversely, two imperfect observations from different positions can provide useful independent constraints if their times and directions are recoverable.

The method is qualitative and structured. It does not calculate an aggregate credibility score or treat every evidential deficit as interchangeable. A missing radar range scale affects inferred distance; a missing interview recording affects the ability to distinguish spontaneous description from prompted detail. Their consequences differ. The register consequently records a case-specific comparative proposition rather than assigning points for witness numbers, professional status or the existence of an official file.

Selection for sustained discussion favours contrasts that can discriminate between interpretations. School cases are compared with one another; material cases with documented testing are compared with poorly located specimens; modern infrared recordings are considered alongside earlier disputes over apparent movement. Cases with evidence supporting ordinary explanations remain in the analysis. They are necessary comparisons for assessing whether a proposed criterion—multiple witnesses, military attention or continuous photographic grain—actually distinguishes extraordinary physical events. Purposive case comparison can clarify mechanisms and challenge generalisations, but it does not make a selected historical collection representative. Seawright and Gerring, 2008.

2.3 Interpretation and analytical safeguards

Three distinctions govern the synthesis. First, a report of a property is separated from a measurement of that property. Second, compatibility with an explanation is separated from a dated identification of the source. Third, a successful explanation of one component is applied only to that component unless the evidence connects it to others. These distinctions prevent a plausible balloon mechanism from becoming an identified flight, and prevent an unresolved visual episode from invalidating a well-supported explanation of a separate recording.

No percentage of genuine, explained or anomalous cases is calculated. Such a statistic would require stable event boundaries, comparable investigative coverage and an agreed outcome measure. The present collection contains individual events, extended waves and proceedings, selected partly for historical prominence and surviving documentation. Its geographical and chronological distribution therefore cannot establish where or when unusual physical phenomena occur most often.

The review also distinguishes absent evidence from negative results. An unavailable sample analysis contributes little to a chemical claim. An inspected analysis reporting ordinary composition directly bears on it. An unsuccessful search has a defined location, duration and detection capability; it cannot be treated as universal proof of absence. Evidence and prose were reviewed separately by the drafting agent. This manuscript has not undergone independent scholarly peer review.

3. Reported similarity and physical similarity

3.1 Shape, scale and apparent movement

Shape categories provide a convenient descriptive index, but weak support for common origin when the viewing conditions differ. A resolved silhouette, a collection of lights and a blurred infrared image are different observations even when each is described as triangular or disc-like. The comparison must retain how the shape was produced. Otherwise, an observer’s inferred outline can be placed beside a photographed boundary as though both were equivalent measurements.

The triangle reports illustrate this difficulty. In the Belgian wave, the admitted fabrication of the Petit-Rechain photograph removes that image from the support available for a physical triangular object. It does not explain the separate visual reports. In the Illinois case, dispatch records help reconstruct a sequence of observations, but discrepancies between a witness’s remembered clock time and the recorded transmission affect calculations of movement between locations. Hudson Valley includes documented aircraft-formation activity; the proportion of the wider wave attributable to such activity depends on matching particular reports with flights. These cases share a visual description more securely than a demonstrated object class. Belgian wave, Illinois, Hudson Valley.

The same problem appears in unusually large size estimates. Arnold’s speed estimate depended on placing the observed formation along a distant geographical path. At Alderney, the duration and partial agreement from another aircraft give the reporting substantial value, but the radar material does not supply a secure range for the visual object. At Montreal, large dimensions depend on remembered angular extent and assumed altitude, while measurements from photographs introduce a different apparent scale. Agreement that a source looked large does not establish agreement about its physical dimensions. Arnold, Alderney, Montreal.

A simple geometrical relationship explains why this matters across otherwise dissimilar cases. For a fixed angular size, physical size increases in proportion to assigned range. Transverse speed inferred from angular movement has the same dependence. Halving the assumed range halves both estimates under the same geometry. The distant and nearby interpretations therefore cannot be compared by retaining the large size from one and the movement model from the other. Photographs at Lake Cote and Calvine present this problem particularly clearly: the image records a form, while its dimensions depend on where that form is placed in the scene. Lake Cote, Calvine.

3.2 Waves and common stimuli

Regional waves require a different comparison. Their geographical spread may reflect a large or distant source, several ordinary sources, shared attention, or some combination. The Yukon reports provide a specific re-entry candidate whose timing and path can be tested against individual witness groups. Petrozavodsk includes an expanding luminous phase compatible with a launch plume. These are stronger comparisons than a general assertion that unusual lights are probably space debris: they identify a mechanism with a dated astronomical or aerospace context. Their remaining disagreements concern particular times, bearings and the association of additional reports. Yukon, Petrozavodsk.

The Canary Islands cases show why even a promising explanatory family needs separate applications. Expanding launch-plume reconstructions address features of the 1976 and 1979 events, but an exact launch attribution requires compatible operational timing and geometry. The May 1980 report presents a distinct radar–visual association problem. Its occurrence in the same region does not identify its source. Within the 1976 case, the reported occupants also require witness-by-witness treatment because the doctor's account cannot be assigned to everyone in the vehicle. Canary Islands, 1976, 1979, 1980.

Phoenix and Cosford further demonstrate the importance of retaining episode boundaries. The later Phoenix stationary-light recordings have a developed flare explanation; earlier moving-formation reports include both aircraft observations and accounts of a continuous body. At Cosford, a re-entry explains the principal lights and trails, while disputed timings in the Shawbury account prevent the entire sequence from being reduced to one securely reconstructed observation. A wave can contain a well-explained component and a residual question without either determining the status of the other. Phoenix, Cosford and Shawbury.

These comparisons establish the limited evidential value of recurrence by description alone. More informative recurrence concerns a mechanism: parallax affecting apparent speed, formation lights producing an inferred outline, or a distant plume occupying a large angular region. Each mechanism generates testable predictions. The relevant question becomes whether those predictions fit the dated observations, including the features that initially made the report unusual.

4. Witness agreement, independence and retrospective detail

4.1 Collective observations

Multiple witnesses can strengthen an account by constraining different aspects of an event. Their evidential contribution is greatest when the records preserve separate positions, acquisition times and descriptions before discussion. A group total alone cannot establish that each person saw the same feature. Some may have observed a light, others an apparent structure, and others only the disturbance among the witnesses.

Ariel is instructive because the early investigator's own publications preserve qualifications often lost in later shorthand. Cynthia Hind reported interviewing approximately ten or twelve older children and selecting 22 of the clearer drawings from a larger group of roughly 30–40. Those materials cannot be treated as 62 separately documented, uniform accounts. Her reports also preserve disagreement about landing or hovering and uncertainty in at least one child's identification of a dark feature as a figure. Selection affects what can be inferred from the surviving visual agreement. These details come from an accessible reproduction of Hind's 1994–95 publications; the present study did not authenticate the original drawings. Hind, reproduced reports.

Broad Haven supplies a useful comparison because its early published witness excerpts differ over a figure, duration and the episode observed. The adult observation later that month is a separate event. Claims that the children's accounts were immediately isolated also encounter a disagreement between an investigator's published account and a later description of the school diary. Without direct inspection of that diary, the discrepancy remains a source question. Treating both publications as independent proof of an identical collection procedure would conceal it. Broad Haven dossier.

Westall introduces a different strength: an early account from a named adult observer. It also contains a comparison between reports of disturbed vegetation and a later unsuccessful inspection, for which location and timing matter. A proposed HIBAL explanation requires a flight path compatible with the event; proximity to a balloon programme supplies a candidate rather than a complete identification. Taken together, the three school cases support attention to early reporting and individual descriptions. They do not support a general rule that school groups are either uniquely reliable or uniformly susceptible to one explanation. Westall dossier.

Research on memory gives these comparisons a firmer methodological basis. Gabbert, Memon and Allan experimentally showed that discussion between co-witnesses who had seen different versions of an event could introduce details unavailable in a participant's own view. The study identifies a mechanism through which agreement can develop after observation. It does not establish that this mechanism produced the Ariel or Broad Haven accounts, nor supply a numerical error rate transferable to them. Gabbert, Memon and Allan, 2003.

Interview research likewise supports preserving the question as well as the answer. Lamb and colleagues' review of the National Institute of Child Health and Human Development (NICHD) interview protocol found substantial advantages in using open invitations before focused or suggestive prompts. Its evidence comes principally from forensic interviewing, with a different subject matter and investigative context. The applicable lesson is procedural: an archive that retains complete interviews permits examination of how a detail emerged. A collection of selected statements cannot provide the same test, regardless of whether the selected statements are impressive or sceptical. Lamb et al., 2007.

4.2 Agreement about presence and disagreement about detail

Adult and professional witnesses present related difficulties. At Father Gill's mission, the small set of named observers describing figures must be distinguished from the larger group endorsing a sighting. Drawings made separately and then compared preserve a different degree of independence from drawings never discussed. Waving followed by apparent movement may reasonably have been experienced as an interaction, but the inference of intention requires more than temporal succession. Father Gill dossier.

The Coyne helicopter case shows why an encounter should not be summarised through a composite observer. The crew's views differed, including a limited view by Jezzi and differences over green illumination. Coyne's report of a climb with the collective-pitch control lowered raises a concrete flight-dynamics question involving attitude, airspeed and control inputs. Those variables are not continuously recorded in the surviving account. Four people aboard an aircraft therefore cannot be counted as four equivalent measurements of an external lifting force. Coyne dossier.

Disagreement can also preserve useful information. A witness who reports only a light, or declines to identify a shape, limits the interpretation available from that position. At Kofu, adult observations of lights do not independently establish the children's account of figures. In the Moreland case, contemporary police and air-force involvement strengthens the documentation of an individual report while leaving its physical interpretation dependent on that observer's account. Comparative analysis should retain these differences instead of using the most detailed statement to complete the others. Kofu, Moreland.

4.3 Later narratives and recovered memory

Encounter narratives that develop over years require a chronology of claims. The Hill case contains material before the 1964 hypnosis sessions, including Betty Hill's dreams and elements of Barney Hill's earlier account. It would be inaccurate to assign the entire narrative to hypnosis. Equally, details emerging through later treatment cannot be counted as independent confirmation of the earlier sighting. The chronology determines what each record can corroborate. Hill dossier.

Allagash provides an especially consequential internal disagreement: Rak retained a light-sighting account while rejecting the abduction interpretation. Later narratives and differing descriptions of the campfire cannot retrospectively supply a measured interval of missing time. At Pascagoula, the secretly recorded conversation is evidence of the men's behaviour and expressed fear in that setting. Its value must be considered alongside later statements about prior agreement and withheld detail, rather than elevated into a test that excludes deception under every circumstance. Allagash, Pascagoula.

Lynn and colleagues' review explains why vividness and confidence after hypnosis cannot serve as independent accuracy checks. Hypnosis can increase accurate recall alongside errors and unwarranted confidence; it is not a dependable method of validating a disputed historical memory. That finding bears on the evidential use of a technique, without diagnosing an individual witness or determining the origin of every reported experience. Lynn et al., 2020.

Walton and Zanfretta also require separation between publicly observable circumstances and privately reported experience. The Walton witnesses' views of the flash were not identical, and the disappearance and later interior account are distinct propositions. In the Zanfretta material, colleagues' observations of distress and reported marks provide a different evidential basis from the encounters narrated later. These comparisons support a disciplined chronology of disclosure: what was said, when, to whom, under which questioning conditions, and with what independent constraint. Walton, Zanfretta.

5. Instruments, images and the reconstruction of performance

5.1 The problem of a common target

Radar–visual cases are often treated as a particularly strong category because instrumentation may constrain errors of visual judgement. The additional evidence is valuable when the detections can be associated. That requires compatible times, positions, detection volumes and continuity. A radar operator and pilot can each accurately describe what they observed while observing different sources. Coordination between them can help establish correspondence, but a shared operational conversation is not itself a measured common trajectory.

The RB-47 case contains specific episodes of reported correspondence, including Piwetz's account of losses and reacquisitions. Those details warrant examination because they could discriminate between unrelated detections and a shared source. The record also contains a timing mismatch and an unsuccessful airborne-radar acquisition. A single seamless track would suppress those differences. Lakenheath–Bentwaters similarly contains a relatively close temporal association among radar, tower and aircraft observations at Bentwaters, while the later interception narrative encounters differences between American reporting and British crew accounts. Each interval needs its own evidential reconstruction. RB-47, Lakenheath–Bentwaters.

At Haneda, the interceptor's identification of a C-54 belongs to a different part of the record from a brief unidentified return. A proposed clock correction would materially improve a radar–visual match, but the correction itself needs documentary support. Assigning the radar range to the observed light also changes the estimated physical size. The same association is therefore doing two jobs: identifying a common target and supplying its dimensions. If the first remains uncertain, the second inherits that uncertainty. Haneda dossier.

Minot offers preserved radar photographs, a substantial advantage over cases dependent entirely on recalled displays. Haines nevertheless states that the photographs he examined do not identify the selected range setting. His preliminary physical estimates use information from McCaslin's later interview. The image positions are inspectable; their conversion into distance requires an additional input. Haines's illustrated frame and explicit qualification demonstrate why possession of a radar photograph does not automatically resolve scale. Haines, revised analysis, pp. 1–3.

The Minot dossier also discusses Shough's reconstruction and its treatment of possible radar artefacts. Its positive assessment of an airborne reflector should remain available to the comparison. It concerns a particular reconstructed sequence, however, and should not be merged with every reported visual observation or earlier extreme-speed estimate. In this case the appropriate disagreement concerns which return persisted, how the aircraft moved and which scale was selected. Those are more precise questions than whether radar is generally reliable. Minot dossier.

Other cases demonstrate the consequences of importing range from an uncertain association. Kaikoura's film-brightness analysis depends on a radar range assigned to the filmed source. In JAL 1628, the radar-processing evidence and disagreement about the large apparent structure weaken a unified interpretation of the encounter. At Stephenville, selected radar returns require justification as a continuous track before they can supply distances for visual size estimates. Independent records add information, but only a defensible connection allows the measurements in one record to describe the source in another. Kaikoura, JAL 1628, Stephenville.

5.2 Nimitz and conditional acceleration estimates

Fravor's 2023 statement describes his visual encounter, the controller's subsequent report of a contact at the combat air patrol point, and Underwood's later recording. These are different evidential stages. Fravor directly observed the close encounter; the reported arrival at the patrol point came through the controller after visual contact had been lost. The later infrared video belongs to another sortie. His statement gives a clear account of why he judged the behaviour extraordinary, but it does not supply a continuous public measurement connecting every stage. Fravor, prepared statement, pp. 1–3.

The analytical importance of this distinction is visible in Knuth, Powell and Reali's peer-reviewed calculations. Their Nimitz radar example uses a reported descent of approximately 8,530 metres in 0.78 seconds, with assigned uncertainties, to estimate extreme acceleration. The mathematics evaluates the consequences of those inputs and a motion model. Its applicability to a physical vehicle also requires reliable endpoints, elapsed time and target continuity. These conditions are not tested merely by sampling numerical uncertainties around the reported values. Their paper explicitly makes assumptions about report accuracy and uses witness-derived information in the absence of detailed radar data. Knuth, Powell and Reali, 2019, §2.3.

The distinction can be expressed without claiming a new measurement. For a rest-to-rest journey modelled as equal periods of constant acceleration and deceleration, the acceleration magnitude is proportional to distance divided by elapsed time squared. Doubling the elapsed time reduces the estimate to one quarter; halving the assigned distance halves it. A mistake about whether the two detections belong to the same target is more fundamental: the journey being calculated may never have occurred. Sensitivity to numerical inputs and uncertainty about the existence of a trajectory must therefore be evaluated separately.

Conditional acceleration estimates leave the visual Nimitz encounter as a separate question. Fravor and Dietrich observed from different positions, and their accounts remain substantive testimony requiring comparison. A public continuous radar record, if recovered, could connect their observations with a measured trajectory. The existing accounts support investigation of that connection; neither the conditional calculation nor a generic perceptual explanation resolves it. Nimitz dossier.

5.3 Infrared recordings and image geometry

GOFAST provides a contrasting outcome because the displayed sensor information constrains the apparent movement. The All-domain Anomaly Resolution Office (AARO) places the object at approximately 13,000 feet and attributes the impression of rapid motion over the water to parallax. Its reported range of approximately 2–41.3 metres per second is wind-compensated speed, not ground speed. The aircraft's unknown heading prevents a unique solution; the source itself remains unidentified. The report's speed plot and input table support a range of possible solutions, constraining the performance claim more closely than the object identification. Apparent rapid movement in the recording does not establish extraordinary performance. AARO, GOFAST report, pp. 1–7.

GIMBAL poses a different optical question concerning glare, image orientation and the camera's rotation. An explanation of GOFAST's apparent speed cannot be transferred to GIMBAL merely because the videos are catalogued together. Nor can either clip settle the broader operational testimony about objects encountered by the squadron. The grouping is useful historically; it does not make the sources, geometries or claims identical. Roosevelt dossier.

Aguadilla shows how the apparent background can mislead a trajectory reconstruction. AARO models the sources as remaining over land, with changing viewing geometry accounting for the impression of water entry and later separation. Its confidence in the absence of extraordinary performance is explicitly higher than its confidence in identifying sky lanterns. The distinction matters: a reconstruction can undermine the transmedium interpretation more strongly than it identifies the source. The published map and early separation frames provide specific points for comparison with competing models. AARO, Puerto Rico report, pp. 1–4.

The remaining Aguadilla dispute should accordingly be pursued through frame timing, line of sight, visibility and the proposed range histories. Disagreement with the lantern identification would not by itself restore the water-crossing trajectory. Conversely, an ordinary reconstruction should reproduce the observed sequence rather than rely only on the broad possibility of lanterns. This separation of explanatory tasks is also useful in Campeche, where the principal infrared light group and radar detections are distinguished in the operator's account, and in the Chilean Pelican case, where aircraft candidates require their own flight and camera-geometry comparisons. Aguadilla, Campeche, Pelican.

5.4 Photographic authenticity and scene reconstruction

Photographic authenticity has several meanings that should not be collapsed. A continuous photographic emulsion or absence of compositing may establish that an image records one exposure. It does not establish the distance or nature of the photographed object. Calvine model experiments illustrate how a small staged scene could preserve continuous grain. Trindade's reported examinations distinguished alteration after exposure from fabrication arranged before or during acquisition; summaries claiming that every form of deception was excluded exceed that distinction. Calvine, Trindade.

McMinnville requires more exacting treatment than a generic model suggestion. The two photographs constrain camera position through scene features, while the object's underside brightness contributes a separate optical argument. A nearby-object reconstruction must account for both geometry and illumination; a distant-object interpretation must justify the atmospheric and surface assumptions used to infer range. The dossier records a substantive disagreement rather than a shared measurement set selecting one distance. A proposed suspension point or bright underside has evidential force only within a model that explains how the corresponding image would be formed. McMinnville dossier.

Scene constraints can narrow possible arrangements even when they do not select one range. Lake Cote's reported original-film examination similarly bears directly on a particular deformation hypothesis. At Gulf Breeze, specific reflection objections and the reported model provide positive grounds for doubt; those grounds concern the production of particular images. These photographic investigations therefore contribute more discriminating evidence than a general judgement about witness reliability. Lake Cote, Gulf Breeze.

6. Physical traces, specimens and associated effects

6.1 Establishing what changed

Physical evidence introduces questions that testimony alone cannot answer. Marks can be photographed, specimens tested and equipment failures investigated. The first comparison concerns what was present and how it differed from an appropriate baseline. A second concerns whether the alteration occurred during the reported event. A third concerns the process that caused it. These questions require connected evidence, but an answer to one does not automatically answer the others.

Socorro preserves prompt investigation and physical observations of impressions and charred vegetation. Those records impose constraints on accounts of the site, while Zamora remains the direct witness to the reported apparatus. Valensole also contains measured marks, with prior visitor access affecting their interpretation and a chronology of disclosure relevant to the reported immobilising device. In both cases the physical and testimonial records deserve examination together, with their different temporal and observational limits retained. Socorro, Valensole.

Trans-en-Provence supplies a more developed laboratory record. GEPAN's sampling chronology distinguishes plants taken near the trace on 9 January from distant comparison plants collected on 23 January. A further distance series was collected on 17 February. The first comparison therefore combines spatial and collection-date differences; the later series improves the spatial comparison but supplies no pre-event baseline. The report's laboratory table also separates analytical findings from proposed interpretations, including mechanical interaction and possible residues. Its conclusion leaves the precise causal relationship to the witness's account unresolved. These features were checked against the original report, including the sampling page and laboratory table. GEPAN, 1983, printed pp. 32, 40 and 65.

The comparison with Nancy is informative because the collection and preservation conditions became part of the laboratory dispute there. A proposed electrical effect on grass and the reported desiccation of flowers are different questions. Tully's flattened reeds and Rosedale's vegetation and water-level claims present less constrained baselines. A vortex or another local process must be tested against the relevant site and weather; the existence of a circular formation cannot select a mechanism by shape alone. Nancy, Tully, Rosedale.

6.2 Material characterisation and origin

Ubatuba permits a definite correction to a widely repeated evidential claim. In the Colorado project's examination, the submitted magnesium was compared with triply sublimed Dow material. The reported zinc concentrations were approximately 500 ± 100 parts per million in the submitted sample and 5 ± 1 in the comparator. Other measured impurities also contradicted exceptional purity in that tested fragment. The isotope comparison did not establish an anomalous magnesium composition. The initial collector remained unidentified. These results concern the material actually examined; they cannot authenticate its alleged aerial origin or determine the composition of earlier fragments no longer available. Craig, Colorado report, Case 4, printed pp. 391–396.

The Ubatuba measurements resolve a chemical proposition despite an incomplete collection history. Establishing material origin requires the additional documented route from collection to testing. A specimen described as unusual also needs an appropriate comparison population and an account of contamination, processing and measurement uncertainty. These requirements explain why a precise laboratory result may remain only one part of a successful identification.

Dalnegorsk and the reported falling fibres at Évora, Florence and Oloron encounter different versions of this problem. At Dalnegorsk, publications discussing samples from different collection sites must not be combined into one material result. Évora contains inconsistent units and incompletely documented preparation. In Florence, a laboratory note with an uncertain collection location cannot automatically authenticate the material allegedly falling at the stadium. At Oloron, the relationship between reported threads and the aerial observation remains disputed, and a promised analysis is not an available result. Their resemblance as material-associated cases is less informative than these distinct collection histories. Dalnegorsk, Évora, Florence, Oloron.

6.3 Injury, equipment failure and causal sequence

Associated health effects require similarly careful chronology. Falcon Lake includes early injury documentation, initially negative radioactivity findings and later discoveries at an accessible site. Combining those findings into one immediate radiological observation would change the evidence. Cash–Landrum contains medical records and disputed exposure interpretations, alongside an unverified helicopter-operation account. The seriousness of an injury warrants attention in its own right; identifying its cause requires a compatible exposure history and findings that discriminate between mechanisms. This study makes no new clinical diagnosis in either case. Falcon Lake, Cash–Landrum.

Vehicle and system failures can be compared by sequence rather than by a common label. Levelland accounts differ over when engines and lights failed or recovered relative to the observed light. Those differences matter for a proposed common interference mechanism. Tehran's reported directional loss and recovery of equipment suggests a more specific relationship to test, but the surviving account lacks the complete system chronology and maintenance evidence needed to choose its cause. A claim that every case exhibits the same electromagnetic effect would discard the very variation needed for diagnosis. Levelland, Tehran.

Malmstrom makes the distinction between mechanism and initiating event especially clear. The Echo missile outage is documented, and testing addressed the susceptibility of relevant circuitry to a pulse. Reproducing a failure pathway does not locate an external source or show how a signal entered the operational system. The Oscar allegations and nearby visual reports need their own records. Nuclear association increases the institutional significance of the subject, but it cannot supply the missing causal connection. Malmstrom dossier.

At Val Johnson, vehicle damage was observed by others, while the connection to the reported light depends on the encounter account. The Nullarbor case contains a more specific material correction: the tested deposit came from the left front wheel area, and tyre damage contributes a concrete mechanical explanation. That finding should not be converted into a claim that every family member saw the same light or that the visual episode has been independently reconstructed. Across this group, damage and distress are investigable outcomes; causal attribution requires evidence at the point where the outcomes are joined to the reported source. Val Johnson, Nullarbor.

7. Institutional response and the production of the record

7.1 Operational action under uncertainty

Searches, interceptions and airport closures are often cited as evidence that an event was substantial. They are substantial actions, with recoverable decision histories. Their diagnostic value depends on why officials acted and what they subsequently found. An organisation responsible for safety may respond before identifying a source. Treating the response as a completed identification reverses that sequence.

Shag Harbour provides a clear example. Contemporary reporting and a search establish that an apparent surface incident was taken seriously. Differences among witnesses over the descent and impact, together with the absence of a recovered object, leave the source unresolved. A later underwater journey requires additional evidence beyond the initial search. At Xiaoshan, the airport disruption is documented more securely than the association between the initiating observation and widely circulated images. A photograph attached to a closure story may acquire an evidential role that its date and provenance do not support. Shag Harbour, Xiaoshan.

Aircraft accidents demand particular precision because the unidentified source can become the organising explanation for a fatal outcome. In Mantell's case, the accident investigation's oxygen-deprivation account has a firmer documentary basis than the exact identity of the pursued object. Valentich's radio exchange records an alarming experience and an engine complaint, but does not continuously establish aircraft attitude or position. The surrounding Bass Strait reports cannot supply those missing variables simply through proximity in time or place. Mantell, Valentich, surrounding reports.

The February 2023 shootdowns require separate treatment of the recovered Chinese balloon and the other objects. A changed detection threshold can change which targets enter an operational system, altering the apparent incidence without a corresponding change in the underlying population. The New Jersey reports of 2024 introduce a related problem of aggregation: official identifications of ordinary activity have a stated scope, while particular military-site reports require their own logs. Neither a broad official reassurance nor a large public report count can replace incident-level matching. 2023 balloon and shootdowns, New Jersey.

7.2 Official documents and testimonial access

An official file preserves information received or produced by an institution. Its contents can include direct observation, hearsay, preliminary interpretation and later correspondence. Their evidential status depends on the document's function. Trindade's American classification, for example, cannot be counted as another independent photographic examination when the relevant file states that photographic copies had not been obtained. Colares similarly requires separation of military observations, patient reports and a doctor's own observations. Institutional custody does not make every statement an institutional finding. Trindade, Colares.

Congressional hearings preserve named testimony and establish public accountability, but witnesses can have different forms of access. In 2023, Fravor described a personal encounter, Graves discussed operational experience and reporting, and Grusch presented allegations based substantially on information supplied by others. The 2024 hearing likewise contains claims resting on different documentary and testimonial relationships. A statement made in the same proceeding does not acquire the same proximity to the underlying event as every other statement. Comparison should follow the witness's stated access and identify the record that could corroborate the particular claim. 2023 hearing, 2024 hearing.

Recovery narratives illustrate how documentary authority can be accumulated without resolving provenance. Roswell's contemporary debris history and later occupant accounts require different sources. At Kecksburg, the fireball trajectory evidence and later descriptions of an acorn-shaped recovery are distinct propositions. In Varginha, alleged links among witnesses, military transport and a death require their own chronological and documentary support. An account of official secrecy may explain why evidence is said to be unavailable, but it cannot substitute for the evidence in a comparative inference. Roswell, Kecksburg, Varginha.

Some records provide positive reasons to reject parts of a narrative. Maury Island includes an admission concerning the supplied fragments and evidence about pre-existing boat damage. Aztec's promoted material claims encounter a failed demonstration and serious problems with the authority of their principal promoters. Kalahari's documentary inconsistencies weaken the alleged crash package. These findings should receive proportionate weight. They do not require a speculative account of every participant's intention, and they do not establish that unrelated reports elsewhere share the same origin. Maury Island, Aztec, Kalahari.

7.3 Reporting conditions and apparent concentration

Public reporting is also an observation process. Medina, Brewer and Kirkpatrick's spatial study of National UFO Reporting Center (NUFORC) reports examines associations with visibility and human activity. Its county-level results concern volunteered reports and acknowledge reporting limitations. They are relevant to the design of comparative research because report concentrations can reflect opportunities to observe and submit accounts. They cannot identify the sources in UAPRAD's historical cases or convert this selected collection into a population dataset. Medina, Brewer and Kirkpatrick, 2023.

The same distinction applies to military and nuclear-site concentrations. Such locations may combine frequent aerial activity, more sensors, trained observers, reporting procedures and restrictions on public access. A claim of unusual attraction requires comparison with exposure and detection opportunity. The present corpus has no uniform denominator of observation hours, sensor coverage or background traffic. Its military cases can establish important operational and documentary questions; their number cannot estimate an excess physical incidence around particular facilities.

8. Comparative findings and explanatory strength

The collection supports five principal findings. First, evidence for an event often has a different strength from evidence for the interpretation attached to it. Contemporary reporting at Socorro, a preserved radar image at Minot and the operational history of Shag Harbour contribute different positive constraints. None should be reduced to the statement that a mystery remains. The comparative task is to specify the proposition each record can sustain.

Second, the evidential advantage of multiple sources depends on their relationship. A second observation can restrict geometry, establish simultaneity or corroborate a particular action. Republication of one interview cannot do so. Nor can a later drawing, informed by discussion, be treated as equivalent to a separately recorded early description. Source dependence is especially consequential when a cumulative argument relies on the apparent number of confirmations. Several accounts derived from one origin remain one chain of transmission, however many institutions later cite them.

Third, uncertainty about identity and uncertainty about performance frequently diverge. GOFAST permits a substantial challenge to extraordinary-speed interpretations without a definite object identification. McMinnville preserves a dispute over scene geometry and illumination that cannot be settled by naming a generic model. Malmstrom documents a system failure while leaving the proposed external trigger unestablished. A common label such as unexplained would obscure the different achievements and remaining questions in these investigations.

Fourth, conventional explanations vary in evidential strength. A mechanism with a dated candidate, compatible geometry and successful reproduction of the observed feature deserves more weight than a resemblance. A release programme near an event is less specific than a matching flight. A laboratory reproduction of susceptibility is less specific than a demonstrated initiating signal. The same standards apply to extraordinary interpretations: an acceleration calculation conditional on reported endpoints does not independently establish that those endpoints belong to a single vehicle.

Fifth, unresolved components do not automatically accumulate into a common extraordinary cause. Their uncertainties arise at different stages: incomplete exposure geometry, missing original media, selective witness preservation, disrupted specimen custody or inaccessible operational records. A common cause would gain support from shared, discriminating physical properties established independently across cases. The reviewed records do not establish such a property across the collection. This finding leaves individual questions open while limiting what can be inferred from their aggregation.

The distinction between levels of explanatory support can be summarised as follows. These are analytical states of particular propositions, not grades assigned to whole witnesses or cases.

A case can occupy several rows simultaneously. This is an advantage for research because it identifies what has already been learned. The remaining work can then address a defined uncertainty instead of repeating the entire debate. It also gives disagreements a common object: the same frame interval, laboratory sample, witness statement or operational episode.

Counterexamples are essential to that synthesis. McMinnville and Minot retain technical questions that deserve more than a general appeal to misperception. Ubatuba and GOFAST demonstrate that analysis can materially narrow a claim. Ariel and Broad Haven show that substantial reporting does not entail uniform testimony. Maury Island illustrates the stronger evidential position created by positive contradictory information. Their coexistence rules out treating the collection as a homogeneous body of either extraordinary vehicles or evidentially empty stories.

9. Research priorities arising from the comparison

The most productive archival work would recover records at the points where consequential associations remain testable. For radar cases, priorities include original plots, range settings, clock conventions, controller logs and aircraft tracks. Minot's scale and track questions and the separated Nimitz episodes have identifiable documentary targets. A further interview can clarify how an operator remembers a display, but it cannot provide the same information as the original recording. Searches should therefore document which record was sought, its expected custodian and the effect of non-recovery on the claim.

Photographic work should preserve acquisition history and publish reproducible scene models. For McMinnville, competing models need common image coordinates, camera positions, illumination assumptions and predicted brightness relations across both exposures. For Calvine and Lake Cote, original-generation media and dependable camera information would materially improve the evaluation of scale and image-formation effects. Repeatedly enlarging copied images cannot recover spatial detail that was never resolved in the available recording.

Witness archives would benefit most from complete early interviews, unselected drawings and documented communication histories. In the school cases, establishing which child produced which account, before or after which discussion, would improve comparison more than another undifferentiated witness total. Interview protocols should invite free description before proposing categories or candidate explanations. Later testimony should be retained with its collection date, including changes and explicit uncertainty, so that the development of the account remains visible.

Material investigations require documented collection locations, sample identifiers, custody, controls and laboratory methods. A new measurement should answer a specified question about a preserved specimen. Where the connection to an event cannot be recovered, the result should remain a material-characterisation finding. Replication across laboratories is especially useful when they test aliquots from the same documented specimen against suitable reference materials, with preparation and analytical blanks distinguished from environmental controls.

Prospective observation can address limitations that historical archives cannot repair. NASA's independent study emphasised calibration, metadata and coordinated observations; it did not adjudicate the historical case collection. Watters and colleagues' observatory proposal similarly connects desired physical parameters to measurement requirements, including triangulation and multiple sensing modes. Their practical contribution here is a design principle: record the information needed to distinguish a local source, a distant source and a sensor effect at acquisition, rather than infer it later from a compelling image. NASA, 2023; Watters et al., 2023.

For UAPRAD, the immediate editorial application is to retain claim-level distinctions within each dossier and make consequential revisions traceable. An updated laboratory finding should identify the specimen it concerns. A newly proposed flight should be matched to an episode. A later witness statement should preserve its relationship to the earlier record. This would allow future cross-analysis to measure actual changes in the evidence rather than changes in the strength of published wording.

10. Conclusion

Comparison across the 123 dossiers establishes a recurring evidential problem: the movement from individually documented observations to a unified causal account. Witness agreement, instrumental detection, physical alteration and institutional response can each contribute substantial information. Their combined force depends on whether the records establish the connections among them. The collection repeatedly makes those connections available for examination, but does not support assuming them from a shared case name.

Several investigations yield substantive conclusions. Geometry can constrain apparent performance; laboratory measurements can refute a claim about a tested specimen; contemporary records can expose later conflation; and specific ordinary mechanisms can account for particular observations. Other cases retain bounded disagreements over scene reconstruction, target association or the development of testimony. These outcomes are compatible because they answer different questions.

The reviewed corpus does not establish a common extraordinary physical origin. It does establish a substantial history of observation, interpretation and investigation in which evidential quality varies within cases as well as between them. Its most defensible contribution is therefore a comparative account of those variations: what was recorded, which inference follows, and where the next evidential connection remains to be demonstrated. Future research will be most informative when it tests those defined connections with records and measurements capable of discriminating among the competing accounts.

Selected scholarly references

Gabbert, F., Memon, A. and Allan, K. (2003). Memory conformity: Can eyewitnesses influence each other's memories for an event? Applied Cognitive Psychology. doi:10.1002/acp.885.

Knuth, K. H., Powell, R. M. and Reali, P. A. (2019). Estimating flight characteristics of anomalous unidentified aerial vehicles. Entropy, 21, 939. Full text.

Lamb, M. E., Orbach, Y., Hershkowitz, I., Esplin, P. W. and Horowitz, D. (2007). Structured forensic interview protocols improve the quality and informativeness of investigative interviews with children: A review of research using the NICHD Investigative Interview Protocol. Child Abuse & Neglect. Full text.

Lynn, S. J., Kirsch, I., Terhune, D. B. and Green, J. P. (2020). Myths and misconceptions about hypnosis and suggestion: Separating fact and fiction. Applied Cognitive Psychology. Accepted manuscript.

Medina, R. M., Brewer, S. C. and Kirkpatrick, S. M. (2023). An environmental analysis of public UAP sightings and sky view potential. Scientific Reports, 13, 22213. Full text.

Seawright, J. and Gerring, J. (2008). Case selection techniques in case study research: A menu of qualitative and quantitative options. Political Research Quarterly, 61, 294–308. Article.

Watters, W. A. et al. (2023). The scientific investigation of unidentified aerial phenomena (UAP) using multimodal ground-based observatories. Journal of Astronomical Instrumentation, 12, 2340006. Author-hosted preprint.

Primary reports and case dossiers are linked at the claims they support. The companion case comparison register covers all 123 catalogue entries.