The operational record
A compilation with stated technical limits
The Comando Aéreo de Defesa Aérea (COMDA), Brazil's Air Defence Command, forwarded its occurrence report on 2 June 1986. Brigadier José Pessoa Cavalcanti de Albuquerque signed it as interim commander. Its opening explains that time and specialist knowledge constrained the investigation and that the command restricted itself to an operational account. This defines the report's remit: it assembled information from controllers and intercepting pilots rather than conducting a complete scientific investigation. Covering letter and introduction, pp. 1–2.
The initial considerations describe radar systems whose primary purpose was air traffic control, manual intervention needed to maintain contacts, and returns too weak to support satisfactory tracking. Controllers therefore used their judgement and experience to direct aircraft towards reported contacts. The distinction between an indication on a display and a continuously tracked target is central to interpreting the subsequent account. A sequence of intermittent indications can suggest movement without establishing that every indication belongs to the same source.
The original report uses times marked “Z”, the aviation designation for UTC. The chronology below preserves that convention, including its transition into 20 May. It should not be read as a sequence of local clock times on 19 May.
The Santa Cruz visual account
First Lieutenant Kleber Caldas Marinho's personal report, dated 28 May 1986, identifies his aircraft as Jambock 17. He describes a light changing position relative to his aircraft and an onboard radar contact at approximately the same bearing, at a displayed range of 10–12 nautical miles. He also describes observing green, red and white colours, with white predominating. His account records an effort to interrupt his visual fixation to reduce night-vision illusion. Marinho report, pp. 1–2.
This is a specific report of correspondence, rather than the general claim that radar and visual observations occurred somewhere during the same night. It deserves separate treatment. The personal statement, however, contains no preserved radar plot sequence or optical record from which that association can be independently reconstructed. Its range belongs to the radar indication; assigning it to the visible light depends on the pilot's reported association.
Marinho's references to flight levels describe the aircraft's altitude during the observation. They cannot automatically be transferred to the light. A source seen above or below the aircraft requires a constrained distance and viewing geometry before its altitude can be calculated. Likewise, apparent position changes during an intercept combine the observer's movement with whatever movement the observed source may have had.
Anápolis acquisition and unsuccessful searches
Captain Armindo Souza Viriato de Freitas's Anápolis report identifies Jaguar 116 and describes repeated airborne radar acquisition without visual contact. He records his aircraft reaching Mach 1.05 during one approach, an increasing indicated separation and subsequent loss of acquisition. That Mach value describes the interceptor. Establishing the other source's speed would require the time history of range, relative direction and both trajectories. Anápolis personal reports, p. 5.
The same packet preserves negative results. Captain Rodolfo da Silva Souza, identified with Jaguar 98, reports unsuccessful visual and radar searches. His account describes difficulty determining a target's heading and speed through a relay from the ground station to the controlling position and then to the aircraft. Captain Júlio Cezar Rozemberg's report, identified with Jaguar 107, also records searches without visual or airborne radar contact. Anápolis packet, pp. 3–4.
These accounts show why a count of aircraft launched cannot serve as a count of independent confirmations. A null search does not identify the source of an earlier ground indication, because visibility, geometry and sensor coverage may differ. It nevertheless prevents the operation from being described as five pilots all confirming the same display.
Thirteen plots and an unconfirmed association
COMDA's account of the second Santa Cruz aircraft describes a red light at the position supplied by controllers. It subsequently records thirteen plots appearing behind the aircraft. A commanded turn produced neither visual nor onboard radar contact with those indications. Occurrence report, pp. 6–7.
The thirteen-plot passage is often an arresting part of the episode's narrative. Its evidential unit remains a reported display indication, with an unsuccessful attempt at additional confirmation. Treating it as thirteen physical objects pursuing the aircraft would require a target association that this passage does not supply. It also should not be combined arithmetically with every light or return mentioned elsewhere: observations can overlap, recur or originate in different processes.
Documentary chronology
- 19 May 1986, 23:15Z: COMDA records notification of the São José dos Campos tower's visual observations and São Paulo approach radar reporting.
- 20 May, 00:08Z: The report records a civilian pilot's observations from PT-MBZ.
- 20 May, 00:39Z: The Santa Cruz alert aircraft is activated; Anápolis activation follows at 01:11Z.
- 20 May, 01:34Z–02:36Z: The report records two Santa Cruz and three Anápolis launches. Its final Anápolis launch is 02:36Z; the controller's separate account gives 02:35Z. The inspected records do not resolve the one-minute difference.
- 20 May, 02:37Z–03:30Z: The report records the five aircraft landing normally.
- 28 May 1986: Marinho's personal statement is dated. The Anápolis packet preserves controller and pilot accounts, but its handwritten covering-letter date remains unresolved here.
- 2 June 1986: COMDA forwards the compiled occurrence report and its concluding opinion.
- 25 September 2009: A retrospective government account dates the public release of that report to this day.
- October 2015: An archive release notice describes sixteen audio files, including eight recordings associated with the episode. Release metadata must be distinguished from a verified tape inventory. Government retrospective, archive release notice.
The command's assessment
COMDA's final considerations describe common characteristics, including radar echoes, varied velocities and abrupt changes of movement. The command inferred solidity and a degree of intelligence from the reported behaviour. Those conclusions belong to the institution that compiled the accounts. Reading them alongside the report's technical limitations and unsuccessful searches makes their inferential basis visible. Final considerations, pp. 7–8.
Several documentary stages repeat the same material. A personal statement incorporated into a command report and later summarised by an archive remains one underlying observation chain. Agreement between the versions can establish faithful transmission; it does not create another witness or another sensor measurement. The copies used here bear archive identifiers ARX.249, ARX.254 and ARX.256, but are hosted by FENOMENUM. Their host's interpretation is separate from the military documents.
Evidence assessment
The documentary record supports a military response to observations that participants could not identify. Evaluating their physical causes requires a comparison of specific observations, because the operation produced different forms of correspondence and several unsuccessful searches. The most useful distinction is between evidence that a report entered the operational record and measurements capable of identifying its source.
Marinho's account supplies a particular visual/radar association worth investigating. Its significance depends on retaining the distinction between his recognition of a corresponding bearing and an independently reconstructed track. Viriato's radar-only account poses a different problem: apparent changes of range must be assessed against the interceptor's movement and the continuity of acquisition. Combining those records into one performance description would conceal the different assumptions each requires.
The negative searches place a useful boundary around the command's synthesis. They show that the response did not produce uniform confirmation across the aircraft, although they cannot identify the causes of contacts reported elsewhere. COMDA's general account of common characteristics therefore needs to be tested observation by observation. The available material supports a documented operational history and an unresolved association problem; it supplies no reproducible basis for a single physical origin across the episode.
A common unidentified physical source
An unidentified physical source is a legitimate working category for a reported visual/radar association. The Marinho and Viriato accounts provide specific observations to examine. Neither supplies sufficient data here to determine shape, dimensions, construction or origin. The frequently repeated total of 21 objects appears in the government retrospective; a counting method linking each observation to a distinct source is not established by this document comparison.
Conventional aircraft or distant lights
Ordinary aircraft or distant lights require a position and viewing reconstruction. Such an explanation should reproduce the relevant sightline and timing rather than merely point out that conventional sources exist. No demonstrated identification of the individual observations is available in the inspected packet.
Radar propagation or target-association errors
Radar propagation, clutter or target-association errors are also appropriate avenues of investigation, given the report's account of weak returns and tracking difficulties. The records do not supply the original plots, instrument configuration or contemporaneous environmental measurements needed to demonstrate a particular radar mechanism. These possibilities therefore remain hypotheses, not findings that the episode was entirely a radar artefact.
A combination of sources
A mixed explanation deserves explicit consideration. Different sorties can encounter different visual sources or radar processes during one alert. A satisfactory assessment must establish the association between them before requiring one explanation to account for every report. Conversely, treating all contacts as unrelated without examining their timing would discard potentially useful evidence. A synchronised reconstruction could distinguish these alternatives.
Assessment snapshot
| Observation | Recorded correspondence | Assessment |
|---|---|---|
| Marinho, Jambock 17 | Visible light and a reported airborne radar indication at a corresponding bearing | A specific participant association; original sensor sequence needed for reconstruction |
| Viriato, Jaguar 116 | Airborne radar acquisition; no visual contact | Radar reporting with unresolved physical identification |
| Second Santa Cruz aircraft | Reported light at controller-supplied position | Association recorded in compilation; underlying personal account merits further comparison |
| Thirteen plots behind that aircraft | No visual or onboard radar contact after the commanded turn | Ground indications remain unconfirmed by the aircraft |
| Jaguar 98 and Jaguar 107 | Unsuccessful visual and onboard radar searches | Material negative results; constrain claims of universal confirmation |
| COMDA final judgement | Synthesis of controller and pilot reporting | Institutional interpretation whose scope exceeds a single independently reconstructed track |
Claim ledger
| Proposition | Supporting record | Current assessment | Further discriminating evidence |
|---|---|---|---|
| Five aircraft were launched | COMDA chronology, pp. 5–7 | Documented operational action; separate from the number of confirmed sources | Complete launch and control logs |
| Marinho associated a light with an airborne radar contact | Personal statement, pp. 1–2 | Supported as his reported association; target identity remains unverified | Synchronised sightlines and original radar records |
| Viriato reported increasing separation while supersonic | Anápolis packet, p. 5 | The Mach figure belongs to the interceptor; source speed is not independently derived | Time-resolved positions and reconstruction of both trajectories |
| Thirteen plots appeared behind the second Santa Cruz aircraft | COMDA, p. 7 | Reported ground indications with no visual or onboard confirmation after the turn | Continuous tracks and physical-source association |
| Jaguar 98 and Jaguar 107 obtained no visual or airborne radar contact | Anápolis packet, pp. 3–4 | Documented negative searches; other contacts remain to explain | Fuller control logs and sensor coverage |
| COMDA inferred solidity and intelligence | Final considerations, pp. 7–8 | Attributed institutional judgement, without a reproduced physical identification | Reproducible observations supporting the proposed properties |
Open questions
Synchronising the observations
The archive identifies surviving communications recordings. Time-stamped listening could establish when instructions, acquisitions and losses occurred, and whether a reported light remained at a corresponding bearing through a radar contact. The audio has not been inspected for this assessment. A recording's catalogue description cannot fill those gaps.
Recovering the measurement basis
Original radar material, if preserved and available, could clarify continuity, track identity and apparent changes of range. Instrument settings and environmental information would allow proposed radar explanations to be tested against particular returns. Pilot aircraft trajectories would also help separate a source's movement from changes produced by the intercept itself.
Further research
Witness record
Pilots whose statements were inspected
Marinho's signed statement describes his Santa Cruz observation. The Anápolis packet preserves the accounts of Viriato, Rodolfo da Silva Souza and Júlio Cezar Rozemberg. These are contemporary operational statements, rather than later interviews. Their different results remain distinct in the assessment.
Controller and institutional reporting
The Anápolis packet includes a controller statement signed by Paulo Roberto Ribeiro do Amaral. Albuquerque signed the command report as its institutional author; the document does not identify him as an eyewitness to the airborne encounters.
Source library
Contemporary records reproduced online
- COMDA occurrence report, 2 June 1986; ARX.249: eight scanned pages covering the operational sequence, radar limitations and institutional conclusion.
- Marinho personal report, 28 May 1986; ARX.254: two scanned pages describing visual observations and reported airborne radar correspondence.
- Anápolis report packet; ARX.256: five scanned pages containing the covering letter, a controller account and three pilot reports. The covering letter's handwritten date remains unresolved in this assessment.
Official retrospective context
- Arquivo Nacional, Série Relatos Extraterrestres: identifies the episode within the archive's UFO holdings and reproduces a pilot-report image.
- Government of Brazil, Official UFO Night in Brazil, 20 May 2022: public account of the episode, official response and report release. Counts and attributed conclusions require comparison with their originating records.
- Arquivo Nacional audio release notice: documents the archive's description of the October 2015 release.
Further primary material
- ArquivoNacionalBrasil, recording 1: metadata identifies BR_DFANBSB_ARX_0_0_0692_m0001de0002. Its cover illustration concerns a separate Bauru report from 1968; it is not imagery of the 1986 episode. Recording content remains outside this assessment.