The observation record
Reports during the approach
The reproduced communications record shows the Boeing crew asking about other traffic, reporting bright lights, and later distinguishing an unidentified light from the known aircraft above them. It also marks an interruption in the recording. This preserves uncertainty at the time of the encounter and gives only a partial sequence through a critical period. The online text is a later reproduction of the official transcript, rather than an audio recording independently authenticated for this dossier. Air–ground transcript, document D1.
A description of something alongside an aircraft is a report of apparent position. Establishing physical proximity would require independently constrained distance. Several observers can agree on a bright light while estimating its location differently, particularly when their own aircraft are moving. The interpretation of close formation therefore carries a greater evidential burden than the occurrence of the light report.
What the statements distinguish
The reproduced statement attributed to captain Jorge Néstor Polanco describes green and orange lights and the loss of airport radio aids. Controller Ramón Alfredo Blanco records runway-light failure while the VOR radio beacon and instrument landing system remained operational. Operations officer Daniel Alberto García reports a brilliant light without an identifiable silhouette and relays the generator technician's alarm before the runway lights went out. Witness statements, documents D2–D4.
These differences matter because a collective account of total equipment failure would erase a specific operational disagreement. A useful reconstruction must retain each person's role and access to the equipment. The statements document an electrical disturbance, but leave the exact extent of the outage less settled than a general description suggests.
Electrical failure and the encounter narrative
The city outage and the airport generator failure require separate causal explanations. Investigator Heriberto Janosch's 2003 assessment attributes the city interruption to a switching error, citing a telephone consultation with engineer Marcelo Etchegoyen on 2 August 1995. That is an investigator's report of a technical explanation; the underlying utility log has not been inspected here. Janosch proposed an overload when runway lighting was switched on as the cause of the later generator stoppage. Janosch, sections III and V.
An approach disrupted by darkness can produce both a go-around and heightened attention to unfamiliar lights. The occurrence of those events together does not determine whether the light source affected the electrical system. A causal claim needs a mechanism supported by equipment records, together with a sequence showing that the proposed cause preceded the failure.
An airport flash is a further component. CEFAE favoured an electrical explanation, but the reported timings differ. Its treatment of that flash should not inherit the confidence attached to the searchlight identification. A reconstruction can explain one visual sequence while leaving the precise origin of an earlier illumination uncertain. CEFAE, printed pp. 73–75.
How the investigation developed
The initial institutional response
A reproduced Argentine Air Force letter dated 5 September 1995 states that witness declarations were taken and that the investigation ended after no incursion by known conventional aircraft was established. It left natural objects, cloud reflections and other optical phenomena among the possibilities. That response records the scope and stopping point of an institutional investigation. Its wording provides neither a positive identification of an extraordinary aircraft nor the later searchlight reconstruction. Air Force correspondence, document D7.
The same collection reproduces a meteorological reply dated 18 March 1996: some cloud, westerly wind and 30 km visibility. These general conditions allow an optical hypothesis to be examined. They cannot fix the cloud surface or scattering conditions for every reported sightline. Meteorological reply, document D6.
From a possible projector to operator testimony
Janosch's April 2004 report explains that early enquiries to nightclub staff had discouraged the projector hypothesis. His May report then presents anonymised accounts from people connected with a nightclub and its lighting equipment. One interviewee recalled testing equipment during an outage; another recalled operation on independent generator power but could not fix the date. Their timing precision and descriptions of control over the equipment differ. April investigation report, May operator report.
This changed the hypothesis materially: there was now testimony concerning equipment in use, rather than a proposed light source without an identified operator. The discrepancies still require assessment. A witness's confidence about an event date, the date actually given in an interview and an investigator's eventual identification should remain distinguishable.
The later official reconstruction
CEFAE's study identifies Juan Carlos Rivero as the operator and credits Janosch's earlier investigation. Rivero's signed statement in annex 2 is dated 8 January 2019, although the file is labelled as the 2018 report. The original event, the earlier enquiries and this later statement consequently belong to different stages of the case's documentary history. CEFAE acknowledgements and annex 2, printed pp. 76–79.
The institutional study extends an existing line of investigation. Counting the earlier articles and CEFAE report as separate confirmations would overstate their independence where they share testimony or an explanatory framework. Their agreement is most useful when the later study contributes an additional record or a test that can itself be examined.
Assessing the searchlight reconstruction
What the model tests
The study aligns Rubén Adolfo Cipuzak's 1996 drawing with a geographical model. It reports that Rivero first demonstrated his remembered pointing directions without seeing the drawing overlay. Directional rods were fixed before the comparison. The model doubles the vertical scale and enlarges the aircraft and cloud representation for visibility. CEFAE reconstruction procedure, printed pp. 63–69.
Withholding the overlay is a useful precaution against fitting the pointing exercise directly to that drawing. Its value is specific to the procedure. It cannot establish that a participant had no prior exposure to published descriptions during the intervening years. The study should therefore be read as a comparison of historical testimony, with a safeguard against one immediate source of suggestion.
The exaggerated height requires a further distinction. Plan-view correspondence can help test whether directions are geographically compatible. Elevation and apparent proximity need their own error estimates. The model's visual clarity helps explain the proposed sequence, but its enlarged components cannot supply the accuracy of an instrumented reconstruction.
Explanatory fit and remaining tests
The proposed display connects changing colour, movement over the lake, upward motion and disappearance with searchlight operation. A projected patch can change apparent position without carrying a vehicle through that trajectory. This makes a light-source explanation relevant to the reported performance, rather than merely an appeal to a familiar object. CEFAE sequence comparison, printed pp. 69–71.
The next discriminating test would examine how much the correspondence changes when uncertain aircraft positions, cloud height and operator directions vary. A match that survives reasonable variations would carry more weight than one requiring a narrow selection of assumptions. The illustrated comparison offers a coherent identification; it does not provide such a sensitivity analysis.
The report explains that additional rods made the sweep clearer in photographs and that the final withdrawal was recorded on video rather than represented with rods. These distinctions separate the comparison procedure from its presentation. Measurements would need to come from the recorded pointing exercise and its geometry, with the original comparison preserved. Photographic persuasiveness alone cannot determine numerical precision. CEFAE, printed pp. 67 and 71.
Evidence assessment
The case supports a substantial conventional explanation for the principal lights, while leaving specific historical and technical questions open. The most useful analytical distinction is between identification of a likely stimulus and reconstruction of every reported detail. That distinction prevents either institutional closure or residual uncertainty from controlling the whole assessment. The inspected record provides no independent basis for attributing the lights to a non-human vehicle.
Searchlight and aircraft lights
Operator testimony and the later geographical comparison give the searchlight explanation positive support. Known aircraft traffic is also relevant to the early light sequence. The combined interpretation should be tested episode by episode: observing one aircraft light earlier does not automatically identify a subsequent coloured display.
A physical object near the Boeing
The reports preserve an impression of proximity and, in some accounts, a structured object. Those descriptions deserve attention at their recorded dates. Establishing a nearby vehicle would require evidence that separates physical structure from interpretation of lights and supplies a dependable range. The inspected material does not settle that proposition.
Electrical interference
The airport interruption has an immediate operational context involving electrical supply and generator failure. An external interference explanation would need evidence beyond concurrence with a light observation. Generator, switching and maintenance records would be more discriminating than further repetition of the encounter narrative.
Assessment snapshot
| Proposition | Assessment | Reason for the distinction |
|---|---|---|
| Unusual lights were reported during the approach | Well supported as a reporting event | Communications and statements preserve accounts close to the event; original audio remains to be audited |
| A searchlight contributed to the main display | Substantially supported | Operator testimony and a later comparison provide an identified source and explanatory fit |
| The model reproduces the precise encounter geometry | Not demonstrated at that precision | Exaggerated height and uncertain historical inputs require separate tolerances |
| The lights caused the electrical interruption | Not established | Concurrence does not identify a mechanism; equipment accounts differ |
| The airport flash has a confirmed electrical cause | Provisional explanation | Its timing and proposed mechanism need an independent equipment reconstruction |
Claim ledger
| Claim | Source function | What further evidence would change the assessment |
|---|---|---|
| Lights appeared to accompany ARG 674 | Contemporary reporting reproduced later: D1 and D2–D4 | Original audio, complete timestamps and separately recorded sightlines |
| Projector operation explains the principal sequence | Operator accounts and retrospective technical study: May 2004 report; CEFAE pp. 63–71 | Original interview records and a reproducible comparison with uncertainty ranges |
| Airport navigation aids all failed | Conflicting participant statements: D2 and D4 | Electrical, navigation-aid and maintenance logs for the relevant interval |
| An anomalous vehicle produced the reported effects | Interpretation combining observations and equipment claims | Independently constrained structure, range or physical effects attributable to the same source |
Open questions
Can the recording interruption be reconstructed?
The reproduced transcript marks a gap. The original recording, copying history and official log could establish its extent more securely. A gap limits continuity; it does not identify the missing events.
How precise is the searchlight match?
The study demonstrates a qualitative correspondence. Recovering the original model dimensions, pointing record and comparison data would allow the effect of uncertain inputs to be tested.
Which airport systems failed?
Participant statements differ. Equipment logs would distinguish loss of lights, loss of power to individual instruments and loss of the radio aids used for the approach.
Are all reported lights necessarily one source?
The sequence includes traffic observations, a changing display and an airport illumination. Each needs a time and sightline before it can be assigned to a common cause.
Further research
Witness record
The principal witness groups include the ARG 674 crew, the GN 705 crew, airport personnel and observers around Dina Huapi. Polanco, Carlos Dortona, Jorge Allende, Juan Domingo Gaitán, Cipuzak, García, Blanco, Nicolás Enoe Araya, Berta Cabral and Silverio Cabral appear in the published investigation inventory. Rivero enters the later record as the proposed light-source operator. Their evidential roles differ, so a single undifferentiated witness total would conceal the relationships most relevant to the assessment. Witness inventory, operator reconstruction.
Source library
Tier 1: Contemporary records in later reproduction
- Air–ground communications, D1: the circulated transcript of the 31 July–1 August UTC sequence. Inspected as text; the original audio and stamped transcript have not been independently authenticated here.
- Witness declarations, D2–D5: reproduced statements attributed to Polanco and airport personnel. These retain important disagreements about equipment and observation; they are not separate confirmations of every component.
- Meteorological and Air Force correspondence, D6–D7: reproduced institutional replies. Read their conclusions within the stated purpose of each document.
Tier 2: Institutional technical analysis
- Argentine Air Force, CEFAE report labelled 2018, Bariloche chapter, printed pp. 51–79: official retrospective case study, illustrated reconstruction and annexes. Annex 2 contains Rivero's statement dated 8 January 2019. The model is an explanatory comparison, not a preserved 1995 sensor record.
Tier 3: Original investigator reports
- Heriberto Janosch, “10 preguntas y algunas respuestas”, 14 September 2003: separates the observations, outages and interpretive claims; also reproduces an earlier investigation.
- Janosch, “Una hipótesis plausible”, 19 April 2004: explains source acquisition and the developing projector hypothesis.
- Janosch, “¿Hacia una solución definitiva?”, 1 May 2004: presents anonymised operator and staff testimony. Interview uncertainty and conflicting details remain part of the source record.
The investigator reports and official study share a research history. Their number should not be used as a count of independent observations. References to earlier broadcasts or underlying documents are research leads wherever the original item has not been inspected for this dossier.