Flights, reports and investigation
The 30–31 December sequence
Captain W. Startup's interview summary records departure from Wellington at 23:46 on 30 December. A handwritten annotation identifies First Officer Guard as also present during the interview. The summary says that the aircraft radar was not used on the southbound leg. Crew observations near the coast were therefore compared with indications supplied by Wellington controllers, rather than with independently acquired airborne targets. Several instructed searches produced no visual observation. Startup interview summary, scan p. 172.
After departure from Christchurch at approximately 02:16 on 31 December, the crew observed a bright light to the north-east and an aircraft radar indication. The interview summary explicitly retains doubt about whether these represented the same source: the light lacked a background against which its distance could be judged. The crew's turn towards it also changed viewing geometry. Neither a displayed range nor an apparent movement can be assigned to the visible source without establishing that association.
The account describes the aircraft's radar indication moving closer and then maintaining a relative position for a period. The crew could not determine whether the range change resulted from their own movement or movement of the indicated target. During this principal return-flight episode, the summary records no unidentified contact reported by either ground radar to the aircraft, while also raising incomplete controller dialogue as a possible communication limitation. Ground reporting earlier in the night cannot automatically confirm this later light. Interview summary, scan pp. 172–173.
An earlier cluster with conflicting dates
Wing Commander J. B. Clements's report groups earlier observations under 20–21 December. Captain J. Randle's interview summary and the heading of his own inflight notes instead identify a 22 December flight. The inspected records do not reconcile those dates. The earlier material is therefore retained as context rather than forced into one exact chronology. Clements report, scan pp. 148–151, Randle records, scan pp. 178–180.
Randle's record also separates inflight notes from later comment and speculation. He describes some ordinary vessel observations alongside lights he considered unusual, and offers possible explanations for his own experience. Those distinctions preserve a participant's reasoning without turning every subsequent suggestion into an identification. His interpretation of his flight also cannot resolve observations he did not share.
The official inquiry's method
Clements signed the report on 11 January 1979. Its introduction states that witnesses were not necessarily interviewed separately and that their statements were taken at face value. Joint discussion may help reconstruct a sequence, but it limits the use of agreement as evidence of independently preserved recollection. The compiled report and its underlying interview summaries belong to a shared testimony chain. Introduction, scan p. 147.
The report proposed atmospheric effects, astronomical sources, surface lights and radar anomalies as explanations. Its conclusion retained some observations as inconclusive pending further investigation. His assessment favoured predominantly conventional causes without demonstrating a physical identification for every reported light. Conclusions, scan pp. 154–155.
What the optical studies examined
Aircraft windows and image structure
N. J. Rumsey's January 1979 Physics and Engineering Laboratory report examined Argosy windows after television screening of David Crockett's aircraft film. Rumsey used a telescope and a nearby torch source, checking windows from another Argosy before the filming aircraft became available. Crew identified the window through which much of the footage had been taken. The report describes horizontal streaks and changing image shapes produced through that window. PEL Report 625, scan pp. 120–122.
Rumsey regarded the photographed streaks as window artefacts and argued that the image could safely provide only an upper limit to angular size. The test was qualitative or at best semiquantitative; it did not reproduce the complete camera, film and viewing conditions. Its result challenges the use of blurred image structure as a physical outline. It does not identify the light outside the aircraft or erase the visual observations.
The separate 3 January ground film
M. J. McDonnell and A. D. W. Fowler's February 1979 PEL Report 632 concerns TV1 footage from the Clarence River area on 3 January. TV1 supplied two samples of the original film, from which five frames were digitised and processed. This is a separate dataset from Crockett's aircraft footage. PEL Report 632, scan pp. 88–95.
The authors concluded that the images were severely out of focus and that estimated brightness and restored angular extent were consistent with Venus. Their reconstruction fitted a model of the optical blur, adjusting parameters to reduce artefacts. The figure sequence shows the relationship between the blurred image, processing and restored result. The authors also retained insufficient resolution to distinguish physical detail. This provides a bounded optical interpretation of the ground film; it cannot identify the aircraft light recorded days earlier.
The published optical disagreement
Range association drives the calculation
Bruce S. Maccabee's August 1979 letter in Applied Optics analysed the bright return-flight source using film density, camera settings and an associated airborne radar range. He acknowledged that individual film frames could not be assigned exact radar distances because the cameraman did not record frame timings. His calculated luminous intensity and physical size therefore depended on linking the filmed light to the reported radar target, as well as on film-response and exposure assumptions. Original letter, p. 2527, continuation, p. 2528, left panel.
The calculation adds a technical treatment of the recorded brightness. Its dependence on source distance remains decisive: apparent brightness measured at the camera becomes a source-output estimate only after distance and intervening transmission are constrained. A precise numerical result cannot resolve an uncertain association in its inputs.
A stationary vessel and competing geometry
W. Ireland and M. K. Andrews's December 1979 comments disputed the exclusion of conventional sources and the inference that the source was airborne. Their reconstruction treated a bright squid-fishing vessel as a possible explanation. It drew principally on Maccabee's earlier detailed account for flight information. They also acknowledged that the relevant vessel movement routes were not known; a possible boat location remained a hypothesis. Comments, p. 3889, right panel, p. 3890 and geometry figure, left panel.
Maccabee's June 1980 reply contested that geometry and revised parts of his original reconstruction. His footnote 6 explicitly identifies the reported three-minute delay before radar operation as his own investigative error: the radar had already been warmed in standby. He also replaced an estimated radar sweep angle with a measured one. These changes matter because they alter the sequence used to associate the light with the radar indication. Reply, p. 1745, right panel, p. 1746, footnote 6 and figure.
The reply's figure makes uncertainty in remembered ranges, angles and radar-loss timing visible. Revised witness interpretation is not an independently recorded trajectory. The disagreement consequently turns on geometry and source association as much as photometry. The journal's decision to close correspondence concerned its subject remit; it did not adjudicate the source's identity. Neither Maccabee's Navy affiliation nor the comments' DSIR affiliation settles that question.
Documentary chronology
- December 1978: Earlier aircraft and controller reports attract official attention. Clements uses a 20–21 December grouping; Randle's relevant summary and notes identify 22 December.
- 30–31 December 1978: The television-accompanied Argosy flies from Wellington to Christchurch and returns north. The principal bright-source filming occurs on the return leg.
- 3 January 1979: TV1 records separate ground footage near the Clarence River.
- 8 January 1979: Ireland observes radar indications at Wellington during a later comparison session; his note is dated 9 January. This session cannot substitute for the December radar record.
- 11 January 1979: Clements signs the official investigation report.
- January–February 1979: PEL Reports 625 and 632 address aircraft-window effects and the ground film respectively.
- 1 August 1979: Maccabee's photometric letter is published.
- 1 December 1979: Ireland and Andrews's comments are published.
- 1 June 1980: Maccabee's reply appears, correcting inputs and maintaining disagreement with a stationary-vessel interpretation.
- December 2010: The reproduced file's NZDF wrapper records the public release. Publication, report and observation dates remain distinct.
Evidence assessment
The records establish reported lights, radar indications, filming and substantive investigation. They also show that association varies by flight segment and that some apparent image structure arises within the imaging system. A defensible explanation must account for a defined observation using its own times, sightlines and instrument conditions. Combining every report under one physical object would discard the distinctions preserved in the contemporary record.
Photometry alone cannot decide whether the principal source moved or remained stationary when its distance depends on an unresolved light/radar association. An assigned range affects the calculated output and the reconstructed physical size, while a changed timing or bearing affects the proposed trajectory. Maccabee's corrected radar sequence therefore reopens the geometry used to interpret the brightness calculation; it supplies no independent measurement resolving that geometry. A vessel comparison and an airborne-source comparison must first establish whether they are using the same constrained observation.
Astronomical sources modified by viewing conditions
Clements proposed Venus and unusual refraction for parts of the reporting. His return-flight discussion acknowledges a timing problem: the observation preceded the planet's ordinary rise time and required enhanced refraction. A quantitative atmosphere and sightline reconstruction would be needed to test that proposal. PEL 632 supplies a separate Venus-compatible result for the later ground film, whose date and viewing position cannot be transferred to the flight.
Bright vessels or coastal lights
A vessel is a substantive candidate because fishing lights could supply strong illumination and because surface sources interact with aircraft motion. Ireland and Andrews developed that possibility rather than documenting a specific vessel at the inferred point. Maccabee disputed both the positional fit and the available vessel evidence. Original movement records and a reconstruction allowing the uncertainty in pilot bearings would distinguish compatibility from identification.
Radar anomalies accompanying unrelated lights
The official file documents recurring radar anomalies and later instrument observations. Ireland's 8 January note distinguishes display modes and records both unidentified echoes and likely ships; he explicitly leaves the effect of equipment modifications unverified. These findings justify examination of radar propagation and processing, but the later session does not establish the cause of a particular December contact. A visual source and an unrelated radar indication could coincide in bearing without sharing range. Radar observation and annotated displays, scan pp. 167–169.
A common unidentified physical source
The crew's perceived association and Maccabee's reconstruction warrant evaluating a common source. Establishing it requires more than a large calculated output conditional on that association. An authenticated sequence connecting film frames, radar returns and aircraft attitude would constrain range, movement and source height. The available document comparison does not determine those properties or establish an extraordinary origin.
Assessment snapshot
| Proposition | Assessment | Basis and limit |
|---|---|---|
| Crew reported unusual lights during the flight | Documented reporting event | Contemporary interview summary and official compilation; some joint interviewing |
| Every light was confirmed by airborne and ground radar | Not established | Instrument use and reported correspondence differ by segment |
| Image streaks describe physical object structure | Unsupported inference | Rumsey's window test reproduces relevant artefacts under different apparatus conditions |
| Ground-film analysis identifies the aircraft source | Incorrect dataset transfer | PEL 632 concerns 3 January ground film |
| Published photometry establishes source size and output | Conditional | Range association, film response and exposure assumptions determine the result |
| Every observation was officially resolved | Not established | Clements retains inconclusive material; published interpretations remain disputed |
Claim ledger
| Claim | Record and locator | Evidential role | Further discriminator |
|---|---|---|---|
| Principal light and airborne return share one source | Startup summary 172–173; optical letters | Participant association and competing reconstruction | Synchronised film, radar, aircraft attitude and timing |
| The source remained stationary | Ireland/Andrews 3890 | Proposed vessel interpretation | Original vessel position and flight geometry with uncertainties |
| Radar began three minutes after sighting | Maccabee 2528, corrected 1746 note 6 | Superseded investigative statement | Reply explicitly records standby/warmed condition |
| Photographed streaks are physical detail | Rumsey 121–122 | Optical-system test | Original-window and camera-specific reproduction |
| Restored ground image is Venus-compatible | McDonnell/Fowler 88–95 | Separate film analysis | Verified ground pointing, timings and original calibration |
| Earlier flight date is settled | Clements 148; Randle 178–179 | Conflicting contemporary labels | Flight logs and original interview dates |
Open questions
Synchronising the film and sensors
The original soundtrack, film sequence and operational records could determine which intervals genuinely coincide. Exact frame-to-time mapping is especially relevant to brightness calculations made using remembered or bracketed radar ranges.
Identifying the camera and lens configuration
Rumsey and Maccabee describe differing lens settings. Establishing which film sections and equipment each author addressed would be necessary before transferring their numerical assumptions between analyses.
Establishing the physical sightline
Aircraft position, attitude and bearings should be reconstructed with explicit errors. A surface-vessel model, astronomical refraction model and airborne-source model must be tested against the same observations rather than against different selections from the narrative.
Further research
Witness record
Aircraft crew and filming
Startup and First Officer Guard appear in the inspected interview summary for the television-accompanied flight. Randle's summary and notes concern earlier reporting and remain a separate testimony record. Crockett is identified as the aircraft cameraman in Rumsey's technical report. Statements attributed to those participants through an investigator retain that documentary mediation.
Investigators and technical authors
Clements authored the official synthesis. Rumsey tested the windows; McDonnell and Fowler analysed the separate ground footage. Ireland participated in later radar observation and, with Andrews, disputed Maccabee's aircraft-film interpretation. Maccabee's letters record his own technical analysis and revisions. Those technical roles differ from firsthand observation aboard the aircraft.
Source library
Contemporary official records
- Clements report and map annexes, scan pp. 147–157: signed inquiry report, scope, chronology and qualified conclusions. The private mirror reproduces an NZDF release; its filename's NARA wording does not establish NARA authorship or custody.
- Startup interview summary, scan pp. 172–173: selected underlying testimony and cross-sensor uncertainties.
- Randle records, scan pp. 178–180: interview summary, inflight notes and subsequent speculation, with a material date discrepancy.
Technical investigations
- Rumsey, PEL 625, scan pp. 120–122: aircraft-window test, January 1979.
- McDonnell and Fowler, PEL 632, scan pp. 88–95: analysis and figures for the separate 3 January ground film, February 1979.
- Ireland radar note and photographs, scan pp. 167–169: subsequent comparison session, 8 January 1979.
Original published exchange
- Maccabee 1979, publisher record: Applied Optics 18(15), 2527–2528, DOI 10.1364/AO.18.002527.
- Ireland and Andrews 1979, publisher record: Applied Optics 18(23), 3889–3890, DOI 10.1364/AO.18.003889.
- Maccabee 1980 reply, publisher record: Applied Optics 19(11), 1745–1746, DOI 10.1364/AO.19.001745.
- Archived author page with all six journal pages: source of the inspected scan reproductions. Later advocacy on the same page is separate from the original letters and supplies no independent observation chain.