1. Why this review was done

The Aurora Dossier / Feature foundation was tested against the existing GBR website and study database. The review showed that the earlier GBR system already contains valuable working structures for observations, events, resources, bibliography, analysis, contributor participation and research synthesis. Rather than replace these, the new Aurora architecture should preserve the accumulated knowledge while separating the concepts into reusable canonical objects.

2. Main architectural result

The central principle is one shared underlying model, many purpose-specific views. Public pages should remain simple while the underlying records retain scientific depth, uncertainty and provenance.

3. What the existing GBR site has confirmed we need

4. Important lessons from the GBR observation records

• A single wide observation row was carrying several different objects: contributor, owner, source, location, observation, media, equipment, processing, rights, assessment, mapping and Study links.
• Contributor, observer, photographer, analyst, mapper and image owner must remain separate roles even when one person performs several of them.
• Visual detectability, image detectability, prominence and identification confidence are different measurements.
• Observation interval, image/frame time, mapped time and derived Feature Instance interval must remain distinct.
• Original, processed, enhanced, mapped and annotated media must retain their derivation history.
• Subtle Feature identifications may be strongly processing-dependent; processing must not overwrite the status of the original evidence.

5. 13 March 2022 multi-observer Event

This Event is a particularly useful end-to-end stress test because it contains geographically separated observations from New Zealand, Victoria and Western Australia, including mapped images, timelapse material and angular boundaries.

The architecture must distinguish same Event, same Feature type, and same physical Feature Instance. Those are not automatically equivalent.

6. Mapping and image-analysis methods

The GBR work has already demonstrated reusable analysis methods including astrometric calibration, traced ray axes, horizon/reference calibration, triangulation, synchronised timelapses, enhancement, tone profiling and mapping.

These should be recorded as Analysis Methods, with derived Feature Traces and Measurements rather than leaving the result only as a saved plot or annotated image.

7. Event context and environmental data

Existing GBR Events link to data from sources such as SDO, AMPERE, OVATION Prime, OMNI/ACE, GOES, ISGI, SOHO/LASCO, SuperDARN, SWMF and WSA-ENLIL.

The Event should link to specific Data Products rather than contain one permanent field for every possible dataset. Environmental coexistence is evidence for analysis and correlation, not automatic evidence of causation.

8. Coronal-hole and disturbance history

The existing Coronal Hole History is more than a resource table. It is a longitudinal record of upstream Driver / Disturbance Episodes, including source location, disturbance intervals, Kp/ap response, solar-wind speed, recurrence and competing CME influences.

9. Bibliography and mechanisms

The bibliography should separate Publication, Study–Publication Link, and Literature Claim. A separate Mechanism library should record candidate or established physical processes, their prerequisites, outputs, predicted signatures, supporting evidence, alternatives and maturity.

10. Provisional blue-ray classification

The current IBR / BTR / TBR / GBR working terminology should be preserved as a provisional research classification, not presented as an established universal taxonomy.

“Blue Rays” is better treated as a Feature Family / working comparison group. Membership based on colour and morphology does not imply a shared genesis.

11. Contributor and submission workflow

The front-end form can remain simple while the backend creates linked Observation, Media, Acquisition, Location, Rights and Contributor-role records.

12. Living Study Summary

The existing Details page performs an important role: it is a dated statement of what the Study can responsibly say at that time. It should become a versioned Living Study Summary rather than a page that is continuously overwritten.

13. Project history

The GBR Study history should be retained as structured Study Milestones, including first awareness, key Events, calls for submissions, Aurorasaurus engagement, Study group formation, analysis-method development, measurement work and publication decisions.

14. Scientific discipline retained throughout

Association must never be silently upgraded to causation, and a Feature can be strongly established observationally while its genesis remains unresolved.

15. Recommended next steps when the GBR work resumes

1. Use the 13 March 2022 Event as the first end-to-end migration/stress test.
2. Map its observations, media, mapped times and angular measurements into the new canonical structure.
3. Preserve original and processed image versions and record analysis methods.
4. Build candidate Feature Instance matching across observers.
5. Link Event-context Data Products rather than copying plots into fixed Event columns.
6. Convert current descriptive GBR statements into parameter values, Claims, Findings or Hypotheses according to evidence status.
7. Retain the current blue-ray classification as provisional and comparison-oriented.
8. Version the current public/pre-publication Study summary before changing it.
9. Do not rebuild successful contribution and assessment workflows unnecessarily; refine their underlying data model instead.

16. Key takeaway

The review has not shown that the older GBR work was wrongly structured. It has shown that the Study had already accumulated many of the pieces now needed by the wider Aurora platform. The opportunity is to separate, link and generalise those pieces so they can be reused across Features, Events, Observations and future Studies.

Model richly, reveal selectively, preserve provenance, and let real Events continue to refine the architecture.