Skip to content

The science of a fin

Fenua FINdex is built on peer-reviewed research showing that a blacktip reef shark's dorsal fin is a natural, lifelong identifier.

The foundation

A natural barcode

The black tip of a blacktip reef shark's dorsal fin is bordered by a distinctive band of pigmentation. Research by the team behind Fenua FINdex — published in the Journal of Fish Biology — tested whether that pattern can reliably identify individuals, using a catalogue of 928 individually identified sharks from Mo'orea.

The findings are the cornerstone of this project: each shark's pattern is unique, and it remains recognisable as juveniles grow, with recaptures between two weeks and nine months confirming the pattern's stability over time.

Unique

each individual carries its own dorsal-fin pattern

Stable

patterns stayed recognisable across recaptures from 2 weeks to 9 months

928

individually identified sharks in the validation catalogue

A crucial detail

Left is not right

The same shark carries different patterns on the left and right sides of its dorsal fin. Treat the two sides as interchangeable and the same animal could enter a catalogue twice — inflating population estimates and corrupting life histories.

Fenua FINdex is designed around this asymmetry: every photograph is recorded with the side it shows, and both profiles are linked to a single individual. It is also why contributors are asked to note which side they photographed.

The technology

AI that assists — and never decides

Comparing one fin against hundreds of catalogued sharks by eye is slow, painstaking work. Researchers at James Cook University have already shown that deep neural networks can shortlist candidate matches in shark photo-identification, dramatically reducing that manual effort.

Fenua FINdex applies the same philosophy to blacktip reef sharks with a firm rule: artificial intelligence proposes, humans dispose. Every AI-suggested match is reviewed by a trained researcher before it enters a shark's record, and each identification carries its confidence and validation status. An uncertain match is never presented as fact.

01

Suggest

AI ranks the most likely catalogue matches for a new photo.

02

Validate

A researcher confirms, rejects or flags each candidate.

03

Record

Confidence and validation status are stored with every match.

The archive

A decade of data behind every match

The project draws on annual photographic collections of Mo'orea's juvenile blacktips gathered since 2016, field monitoring logs reaching back to 2013, and mapping of the island's nursery habitats.

These datasets remain the property of the research institutions that created them. Fenua FINdex integrates them under strict data-governance agreements — with EU-based hosting, encryption, role-based access and a firm rule that project data is never used to train third-party AI models.

Key research

Fenua FINdex builds directly on peer-reviewed work by project scientists and collaborators, including:

  • Lionnet L., Debaere S., Heuls H., Mourier J., Planes S. & Rummer J.L. — “A blacktip's black tip: the reliability of using dorsal-fin patterns for photo identification of blacktip reef sharks.” Journal of Fish Biology.
  • Lonati et al. — deep neural-network photo-identification of epaulette sharks across life stages, James Cook University.