New Insights into Geometric Morphometry Applied to Fish Scales for Species Identification

Area of Science:

• Ichthyology
• Morphometrics
• Bioinformatics

Background:

• Accurate fish species identification is crucial across various scientific and commercial contexts.
• Dermal scale analysis offers a potentially rapid and cost-effective method for fish identification.
• Geometric morphometry presents promising techniques for analyzing scale morphology, yet comparative studies are limited.

Purpose of the Study:

• To compare the efficacy of two geometric morphometry methods for fish species recognition using dermal scales.
• To evaluate landmark-based and outline-based approaches on five distinct teleost species.
• To investigate the potential for automation in fish scale identification using these morphometric techniques.

Main Methods:

• Applied landmark-based geometric morphometry using the R package 'geomorph'.
• Developed a novel approach combining landmarks and semilandmarks for fish scale analysis.
• Utilized outline-based geometric morphometry with the R package 'Momocs'.
• Analyzed scale datasets from *Danio rerio*, *Dicentrarchus labrax*, *Mullus surmuletus*, *Sardina pilchardus*, and *Sparus aurata*.

Main Results:

• Both geometric morphometry methods successfully clustered the five teleost species.
• The landmark-based method generally yielded higher R² values for species clustering.
• The outline-based method distinguished between more species pairs than the landmark-based method.
• The outline-based method showed potential for automation due to its independence from landmark placement.

Conclusions:

• Geometric morphometry, particularly outline-based analysis, is a viable tool for fish species identification from scales.
• The outline-based method demonstrates higher potential for future automated fish identification systems.
• Further research with larger datasets could enhance the performance of outline-based geometric morphometry for fish scale analysis.