Reef Fish

Quantifying fish community composition, biomass, and trophic structure

Introduction

Reef fish are key indicators of ecosystem health, providing insights into marine productivity, biodiversity patterns, and human impacts. The Pristine Seas fish belt transect (BLT) methodology provides standardized protocols for quantifying fish community structure, biomass distributions, and trophic organization across different habitats and depths. These surveys form a cornerstone of our ecosystem assessments, enabling quantitative comparisons of fish assemblages across geographies.

Method Overview

The fish belt transect method employs a standardized visual census approach to systematically document fish abundance, size structure, and species composition. This method has been refined over decades of reef monitoring and provides robust, quantitative data suitable for global comparisons and trend analyses.

Key Features

  • Paired divers: Two fish specialists conducting parallel surveys
  • Multi-size census: Dedicated passes for large and small-bodied fishes
  • Precise size estimation: Length measurements to nearest centimeter
  • Complete taxonomic assessment: All visible fish species recorded
  • Biomass calculation: Size-based biomass estimates for trophic and functional analyses
  • Depth stratification: Consistent sampling at standard depth strata
  • Replicated design: Multiple transects per station for statistical robustness

Field Implementation

Team Structure

Fish surveys require dedicated specialists who focus exclusively on fish observations. Unlike the benthic team, which divides labor by taxonomic expertise, the fish team comprises:

  • Two fish divers working independently at each depth stratum
  • Each diver conducts three replicate transects
  • Divers typically alternate between depth strata on successive dives

This approach maximizes sampling efficiency while ensuring robust sample replication.

Fish Belt Transect Specifications
  • Transect dimensions: 25 m × 4 m for large fish (≥20 cm TL); 25 m × 2 m for small fish (<20 cm TL)
  • Replication: 3 transects per diver per depth stratum
  • Depth strata: Typically 10 m and 20 m (with 5 m stratum added when appropriate)
  • Direction: Parallel to shoreline at constant depth
  • Separation: 5+ m between adjacent transects
  • Total area: 300 m² for large fish and 150 m² for small fish per station

Collection Procedure

  1. Site preparation:
    • Select site with appropriate habitat at target depth
    • Ensure sufficient separation from benthic team operations
  2. Transect deployment:
    • Deploy 25 m transect tape parallel to shoreline
    • Maintain consistent depth throughout transect length
  3. Survey execution:
    • First pass (outward swim):
      • Target all fish ≥20 cm TL within 2 m on either side of the transect (4 m total width)
      • Swim at a consistent pace (~8-10 minutes per transect)
      • Identify all fish to species level
      • Estimate total length to nearest cm
      • Count individuals and record observations on slate
    • Second pass (return swim):
      • Target all fish <20 cm TL within 1 m on either side of the transect (2 m total width)
      • Maintain consistent pace with careful attention to substrate
      • Identify all fish to species level
      • Estimate total length to nearest cm
      • Count individuals and record observations on slate
  4. Repeat process:
    • Complete three replicate transects at the same depth stratum
    • Maintain adequate separation between transects (minimum 5 m)
    • Ensure transects sample representative and consistent habitat
  5. Photo documentation:
    • Photograph unusual or difficult-to-identify species
    • Document general habitat conditions
    • Capture images of notable aggregations or behaviors
  6. Survey completion:
    • Conduct general site exploration if time permits
    • Note presence of significant species not observed on transects
    • Join the benthic team and exit the water together

Accurate size estimation is critical for biomass calculation and population structure assessment. Fish divers use several strategies to maintain accuracy:

  • Regular calibration: Practice with objects of known size before each expedition
  • Reference standards: Use dive slate markings for direct comparison
  • Progressive bracketing: Categorize fish into broad size groups initially, then refine by specific cm
  • Consistency checks: Compare size estimates between divers periodically
  • Photo validation: Use photography to validate sizes of unusual specimens

Taxonomic Framework

Precise species identification and classification are fundamental to the fish survey methodology. Pristine Seas uses a comprehensive taxonomic system that balances field practicality with scientific rigor. This framework connects underwater observations to global taxonomic databases, with the World Register of Marine Species (WoRMS) AphiaID serving as the authoritative identifier throughout all datasets.

Identification Process

Our divers achieve species-level identification for over 90% of individuals in real-time, using a structured approach for challenging identifications:

  1. Field observation: Record distinguishing features when species identity is uncertain
  2. Reference consultation: Compare observations with identification guides post-dive
  3. Team verification: Review difficult identifications collectively
  4. Photo documentation: Capture images of unusual specimens for later analysis
Identification Resources

Field teams use multiple resources to support accurate species identification:

  • Pre-expedition training: Review regional species lists and identification challenges
  • Field guides: Region-specific printed and digital references available onboard
  • Digital libraries: Offline mobile applications with comprehensive visual databases
  • Photography: Systematic image collection of uncertain identifications
  • Post-dive consensus: Regular team discussions to standardize identification approaches

Field Code System

For efficient underwater data collection and rapid at-sea data entry, Pristine Seas employs a specialized field code system. These codes serve as temporary identifiers that link field observations to formal taxonomy.

Taxon Code Structure

Our field code (taxon_code) system uses a consistent format that makes underwater data recording efficient while maintaining taxonomic precision.

Species-Level Codes
GEN2.SPEC4 — First 2 letters of genus + first 4 of species (uppercase)

  • Acanthurus tristisAC.TRIS
  • Apogon tricolorAP.TRIC
  • Anthias tricolorAN.TRIC

Handling Duplicates
When multiple species would share the same code:

  • Most common taxon keeps the default code
  • Others extend the genus or species portion to ensure uniqueness

Examples:

  • Apogon tristisAP.TRIS
  • Aplodactylus tristisAPL.TRIS
  • Labroides bilineatusLA.BILI
  • Labroides bilinearisLA.BILIN

Genus and Family Codes
Used when species-level identification isn’t possible:

  • Genus → GEN4.SP (e.g., Labroides sp. → LABR.SP)
  • Family → FAM4.SPP (e.g., Labridae spp. → LABR.SPP)

Hybrids
Hybrid taxa use extended format: GEN2.SPxSP

Examples:

  • Acanthurus achilles × nigricansAC.ACxNI
  • Acanthurus olivaceus × nigricansAC.OLxNI

These conventions ensure clean data joins, traceability, and consistent taxonomy across all datasets.

Reference Database

Our two-tier database system translates field observations into analysis-ready data:

  1. taxonomy.uvs_fish_codes: Maps field codes to formal taxonomy
    • Connects temporary taxon_code to accepted scientific names
    • Links each code to its authoritative AphiaID
    • Includes regional variants and handles identification edge cases (e.g., hybrids)
    • Maintains historical continuity across expeditions
  2. taxonomy.fish: Comprehensive species trait database with AphiaID as the primary key
    • Taxonomic identifiers: AphiaID (WoRMS), SpecCode (Fishbase), SIS_ID (IUCN)
    • Taxonomic classification: Complete tree from phylum to species
    • Common names: Species and family common names (in English)
    • Trophic information: Trophic group and level, feeding guild, and diet composition
    • Morphometrics: Length-weight relationships, maximum size, and growth parameters
    • Ecological context: Habitat associations and fishery importance
    • Conservation status: IUCN categorization

This integrated system enables seamless workflow from underwater observations to sophisticated ecological analyses while maintaining taxonomic precision and compatibility with global biodiversity databases.

Ecological Classifications

All fish species in the Pristine Seas database are assigned to standardized trophic groups and functional categories based on their ecological roles. This classification system provides a foundation for analyzing reef community structure and ecosystem function.

Trophic Groups

We use five primary trophic categories that reflect feeding ecology and position in the food web:

Herbivore/Detritivore
Species that primarily consume plant material or detritus
Examples: Parrotfishes (Scaridae), surgeonfishes (Acanthuridae), rabbitfishes (Siganidae)

Planktivore
Species that primarily consume planktonic organisms
Examples: Fusiliers (Caesionidae), chromis (Pomacentridae), anthias (Anthiadinae)

Lower-Carnivore
Species that consume mainly invertebrates and small fishes
Examples: Wrasses (Labridae), snappers (Lutjanidae), emperors (Lethrinidae)

Top-Predator
Non-shark predatory species primarily consuming other fishes
Examples: Groupers (Epinephelidae), barracuda (Sphyraenidae), jacks (Carangidae)

Shark
All sharks, regardless of feeding ecology
Examples: Reef sharks (Carcharhinidae), nurse sharks (Ginglymostomatidae), hammerheads (Sphyrnidae)

These categories allow us to analyze trophic structure across reef ecosystems and assess the impacts of fishing pressure, which typically affects top predators and sharks first, followed by lower trophic levels.

Herbivore Functional Groups

Herbivorous fishes play critical roles in coral reef resilience by controlling algae that compete with corals. We further classify herbivores into four functional groups based on their feeding mode:

Browsers
Selectively feed on macroalgae, removing entire plants
Examples: Naso unicornis, many Kyphosus species

Grazers
Crop algal turf, creating patches of cropped substrate
Examples: Many Acanthurus species, Zebrasoma

Scrapers
Remove algae and associated sediment and detritus
Examples: Many Scarinae (parrotfishes)

Excavators
Remove chunks of substrate while feeding, creating space for coral recruitment
Examples: Large-bodied Scarinae, Chlorurus species

This functional classification enables detailed analysis of how different types of herbivores contribute to reef processes, particularly coral-algal dynamics and reef recovery potential after disturbances.

Data Workflow

The Pristine Seas fish survey methodology follows a systematic workflow from field collection to analysis, ensuring data quality and scientific rigor at each stage.

Data Entry

Fish survey data are recorded on underwater slates during dives and transferred to standardized Excel fieldbooks on the same day. This immediate transcription preserves data accuracy while memories are fresh. The ISO3_YEAR_fish_fieldbook.xlsx serves as the primary data repository for each expedition with four key components:

Fieldbook Structure
  • Readme: Expedition details, data entry protocols, and guidelines
  • Stations: Site information including depth, survey effort, and habitat type
  • Observations: Individual fish records with taxonomic codes, counts, and size estimates
  • Species: Reference list of fish species with taxonomic information

The observations sheet includes built-in validation rules that flag potential errors such as:

  • Species codes not in the reference list
  • Size estimates exceeding known maximum lengths
  • Missing or inconsistent data fields

This real-time validation enables immediate correction while field observations remain clear in divers’ memories.

Processing Pipeline

Our data processing transforms raw field observations into analysis-ready datasets through four sequential phases:

1. Consolidation and Standardization

First, we combine data from multiple divers and expedition legs into a unified dataset:

  • Convert field codes to accepted scientific names and AphiaIDs
  • Standardize size measurements and abundance records
  • Add spatial metadata from the site documentation
  • Join species traits from the taxonomy.fish reference table using AphiaID
  • Create a complete dataset that preserves all original observations

2. Quality Control

Next, we verify data quality through a systematic review process:

  • Analyze length distributions to identify unlikely size measurements
  • Compare species lists between divers and stations for consistency
  • Check geographic ranges against known species distributions
  • Verify taxonomic classifications against global databases

Potential issues are flagged for expert review rather than automatically removed, preserving data integrity while ensuring scientific accuracy.

3. Biomass Calculation

We then transform fish lengths into biomass estimates using the standard length-weight relationship:

\[W = a \times L^b\]

Where \(W\) is weight in grams, \(L\) is total length in centimeters, and \(a\) and \(b\) are species-specific constants from our reference database. For species without specific parameters, we use coefficients from closely related species or family-level relationships.

A critical step in this process is the addition of zero values for species that were not observed on a particular transect but are known to occur in the region. This ensures that absence data is properly incorporated into average calculations, preventing biased estimates of abundance and biomass. Zero values are applied:

  • After taxonomic standardization but before aggregation
  • Based on the regional species pool for each location
  • Only for species that could reasonably occur in the sampled habitat type
  • Prior to calculating transect-level and station-level metrics

This process generates standardized metrics including:

  • Fish abundance (individuals/m²) by species and functional group
  • Biomass (g/m²) at various taxonomic and ecological levels
  • Community structure indicators like diversity indices

4. Database Integration

Finally, we format the processed data for ingestion into our central database, creating four core output files:

  1. blt_observations: Individual-level records with taxonomic classification and biomass
  2. blt_biomass_by_taxa: Abundance and biomass aggregated by species and station
  3. blt_stations: Site-level metadata and summary statistics
  4. blt_taxa_summary: Regional patterns of occurrence, abundance, and biomass

These standardized outputs allow consistent analysis across expeditions and regions while maintaining connections to the original field observations.

Exploratory Data Analysis

Our analysis examines fish communities through multiple complementary perspectives:

Community Structure

We analyze biodiversity patterns through various metrics including:

  • Species richness and diversity indices
  • Community composition visualizations (NMDS, cluster analysis)
  • Dominant species identification and rank-abundance patterns

These analyses provide insights into ecosystem complexity and community organization at different sites.

Biomass and Trophic Organization

We examine how biomass is distributed across trophic levels to understand ecosystem function. Key analyses include:

  • Proportional biomass of different trophic groups
  • Contribution of top predators and sharks to total fish biomass
  • Comparison with baseline values from remote, unfished locations
  • Assessment of key species for local economies and food security

These patterns reveal fishing impacts and help evaluate ecosystem integrity.

Size Structure

Fish size distributions provide insights into population health and fishing pressure:

  • Length-frequency distributions for target species
  • Mean size comparisons across protection levels
  • Size spectra analysis to detect fishing impacts
  • Growth and recruitment pattern assessment

Size structure analysis is particularly valuable for identifying overfishing signals and evaluating protection effectiveness.

Spatial Patterns

By analyzing how fish communities vary across space, we identify:

  • Biomass hotspots and priority conservation areas
  • Habitat associations for key species
  • Depth-related changes in community structure
  • Environmental correlates of fish distribution

Cross-Method Integration

We combine fish data with other survey components to reveal ecosystem relationships:

  • Correlations between coral cover and fish diversity/abundance
  • Relationships between herbivorous fish and algal communities
  • Connections between benthic habitat complexity and fish biomass
  • Potential trophic cascades linking predators, herbivores, and reef algae

These integrated analyses provide a holistic understanding of reef ecosystem function and reveal the ecological processes that drive reef health and resilience.

Through these complementary analytical approaches, we transform field observations into ecological insights that inform conservation planning and contribute to our understanding of marine ecosystems—from individual reefs to global patterns.

Limitations

While belt transects are an efficient and widely used method for assessing reef fish communities, several limitations should be considered:

  • Diver influence: Some fish species actively avoid or are attracted to divers
  • Cryptic species: Small, hidden, or nocturnal fish are typically underrepresented
  • Spatial variability: Fish distribution patchiness requires adequate replication
  • Temporal variation: Fish communities change throughout the day and seasons
  • Observer differences: Individual divers may vary in species identification and size estimation
  • Depth constraints: Standard SCUBA limits surveys to relatively shallow depths
  • Coverage limits: Transects sample only a small portion of the total habitat area

We address these limitations through standardized protocols, thorough diver training, adequate replication, and by combining fish surveys with complementary methods like environmental DNA and baited remote underwater video systems.

References

Caldwell ZR, Zgliczynski BJ, Williams GJ, Sandin SA (2016) Reef Fish Survey Techniques: Assessing the Potential for Standardizing Methodologies. PLOS ONE 11(4): e0153066. https://doi.org/10.1371/journal.pone.0153066

Friedlander AM, Sandin SA, DeMartini EE, Sala E (2010) Spatial patterns of the structure of reef fish assemblages at a pristine atoll in the central Pacific. Marine Ecology Progress Series 410: 219-231. https://doi.org/10.3354/meps08634

MacNeil MA, Graham NAJ, Cinner JE, Wilson SK, Williams ID, Maina J, Newman S, Friedlander AM, Jupiter S, Polunin NVC, McClanahan TR (2015) Recovery potential of the world’s coral reef fishes. Nature 520(7547): 341-344. https://doi.org/10.1038/nature14358

Morais RA, Bellwood DR (2018) Global drivers of reef fish growth. Fish and Fisheries 19(5): 874-889. https://doi.org/10.1111/faf.12297