Reef fish abundance, size and biomass from Aldabra Atoll (Seychelles) between 2014 and 2021 collected by the Seychelles Islands Foundation.

Data ownershipAll data is owned by the Seychelles Islands Foundation (SIF), the management authority of Aldabra Atoll. SIF must be cited alongside all authors in any works using all or parts of this data.
Data collectionData was collected at Aldabra Atoll (Seychelles) by SIF staff and volunteers as part of the SIF Aldabra Reef Monitoring (ARM) programme at 12 monitoring sites between 2014 and 2021. Nine sites are on the seaward reef with permanently marked transects at both 5 m and 15 m water depth; three sites are inside the lagoon with permanently marked transects at 2 m water depth. Fish visual census surveys were conducted along one to three 50 × 4 m belt transects (i.e., one transect = 200 m2; 10 m gap between replicate transects) marked by surveyors with tape measures. The first transect was always the permanently marked transect; subsequent replicate transects were surveyed along the same compass bearing and following the same depth contour as the permanent transect. All transects of the same depth at each site were surveyed during the same dive.
Two survey methods were used:
Indicator surveys: in 2014, 2016–2019 and 2021 all fishes of a pre-defined list of 84 taxa (63 of which were identified to species-, six to genus- and 14 to family/sub-family level) were counted and classified into one of five size classes based on total length (i.e., < 10 cm, 10–19 cm, 20–29 cm, 30–39 cm, > 39 cm). Selection of taxa for these ‘indicator’ surveys followed Obura and Grimsditch (2009) and included taxa that represent major functional groups, and are sensitive to either anthropogenic pressures, e.g. fishing, or to environmental change, e.g. coral bleaching.Extended surveys: In 2015 and 2020, all diurnally active, reef-associated fishes were identified to species level whenever possible (198 taxa in total, 182 of which were identified to species-, six to genus- and 10 to family-level), counted and estimated to the nearest cm when ≥ 8 cm.With both survey methods, large, more mobile fish were surveyed on the first pass of the transect, while smaller more site-attached fish were surveyed on the return pass. Each year, surveys were done by different surveyors but all surveyors underwent basic (indicator surveys) or extensive training (extended surveys; whereby the 2015 surveyor had substantial prior survey experience). Training for both methods included desk-based fish identification training, a written fish identification exam, and in-water practical survey training of fish identification, size estimation and surveying, which was done more frequently and intensively for the extended surveys.
Data processingFish taxa were assigned to one of six functional groups based on broad feeding habits (piscivores, omnivores, invertivores, corallivores, planktivores, herbivores) following Samoilys et al. (2018) and Froese and Pauly (2022).Fish biomass was estimated using the length-weight equation W = a x Lb (Cren, 1951). W is weight, L is length and a and b are published species-specific constants (Froese and Pauly, 2022). For data obtained through the indicator surveys, the midpoint of each size class was used as length estimation. To approximate the length of fishes within the largest size class (i.e., > 39 cm), the median size within that size class was calculated from the size estimates of the extended surveys. To obtain the species-specific constants (a and b) for fish that could not be identified to species level but only to genus or family, constants from all species of that genus or family present at Aldabra (based on Friedlander et al., 2015) were averaged.AcknowledgmentsWe thank current and former staff members of the Seychelles Islands Foundation for their work and support in data collection. Specifically, we thank Jude Brice, Heather Richards, Samuel Basset, Rebecca Filippin, Joel Bonne, Marvin Roseline, Leeroy Estrale, Ronny Marie, Lorraine Cook, Selwyn Laurence, Albert Belmont, Jeremy Raguain, Janske van de Crommenacker, Germano Soru, Sophie Adams, Julio Agricole, Mikael Esparon, Guilly Mellie, Jilani Suleman, Martin van Rooyen, Bruno Mels, Edward Constance, Jessica Constance, Alex Rose, Rickpert Woodcock, Frances Benstrong and Veronique Banane for support in the field. We further thank the Global Environment Facility for funding diving and research equipment and the development of the Aldabra Reef Monitoring programme. We also thank Anthony Bernard and Nick Riddin from the South African Institute for Aquatic Biodiversity and SIF staff for assistance in establishing the permanent benthic transects; and Rebecca Klaus for advice on benthic transect set-up.
FundingThe development and set-up of the Aldabra Reef Monitoring (ARM) programme was funded by the Global Environmental Facility (GEF Project ID 3925). The annual ARM surveys were funded by the Seychelles Islands Foundation.
ReferencesCren, E. D. L. (1951). The Length-Weight Relationship and Seasonal Cycle in Gonad Weight and Condition in the Perch (Perca fluviatilis). The Journal of Animal Ecology 20, 201. doi: 10.2307/1540Friedlander, A. M., Ballesteros, E., Beets, J., Brown, E. K., Fay, J., Haupt, P., et al. (2015). Biodiversity and ecosystem health of the Aldabra Group, southern Seychelles: Scientific report to the government of Seychelles. National Geographic Pristine Seas. Available at: https://media.nationalgeographic.org/assets/file/PristineSeasSeychellesScientificReport.pdf.Froese, R., and Pauly, D. (2022). FishBase. World Wide Web electronic publication. Available at: www.fishbase.orgObura, D., and Grimsditch, G. (2009). Resilience assessment of coral reefs: assessment protocol for coral reefs, focusing on coral bleaching and thermal stress. Available at: http://cmsdata.iucn.org/downloads/resilience_assessment_final.pdfSamoilys, M., Roche, R., Koldewey, H., and Turner, J. (2018). Patterns in reef fish assemblages: Insights from the Chagos Archipelago. PLoS ONE 13, e0191448. doi: 10.1371/journal.pone.0191448Correction October 2024 Please note that the previously available datafiles had wrong biomass values for Caranx melampygus. These values have been corrected in the newly uploaded datafiles.