Data archive: Spatio-temporal changes in fish assemblages in the Bay of Biscay and Celtic Sea.

The  biological data sets came from the French international bottom trawl  survey (EVHOE) carried out annually during autumn to evaluate demersal  fish resources in the Bay of Biscay (BoB) since 1987 and the Celtic Sea  (CS) since 1997 (Laffargue et al. 2021). The BoB, which stretches from  Spain to Armorica, is an intracontinental sea that is largely open to  the Atlantic Ocean. The French part of the BoB continental shelf (80 000  km²) is narrow in the south and becomes broader in the north mainly  influenced by the warm water of the Gulf Stream (Palter, 2015). The  epicontinental CS is open to the Atlantic Ocean, stretching between  Ireland, Wales, British Cornwall and Armorican Brittany. We compiled  the presence/absence and abundance data for the period 1997-2018, the  most homogenous time series in terms of research vessel, taxonomic  identification and gear (GOV 36/47, the opening is 20 m horizontally at  the wings and 4 m vertically). The time series is continuous, except for  2017 due to a technical problem, and the number of sampling stations  varied between 119 and 158 per year (n= 2957  in total). Our data set included 180 fish (Actinopterygii) species or  genera after grouping taxa that could not be unambiguously identified at  the species level for the whole time series. For example, Trachurus mediterraneus and Trachurus trachurus were merged into Trachurus sp. We analysed taxonomic diversity for 171 benthic and demersal species/genera. We excluded from diversity calculations, nine of the most abundant small and medium-sized pelagic species (Alosa  alosa, Alosa fallax, Atherina presbyter, Engraulis encrasicolus,  Sardina pilchardus, Scomber japonicus, Scomber scombrus, Sprattus  sprattus, Trachurus sp.)  because the bottom trawl used in EVHOE has a 4 m vertical opening,  which leads to low catchability and thus unreliable spatial patterns  (Laffargue et al., 2021). We  used the ICES statistical rectangle resolution (1° longitude x 0.5°  latitude, ICES, 2019) to analyse spatial patterns (74 rectangles) and the full data set included 1242 ICES rectangles sampled from 1997 to 2018. ICES rectangles corresponded to the highest spatial resolution available for fishing data. We controlled for the temporal imbalance  in sampling effort (1 to 10 stations per rectangle per year) and the  increasing number of stations over the years (i.e. increasing sampling effort over time, rSpearman = 0.55, p = 0.01), which biases the temporal trends of biodiversity indices (e.g. creates an artificial increase in species richness). We used a sample-based rarefaction approach consisting of randomly sampling 1 station per ICES rectangle for each year and repeating this process 100 times to calculate average biodiversity indices.

References:

Laffargue,  P., Delaunay, D., Badts, V., Berthele, O., Cornou, A-S., & Garren,  F. (2021). Fish and cephalopods monitoring on the Bay of Biscay and the  Celtic Sea continental shelves. Earth System Science Data Discussion, [preprint] https://doi.org/10.5194/essd-2021-146,  in review, 2021. 

Palter, J.B. (2015). The role of the Gulf Stream in European climate. Annual review of marine science, 7, 113–137. https://doi.org/10.1146/annurev-marine-010814-015656