Minimum level of taxonomic identification of benthic invertebrates by-catches on Ifremer's fisheries vessel
|Author(s)||Hazevis Gaëtan1, Baudrier Jerome1, Vaz Sandrine2|
|Affiliation(s)||1 : Ifremer Nantes, Département Océanographie et Dynamique des Ecosystèmes, Service Valorisation de l’Information pour la Gestion Intégrée Et la Surveillance
2 : Unité MARine Biodiversity Exploitation and Conservation, Ifremer Sète, Université de Montpellier, IRD, CNRS
|Keyword(s)||Marine Strategy Framework Directive, benthic fauna, fisheries research surveys, taxonomic level, identification|
Ifremer conducts numerous fisheries surveys dedicated to benthic and demersal populations (commercial / non-commercial fishes and invertebrates). For several years, in application of the ecosystem approach, all benthic invertebrate fauna collected in fishing gear has been systematically monitored: megabenthic invertebrates captured have been sorted, identified, counted and weighted. All these surveys are based on fixed or random stratified sampling strategy with varying intensity depending on the covered survey area. These data are stored, in historical access-based databases or for the most recent years in the centralised “Harmonie” database held in the Ifremer Fishery Information Systeme (SIH). The species nomenclature used was standardized using WoRMS database.
Taxa caught at least once a year are listed for each monitoring area on the basis of already available data series. In order to facilitate the identification of individuals sampled on board vessels and to improve the training of onboard scientists, the present work aims to define the minimum level of identification for each of them. The analysis identifies taxa that appears recurrently on available historical series or gathers them on less precise taxonomic levels if this is not the case, which may indicate potential identification difficulties. The following procedure was used: all taxa expressed at the species level were first aggregated at genus level if they occurred less 90% of the years over the available time series. For MEDITS, EPIBENGOL and ORHAGO, the occurrence threshold was set to 70% and to only 50% for NOURMONT because the datasets were less than 10 years long. Then to be kept at that taxonomic level, a given genus had to be observed over 90% of the time (for example over at least 9 years if the dataset contains 10 years). Otherwise it was iteratively regrouped into a higher taxonomic level (family, order, class, division) following the same criteria (Foveau et al, 2017). For instance, for the NOURSEINE survey, this resulted into the aggregation of the 103 origin taxa into 35 taxonomic groups. The name of the final taxon after data processing represents the minimum level of identification defined by the analysis. However, these results are very theoretical. This is why they were sent to scientists who embark regularly in order to refine the level of taxonomic identification with field experience.
The first dataset is composed of 8 tables relevant to the different vessel surveys. The first column of each table represents the permanent code of the taxon in the Ifremer taxonomic referential, the second the systematic number and the third the species abbreviated code. The other columns are the different taxonomic levels of the taxon. The minimum level of identification at sea defined by the data processing appears in blue. The level determined by feedback of scientist’s field experience, which is the one to use at sea, appears in green. The second dataset summaries the results detailed in the first table and indicates directly for each taxon identified to far, the minimum level of identification required for the benthic invertebrates by-catch of each fisheries surveys studied.
|Utilisation||Data are published without any warranty, express or implied. The user assumes all risk arising from his/her use of data. Data are intended to be research-quality, but it is possible that these data themselves contain errors. It is the sole responsibility of the user to assess if the data are appropriate for his/her use, and to interpret the data, data quality, and data accuracy accordingly. Authors welcome users to ask questions and report problems.|
|Acknowledgements||The authors want to thank all the scientific staff and vessel crews who collect data used in this study and everyone who contributed to the proposition of the minimum level of taxonomic identification.|
Data provided in these files are raw data processed. Raw data are collected by beam trawl on ORHAGO and nursery surveys, Very High Vertical Opening (VHVO) bottom trawls (or “GOV”) on IBTS, CGFS, CAMANOC, EVHOE, MEDITS, EPIBENGOL surveys and dredge on COMOR survey. Fisheries vessel surveys are carried out either on board of Research Vessels or professional fishing vessels where sampling gears are well adapted for catching demersal species (in particular fish and cephalopods). Temporal series used on this study are the following ones: