EMODnet Chemistry aims to provide access to marine chemistry data sets and derived data products concerning eutrophication, ocean acidification, contaminants and litter. The chosen parameters are relevant for the Marine Strategy Framework Directive (MSFD), in particular for descriptors 5, 8, 9 and 10. The datasets contain standardized, harmonized and validated data collections from seafloor litter. Datasets concerning seafloor litter data are loaded in a central database after a semi-automated validation phase. Once loaded, a data assessment is performed in order to check data consistency and potential errors are corrected thanks to a feedback loop with data originators. EMODnet seafloor litter data and database are hosted and maintained by ‘Istituto Nazionale di Oceanografia e di Geofisica Sperimentale, Division of Oceanography (OGS/NODC)’ from Italy. For seafloor litter, the harmonized datasets contain all unrestricted EMODnet Chemistry data on seafloor litter data. Data are formatted following Guidelines and forms for gathering marine litter data, which can be found at: https://dx.doi.org/10.6092/15c0d34c-a01a-4091-91ac-7c4f561ab508 The updated vocabularies of admitted values are available at: https://vocab.seadatanet.org/search https://vocab.ices.dk/ The harmonized datasets can be downloaded as EMODnet Sea-floor litter data format version 1.0, which is a csv file, tab separated values.

210Pb, 226Ra and 137Cs were measured by non-destructive gamma spectrometry on marine sediment cores, collected during RIKEAU 2002 cruise on board r/v Thalia, on the shelf of the Bay of Biscay

Numerous reef-forming species have declined dramatically over the last century. Many of these declines have been insufficiently documented due to anecdotal or hard-to-access information. The Ross worm Sabellaria spinulosa (L.) is a tube-building polychaete that can form large mostly subtidal reefs, providing important ecosystem services such as coastal protection and habitat provision. It ranges from Scotland to Morocco and into the Mediterranean as far as the Adriatic, yet little is known about its distribution outside of the North & Wadden Seas, where it is protected under the OSPAR & HELCOM regional sea conventions respectively. As a result, online marine biodiversity information systems currently contain haphazardly distributed records of S. spinulosa. One of the objectives of the REEHAB project (http://www.honeycombworms.org) was to combine historical records with contemporary data to document changes in the distribution and abundance of the two Sabellaria species found in Europe, S. alveolata and S. spinulosa. Here we publish the result of the curation of 555 S. spinulosa sources, gathered from literature, targeted surveys, local conservation reports, museum specimens, personal communications by authors  their research teams, national biodiversity information systems (i.e. the UK National Biodiversity Network (NBN), www.nbn.org.uk) and validated citizen science observations (i.e. https://www.inaturalist.org). 56% of these records were not previously referenced in any online information system. Additionally, historic samples from Gustave Gilson were scanned for S. spinulosa information and manually entered.   The original taxonomic identification of the 40,261 S. spinulosa records has been kept. Some identification errors may however be present, particularly in the English Channel and Mediterranean where intertidal and shallow subtidal records can be mistaken for Sabellaria alveolata. A further 229 observations (16 sources) are recorded as ‘Sabellaria spp.’ as the available information did not provide an identification down to species level. Many sources reported abundances based on the semi-quantitative SACFOR scale whilst others simply noted its presence, and others still verified both its absence and presence. The result is a curated and comprehensive dataset spanning over two centuries on the past and present global distribution and abundance of S. spinulosa. Sabellaria spinulosa records projected onto a 50km grid. When SACFOR scale abundance scores were given to occurrence records, the highest abundance value per grid cell was retained.

This dataset contains all satellite altimeter wave heights above 9 m, from the following satellite missions: ERS-1, ERS-2, Topex-Poseidon (Topex only), Envisat, SARAL, Jason-1, Jason-2, Jason-3, Sentinel-3A, Sentinel-3B, Sentinel-6A, Cryosat-2, CFOSAT, SWOT. Storm event identification used the DetectHsStorm package developed by M. De Carlo and F. Ardhuin (  https://github.com/ardhuin/) . This data can be combined with modeled storm tracks (see F. Ardhuin, M. De Carlo, Storm tracks based on wave heights from LOPS WAVEWATCH III hindcast and ERA5 reanalysis, years 1991-2024, SEANOE (2025). doi: 10.17882/105148 )

This visualization product displays the size of litter in percent per net per year from specific protocols different from research and monitoring protocols. EMODnet Chemistry included the collection of marine litter in its 3rd phase. Before 2021, there was no coordinated effort at the regional or European scale for micro-litter. Given this situation, EMODnet Chemistry proposed to adopt the data gathering and data management approach as generally applied for marine data, i.e., populating metadata and data in the CDI Data Discovery and Access service using dedicated SeaDataNet data transport formats. EMODnet Chemistry is currently the official EU collector of micro-litter data from Marine Strategy Framework Directive (MSFD) National Monitoring activities (descriptor 10). A series of specific standard vocabularies or standard terms related to micro-litter have been added to SeaDataNet NVS (NERC Vocabulary Server) Common Vocabularies to describe the micro-litter. European micro-litter data are collected by the National Oceanographic Data Centres (NODCs). Micro-litter map products are generated from NODCs data after a test of the aggregated collection including data and data format checks and data harmonization. A filter is applied to represent only micro-litter sampled according to a very specific protocol such as the Volvo Ocean Race (VOR) or Oceaneye. To calculate percentages for each size, formula applied is: Size (%) = (∑number of particles of each size)*100 / (∑number of particles of all size) When the number of micro-litters was not filled or was equal to zero, it was not possible to calculate the percentage. Standard vocabularies for micro-litter size classes are taken from Seadatanet's H03 library (https://vocab.seadatanet.org/v_bodc_vocab_v2/search.asp?lib=H03 ). Different protocols with different degrees of precision were used to classify the sampled micro-litters. Consequently, on the map, the distribution of micro-litter in the size classes depends on the protocol applied during the survey. Warning: the absence of data on the map does not necessarily mean that they do not exist, but that no information has been entered in the National Oceanographic Data Centre (NODC) for this area.

This visualization product displays the spatial distribution of the sampling effort (based on the start position of the sampling tow) over the six-years' period 2017-2022 from other specific protocols. EMODnet Chemistry included the collection of marine litter in its 3rd phase. Before 2021, there was no coordinated effort at the regional or European scale for micro-litter. Given this situation, EMODnet Chemistry proposed to adopt the data gathering and data management approach as generally applied for marine data, i.e., populating metadata and data in the CDI Data Discovery and Access service using dedicated SeaDataNet data transport formats. EMODnet Chemistry is currently the official EU collector of micro-litter data from Marine Strategy Framework Directive (MSFD) National Monitoring activities (descriptor 10). A series of specific standard vocabularies or standard terms related to micro-litter have been added to SeaDataNet NVS (NERC Vocabulary Server) Common Vocabularies to describe the micro-litter. European micro-litter data are collected by the National Oceanographic Data Centres (NODCs). Micro-litter map products are generated from NODCs data after a test of the aggregated collection including data and data format checks and data harmonization. A filter is applied to represent only micro-litter samplings carried out according to a very specific protocol such as the Volvo Ocean Race (VOR) or Oceaneye. The spatial distribution was then determined by calculating the number of times each cell was sampled during the period 2017-2022, only taking into account the start position of the tows. The corresponding total distance (kms) sampled in each cell is also provided in the attribute table. Information on data processing and calculation are detailed in the attached methodology document. Warning: the absence of data on the map does not necessarily mean that they do not exist, but that no information has been entered in the National Oceanographic Data Centre (NODC) for this area. This work is based on the work presented in the following scientific article: O. Gerigny, M. Brun, M.C. Fabri, C. Tomasino, M. Le Moigne, A. Jadaud, F. Galgani, Seafloor litter from the continental shelf and canyons in French Mediterranean Water: Distribution, typologies and trends, Marine Pollution Bulletin, Volume 146, 2019, Pages 653-666, ISSN 0025-326X, https://doi.org/10.1016/j.marpolbul.2019.07.030.

This dataset contains the biological outputs of a global ocean simulation coupling dynamics and biogeochemistry at ¼° over the year 2019. The simulation has been performed using the coupled circulation/ecosystem model NEMO/PISCES (https://www.nemo-ocean.eu/), which is here enhanced to perform an ensemble simulation with explicit simulation of modeling uncertainties in the physics and in the biogeochemistry. This dataset is one of the 40 members of the ensemble simulation. This study was part of the Horizon Europe project SEAMLESS (https://seamlessproject.org/Home.html), with the general objective of improving the analysis and forecast of ecosystem indicators.   See Popov et al. (https://os.copernicus.org/articles/20/155/2024/) for more details on the study.

The present repository makes available the model, material and outputs of the ISIS-Fish modeling work showcased in the peer-reviewed scientific article by Bastardie et al. 2025. As part of the SEAwise research project (seawiseproject.org), we used an ISIS-Fish database (Mahevas et al 2003, Pelletier et al. 2009, isis-fish.org) previously developed within the MACCO project which describes the mixed demersal fishery in the Bay of Biscay. For this application, the spatial extent of the fishery is the Bay of Biscay, defined here by ICES divisions 8a, 8b and 8d and the resolution chosen is 1/16 ICES statistical rectangle. The biological module (Vajas et al. 2024) includes 7 species of economic interest in the mixed demersal fishery: European hake (Merluccius merluccius), common sole (Solea solea), Norway lobster (Nephrops norvegicus), megrim (Lepidorhombus whiffiagonis), anglerfish (Lophius piscatorius) and two ray species (Raja clavata, Leucoraja naevus). The fishing activities module (Mahevas et al. 2024) is made up of 41 demersal fleets (including all French vessels < 12 meters and > 12 meters fishing in this area, Spanich, UK and Belgium fleets) and 431 métiers (combination of a gear, location and mix of target species) catching these 7 species, as target or bycatch. Monthly effort of a fleet distributes among the possible métiers (those historically practiced). The biological and fishing activity modules are identical to the published version. The original model used here has been calibrated on historical catch data 2015-2018 by tuning accessibility and catchability parameters. In the present application the Bay of Biscay model is used to investigate the spatial- and effort- based fisheries management strategies. Consistently with for a task of the SEAwise project (Bastardie et al. 2024) simulations were conducted from 2021 onwards, projecting the effect of an implementation of 3 different closures from 2022 to 2050, under current fishing effort conditions or in a context of fishing effort reduction. Outcomes of these simulations are averaged over short/medium (10 year horizon) and long-term period (20 year horizon). The data project includes: 1) the database including the biological module and fishing activity module; 2) 8 .properties files, each corresponding to one combination of management measure and closure, to restore the simulations parameters in the ISIS-Fish interface and reproduce the simulation runs; 3) the .java scripts to force effort dynamics and simulate spatio-temporal closures, as well as generate the main output files - they will be called by the ISIS-Fish software once the simulations restored 4) the .rds containing the main outputs of the simulations and the associated .html document displaying the R code to compute the indices of interest at different levels of aggregation and reproduce the figures in Bastardie et al. 2025. All files are provided in the Zip. Associated with this material, a study summary and a readme .docx are provided. The first one provides context on the present work and describes the model and simulations' design. The second provides guidelines to reproduce the simulations and their derived outcomes from the data project material made available in this repository. They are both directly downloadable from this repository and are also copied to the zipped folder containing the data project. All the data are reproducible using isis-fish-4.4.8.1 (isis-fish.org; available at forge.codelutin.com) and R 4.2.0.

This daily High-Resolution (HR) Level 3 gridded wind product is derived from Copernicus Sentinel-1 SAR (Synthetic Aperture Radar) observations, over the North Western Atlantic ("ATL" area). It is based on the European Space Agency (ESA) Level-2 OCN products at the highest available resolution. Although L2-OCN products already contain wind vectors, those are calculated using the CMOD5.n Geophysical Model Function (GMF) applied to the co-polarized (co-pol) VV channel (emitting in Vertical polarization and receiving in Vertical polarization). This VV GMF was mapped from scatterometer sensors (Hersbach et al., 2007) which are only able to use co-pol measurements. However, these co-pol GMF are known to lose sensitivity for wind above 20 m/s. Therefore, wind based on such GMF alone, are known to under-estimate wind speed (Polverari et al., 2022). For the L3 products winds based on SAR, we take advantage of the available cross-polarized (cross-pol) VH channel (emitting in Vertical polarization and receiving in Horizontal polarization) for which GMF were specifically derived based on C-Band SAR (Mouche et al., 2017, Mouche et al., 2019). Winds estimated from the combination of both the co-pol and cross-pol channels are referred to as dual-polarization (or dual-pol) winds. As shown in Mouche et al. (2019), taking advantage of the dual polarization strongly improves the wind estimation for high wind conditions thanks to the much greater VH channel sensitivity compared to VV. These new wind estimations are then gridded with a 0.012 degree resolution (between 0.5 and 1.2 km in zonal direction depending on the latitude and 1.3 km in meridional direction) using a cylindrical equidistant projection, independently for ascending and descending satellite passes and for each satellite (so 4 wind fields are available per day for two satellites). This dataset is generated over all Sentinel-1 mission time series starting from March 2018 and updated in delayed mode with a 4-months delay. It is also produced for 4 other different European areas. This dataset is produced and disseminated in the frame of Copernicus Marine Service.