Ifremer

For five decades, Mussel Watch has provided an unparalleled record of coastal contamination using bivalve sentinels. However, contemporary marine pollution is increasingly complex, involving chemical mixtures modulated by climate change and tightly linked to ecosystems and human health. Traditional pollutant-by-pollutant monitoring approaches, while essential for tracking regulatory contaminants, are poorly suited to capture mixture effects, emerging contaminants, and cumulative exposures, which characterize real-world environmental risks. This Perspective paper reassesses Mussel Watch through the lenses of the chemical exposome and One Health frameworks. We argue that the program could evolve from a primarily descriptive monitoring system into a mechanistic observatory capable of linking environmental exposures to biological responses and ecosystem-level processes. Achieving this transition requires integrating emerging tools-including non-traditional stable isotopes, high-resolution non-target screening, and omics-based biomarkers-to better characterize complex exposure patterns and their biological consequences. We propose a dual-track architecture in which long-term routine monitoring is maintained to preserve the historical record, while a complementary research track supports hypothesis-driven investigations into biogeochemical transformation, exposure pathways, mixture effects, and biological responses. By combining these approaches, Mussel Watch could strengthen its capacity to detect emerging risks, support mechanistic understanding, and enhance the policy relevance of coastal monitoring in a rapidly changing environment.

Abstract Since 1995, European fisheries of Pecten maximus have faced the presence of Pseudo-nitzschia species, which are able to produce the neurotoxin domoic acid responsible for Amnesic Shellfish Poisoning (ASP). As filter-feeders, scallops can accumulate and retain domoic acid much longer than most other bivalves, from months to years. When concentrations exceed the regulatory threshold, fisheries are closed leading to economic concern. Inter-individual variability increases the difficulty to predict the depuration dynamics. Quantifying the correlations between domoic acid depuration in P. maximus and individual physiological traits, particularly body size, could improve the understanding of contamination and depuration. We analysed toxin dynamics in organs and assessed the effects of body size and growth. This analysis was based on two datasets from an experimental and an in situ depuration monitoring of P. maximus exposed to a natural bloom of toxic P. australis . Results showed that the distribution of domoic acid shifted among organs between contamination and two months of depuration. Toxin concentrations correlated negatively with body size during contamination and after two months of depuration, but shifted to a positive correlation after 7 months of depuration. This suggested that smaller scallops both accumulate more domoic acid and depurate it more rapidly. Dilution by growth appeared to explain the inversion of the correlation between domoic acid and body size throughout depuration. These results yield useful information for modelling these mechanisms, thus providing valuable tools for scallop fishery management facing ASP. Graphical Abstract Highlights Experimental and in situ datasets allowed to quantify DA proportion dynamics in organs in P. maximus DA concentration and body size are negatively correlated during contamination phase, but positively after a 7-month depuration Considering dilution by growth is important for young scallops to assess DA depuration dynamics Both depuration rate and dilution by growth need to be considered to model DA depuration over the whole scallop size range

Data Terra, a French national research infrastructure for Earth system observations, brings together five disciplinary poles—FORMATER (solid earth), AERIS (atmosphere), ODATIS (ocean), THEIA (continental surfaces), and PNDB (biodiversity)—to provide access to diverse environmental data and associated services. Several of these hubs support field and observational programs, yet researchers often face uncertainty about the tools, services, and resources available to assist them before, during, and after their field program. This lack of visibility leads to fragmented practices and repetition of efforts across domains.To address this challenge, we initiated an inter-pole collaboration to develop a catalog of services relevant to the needs of scientists for the entire life cycle of their field program. These services include software tools, applications, data management solutions, user support, technical assistance, and training resources. The goal is to provide a coherent and easily navigable overview of the resources offered by Data Terra and its collaborators, free from disciplinary boundaries.This effort is built on a GeoNetwork-based catalog, expanding it to accommodate cross-domain field program needs. A new dedicated thesaurus has been created to classify the diverse resources, ensure consistent tagging, and allow an eased search and findability of the resources. By structuring needs and services in a shared semantic framework, we aim to enhance discoverability, foster interoperability between poles, and better support the scientific communities conducting environmental field programs.

Mode I fracture toughness, KIc, is a key measure of rock strength. In volcanology, KIcis particularly relevant for dike propagation, as it quantifies the critical stress intensity factor required for fracture propagation under tensile stress. KIc has been extensively studied for rock types and construction materials relevant to civil engineering, mining, and hydrocarbon-related applications. However, there is a paucity of data for volcanic rock. In this study, we present KIcvalues for basalt samples of varying porosity and texture from four different lava flows on Mt. Etna (Italy) and investigate the influence of key microstructural parameters on KIc.We conducted 12 mode I fracture toughness experiments under dry, ambient pressure conditions on Cracked Chevron-notched Brazilian Disc specimens and determined KIc using a standardised method. Additionally, we characterised rock physical properties including porosity, elastic wave velocities, and permeability, and analysed thin sections to determine mineralogical composition and rock texture. We compared the physical and microstructural properties of the four lavas and then assessed those properties regarding any correlations with KIc.The fracture toughnessmeasurements were successful for 10 of the 12 specimens, yielding KIcvalues of 0.6–1.3 MPa·m1/2. Average connected porosity varied between 9­ and 17%. P-wave velocities varied from 3.1 to 3.8 km/s, while permeability varied from 6.7·10-17 to 6.3·10-12 m2.Our fracture toughness data are consistent with experimental data from the literature, fitting the general trend of high KIc typically corresponding to low porosity. However, within our small data set of rather heterogenous porosity characteristics and rock textures, we observe no strict inverse correlation of KIc,and porosity, since the porosity range of our samples is only moderate and other microstructural factors (e.g. pore size and shape) can dominate fracture behaviour in individual cases. We observe no systematic relationship between elastic wave velocities and KIc.

The Argo Program is a global network of 4,000 autonomous drifting floats that provide essential, real-time data on the upper 2,000 meters of the ocean. By measuring temperature and salinity, Argo has become the primary source of information for monitoring ocean warming, sea-level rise, and climate variability. However, the massive volume of data generated—totaling millions of profiles—presents a significant challenge for Quality Control (QC).Traditionally, delayed-mode quality control has relied heavily on human expertise and the "trained eye" of scientists to identify instrumental drifts and sensor malfunctions. To address the cost and limitations of manual inspection, we introduce Argo-YOLO, an innovative approach that transposes computer vision techniques into the field of physical oceanography.By converting oceanographic profiles into graphical representations, our system utilizes the YOLO (You Only Look Once) deep learning architecture to "scan" the data, mimicking the visual diagnostic capabilities of expert oceanographers. This method enables high-speed, systematic detection of instrumental drifts, sensor malfunctions, and profile anomalies across the entire Argo dataset while maintaining the nuanced precision of human analysis.Initial results demonstrate that Argo-YOLO faithfully reproduces expert visual diagnostics with high performance: 97% accuracy in identifying valid profiles with only 3% false alarms, and 96% success in detecting anomalous profiles with 4% missed detections.These results confirm the viability of computer vision for operational oceanographic quality control.Argo-YOLO demonstrates how computer vision can be successfully adapted to oceanographic challenges, representing a major step toward automated, scalable quality control in global ocean observing systems and ensuring the integrity of long-term climate records in an era of "Big Data" oceanography.

The increasing volume and diversity of Earth Observation (EO) and climate data produced by Copernicus missions and the Destination Earth (DestinE) initiative pose a major challenge for interoperability and large-scale analysis. Today, global datasets are distributed on heterogeneous spatial grids, forcing users to repeatedly perform ad-hoc regridding steps, which are costly, error-prone, and difficult to reproduce.The GRID4EARTH project addresses this issue by promoting a common Discrete Global Grid System (DGGS) as a foundation for analysis-ready Earth data. In this context, HEALPix emerges as a strong candidate due to its equal-area property, hierarchical structure, and long-standing adoption in global modelling and large-scale data analysis. These properties enable efficient multi-resolution workflows, scalable data access, and natural integration with modern cloud-native formats such as Zarr.However, a key limitation remains: HEALPix is formally defined on the sphere, whereas EO are naturally referenced to the WGS84 ellipsoid. While often ignored at coarse resolutions, this mismatch introduces non-negligible area distortions at Copernicus resolutions and may bias zonal or regional analyses.To overcome this limitation, GRID4EARTH explores anextension of HEALPix to the WGS84 ellipsoid using the authalic sphere associated with the ellipsoid. By preserving equal-area properties on the ellipsoid, this approach provides a consistent spatial framework bridging spherical climate models and ellipsoidal EO. It enables a unified representation for DestinE model outputs and Copernicus satellite data, while remaining compatible with existing HEALPix-based tools and workflows.This contribution presents the motivation, principles, and expected benefits of ellipsoidal HEALPix within GRID4EARTH, and discusses its role as a practical and scalable DGGS for next-generation Earth system data infrastructures.

Amongst the Thematic Assembly Centers (TACs) of the Copernicus Marine Service, the Wind TAC is in charge of providing L3 and L4 sea surface wind and wind stress observation products at a global and regional scale. From 2015, the Wind TAC has been successful in providing data coming from scatterometer observations both Near Real Time (NRT) and reprocessed (Multi Year, MY) data. Since the new phase of the project, starting in 2025, high-resolution (i.e. 1 km) data from C-band Synthetic Aperture Radar (SAR) Satellite have been included in the Wind TAC. As the Copernicus Marine Service evolution is mainly user-driven, we take this opportunity to announce that L3 daily products of Sentinel-1A winds are operationally processed and delivered in Near Real Time since June 2024. The associated Multi Year database of L3 daily products has also been uploaded to the Copernicus Marine Service and covers data from 2018 to 2025. To obtain wind data from SAR observations the ocean surface roughness measured by the satellite is transformed into wind speed and direction using an empirical inversion method called the "dual-pol" (Mouche et al. 2017, Mouche et al. 2019) and an a priori wind helps constrain the possible values. To this purpose, the SAR Wind Production Unit uses the L2 files produced by ESA's Mission Performance Center for Sentinel-1 (MPC-SAR) both for NRT and MY processing. Then, the only difference between the two processing chains (NRT and MY) lies in the a priori wind used to constrain the computed wind speed: while the NRT chain uses wind model from the European Center for Medium-range Weather and Forecasting (ECMWF) operational Integrated Forecast System (IFS), the MY chain uses wind model from ECMWF ReAnalysis v5 (ERA5). To quantify the impact of the input a priori wind, the NRT and MY data are compared over their common period. More generally, the SAR wind data from Wind TAC has been qualified through comparisons against Numerical Weather Forecast (NWP) models, in situ measurements from buoys and co-located measurements from scatterometer. The results of these comparisons are shown here, illustrating some limitations in the methodology and ways to overcome them. Overall, these global comparisons highlight the complementarity of high-resolution SAR wind data with other wind observations. In the near future, the integration (in 2026 and 2027 respectively) in the SAR Wind TAC of the recently launched Sentinel-1C and Sentinel-1D satellites will allow for better coverage, and the complementarity between observation should be further investigated through triple-collocation between SAR, scatterometer and buoy data.

This study investigates the coastal dune deposits of "Barrier III" along the southern Brazilian coast, and in particular along the coastal area close to Osório and Paranaguá. While traditionally attributed to the Marine Isotope Stage (MIS) 5e highstand, emerging evidence suggests this barrier is a complex system incorporating younger deposits from MIS 5a or even MIS 3. The analyzed sequences consist of aeolianites outcropping up to a few meters above present sea level. To ensure high-resolution spatial accuracy, proxy elevations were measured using a GNSS RTK station and referenced to the local geoid model (MAPGEO2015), maintaining a vertical error margin of a few centimeters.Through geomorphological, granulometric, and morphoscopic analyses, this research re-evaluates the geochronological framework of these deposits by presenting 19 new Optically Stimulated Luminescence (OSL) ages. The findings address a critical scientific debate: while a broad consensus has long placed sea levels during MIS 5a and 5c several meters below present, recent studies increasingly suggest that sea level at that time may have been closer to, or even above, modern levels in certain regions. This study examines the extent to which eustatic oscillations during these substages, and potentially MIS 3, drove sedimentary dynamics capable of shaping dune systems in positions proximate to the modern shoreline.