Short-term forecasting (several days in advance) of underwater visibility range is needed for marine and maritime operations involving divers or optical sensors, as well as for recreational activities such as scuba diving (e.g. Chang et al 2013). Underwater visibility mainly depends on water turbidity, which is caused by small suspended particles of organic and mineral origin (Preisendorfer 1986). Modelling the fate of these particles can be complex, encouraging the development of machine learning methods based on satellite data and hydrodynamic simulations (e.g. Jourdin et al 2020). In the field of forecasting visibility, deep learning methods are emerging (Prypeshniuk 2023). Here, in continuation of Vient et al (2022) on the interpolation purpose, this work deals with forecasting subsurface mineral turbidity levels over the French continental shelf of the Bay of Biscay using the deep learning method entitled 4DVarNet (Fablet et al 2021) applied to ocean colour satellite data, with additional data such as bathymetry (ocean depths) and time series of main forcing statistical parameters like wave significant heights and tidal coefficients. Using satellite data alone, results show that 2-day forecasts are accurate enough. When adding bathymetry and forcing parameters in the process, forecasts can go up to 6 days in advance. References Chang, G., Jones, C., and Twardowski, M. (2013), Prediction of optical variability in dynamic nearshore environments, Methods in Oceanography, 7, 63-78, https://doi.org/10.1016/j.mio.2013.12.002 Fablet, R., Chapron, B., Drumetz, L., Mémin, E., Pannekoucke, O., and Rousseau, F. (2021), Learning variational data assimilation models and solvers, Journal of Advances in Modeling Earth Systems, 13, e2021MS002572, https://doi.org/10.1029/2021MS002572 Jourdin, F., Renosh, P.R., Charantonis, A.A., Guillou, N., Thiria, S., Badran, F. and Garlan, T. (2021), An Observing System Simulation Experiment (OSSE) in Deriving Suspended Sediment Concentrations in the Ocean From MTG/FCI Satellite Sensor, IEEE Transactions on Geoscience and Remote Sensing, 59(7), 5423-5433, https://doi.org/10.1109/TGRS.2020.3011742 Preisendorfer, R. W. (1986), Secchi disk science: Visual optics of natural waters, Limnology and Oceanography, 31(5), 909-926, https://doi.org/10.4319/lo.1986.31.5.0909 Prypeshniuk, V. (2023), Ocean surface visibility prediction, Master thesis, Ukrainian Catholic University, Faculty of Applied Sciences, Department of Computer Sciences, Lviv, Ukraine, 39 pp, https://er.ucu.edu.ua/handle/1/3948?locale-attribute=en Vient, J.-M., Fablet, R.;, Jourdin, F. and Delacourt, C. (2022), End-to-End Neural Interpolation of Satellite-Derived Sea Surface Suspended Sediment Concentrations, Remote Sens., 14(16), 4024, https://doi.org/10.3390/rs14164024

Monitoring optical properties of coastal and open ocean waters is crucial to assessing the health of marine ecosystems. Deep learning offers a promising approach to address these ecosystem dynamics, especially in scenarios where gap-free ground-truth data is lacking, which poses a challenge for designing effective training frameworks. Using an advanced neural variational data assimilation scheme (called 4DVarNet), we introduce a comprehensive training framework designed to effectively train directly on gappy data sets. Using the Mediterranean Sea as a case study, our experiments not only highlight the high performance of the chosen neural network in reconstructing gap-free images from gappy datasets but also demonstrate its superior performance over state-of-the-art algorithms such as DInEOF and Direct Inversion, whether using CNN or UNet architectures.

Neural mapping schemes have become appealing approaches to deliver gap-free satellite-derived products for sea surface tracers. The generalization performance of these learning-based approaches naturally arises as a key challenge. This is particularly true for satellite-derived ocean colour products given the variety of bio-optical variables of interest, as well as the diversity of processes and scales involved. Considering region-specific and parameter-specific neural mapping schemes will result in substantial training costs. This study addresses generalization performance of neural mapping schemes to deliver gap-free satellite-derived ocean colour products. We develop a comprehensive experimental framework using real multi-sensor ocean colour datasets for two regions (the Mediterranean Sea and the North Sea) and a representative set of bio-optical parameters (Chlorophyll-a concentration, suspended particulate matter concentration, particulate backscattering coefficient). We consider several neural mapping schemes, and we report excellent generalization performance across regions and bio-optical parameters without any fine-tuning using appropriate dataset-specific normalization procedures. We discuss further how these results provide new insights towards the large-scale deployment of neural schemes for the processing of satellite-derived ocean colour datasets beyond case-study-specific demonstrations.

We introduce an end-to-end deep, physics-informed learning framework, 4DVarNet, for reconstructing high-resolution spatiotemporal fields of suspended particulate matter (SPM) in coastal seas by synergistically combining numerical models and sparse CMEMS observations. The approach employs a novel two-phase transfer learning strategy: (1) pre-training on Observing System Simulation Experiments (OSSEs) where gap-free model outputs are masked with synthetic cloud patterns, and (2) fine-tuning on Observing System Experiments (OSEs) using sparse satellite data and an additional independent validation mask. This design enables the network to transfer the physical dynamics learned from the models to observation-driven reconstructions. The architecture embeds a trainable dynamical prior and a convolutional LSTM solver to iteratively minimize a cost function that balances data agreement with physical consistency. Applied to the German Bight in 2020, the framework demonstrates robust performance under operational conditions, outperforming DInEOF, eDInEOF with a 70% reduction in RMSE and correlations up to R2=0.975. Reconstructions preserve fine-scale spatial patterns while maintaining accuracy, with the structure similarity index increased by 50% compared to the EOF approaches. Half of the errors are within ± 0.2 mg/L, even when 27% of days lack any observations. Sensitivity experiments reveal that removing available data increases RMSE and smooths fine-scale SPM spatial features. Increasing the assimilation window length degrades data variability. This work establishes that neural networks can successfully bridge model-based and observation-based systems, with immediate applications for coastal monitoring. It also highlights the need to incorporate tidal dynamics and sub-daily variability into future implementations, particularly for applications targeting real-time sediment transport forecasting.

Monitoring optical properties of coastal and open ocean waters is crucial to assessing the health of marine ecosystems. Deep learning offers a promising approach to address these ecosystem dynamics, especially in scenarios where gap-free ground-truth data is lacking, which poses a challenge for designing effective training frameworks. Using an advanced neural variational data assimilation scheme (4DVarNet), we introduce a comprehensive training framework designed to effectively train directly on gappy datasets. Using the Mediterranean Sea as a case study, our experiments not only highlight the high performance of the chosen neural network (NN) in reconstructing gap-free images from gappy datasets but also demonstrate its superior performance over state-of-the-art algorithms such as data interpolating empirical orthogonal function (DInEOF) and end-to-end neural mapping scheme-based CNN or UNet architectures.

Underwater gliders equipped with current profilers and optical turbidity sensors offer a low-energy solution for high-resolution measurements of currents, suspended particle properties, and sediment transport in coastal waters. Because the spatial structure of hydrosedimentary processes often changes on short time scales (hours to weeks), especially in coastal areas, validating the distribution of glider observations is required to assess our capacity to represent hydrosedimentary processes. Here we propose to validate in a shelf tide-dominated environment, both (i) glider-based currents, and (ii) glider-based acoustic backscatters and optical turbidities in full resolution delayed mode, using in situ collocated and synchronous ancillary observations. The deployed glider system correctly measures the periodic pattern of the tidal current, with a RMSD of O(3 cms-1), demonstrating the system's ability to accurately capture tidal variability. Glider optical turbidities highly correlate with the ancillary observations (R2 up to 0.83). They also correlate well with their glider acoustic counterpart for most of the campaign period (R2=0.76), allowing an estimation of suspended particulate matter concentrations from acoustic measurements. Hence, the glider could observe not only the presence of bottom nepheloid layers of several mgL-1 but also residual fluxes of the order of 1 gm-1s-1 on the shelf. These results highlight the potential of gliders for quantifying sediment fluxes and advancing our understanding of coastal hydrosedimentary processes.

In the ocean, bioluminescent organisms are ubiquitous (e.g. Martini and Haddock 2017) anddeeply related to the ocean dynamics at multiple-scale (e.g. Piontkovski et al 2023).Bioluminescent organisms range from bacteria (bioluminescent bacteria have ecologicalimportance in the biological carbon pump, e.g. Tanet et al 2020) to fishes, not forgetting inparticular many zooplankton species and most dinoflagellates (these bioluminescent organismsemit light through mechanical stimulation allowing in situ sensing of these biological tracers).BIOLUMOPS (“BIOLUminescence Marine, Observations spatio-temporelles in situ par PlaneurSous-marin”) is a project running for 3 years from January 2024 to December 2026. The studyfocuses on the Gulf of Lion, in the Mediterranean Sea, which is known to be an area where winterdeep convection occurs recurrently, in correlation with bioluminescence signals (Martini et al 2014).The project aims at observing the bioluminescence and their related physical and biogeochemicalvariables at multiscale: from the fine scale vertical sawtooth paths sampling of an ocean glider tothe large scale surface ocean colour sampling of satellite remote sensors, not mentioning thediscreet sampling of the ship measurements. The four main tasks of this project aremultidisciplinary: 1. integrating two (reference and innovative) bioluminescence sensors on a sameglider; 2. deploying ship and glider in the Gulf of Lion over three surveys; 3. data processing usingclassification of organisms, in relation with biochemical and hydrodynamic variables; 4. validatingocean colour satellite image of dinoflagellates in the highly dynamic waters of the Gulf of Lion.ReferencesMartini, S., Nerini, D., Tamburini, C. (2014), Relation between deep bioluminescence and oceanographicvariables: A statistical analysis using time–frequency decompositions, Progress in Oceanography, 127, 117-128, https://doi.org/10.1016/j.pocean.2014.07.003Martini, S., Haddock, S. (2017), Quantification of bioluminescence from the surface to the deep seademonstrates its predominance as an ecological trait, Sci Rep 7, 45750, https://doi.org/10.1038/srep45750Piontkovski, S. A., Melnik, A. V., Serikova, I. M., Minsky, I. A., Zhuk, V. F. (2023), Bioluminescent eddies ofthe World Ocean, Luminescence, 38(4), 505, https://doi.org/10.1002/bio.4475Tanet, L., Martini, S., Casalot, L., and Tamburini, C. (2020), Reviews and syntheses: Bacterialbioluminescence – ecology and impact in the biological carbon pump, Biogeosciences, 17, 3757–3778,https://doi.org/10.5194/bg-17-3757-2020.

In marine sciences, accurately mapping Suspended Particulate Matter (SPM) in the ocean is important in understanding the dynamics of primary production and managing its ecological impact. In recovering SPM spatio-temporal distribution, deterministic methods like the end-to-end neural variational data assimilation system, 4DVarNet, while robust, often fall short in capturing probabilistic uncertainty due to deterministic outcomes and are limited in handling the extreme values of overly complex real scenarios. This study adopts a novel application of score-based generative diffusion models to address these challenges, presenting a comparative analysis against 4DVarNet, UNet and DInEOF. Our results demonstrate that when diffusion conditioned with 4DVarNet and DInEOF (meaning these twos are input together with a mask), diffusion behaves better than DInEOF and UNet, even with a small number of samples. It is slightly inferior to 4DVarNet; however, the purpose here is not to outperform 4DVarNet, as 4DVarNet is a powerful scheme for a deterministic approach. Instead, the aim is to better capture the distribution by providing different samples instead of one deterministic outcome. In terms of extreme values, where other neural physical approaches like 4DVarNet show limited performance, our results show that diffusion has the potential to generate the tail of the distribution, which may help in better addressing these extremes. Furthermore, unlike typical 2D diffusion models, this study employs a 3D approach, incorporating 2D spatial and 1D temporal dimensions, allowing the model to capture dynamic physical changes over time and enhance the accuracy of probabilistic predictions.

4DVarNet algorithm is an AI based variational approach that performs spatiotemporal time-series interpolation. It has been used with success on Ocean Color satellite images to fill in the blank of missing data due to e.g., the satellites trajectories or the clouds covering. 4DVarNet has shown impressive interpolation performances compare to other classical approaches such as DInEOF.We propose to show that 4DVarNet is a flexible model that learns global dynamics instead of local patterns, thus enabling it to interpolate different type of data, i.e., data from different spatio-temporal domain and/or representing different variables, using the same pre-trained model. The core of our technique involves extrapolating the learned models to other, somewhat larger geographical areas, including the entire Mediterranean and other regions like the North Sea. We achieve this by segmenting larger areas into smaller and manageable sections, and then choosing a section to train the model. Finally the trained model is applied to each segment and seamlessly integrating the prediction results. This method ensures detailed and accurate coverage over extensive areas, significantly enhancing the predictive power of our models while maintaining low computational costs. Our results demonstrate that this approach not only outperforms traditional methods in terms of accuracy but also provides a scalable solution, adaptable to various geographical contexts. By leveraging localized training and strategic extrapolation, we offer a robust framework for ocean monitoring, paving the way for advanced satellite image applications in diverse settings.

This study presents an enhanced approach to ocean colour L4 product generation through the Adaptive Spatial and Multi-Variable Generalization of 4DVarNet - an innovative integration of deep neural networks with variational data assimilation proposed in [1]. We explore the model’s capabilities in generalizing across various geographical regions and bio-optical variables using datasets of the North Sea and the Mediterranean Sea. Our analysis and visualization show that 4DVarNet demonstrates a notable ability to adapt and scale, reducing significantly training cost thanks to this generalization ability.

Optical remote sensing is increasingly used to assess various sea surface biogeochemical parameters (e.g., Chl-a, turbidity). If today’s systems offer better spatiotemporal coverage, the space-time sampling depends on both the satellite orbit and the cloud cover. The resulting sea surface observations generally present large proportions of missing data, making their completion challenging.Here, we explore neural interpolation schemes as an approach for image gap filling, and their training from observation-only datasets with large missing data rates (with a mean of 65% of missing data and up to 100% for the worst days of the time-series), i.e., when no reference gap-free data are available to run a classic supervised learning approach. We propose and assess different strategies based on real or simulated missing data patterns to discard parts of the available data for learning. We combine these learning strategies with 4DVarnet schemes, which are state-of-the-art neural interpolation schemes backed on a variational data assimilation formulation. The approach was tested in a turbidity reconstruction context, using a multi-modal satellite dataset (CMEMS product: oceancolour_med_bgc_l3_my_009_143) at 1km spatial resolution with daily images from year 2019 to 2021, off the French coast in the western Mediterranean Sea.Our learned variational algorithm significantly outperforms state-of-the-art interpolation techniques, including optimal interpolation and DINEOF, with a 37% gain in RMSE reached in preliminary tests.

Optical remote sensing is increasingly used to assess various sea surface biogeochemical parameters (e.g., Chl-a [1] , turbidity [2] ). If today's systems offer a better spatiotemporal coverage, it still depends on both the satellite revisit period and the cloud cover at the time of the acquisition. The resulting sea surface observations generally present large proportions of missing data, limiting their use. Typically, in our experiments, we worked with datasets containing up to 98% of missing data.

Characterization of suspended sediment dynamics in the coastal ocean provides essential information for scientific studies and operational challenges concerning, among others, turbidity, water transparency and the development of microorganisms using photosynthesis, which is critical for primary production. The complexity of the processes involved in sediment dynamics makes it difficult to predict surface dynamics. In the continuity of previous experiments, the 4DVarNet model having shown encouraging results with SSSC interpolations, it is tested in a 20-day forecasting problem. In addition to the learning architecture including the missing observation data, a protocol has been conceptualized to integrate different types of forcing to improve the reconstructions. The results of the method show that it is possible to produce satisfactory results. The results of the method show that it is possible to produce satisfactory results. The contribution of the input forcing is notable improving of 20% precision horizons. The study also highlights a characterization of the different input forcing and their effect on the system.

Estimating the diffuse attenuation coefficient of the Photosynthetically Available Radiation $(K_{\mathrm{PAR}})$ allows to monitor primary production, dissolved organic matters, coastal suspended sediments and water transparency. The latter aim, especially for military purposes, may be efficiently achieved with the use of underwater gliders. The present study aims at estimating the $K_{\mathrm{PAR}}$ in the the Bay of Biscay (North-East Atlantic), during a sea campaign which took place in February 2021, in fairly transparent waters mainly containing non-living suspended material. The sea survey involved a SEAEXPLORER glider equiped with an ocean color radiometer. The glider measurements were in agreement with those from shipboard CTD-PAR casts (with a mean relative difference of about 11%). VIIRS L2 and MODIS L4 satellite products were validated with the glider data. Accordingly, a bias correction has been proposed for the ocean color satellite $K_{\mathrm{PAR}}$ algorithm.

The characterization of suspended sediment dynamics in the coastal ocean provides key information for both scientific studies and operational challenges regarding, among others, turbidity, water transparency and the development of micro-organisms using photosynthesis, which is critical to primary production. Due to the complex interplay between natural and anthropogenic forcings, the understanding and monitoring of the dynamics of suspended sediments remain highly challenging. Numerical models still lack the capabilities to account for the variability depicted by in situ and satellite-derived datasets. Through the ever increasing availability of both in situ and satellite-derived observation data, data-driven schemes have naturally become relevant approaches to complement model-driven ones. Our previous work has stressed this potential within an observing system simulation experiment. Here, we further explore their application to the interpolation of sea surface sediment concentration fields from real gappy satellite-derived observation datasets. We demonstrate that end-to-end deep learning schemes-namely 4DVarNet, which relies on variational data assimilation formulation-apply to the considered real dataset where the training phase cannot rely on gap-free references but only on the available gappy data. 4DVarNet significantly outperforms other data-driven schemes such as optimal interpolation and DINEOF with a relative gain greater than 20% in terms of RMSLE and improves the high spatial resolution of patterns in the reconstruction process. Interestingly, 4DVarNet also shows a better agreement between the interpolation performance assessed for an OSSE and for real data. This result emphasizes the relevance of OSSE settings for future development calibration phases before the applications to real datasets.

Due to complex natural and anthropogenic forcings, the dynamics of suspended sediments within the ocean water column remains difficult to monitor. Nowadays however, more and more available information is coming from in situ and satellite measurements, as well as from simulation models. Data assimilation methods propose to combine all this information to produce the most precise results, allowing better analyzes of the processes in play. Here a comparison of data-driven methods is presented. Optimal Interpolation (OI), Empirical Orthogonal Function (EOF) based and Kalman Filter based methods are compared to a new one using neural networks. The latter is a Data Interpolation method based on convolutional AutoEncoders (DinAE). Present results show that DinAE better performs compared to other methods, having the lowest error budget and the highest learning of high frequency events.

Due to complex natural and anthropogenic interconnected forcings, the dynamics of suspended sediments within the ocean water column remains difficult to understand and monitor. Numerical models still lack capabilities to account for the variabilities depicted by in situ and satellite-derived datasets. Besides, the irregular space-time sampling associated with satellite sensors make crucial the development of efficient interpolation methods. Optimal Interpolation (OI) remains the state-of-the-art approach for most operational products. Due to the large increase of both in situ and satellite measurements more and more available information is coming from in situ and satellite measurements, as well as from simulation models. The emergence of data-driven schemes as possibly relevant alternatives with increased capabilities to recover finer-scale processes. In this study, we investigate and benchmark three state-of-the-art data-driven schemes, namely an EOF-based technique, an analog data assimilation scheme, and a neural network approach, with an OI scheme. We rely on an Observing System Simulation Experiment based on high-resolution numerical simulations and simulated satellite observations using real satellite sampling patterns. The neural network approach, which relies on variational data assimilation formulation for the interpolation problem, clearly outperforms both the OI and the other data-driven schemes, both in terms of reconstruction performance and of a greater ability to recover high-frequency events. We further discuss how these results could transfer to real data, as well as to other problems beyond interpolation issues, especially short-term forecasting problems from partial satellite observations.

The capacity to monitor suspended sediment concentrations (SSCs) in the ocean, from surface to bottom, using data acquired by the future Meteosat Third-Generation (MTG)/flexible combined imager (FCI) satellite sensor has been quantified by observing system simulation experiments (OSSEs). The "true" ocean state for these experiments is based on a 15-month numerical simulation of hydrodynamic and sediment transport, configured to represent the highly dynamical waters of the English Channel under the influences of tides and waves. Simulated MTG/FCI hourly averaged acquisitions at a given location near the Isle of Wight have been processed via hidden Markov model combined with a statistical classification—based on self-organizing maps—of predicted vertical SSC profiles. The resulting experiments demonstrated that MTG/FCI images, despite their high temporal resolution, and because of many gaps due to nights and clouds over the English Channel, still require spatial interpolations to enhance the amount of information available at a given location. For an accurate determination of particle concentrations, time series of the main forcing (wind, tides, and waves) need to be included in the process: 1) as a crucial parameter correlated with the dynamics of large particles (sands) and 2) as an equally important parameter as satellite data themselves in the correlation with the dynamics of fine particles (silts).

The ascending/descending (or ACDC) horn is a single horn with ascending high E-flat and descending F valves. Its design is simple and composed of a standard 3-valve single B-flat horn fitted with these additional E-flat alto and F valves. This horn can be set to A stopping functionality by pressing simultaneously on both these valves. This latter incidental discovery allows us to propose the first 5-valve single horn which admits a complete range of tonalities, including stopped harmonics, without the need of any additional slide or clutch valve, and with fingerings similar to the standard F/B-flat double horn. With this model, we think the single horn becomes now a serious alternative to the standard double horn. (C) 2019 Elsevier Ltd. All rights reserved.

We present the hydroacoustic inversion tool HYDRAC. This versatile open-source software, developed in Python 3, offers the capability to read hydroacoustic data from widely-used instruments (eg. ABS, ADCP,…) and perform acoustic inversions following several advanced methods found in the literature to estimate the Suspended Particulate Matter mass concentration and particle size characteristics. This software, designed for long-term community-based developments, includes a specific module for modelling the SPM scattering properties, from organic to mineral particles. The originality of this tool lies in its high technological readiness level (TRL 5), towards the emergence of a performant SPM technology for operational use.