Research
My main research area is the development of ocean modelling systems that couple the physics, biogeochemistry and biology and predict the weather and climate time scales with a focus on seven axes.
Ocean deoxygenation
For anyone
Marine life depends on oxygen dissolved in seawater, much as we depend on the air we breathe. In many parts of the ocean, oxygen levels are falling; a change driven by warming waters and nutrient pollution. This research asks where that is happening, why, and what it means for the ecosystems that live there.
For peers
Coupled hydrodynamic-biogeochemical models are used to quantify oxygen dynamics from seasonal shelf hypoxia to deoxygenation, with a main focus on the Black Sea. The work connects physical transport, benthic–pelagic coupling and biogeochemical cycling to the oxygen budget, and supports international efforts to monitor and assess ocean oxygen loss.
Ocean stochastic forecasting, AI emulation & data assimilation
For anyone
Forecasting the ocean works like forecasting the weather: models are run forward in time and continuously corrected with observations from satellites, ships and autonomous floats. Because no forecast is ever certain, we run many slightly different versions of the model at once, so that a prediction comes with an honest measure of its own confidence. We also train artificial intelligence to imitate the heaviest models, which makes it possible to explore many more scenarios, much faster!
For peers
MAST develops ensemble-based, stochastic forecasting for coupled physical-biogeochemical systems, with the Black Sea and the NEMO-BAMHBI model as its main asset: systematic perturbation of key processes, probabilistic validation against observations and predictability studies (ODESSA). It works on multiscale, non-Gaussian and AI-assisted data assimilation for the Copernicus Marine Service forecasting centres (COMEDI), on deep-learning emulators of ocean deoxygenation (MITHO), and on the model representation of biogeochemical Essential Ocean Variables (BioGeoSea).
Towards an Earth System approach
For anyone
The sea is not a closed box. Rivers bring it water and nutrients, the atmosphere drives its currents and exchanges heat and gases with it, and its sediments store and release what settles on the sea floor. To predict the ocean, and above all the coast, our models increasingly have to treat all of these together, from the river catchment to the open sea, as one connected system.
For peers
This line of work extends coupled hydrodynamic-biogeochemical models to the other components of the Earth system: seamless coupling of the river-estuary-ocean-atmosphere continuum with applications to the North Sea and the Scheldt (FNRS FullContinuum), benthic-pelagic coupling and sediment processes, and regional Earth-system model configurations for European seas. It contributes to the UN Ocean Decade programme CoastPredict, which aims to transform how the global coastal ocean is observed and predicted.
Climate projections
For anyone
Global climate models tell us how the planet is warming, but they are far too coarse to say what will happen in a given sea or along a given stretch of coast... Our work focuses on this specific part of climate change: what the Baltic, the Black Sea or the Mediterranean may look like in ten, thirty or eighty years, and what that means for the life they hold and the people who depend on them.
For peers
Downscaling of climate change to regional seas and coasts using high-resolution regional Earth-system, ocean and marine-ecosystem models: multi-model decadal predictions and multi-decadal projections for European seas within RIVIERADE (bridging the CORDEX and Copernicus Marine communities), downscaling of past changes and climate projections for coastal services (CE2COAST), and the ecosystem response of the Black Sea to multiple stressors (BRIDGE-BS). Copernicus Marine developments in NECCTON extend the same modelling chain to trophic networks and benthic habitats.
Blue growth assessment effect on biology
For anyone
Europe's seas are working seas: offshore wind farms, fishing and aquaculture all use them, and all leave a mark on the life below the surface. Numerical models let us ask what those activities do to marine ecosystems, alone and combined with climate change, before and after decisions are taken, so that a growing blue economy does not come at the expense of the sea's health...
For peers
Model-based assessment of the effects of blue-economy activities on marine biology and ecosystem functions: coupling physical-biogeochemical models with functional biodiversity, benthic habitats and higher trophic levels (NECCTON's Copernicus capability for trophic networks and benthic habitat modelling), the evaluation of thresholds and opportunities for coastal services under climate change (CE2COAST), and decision-ready indicators for ocean health and a sustainable blue economy at basin, regional and local scales (RIVIERADE).
Science for marine policy
For anyone
Science only helps if it reaches the people who decide. Contributing to international expert groups and assessments means that what we know about the changing ocean can inform how it is governed.
For peers
Contributions to international programmes and expert networks on ocean health and deoxygenation, among them the IOC-UNESCO Global Ocean Oxygen Network (GO2NE) and the UN Ocean Decade programme GOOD, and to environmental assessment at European and international level.
Commitment to the Academic community
For anyone
Beyond my own science, I act for the Academic community, as spokesperson of ADERE at the University of Liège, and as founder of UNIFRA, that opens a public conversation about the Academic world's future.
For peers
ADERE, the Association of Research and Teaching Team Directors at ULiège, represents professors and researchers who lead teams and works to improve their working conditions, for instance by increasing daily support and reducing bureaucratic overhead. UNIFRA, the Union pour les Universités Francophones, is an independent inter-university movement founded to initiate discussion about the 2025-2026 Belgian government reforms and their consequences for universities, especially about Academic pensions.
Explore My Projects - Click on the filters to narrow down my work
- BioGeoSeaEnhancing Biogeochemical Essential Ocean Variables for European and Global AssessmentsA Horizon Europe project closing research gaps on biogeochemical Essential Ocean Variables: improving how they are observed, validated and represented in models, and turning them into indicators of ocean acidification, deoxygenation, the biological carbon pump and greenhouse-gas fluxes for European and global assessments. The University of Liège is a partner and I am the Principal Investigator for ULiège.
- NECCTONNew Copernicus Capability for Trophic Ocean NetworksA Horizon Europe project building new Copernicus Marine capabilities to model and forecast ocean trophic networks, extending operational biogeochemical modelling toward ecosystems. MAST contributes as a partner.
- RIVIERADEImproving Modelling Methods to Produce Climate Services for Resilient European Seas and Coasts in a Decadal to Multi-Decadal HorizonA Horizon Europe project bringing together the CORDEX and Copernicus Marine communities to produce, downscale and deliver decadal predictions and multi-decadal projections of climate change and its impacts on marine ecosystems for the Baltic, the Black Sea and the Mediterranean, using high-resolution regional Earth-system, ocean and ecosystem models. It delivers climate-service demonstrators at basin, regional and local scales for ocean health, the blue economy and coastal risks. I am the Principal Investigator for the University of Liège.
- MITHOMultiple Threats on Ocean HealthAn ESA project developing Earth-observation-based methods to understand, monitor and predict the health of the ocean under multiple, cumulative climate threats: hazard indices for extremes in warming, pH, winds, river inputs and oxygen, integrated with ecosystem services such as biodiversity and fish biomass. At the University of Liège, I am the Principal Investigator; MAST builds a deep-learning emulator of ocean deoxygenation, in collaboration with Prof. Gilles Louppe.
- COMEDICopernicus Marine Services with Enhanced Data Assimilation and Improved Interoperability
- ODESSAOptimal Design of Ensemble-based Simulation Systems for Applications to Marine Forecasting
- O2CHThe Ocean Oxygen to Carbon Heat Nexus in a Warming Climate
- GOODGlobal Ocean Oxygen Decade
- BRIDGE-BSAdvancing Black Sea Research and Innovation to Co-Develop Blue Growth within Resilient Ecosystems
- CE2COASTDownscaling Climate and Ocean Change to Services: Thresholds and Opportunities
- FORCOASTEarth Observation Services for Fishery, Bivalves Mariculture and Oyster Ground Restoration
- GO2NEGlobal Ocean Oxygen Network
- GO2DATGlobal Ocean Oxygen Database and Atlas
- CoastPredictObserving and Predicting the Global Coastal Ocean
- ADEREAssociation of Research and Teaching Team Directors, University of Liège
- UNIFRAUnion pour les Universités Francophones
- FullContinuumSeamless Coupling of the Ocean and River System