Thesis Modelling Abyssal Turbulent Currents Over The Mid-Atlantic Ridge M - F H/F Ifremer

Deadline for applications: 20/04/2026

Who are we ?

The Laboratoire d'Océanographie Physique et Spatiale (LOPS, www.umr-lops.fr) is a Joint Research Unit under the supervision of the Centre National de la Recherche Scientifique (CNRS), the French Research Institute for Exploitation of the Sea (Ifremer), the Institut de Recherche pour le Développement (IRD), and the Université de Bretagne Occidentale (UBO). LOPS develops and participates in oceanography research programs that contribute to advancing knowledge of ocean dynamics. These programs also explore the ocean's interactions with other components of the Earth system, such as the atmosphere, ice, and the biosphere.

The Oceanic Scale Interactions (OSI) team at LOPS relies on a combination of high-resolution models and observations to investigate the diversity and dynamics of fine oceanic scales and their connections to large-scale processes. The team has developed an expertise in the turbulent dynamics of the oceans, grounded in both the theoretical analysis of geophysical fluid dynamics and the development and use of numerical models.

LOPS is a research unit from the Oceanography and Ecosystem Dynamics (ODE) research unit at Ifremer.

What is the topic of the thesis?

Background and scientific context

The deep ocean is critical for Earth's climate, storing heat, carbon, and other compounds over centuries to millennia. Yet it remains under-sampled compared to the surface ocean due to two factors: (1) its historical perception as a stagnant pool, unlike the atmosphere-driven surface, and (2) technological challenges limiting systematic sampling, with knowledge long reliant on sparse cruises. Recently, however, modelling and dedicated expeditions have revealed energetic submesoscale (0.1-10 km) and tidal currents featuring strong interactions with the seafloor topography. Concurrently, there has been a renewed interest in a long-standing problem in physical oceanography: how do deep waters, formed via high-latitude buoyancy loss, return - are upwelled - to the surface?

Munk (1966) proposed that turbulent mixing from breaking internal waves drives homogeneous diapycnal upwelling in the ocean interior. A newer paradigm, based on turbulence observations, instead describes a thin upwelling in a bottom boundary layer (BBL) beneath a thicker downwelling occurring in a stratified mixing layer (SML). This paradigm also predicts an upslope BBL current. However, these 1D mixing-driven theories neglect 3D currents interacting with topography. Capó et al. (2024) extended this framework using 3D simulations of the dynamics over continental slopes, revealing two key findings: (1) low-frequency currents are topostrophic (aligned with topographic Rossby waves), and generate downhill near-seafloor flows in frictional boundary layers, thus contradicting mixing-driven models, and (2) the predicted two-layer upwelling/downwelling system persists but is more confined near the seafloor, with an added third upwelling layer above. While Schubert et al. (2025) highlighted the global relevance of time-mean downhill flows over large-scale topography, small-scale features (e.g., fracture zones) introduce further complexity.

We propose to focus our study on the Mid-Atlantic Ridge (MAR) sector within the subtropical gyre (18°N-29°N) for several reasons. First, it encompasses all key topographic features of mid-ocean ridges: an axial valley, abyssal hills, and a deep fracture zone, enabling us to investigate the feature-specific dynamics. Second, this sector includes the exploration license area granted to Ifremer by the International Seabed Authority (ISA) and has been extensively surveyed, providing robust hydrological datasets to constrain our simulations. Third, our modelling results will directly support interdisciplinary research by Ifremer's BEEP and GEO-OCEAN units.

Scientific objectives

Recent studies showed that the near-seafloor balance between advection and mixing can occur through a very wide range of time and space scales. Those processes have contrasted impacts: time-mean and turbulent currents would not necessarily compensate each other, which leads to a net effect of turbulent currents to drive upwelling and downwelling. These turbulent currents comprise tidal and submesoscale currents. Nonetheless, their exact nature and effects are still unknown and are seemingly constrained by the scale and shape of the underlying seafloor topography - this remains to be uncovered to be able to draw a consistent picture of the ocean circulation over a mid-ocean ridge. Furthermore, both numerical idealized studies and in-situ observations suggest that non-hydrostatic processes in the BBL could have been underestimated and would drive diapycnal upwelling. A more systematic estimation of their impacts needs to be addressed.

Based on the identified gaps in knowledge listed above, the specific objectives are:

- Objective 1 (O1): Disentangle the turbulent dynamics - tidal and low-frequency (subinertial) currents - at play in the BBL and their impact on near-seafloor upwelling and downwelling.

- Objective 2 (O2): Assess the role of non-hydrostatic processes occurring in the BBL and how they can modify mixing and the balance between advection and mixing.

- Objective 3 (O3): Leaning on the two first objectives, draw a refined picture of the diapycnal transformation of near-seafloor water and the abyssal circulation.

What will your mission and activities be?

We resort principally to high-resolution numerical modelling of the ocean circulation using the Coastal and Regional Ocean Community model (CROCO). The strategy is to embed finer and finer grids within parent grids, progressively increasing the horizontal and vertical resolution. We will start from the GIGATL suite of simulations, designed and ran at LOPS. To address O1, we will need to reach a horizontal grid spacing dx<1 km to be able to resolve a wide range of submesoscale processes. We will then run sensitivity simulations with and without tidal forcing to assess the impact of tides on short time-scale advection. A buoyancy budget near the seafloor will be performed to separate the effect of advection and mixing. To address O2, we will set up a second-level simulation embedded in the previous one to reach dx~100 m, where non-hydrostatic processes are likely to emerge spontaneously. We will use the non-hydrostatic option available in CROCO to run sensitivity experiments that will allow us to estimate the significance of non-hydrostatic processes in buoyancy budgets. Last, O3 will be achieved through integrating the first two objectives over the ridge. Tracer release experiments will further help to gain insight on the net effect of advection and mixing.

Key Words :

Deep ocean, Mid-Atlantic Ridge, tides, submesoscales, turbulence, mixing, high-resolution numerical modelling.