Modeling Turbulent Dispersion in the Deep Ocean. Particle Transport near the Fåvne Hydrothermal Vent Field

Hydrothermal vents release hot water into the deep ocean and induce strong local mixing. This mixing affects the transport of particles, larvae, minerals, and biological material. Understanding these processes is important for assessing ecosystem connectivity and the environmental footprint of industrial activities near vent sites.

This thesis investigates particle dispersion near the Fåvne hydrothermal vent field on the Mohns Ridge. The study compares an analytical plume model and numerical Lagrangian particle-tracking simulations. The Morton-Taylor-Turner (MTT) model is used to examine how source area, temperature difference, and buoyancy flux affect plume behavior. The model is parameterized using observations collected during a research cruise to Fåvne in August 2025. In addition, particle dispersion is simulated using OpenDrift, with velocity and turbulence fields from the Finite-Volume Coastal Ocean Model (FVCOM). Different turbulence parameterizations, model resolutions, and plume representations were compared.

The results show that the MTT model is sensitive to buoyancy flux and stratification, and it generally fails to reproduce the observed plume characteristics at Fåvne. The discrepancy suggests that the assumed heat flux is likely an overestimation. The heat flux was itself derived from the MTT model, with the observed plume height of $500$ m and a constant stratification N as inputs. This introduces circularity and makes the input unreliable. Uncertainties in the N profile and in whether the measurements were taken within the plume may also contribute to errors. The numerical dispersion experiments with the Lagrangian particle model show a consistent result across all investigated scenarios: particles released near the seafloor rise rapidly to a depth of approximately $-2500$ m within the first few days. This rise is most likely due to vertical velocities generated by the complex bathymetry of the region.

The empirical vertical eddy diffusivity profile was estimated from observations during the TURBOT cruise. The profile produces stronger vertical dispersion than the spatially and temporally varying diffusivity simulated by FVCOM. However, all scenarios show a similar southwestward transport pathway along the ridge. The choice of diffusivity parameterization or plume representation therefore appears to play a secondary role in the overall horizontal transport pathway.

The results indicate that bathymetry-induced vertical motion may be as important or more important than buoyancy-induced plume rise for lifting particles from the seafloor. This suggests that particles observed up to approximately 500 m above the seafloor may not necessarily have been transported there by the hydrothermal plume itself. As a result, plume-height estimates based solely on observed particle elevations may be unreliable at Fåvne.

Overall, the results highlight the difficulty in interpreting hydrothermal plume dynamics from observations alone in regions of complex bathymetry. This underscores the importance of combining analytical, numerical, and observational approaches when studying particle transport near deep-sea hydrothermal vents.