The remarkable longevity of submarine plumes: Implications for the hydrothermal input of iron to the deep-ocean

•We present laboratory experiments simulating hydrothermal submarine plumes.•Sedimentation from submarine clouds is driven by particle diffusive convection.•This process can extend the longevity of event plumes to 2 yr.•Complete dissolution of fine sulfide minerals occurs prior to their sedimentation.•Hydrothermal clouds could provide 75% of the global dissolved iron in the deep-ocean.

The longevity of submarine plumes generated at sea-floor hydrothermal systems constrains the hydrothermal input of chemical species into the deep-ocean. Decades of observations of episodic “event plumes” suggest that a key process governing the dynamics of an hydrothermal cloud spreading out laterally from a buoyant rising plume is the production of internal layering. Here, we use analog experiments on turbulent, hot particle-laden plumes and clouds to show that this layering occurs where particle diffusive convection driven by the differential diffusion of heat and small mineral precipitates gives rise to a large scale double diffusive instability. Where hydrothermal clouds are enriched in fine minerals, this “particle diffusive convection” can extend the longevity of an event plume to 2 yr after its emplacement. The very long residence time imposed by diffusive convective effects enables complete dissolution of fine sulfide and sulfate minerals. We develop a new theoretical model that includes both sedimentation and dissolution processes to quantify the potential amount of iron produced by the dissolution of iron-sulfide minerals settling through the cloud by diffusive convection. A key prediction is that the concentration of dissolved iron in hydrothermal clouds can reach up to 19±3 nM, which represents about 5% of the global hydrothermal discharge. If these results are representative of all hydrothermal vent fields, hydrothermal systems could provide 75% of the global budget of dissolved iron in the deep-ocean. Regionally, this flux is expected to scale in magnitude with mid-ocean ridge heat flow, consistent with observations and global ocean models.