Oceans: Stratified Flows

Researchers: Professor Colm Caulfield

The Problem

Small-scale turbulent mixing plays a key role in creating and disrupting ocean stratification, which in turn influences large-scale ocean circulation.

Work Needed

We need to better understand how stratification is affected by small-scale processes, as currently global climate models fail to capture these potentially extremely significant processes.

Our Work

By combining laboratory experiments, numerical simulations, and data-driven methods, we are investigating how waves and turbulence influence global ocean stratification.

Background

The ocean is stratified into layers based on differences in temperature and salinity. The ocean is a huge reservoir of heat for the global climate system and its stratification controls the exchange of heat, carbon and nutrients between the surface and deep ocean. But the ocean is not static; turbulent mixing – a chaotic process driven by forces such as internal waves and currents – plays a key role in creating and disrupting this stratification. Mixing processes can occur on very small spatial scales, yet their effects can cascade to influence large-scale ocean circulation.

Understanding how stratification is maintained and affected by small-scale processes such as turbulent mixing is essential to having high quality predictions about the climate and about extreme weather events like hurricanes. Current knowledge gaps mean that global climate models fail to capture the impacts of these processes.

Our Work

We need to better understand how waves and turbulence interact with density variations, and hence the buoyancy force, to set the distribution of heat and salt in the world’s oceans. To do this we are combining laboratory experiments, numerical simulations, physical modelling constrained by observations, and increasingly, data-driven methods to gain a better understanding.

We have been employing a multi-pronged approach to this challenge:

  1. We have been doing experiments to understand the very small scale- the last stage of mixing- to consider how diffusion actually causes the properties of a fluid parcel to change irreversibly, i.e. when does stirring stop and mixing start?
  2. We have reanalysed observations using unsupervised machine learning to identify different types of mixing events in the ocean. In particular we are trying to understand how important highly unusual, extreme mixing events are to determining the overall budget of the stratification.
  3. We have also been doing high resolution numerical simulations to explore how the very small-scale processes, dominated by diffusion, can feedback on the large-scale turbulent motions which arise when waves break in the interior of the ocean.

Future Directions

We will continue working towards gaining a better understanding of the fundamentals through these idealised modelling approaches. We also plan to collaborate with earth scientists and climate modellers to improve the quality of large-scale climate models by embedding this new idealised understanding into improved model parameterisations.