Why's the project called OSMOSIS?

The project is called OSMOSIS as it stands for Ocean Surface Mixing - Ocean Sub-mesoscale Interaction Study. It does not have anything to do with the biological process of osmosis, except in the sense that both are associated with mixing.

The surface ocean is mixed?

Temperature profiles of the upper ocean typically show a layer near the surface (often 50-100 m thick) where the temperature is the same throughout, before falling sharply in the cold abyss below. The ocean is only gaining or losing heat from/to the atmosphere in a thin layer right at the surface, so we would expect this layer to have a different temperature than the water below. Therefore something else must be acting to mix this more evenly across a larger layer.

So what causes this mixing?

There are two ways that the difference in temperatures can be removed. The first is by 'diffusion' - that is heat is transferred from one drop to the next in the same way that an iron rod placed in a fire will slowly become hot at the end away from the flames as the heat travels along it. In this process, the transfer of heat happens without the rod itself moving.

However, the rate at which heat passes through water can be measured. These experiments have shown that the diffusion happens too slowly to explain the uniform temperatures, so another process must be involved.

The second way the temperature difference can be removed is by the water moving up or down to levels with different temperature - a process known as 'advection'. Measurements of the upper ocean have shown it to be full of processes which generate this kind of movement such as the effects of surface waves. Because of this, oceanographers have long identified advection as the main reason for the often uniform properties of the surface layer. Untangling which advective processes are active and their relative importance has been rather more challenging, however. OSMOSIS takes its place among many other pioneering studies conducted over the decades to help address this mystery.

It is important to note that while advection is very effective at stirring the fluid, only diffusion can mix it in the sense of changing the properties such as temperature of the individual molecules. Advection is effective because it stirs the fluid so that the local temperature differences become very large, which finally makes diffusion effective. This is the same reason that we stir coffee when we pour cream into it, as the coffee would be very cold if we waited for the blob of milk to diffuse molecule-by-molecule. Instead we stir it so as to spread it through the cup and make the mixing much more efficient.

What's this other sub-mesoscale thing about then?

The second part of the study is the investigation of sub-mesoscale interactions in the ocean. To get a grip on sub-mesoscales, we first have to understand what mesoscales are (pronounced as 'miso' like a soup or 'mezzo' like a soprano according to your preference).

The mesoscale is a shorthand used by oceanographers for flows which have a medium (or 'meso') length. This excludes flows which take up the whole of an ocean basin (such as the 'gyre' circulation) as well as 'finescale' flows such as those associated with surface waves. This mesoscale is mainly made up of rotating vortices, which are the ocean equivalent of the high and low pressure systems of the atmosphere. The advent of satellite instruments over the last 20 years has highlighted the importance of these flows - as well as the beautiful swirling structure they give rise to.

Sub-mesoscales are then flows which are smaller (1-10 km) than the mesoscale flows (10 - 500 km). If the mesoscales are the ocean equivalent of high and low pressure weather systems, then sub-mesoscales can be compared to the warm and cold fronts which often associated with the most dramatic weather.

Do the sub-mesoscales do anything?

This is a question which is generating plenty of debate amongst oceanographers today. The answer seems to be a tentative 'yes', though there is a lot of work to do in order to make that answer more resounding.

The reason for this caution is that sub-mesoscales are uniquely difficult to observe in the real ocean or to model with computers. They are difficult to observe as they are too small and move too quickly to be observed clearly using satellites. It is not much easier to observe them using instruments deployed in the ocean, as the precise location and time where they develop cannot be predicted with certainty. The instruments must also be deployed close together to be able to observe their small-scale structure. This is more challenging when you consider that you have to put cables in the water that need to be 5 km long to anchor the instruments on the bottom - but these have be placed just 1 km apart horizontally. This creates a risk of the cables getting tangled underwater, the risk of which cannot be eliminated entirely even by careful work on the ship as we don't know what direction the flow at depth is.

What's the difference between these sub-mesoscales and the surface mixing?

While both aspects of the project related to flows which are somehow changing the properties of the surface ocean, the main difference is that the surface mixing flows are primarily vertical while the sub-mesoscale flows are mainly horizontal. The surface mixing flows also generally occur faster, as they develop over hours whereas the sub-mesoscale flows develop over a few days.