Lay summary
Nearly all “standing waters”, such as ponds, reservoirs, lakes and oceans, that are deeper than a few meters, experience cycles of density stratification and de-stratification. Mixing of heavier water from greater depth with lighter water from the surface implies that parcels of different densities are exchanged in the vertical direction. It is evident that mechanical energy is needed to move these water parcels against the prevailing density gradient, which forces lighter water up and heavier water down. Depending on how much wind energy is available for stirring the water column - more in the oceans, less in lakes - and how strong (resistant) the water column is stratified, the vertical exchange of water parcels can be quite restricted. Studying the properties of these limited upward or downward fluxes (against the stratification / gravity) is the subject of “small-scale stratified turbulence”.

Many different techniques have been developed studying these sporadic turbulent eddies and their manifestations as small-scale fluctuations in currents or temperature (or other natural or artificial tracers, including dye). In this project we envisage to measure such fluxes directly and more than 10 times per second at one singe point (a small volume of less than one cubic-centimetre) within the water column. This so-called eddy correlation technique (ECT) is based on the simultaneous measurements of the tur¬bulent fluctuations of the vertical velocity W’ and the fluctuations of the dissolved oxygen content O’. After several ten minutes (i.e. many 10,000s of samples) a stable average flux should result.

Although, we have been able to make first estimates with encouraging comparison with other oxygen flux measurements (determining the rate of oxygen consumption in the sediment), the appli¬cation of this ECT is still novel and many difficulties needs to be challenged during this two-year project, such as the sensor stability, (electronic) noise, partly low signal in well-mixed layers, oxygen response time, temporal sensor mismatch, and others.

Once this method is developed to maturity, we can study fluxes at any point in the water body or at any time interval of interest. In a run-of-the-river reservoir we could - as an example - observe the oxygen production by the algae sitting at the sediment surface or the use of oxygen by the micro-organisms in the sediment. But also turbulence properties, such as the intermittency (“on-off” changes) of fluxes or the dependencies on external forcing can then be studied. Compared to tracer studies, which need much labour force and time to collect and analyse samples, this device is small and light and can be deployed without much effort.