How to find an eddy

Our survey began with the deployment of the underway CTD. This is a small device of just 4 kg that is attached to the ship through a---what looks to be waaay too thin---Kevlar line of just 3 mm thickness. The device is dropped from the stern on a winch with the Kevlar line set to free spooling mode. On its way down, the device records temperature, salinity, and pressure, from which oceanographers love to identify different bodies of water. We let it drop for 90 seconds during which it went down to about 220 m before it was pulled back with the winch. This was repeated every 5 minutes for 3 hours. At the halfway mark of 1.5 hours, we had to take a break for 10 minutes, because the ship went through a thick band of ice. This was a first hopeful sign that the front was still where the satellite image 3 days prior had suggested it to be based on the long thin stretch of ice. We, therefore, collected our first section across the front, from 15 km on one side of the feature to 15 km on the other side of it.

Then, it was time to pull the UCTD back onto deck and look at the data it had recorded. After some quick calculations, the ocean had revealed its structure along this line: Underneath the long line of ice was a dome of much warmer water that had possibly been raised from greater depth. Such upward motion of water may bring nutrient rich water into depths where there is still enough light for phytoplankton to grow and can also drag copepods and other zooplankton up. This is why the oceanographic pattern is also of great biological interest. In respect to this we carried out a number of different biological surveys in the area of the front. Along the transects we used the underway filtration device AUTOFIM to collect water samples every ~15 min for cutting edge molecular genetic studies of composition and distribution of phytoplankton and other marine microbes across the front. AUTOFIM is installed deep down in the bug of the ship 10 m below sea surface. All steps related to sampling and filtrations happen automatically and computer-controlled. Due to the full automation of the sampling process the time needed to take an individual sample is reduced by at least a factor of three and allows sampling with higher spatial and temporal resolution, than manual filtration. This approach, combining AUTOFIM and molecular genetic analyses allows us to complement the front study with meaningful high-resolution information on biological communities. In addition to the automated underway sampling we selected a few close locations within the center of the feature, at its rim and outside of it as target stations for complementary biological water column sampling.