Deployment Notes: Kivik Coastal Shelf, October 2023
The wind was biting as we pushed off from the harbor, a sharp, salt-laden breeze typical of a Swedish autumn. I remember looking at the water surface—choppy, a slate-grey that blended into the horizon—and knowing exactly why this site is a nightmare for the unprepared. We weren't fighting massive tides; those are non-existent here in the Baltic. Instead, we were fighting the 'salt wedge.' In Kivik, the interaction between freshwater runoff from the Swedish interior and the denser, saline inflows from the North Sea creates a stratified mess. If you don't account for the pycnocline, your data is basically fiction.
The bathymetry around the southern tip of Skåne is erratic. One minute you're over a sandy shallow, the next you've dropped into a submarine channel. It's a chaotic landscape. The water temperature was plummeting, and the stratification was aggressive. I could tell just by the way the surface chop behaved that we had a strong density gradient sitting just a few meters down. This layering creates vertical shear. You can have surface currents sprinting east while the bottom water is practically stagnant or even crawling in the opposite direction.
What We Found
The most striking data point came back during the first sanity check: we saw a massive velocity reversal at the 12-meter mark. It was wild. The wind-driven Baltic Current was pushing surface waters toward the coast, but the deeper, saltier layer was reacting to the shoreline geometry and pushing back. I've seen this before in brackish environments, but the shear here was particularly sharp. We recorded surface velocities hitting 0.6 m/s, while the bottom bins showed nearly zero movement. It's a classic setup where the wind forces a coastal pile-up, which then drives a compensatory offshore flow at depth.
We also caught some localized eddies swirling around the rocky outcrops. These weren't large-scale gyres, but small, tight vortices that caused significant bin contamination in the ADCP. When I looked at the raw backscatter, I could see exactly where the flow hit the seabed transitions and tumbled. Most people just average this out in post-processing, but that's a mistake. Those eddies tell you everything about how the coastal shelf is actually shaping the current. Honestly, if you ignore the micro-topography of the Kivik coast, you're missing half the story.
Equipment Performance
I opted for a 600kHz ADCP for this run, and it was the right call. A 300kHz unit would have had bins too large to capture that sharp shear layer—you'd just get a blurred average that hides the physics. The 1200kHz would have been too shallow. We bottom-mounted the unit on a heavy tripod to avoid the heave and pitch of the surface chop. Vessel-mounted units are useless in these conditions; the motion correction can't always keep up with the Baltic's erratic surface state. The signal was clean for the most part, though we did hit a patch of noisy data around the 15-meter mark. It wasn't turbulence. It was signal attenuation caused by a sudden temperature drop. I've seen it happen a dozen times when the thermocline is this aggressive.
The blanking distance was the real battle. Because of the salinity gradient, the acoustic signals refract. If you set your blanking distance too short, you get 'ringing' from the seabed or surface interface; too long, and you lose the most critical data in the upper water column. We spent an hour tweaking the settings on the deck before the drop to ensure we weren't blinding ourselves to the surface flow. It's a delicate balance. The unit held steady on the seabed, and the tripod didn't shift, which is a win in my book given the local current spikes.
Recommendations for Future Deployments
If you're heading back to the Kivik sector, don't wing it. The brackish nature of the Baltic makes this a high-refraction zone. You need to be precise with your configuration or you'll spend three weeks in the office trying to figure out why your velocity profiles look like a zigzag.
- Use a 600kHz ADCP for the best balance of vertical resolution and range in these specific depths.
- Stick to bottom-mounting on a tripod; the surface heave in the Baltic is too erratic for reliable vessel-mounted profiling.
- Calibrate your blanking distance specifically for the current salinity profile to avoid losing data in the upper 5 meters.
- Perform a ground-truthing check with a CTD cast immediately before deployment to locate the pycnocline.
- Position the transducer away from known rocky outcrops to minimize bin contamination from localized eddies.
The key is acknowledging that Kivik isn't the open ocean. It's a transition zone. You have to treat the water column as a series of layers rather than a single body of moving water. Once you accept that the 'salt wedge' is the boss of the environment, the data starts making sense.
Field report by Sarah Jenkins. Sarah is a specialist in underwater acoustics and oceanographic instrumentation with two decades of experience profiling continental shelf currents.
Field Deployment Report: Bottom-Mounted ADCP Profiling off Kivik, Skåne