Deployment Notes: São Sebastião Coast, November 2023
The humidity hit us the moment we stepped off the dock at dawn. We were pushing out toward the channel, fighting a choppy surface swell that felt completely disconnected from the depths below. As we approached the drop site, the water shifted from a deep navy to a murky, sediment-heavy green—a clear sign that runoff from the Serra do Mar mountains was pushing hard into the coastal strip. It is a volatile stretch of water. You can feel the energy of the Brazil Current slamming into the jagged coastline, creating a chaotic mix of eddies and shears that would make any standard point-measurement tool useless.
The wind was gusting from the southeast, pushing the surface layer toward the shore while the deeper water continued its southward march. This is the fundamental problem with São Sebastião. It is a hydrodynamic collision zone. If you aren't capturing the entire water column, you're essentially guessing. I've spent years tracking salt wedges and estuarine flows, but this specific corridor is a nightmare for modeling because the vertical velocity shear is so aggressive. One meter of depth can be the difference between a southward surge and a northward counter-current.
What We Found
The data coming off the ADCP was startling. We caught a massive velocity shear that defied our initial projections for November. While the surface currents were behaving predictably under the influence of the South Atlantic High, the deeper layers were doing something entirely different. We recorded a distinct 'counter-current' effect near the seabed, where the water was actually pushing back against the primary flow of the Brazil Current. It was a clean signal, but a surprising one. The magnitude of the shift happened over a vertical distance of less than five meters. This confirms why previous attempts to model sediment transport in the industrial ports here have been so wildly inaccurate; they were using surface data to explain bottom-layer physics.
We also saw significant tidal asymmetry. The flood and ebb cycles aren't mirror images here. The ebb flow showed a much higher peak velocity, likely squeezed by the rugged bathymetry and the deep inlets that characterize this coastline. It's a high-energy environment. I noticed a few spikes in the data that looked like noise, but after a sanity check against the tide gauges, it turned out to be genuine, short-lived turbulent bursts. These eddies are small, fast, and violent. They swirl through the narrow channels and vanish, leaving a trail of suspended particulate matter in their wake. It's a fascinating, if frustrating, puzzle for anyone trying to map plume dispersion for the local shipping terminals.
Equipment Performance
The ADCP held its own, but it wasn't a walk in the park. We faced some serious signal attenuation during a heavy rain event mid-deployment. The runoff from the hills turned the coastal strip into a slurry of organic matter and silt. I saw the signal-to-noise ratio plummet as the acoustic pulses were absorbed by the turbidity. Honestly, if we hadn't tuned the blanking distance and sampling intervals before the drop, we would have ended up with a dataset full of holes. The 300kHz unit provided the depth we needed, but the bin contamination near the seabed was a constant battle. We had to be aggressive with the data filtering to strip out the 'ringing' caused by the hard, rocky bottom of the channel. Once we cleaned the data, the results were rock solid, but the raw files were a mess.
Recommendations for Future Deployments
To get a reliable read on this corridor, you can't just drop and hope. You need a strategy that accounts for the extreme stratification of the São Sebastião waters.
- Increase the sampling frequency during the spring tide to capture the peak asymmetry of the ebb flow.
- Use a higher-frequency transducer if the deployment is limited to the upper 50 meters to avoid the attenuation issues common during the rainy season.
- Deploy a secondary CTD string to ground-truth the pycnocline; without salinity data, the velocity shifts are just numbers without a cause.
- Adjust the blanking distance to at least 1.5 meters to avoid seabed reflection interference in the shallow inlets.
Field report by Dr. Alistair Vance. Dr. Vance is a specialist in underwater acoustics and oceanographic instrumentation with a focus on high-energy estuarine dynamics.
Field Deployment Report: Bottom-Mounted ADCP Profiling in the São Sebastião Channel