Deployment Notes: Machilipatnam Coast, September 2023
We hit the shoreline at Machilipatnam just as the dawn light broke over the Bay of Bengal. The air was thick, smelling of salt and drying fish from the harbor. My first look at the water told me exactly why this site is a nightmare for acoustics: the Krishna River was dumping a massive load of sediment into the coast. The water wasn't blue; it was a murky, opaque brown. This kind of turbidity kills your signal-to-noise ratio if you aren't careful with your frequency selection.
The wind was pushing hard from the southwest, typical for this time of year. I watched the surface chop—short, aggressive peaks that signaled a strong surface current. We were operating right at the interface where the river's freshwater plume clashes with the saline wedge of the Bay. It's a volatile mix. One minute you're in a calm pocket, the next, a tidal surge pushes a wall of seawater back into the delta.
What We Found
The data came back with a shock. We saw a massive velocity shear in the upper three meters. While the surface was ripping along at nearly 1.1 m/s driven by the monsoon winds, the water just five meters down was practically stagnant or even reversing. This kind of vertical stratification is wild. It proves that the freshwater discharge from the Krishna River creates a floating layer that slides right over the denser seawater. We call this estuarine circulation, but seeing it in the raw data is another thing entirely.
I noticed some weird spikes in the velocity bins during the semi-diurnal tidal shifts. I suspected bin contamination from the seabed, but after a sanity check against the bathymetry charts, it was actually internal waves. The density gradient here is so sharp that it creates these subsurface oscillations. Honestly, if we had used a lower-frequency unit, we would have missed the fine-scale structure of these currents entirely. The 600kHz ADCP gave us the resolution we needed to see the river's influence fighting the tide in real-time.
Equipment Performance
The bottom-mounted ADCP held its ground, but the sediment load was a real problem. We found significant 'noisy data' in the first few bins during the peak flood tide. The silt is so dense here that it scatters the acoustic pings. I had to tweak the correlation threshold to filter out the garbage, or the software would have tried to interpret a clump of river mud as a current vector. Once we dialed in the settings, the signal cleaned up. The tripod mount stayed stable despite the shifting sandy bottom (which is softer than the charts suggest), and the battery life held up through the full lunar cycle.
Recommendations for Future Deployments
If you're heading back to the Andhra Pradesh coast, don't just wing it with standard settings. The river influence changes everything.
- Use a 600kHz or 1200kHz transducer to maintain precision in the shallow, turbid zones near the delta.
- Increase the blanking distance to avoid surface noise from monsoon-driven wave action.
- Deploy a CTD (Conductivity, Temperature, Depth) sensor alongside the ADCP to ground-truth the salinity gradients.
- Double-check your mooring weights; the sandy seabed here can lead to 'sinking' units if the footprint isn't wide enough.
The key is understanding the river. You aren't just measuring the ocean; you're measuring a battle between the Krishna River and the Bay of Bengal. If you ignore the freshwater plume, your data is useless.
Field report by Capt. Marcus Thorne. Capt. Thorne is a specialist in underwater acoustics with twenty years of experience deploying hydrographic instrumentation in complex littoral zones.
Field Deployment Report: Bottom-Mounted ADCP at the Krishna River Delta, Machilipatnam