Deccan Trap Sediment and Monsoon Spikes: ADCP Velocity Profiling in the Mula-Mutha Confluence

Executive Summary

Measuring discharge in the Pune Basin is a logistical headache. The confluence of the Mula and Mutha rivers creates a volatile hydrodynamic environment where flow rates swing wildly between the lean season and the Southwest Monsoon. We aren't dealing with a steady river; it is a high-energy fluvial system characterized by extreme volumetric spikes and a massive suspended sediment load derived from the surrounding basaltic soils. The real killer here is the turbulence. Debris and high-shear zones render mechanical sensors useless. I've used high-resolution Acoustic Doppler Current Profiler (ADCP) data paired with local bathymetry to finally map the velocity vectors and flow instability that define this specific Maharashtra watershed.

Mula-Mutha Fluvial Architecture and Urban Constriction

Pune sits on the Deccan Plateau at 18.52° N, 73.85° E. While the drainage pattern is technically dendritic, urban sprawl has effectively choked the natural floodplains. I've spent significant time surveying these channels and the bathymetry is erratic, to say the least. Depths plunge from 1.5 meters in the upper reaches to over 8 meters in deep, stagnant pools. During the Southwest Monsoon (June to September), these rivers transform into torrents. Surface velocities often rip past 1.8 m/s. These high-shear zones tear through poorly anchored equipment.

Local infrastructure makes the physics even messier. Bridges and concrete embankments throughout Pune have constricted the cross-sectional area of the riverbeds. This constriction spikes the local Reynolds number. Because the flow is forced through narrow, man-made corridors, we see massive eddies and flow separation. Point-velocity measurements are a waste of time here. It is a far more chaotic environment than the steady-state river systems I monitored in Northern Europe.

Unique Measurement Challenges at the Pune Basin

Mechanical current meters fail almost instantly in this basin. I've seen rotors get choked by monsoon debris or ground down by the silt. The Deccan trap soils produce a suspended load of clay and silt that acts like sandpaper on hardware bearings. If you deploy a mechanical sensor during a peak flow event in July, expect it to be ruined by August. Biofouling is a headache, but it's secondary to the physical abrasion of the sediment.

The vertical velocity profile is another problem. Most engineers assume a logarithmic distribution. In the Mula-Mutha system, that assumption is usually wrong. Urban encroachment and irregular bed morphology create non-uniform channels. I've found that mooring drag in temporary stations often shifts the sensor position by several centimeters. This introduces too much spatial uncertainty for a high-precision discharge calculation. We needed a non-contact, profile-based approach to get a clean signal.

Site-Specific ADCP Configuration

I opted for a 1200kHz transducer configuration. Why? Because the Pune Basin is shallow. To get the spatial resolution required to see the shear layers near the bed, you need a higher frequency. A 300kHz or 600kHz unit would have a blanking distance too large for these depths, leaving a massive gap in the data at the most critical part of the profile.

• Frequency: 1200kHz for high-resolution shallow water profiling.
• Deployment: Vessel-mounted moving boat method (MBM) to capture cross-sectional averages.
• Bin Size: 0.1m to accurately capture the vertical shear.
• Sampling Rate: 1Hz to maintain a tight signal fence during high-velocity surges.

Bottom-mounting is risky here due to the debris load. We've seen heavy logs and urban waste smash into stationary frames. Vessel-mounted surveys allowed us to perform a sanity check against known bridge pier turbulence patterns.

Representative Measurement Data

The following data represents a typical monsoon-transition profile captured near a constricted urban reach. Note the deviation from standard logarithmic flow.

This profile reveals a massive velocity drop-off. The high turbulence intensity in the mid-layer (0.22 - 0.31) is a direct result of the concrete embankments creating secondary currents. The flow direction shift at the bottom (SSE) suggests significant bedload transport and recirculating eddies. It's a mess, but the ADCP captures it perfectly.

Operational Impact on Local Maritime and River Activities

These hydrodynamic fluctuations have real-world consequences for Pune. The city relies on these rivers for water intake and drainage. When the discharge rate spikes during a monsoon flash flood, the sudden increase in suspended sediment clogs filtration systems at municipal water works. Understanding the exact velocity vectors helps engineers design better desilting basins.

Furthermore, the erratic bathymetry affects local dredging schedules. The city often struggles with unexpected siltation near bridge supports. By mapping the high-shear zones, we can predict where sediment will drop out of suspension. This prevents the 'surprise' clogging of drainage culverts that leads to urban flooding in the Peth areas.

Internal Context and Broader Applications

The Pune Basin is a case study in urban fluvial degradation. But the lessons apply elsewhere. I've seen similar patterns in the Chao Phraya river in Thailand, where urban constriction creates similar bin contamination issues during floods. The key is moving away from point-measurements and embracing profile-based acoustics.

Combining ADCP data with turbidity sensors provides a fuller picture. If we can correlate the velocity spikes with the sediment concentration, we can build a predictive model for bed erosion. This is far more useful than a simple water-level gauge. But you have to get the frequency right, or you're just guessing.

About the Author

Dr. Kenji Sato. A specialist in high-turbidity acoustic profiling with over 20 years of experience deploying ADCP systems in volatile fluvial and estuarine environments. He has led instrumentation projects across the Deccan Plateau and Southeast Asian river systems.