Merging optical turbidity and ADCP data for sediment studies
Australia's waterways carry some of the most variable sediment loads on the planet, from the red-soil plumes of the Wet Tropics to the carbonate sands churned up along the Ningaloo coast. Understanding how those particles move through estuaries, harbours and irrigation networks is central to protecting the Murray-Darling Basin, managing port dredging at Newcastle, and keeping the Great Barrier Reef clear of agricultural runoff. Optical turbidity sensors and acoustic Doppler current profilers each capture part of this story, and joining their datasets has become a standard approach for sediment transport researchers working in Australian conditions.
Optical turbidity instruments rely on a near-infrared beam and a photodetector arranged at a fixed geometry. Particles scatter or absorb light in proportion to their concentration and size, giving a nephelometric reading in formazin nephelometric units or a direct estimate of suspended sediment concentration when calibrated against gravimetric samples. The technique responds quickly, handles sub-second sampling, and works in shallow streams where depth precludes heavier instrumentation. Optical backscatter sensors are also compact enough to mount on buoys, autonomous underwater vehicles, or fixed frames beside a jetty pile.
Acoustic Doppler current profilers, by contrast, emit a sequence of pings along multiple beams and measure the frequency shift of returning echoes scattered by suspended particles. The shift gives water velocity through the water column, while the echo intensity itself carries information about the concentration of scatterers. Modern profilers resolve velocity in cells as small as a few centimetres, mapping currents and acoustic backscatter simultaneously. When the particles in suspension are fine enough to track the flow faithfully, the profiler essentially sees the same sediment cloud the optical sensor detects, but from the other side of the physics.
Used together, the two methods cross-validate each other. Optical sensors give a precise, point measurement of concentration, while ADCPs add the spatial extent and the velocity field needed to turn that concentration into a flux. Where one instrument drifts or fouls, the other can flag the anomaly. This redundancy is particularly valuable during flood events on the Fitzroy or the Derwent, when rapid changes in grain size, organic colour, and air-bubble content can temporarily confuse either sensor on its own.
Complementary sensing principles
The optical and acoustic approaches rely on different physical mechanisms, and recognising those differences is the first step toward a sensible integration plan. Optical sensors respond to the projected area of particles within a small sensing volume, typically a few cubic centimetres. Their sensitivity is greatest for silt- and clay-sized material in the one to fifty micrometre range, which dominates suspended load in most Australian rivers and coastal waters during baseflow. ADCP acoustic backscatter, in contrast, depends on the density contrast between particles and water and the acoustic wavelength used, commonly around one megahertz for coastal work.
Because the two techniques weight particles differently, their raw outputs rarely agree numerically. A sample loaded with coarse sand reads as a strong acoustic target but a moderate optical signal, whereas a clay-rich plume after a tropical downpour can saturate the optical sensor while leaving a relatively modest acoustic return. Researchers therefore treat optical readings as a calibrated surrogate for suspended sediment concentration and ADCP backscatter as a relative index that needs its own site-specific transformation.
Combining the two lets a project fill in gaps that neither instrument covers alone. ADCPs provide velocity profiles through the full water column, including the bottom boundary layer where optical sensors cannot easily reach. Optical instruments, meanwhile, resolve the near-surface and near-bed detail that acoustic beam geometry blurs, and they continue to deliver usable data when bubbles from breaking waves or propeller wash corrupt the acoustic record.
The result is a layered dataset where the optical sensor anchors the concentration estimate and the profiler supplies the flow structure. This pattern is now standard practice for plume tracking off the Pilbara, for dredging monitoring at the Port of Gladstone, and for process studies in the Coorong estuary at the mouth of the Murray.
Calibration across diverse water types
Calibration is where most Australian deployments succeed or fail. A single regression between voltage and milligrams per litre derived from one river rarely survives a move to a different catchment, because particle size distribution, mineralogy, organic content and colour all shift. In the Mitchell catchment of Far North Queensland, for instance, volcanic soils produce dark, fine sediments with a high optical yield per unit mass, whereas the sandy bedload of the Fitzroy gives a much weaker optical signature.
Site-specific gravimetric calibration remains the gold standard. Teams collect water samples spanning the expected concentration envelope, analyse them in the laboratory, and fit a curve relating sensor output to suspended sediment concentration. Where the range is wide, a second-order polynomial or a piecewise function is often needed. A useful working target is fifteen to twenty calibration points spanning low, medium and event-driven high flows, with replicate samples at each level.
Biofouling adds a slow drift on top of any calibration uncertainty. Warm tropical waters around Cairns and Darwin accelerate biofilm growth, while estuarine sites near Brisbane experience barnacle settlement within weeks of deployment. Wiper-equipped optical probes reduce but do not eliminate the problem, and regular reference readings in clean water remain essential. Some operators log a daily clean-water check, while others rely on the diagnostic channels built into modern instruments to flag drift automatically.
Acoustic calibration presents its own subtleties. Beam attenuation increases with both particle concentration and air entrainment, and a profiler operating near its noise floor in clean water may saturate rapidly during a flood. Operators often collect a background profile in still, clear conditions to establish a baseline, and use the relative change in backscatter rather than absolute values when estimating concentration trends.
Australian deployment contexts
Australian field sites span tropical, temperate and arid climates, and the choice of mounting and integration strategy changes accordingly. In the Pilbara, iron ore export terminals at Port Hedland face continuous resuspension from berthing ships and tidal exchange, and operators favour fixed frames on dolphins with both an upward-looking ADCP and a side-mounted optical sensor a metre above the bed. The combined record supports both environmental compliance and operational dredging planning.
On the east coast, the Great Barrier Reef Marine Park Authority requires ongoing monitoring of catchments that drain into reef waters. Projects on the Burdekin and the Tully use paired instrumentation during the wet season to capture the brief turbidity peaks that drive light limitation on inshore reefs. The optical sensor provides a continuous record at a fixed point, while the ADCP, mounted on a bottom frame or a small vessel, maps the plume as it moves past. Together they yield flux estimates that link river discharge directly to reef exposure.
Inland deployments tell a different story. The Murray-Darling Basin contains long, slow rivers where sediment moves as much in the bottom boundary layer as in the water column. Researchers at several sites along the Murray use an ADCP in a fixed upward-looking mode to track near-bed transport, supplemented by optical sensors at multiple depths to resolve the suspended fraction. This configuration has helped resolve long-standing questions about how irrigation return flows redistribute fine sediment across the basin.
Smaller urban waterways around Melbourne and Sydney present their own challenges. Stormwater plumes arrive suddenly, last only hours, and carry a mixture of organic matter, litter-derived particles and street dust. Event-based monitoring with rapid-response sensors has become the norm in these settings, and a recent overview of automated sampler integration outlines how the optical reading can trigger sample collection during the rising limb of a storm.
Synchronising data streams and metadata
A combined dataset is only as useful as the metadata that joins it. Optical turbidity loggers and ADCPs run on different internal clocks, and a few seconds of drift between them can misalign a sharp concentration peak with the wrong velocity bin, scrambling the flux calculation. Modern instruments accept a GPS pulse or a network time protocol feed, and deploying a small network time server on the rig is a reliable way to keep both clocks within a tenth of a second across a campaign.
File formats and naming conventions also need attention. ADCPs typically output binary ensemble files, while optical sensors produce ASCII or CSV logs. A consistent filename schema with site, date, instrument and version tags makes downstream processing far simpler. Many projects adopt the conventions described in the technical FAQ, which covers standard naming and quality-control flags.
Quality control is a continuous process. Optical records show spikes from air bubbles, leaf fragments, or fish passages, and acoustic records suffer from interference from surface waves, marine growth on transducers, and electrical noise. Automated filters handle the obvious cases, but a weekly visual review by an experienced operator still catches the subtle artefacts. Logging diagnostic channels such as supply voltage, wiper cycles, and beam correlation helps separate instrument problems from real environmental signals.
Storage and archiving matter too. A month of combined optical and acoustic data at one-minute resolution occupies modest disk space, but a multi-year monitoring program at multiple sites can run into terabytes. Cloud-based archives with clear version control and a documented processing chain have become the standard for Australian government and research programs, allowing data to be shared between the Bureau of Meteorology, state agencies and university partners without confusion over which filter was applied.
Field guidance for combined sensor programs
A combined optical and acoustic program rewards careful planning as much as careful instrumentation. Before deployment, a site visit at a representative range of tides and flows helps identify the best mounting points and reveals hazards such as snags, shallow rock bars, or strong currents that could stress the rig. Choosing a location where both instruments can see the same water parcel, with at least one metre of vertical separation from boundaries, keeps the records comparable.
Once the instruments are recovered, the work moves to the desk. A reproducible processing pipeline that ingests raw files, applies calibration, aligns time stamps, and outputs flux estimates at a standard interval lets a team hand the analysis to new staff without losing continuity. Documenting the pipeline in plain language alongside the code, and linking it to the broader product range used by Australian hydrologists, ensures that replacement sensors and firmware updates can be matched across a network without ambiguity.
The habits below have proven useful across a range of Australian field campaigns, from tropical reef catchments to temperate urban creeks.
- Calibrate the optical sensor with site-specific gravimetric samples covering the full concentration range, including event-driven peaks, and recheck the calibration after major flow events.
- Mount the optical sensor at a depth representative of the main sediment transport layer, avoiding both the surface bubble zone and the benthic boundary layer where possible.
- Pair every optical reading with an acoustic beam-check at deployment to confirm that the ADCP sees a consistent backscatter signal, and log the result for later reference.
- Keep a daily clean-water check on site, particularly in warm or biologically active waters around Darwin, Cairns and the Swan River, to catch biofouling before it skews the record.
- Standardise filenames, time stamps and quality flags across the optical and acoustic datasets before leaving the field, since retrospective fixes are tedious and error-prone.
- Document mounting geometry, water depth and surrounding infrastructure so that later staff can interpret velocity bins without ambiguity.
- Build a simple processing pipeline that produces a daily flux plot within twenty-four hours of collection, allowing rapid response to instrument drift or unexpected events.