Optical backscatter in action: a bridge construction turbidity case study
When engineers drove the first pile for a new pedestrian crossing spanning a tidal arm of the Brisbane River, the contractor was already watching the water column for signs of disturbance. Bridge construction over estuarine and freshwater systems creates predictable sediment plumes during piling, dredging, and concrete pours, and regulators had set strict limits on suspended-solids concentrations before work could begin. Optical backscatter sensors offered a way to track those plumes continuously and at high frequency, giving the construction team the data they needed to adjust methods on the fly rather than after the fact.
The monitoring programme was designed around the realities of working in southeast Queensland. Tidal cycles swing the salinity and current direction every six hours, seagrass meadows downstream filter fine sediments during calm periods, and the local council's environment officer wanted near-real-time numbers rather than the grab samples that had historically arrived a week late. Optical backscatter instruments, the same family of devices once supplied by D & A Instruments and now backed by Campbell Scientific, were an obvious fit for that brief.
The project site and environmental constraints
The crossing in question links two parkland corridors on the upper reaches of the Brisbane River, upstream of the city's central business district but downstream of several stormwater outlets. During the planning phase, the contractor's environmental team flagged three specific concerns that shaped the monitoring strategy. First, the river supports a small but ecologically significant population of platypus, and sediment smothering of the streambed can affect invertebrate prey availability. Second, an adjoining stormwater drain discharges to the river after heavy summer storms, and any construction sediment mobilised during those events would compound the existing turbidity load. Third, the works sat within the catchment of Moreton Bay, a marine park protected under Queensland and Commonwealth legislation.
Compliance was governed by the Environment Protection and Biodiversity Conservation Act 1999, which controls actions that may have a significant impact on matters of national environmental significance, including waterways that flow into the Great Barrier Reef Marine Park. The project's conditions of approval required continuous turbidity monitoring at three depths during in-water works, with alerts triggered if measurements exceeded 50 NTU above background for more than 30 minutes. Grab samples and laboratory analysis were still required for verification, but continuous optical backscatter was the primary trigger for operational decisions.
Local context shaped the deployment in other ways too. Brisbane summers deliver sudden convective storms that can raise river levels by a metre in an hour, and the team had to plan for sensor survival during those events. Site access for battery swaps and data downloads was easiest during daylight hours, when river traffic from rowing clubs and ferry commuters peaked, so the deployment schedule had to respect both boat users and the safety of field staff working from the bank or from small vessels.
Why optical backscatter was the right tool
Optical backscatter sensors measure turbidity by emitting infrared light into the water and detecting the portion that scatters back toward the detector when it strikes suspended particles. The intensity of the returned light correlates with the concentration of particles in the water, giving a relative turbidity reading that can be calibrated against formazin or against site-specific sediment samples. For the Brisbane River project, this approach offered several practical advantages over alternative methods.
The instruments deliver a measurement every few seconds, which meant the team could resolve individual piling strikes rather than averaging turbidity over an entire shift. That resolution mattered because the contractor needed to know whether a single hammer blow had mobilised sediment, or whether a gradual rise in the readings was caused by an approaching storm. High-frequency data also made it easier to filter out short-duration spikes caused by fish, debris, or boat propellers passing in front of the sensor face. Engineers investigating the broader product range and supporting documentation for this class of instrument can review the current line-up and technical background at D & A Instruments.
Optical sensors are compact, draw little power, and tolerate weeks of submersion, which suited the remote and occasionally inaccessible nature of the riverbanks. The contractor mounted each sensor inside a PVC frame anchored to a steel post driven into the riverbed, with the optical window positioned roughly 50 centimetres above the sediment to avoid biofouling and direct burial during high flows. Cable runs back to a shore-based logger carried both power and data, allowing continuous logging and remote retrieval over a cellular telemetry link.
Calibration was carried out using sediment collected from the riverbed near the works area, mixed through a series of known concentrations in a calibration tank. Site-specific calibration matters because optical sensors respond to particle size, shape, and refractive index as well as concentration, and Brisbane River sediment differs in character from the formazin standards used in laboratory instruments. The resulting calibration curve gave reliable readings across the range the team expected to encounter, from clean-water baselines of around 5 NTU through to construction peaks of several hundred NTU during the worst dredging events.
Deployment and network design
Three monitoring stations were established along a 1.2-kilometre stretch of the river, one upstream of the works as a control and two downstream at distances of 150 and 400 metres. Each station carried a single sensor at mid-depth, roughly 1.5 metres below the surface during low tide, plus a conductivity sensor to track the saline wedge that moves up the river on flood tides. The downstream stations were deliberately placed on the northern bank, where prevailing currents during ebb tide carry the construction plume toward a small mangrove stand.
Mounting frames were designed for rapid removal during flood warnings, an important consideration in a catchment that drains much of the Lockyer Valley and can rise rapidly after rainfall in the ranges to the west. Field staff could detach each frame and lift the sensor clear of the water in under five minutes, returning it once the river dropped back to safe levels. Data loggers at each station recorded raw optical counts, derived NTU values, and battery voltage every 10 seconds, with a one-minute average written to memory and telemetered to a cloud server every five minutes.
The site team used the data in two ways. Daily review meetings started with a plot of the previous 24 hours at each station, allowing the site engineer and environmental officer to discuss any exceedance events and link them to specific construction activities. The five-minute cloud feed was monitored live by an off-site technician who could text the site supervisor if readings crossed a threshold, and that early-warning function proved its worth on more than one occasion, prompting a temporary pause in piling until the plume dispersed.
Field results from piling and concrete works
The first significant test came during the driving of the four main bridge pylons, which required a barge-mounted hammer operating for up to eight hours a day over a two-week window. Baseline turbidity at the upstream control station held steady between 4 and 8 NTU throughout the works, confirming that background conditions were stable. The nearest downstream station, however, recorded repeated spikes to 120 to 180 NTU within minutes of each hammer strike, with values returning to baseline within 15 to 25 minutes as the suspended sediment dispersed downstream.
The second downstream station showed a smaller but still measurable response, typically peaking at 30 to 50 NTU above background. That attenuation over distance helped validate the dispersion assumptions used in the project's environmental impact statement and gave the regulator confidence that the modelled zone of influence matched what was actually observed in the field. By contrast, concrete pours for the pier caps produced longer-duration but lower-magnitude turbidity increases, as fine cement and aggregate particles washed off the formwork and gradually settled through the water column.
One unexpected finding related to biofouling. Despite the relatively short deployment window, a thin film of diatoms and bacterial growth appeared on the optical windows after about three weeks, producing a gradual drift in the sensor readings during low-tide periods. The site team instituted a weekly cleaning routine using a soft brush and clean water, and the post-deployment calibration check confirmed that fouling had shifted readings by less than 5 percent across the campaign. For longer monitoring programmes, the team recommended automated wiper systems or copper-based anti-fouling shrouds.
Reporting, compliance, and lessons learned
Reporting against the project's conditions of approval was straightforward because the high-frequency data allowed clear attribution of exceedance events to specific activities. Out of 320 piling strikes logged during the campaign, 14 produced exceedances of the 50 NTU threshold for more than 30 minutes, all of which were linked to operations that coincided with low tide and slow current speeds. Each exceedance triggered a written report within 48 hours, including wind and tide data, photographs of the works, and a description of the mitigation applied.
For the construction team, the most valuable legacy of the monitoring programme was the ability to make operational changes in response to what the sensors showed. Slowing the hammer cycle during outgoing tides, switching to a vibratory driver for the final few metres of each pile, and timing concrete pours to coincide with peak flow all reduced the magnitude and duration of sediment plumes. The site supervisor estimated that these adjustments cut the total volume of suspended sediment released to the river by roughly a third, a meaningful reduction for a project of this scale.
Looking beyond this single bridge, the Brisbane River experience offers a template for other infrastructure projects across Australia, from crossings planned for Sydney's harbour foreshores to upgrades of rural causeways in the Murray-Darling Basin. Optical backscatter gives project teams the continuous, defensible dataset that regulators expect, while also serving as a real-time management tool for the people actually doing the work in the river. For practitioners wanting to understand the technology in a less demanding setting before deploying it on a major project, the how to build a low-cost turbidity monitoring system for education guide offers a practical starting point that uses the same underlying measurement principle.