TL;DR
This structural monitoring dam case study follows ISS through the design, installation, and operation of an automated deformation monitoring system on a high-consequence tailings storage facility (TSF) at a Bowen Basin coal operation in Queensland. Periodic GNSS campaigns had flagged 11 mm of crest settlement over two wet seasons but could not say whether the rate was accelerating, so the operator needed continuous, alarmed surveillance scaled to the dam's ANCOLD consequence category. ISS commissioned a permanent robotic total station reading a 38-prism array to ±1.2 mm every two hours, tied to a deep-founded reference network, with SMS alarming against a green/amber/red trigger-action-response plan — turning an ambiguous trend into defensible, real-time data the geotechnical engineer could act on.
Key takeaways
- Periodic monitoring told the operator the embankment had moved (11 mm over ~20 months) but not whether the velocity was stable or accelerating; an automated robotic total station resolving each prism to ±1.2 mm every two hours converted that ambiguity into a measurable rate of change.
- The programme was scoped against ANCOLD Guidelines on Dam Safety Management and the Global Industry Standard on Tailings Management for a "High C" consequence-category facility — not a single prescriptive Australian standard, which is why a documented baseline and methodology mattered more than any one instrument.
- A stable reference frame was the make-or-break decision: four deep, grout-anchored pillars founded on competent rock well outside the zone of influence, observed redundantly and least-squares adjusted, with network stability re-checked every epoch in GDA2020/MGA2020 Zone 55 and AHD heights.
- The installed automated system cost AUD ~$96,000 plus an ongoing monitoring fee — trivial against the consequence of an undetected breach, and the figure that drove the business case was risk, not throughput.
- Within the first wet season the system flagged a localised 4.8 mm/week movement on the downstream toe that periodic survey would have missed entirely, allowing the engineer to mobilise toe-drain remediation before the amber trigger was reached.
The challenge
The asset was a downstream-raised tailings storage facility serving a metallurgical coal operation in the Bowen Basin, classified "High C" under ANCOLD consequence categorisation — a population-at-risk and environmental profile that mandates documented deformation surveillance proportionate to the consequence.
The operator had been monitoring the embankment with quarterly GNSS campaigns and precise levelling on a sparse stud array. Over two wet seasons, the crest had recorded approximately 11 mm of cumulative settlement and 6 mm of downstream lateral movement. In isolation, those magnitudes were within the dam's design envelope. The problem was resolution in time, not space: four readings a year cannot distinguish a benign 1 mm/month creep that is decelerating from a 1 mm/month creep that is about to accelerate. With a consequence category that high, the geotechnical engineer of record could not sign off on a surveillance programme that left a three-month blind spot through the wet season — the exact window when pore pressures peak and embankments move.
The Queensland mining regulator's geotechnical management requirements and the operation's own dam safety management system both pointed to the same gap: the facility needed continuous, alarmed monitoring with a defensible trigger-action-response plan (TARP), and it needed a baseline that could withstand audit.
Our approach
ISS does not sell instruments, so the programme was built around the dam's risk rather than around hardware. The work followed the same disciplined sequence ISS applies to every monitoring programme: risk-informed plan, stable control, considered point layout, a doubly observed baseline, then recurring measurement and reporting against agreed triggers.
Risk assessment and monitoring plan
Working with the operator's geotechnical engineer and dam safety team, ISS defined the expected deformation modes — crest settlement, downstream lateral creep, and toe bulging — and set the required precision and read cadence against the TARP. Green, amber, and red trigger levels were agreed for both cumulative displacement and, critically, velocity: a movement still inside the cumulative amber band but accelerating sharply was defined as its own alarm condition.
Reference control establishment
The single most important design decision was the reference frame. Four monitoring pillars were grout-anchored into competent rock at distances of 180–340 m from the embankment, beyond any credible zone of influence. The network was observed with redundant robotic total station rounds and static GNSS, least-squares adjusted, and georeferenced to GDA2020/MGA2020 Zone 55 with heights on AHD. Reference stability is re-verified every epoch — a reference pillar that quietly moves would corrupt every reading derived from it.
Monitoring point installation
A 38-point array of forced-centring Leica GPR1 and GMP104 mini-prisms was fixed to the crest, mid-slope, downstream berm, and toe to capture the full expected deformation pattern, supplemented by the existing levelling studs for an independent vertical cross-check and tie-ins to the site's vibrating-wire piezometers so movement could be read against pore pressure.
Baseline and automation
ISS measured the baseline twice on separate days to quantify the system's own noise floor — you cannot interpret a 2 mm movement without knowing your measurement uncertainty — then commissioned the permanent system over a two-week verification period before going live.
Equipment
The instrument selection prioritised repeatability and unattended endurance over raw range, consistent with monitoring-grade practice and ISS's ISO 17025 annual calibration regime.
- Leica Nova TM60 monitoring total station — a 0.5″ angular, 0.6 mm + 1 ppm distance instrument with ATRplus automatic target recognition, housed in a weatherproof, climate-managed enclosure. In automated mode it cycles the full 38-prism array on a two-hourly schedule, day and night, with sub-second pointing consistency that eliminates the human error dominating manual rounds.
- Leica GPR1 / GMP104 prisms on forced-centring brackets, selected for long-term stability in a dusty, high-UV environment.
- Leica LS15 digital level with invar staff for the independent ±0.3 mm vertical settlement cross-check on the levelling-stud array.
- Leica Viva GS18 GNSS receivers for the reference-network observation and as a continuous broad-area check on the crest pillars to ±3–5 mm.
- Leica GeoMoS (Monitor + Adjustment) running the automated schedule: it applies real-time atmospheric (temperature/pressure) corrections, runs statistical significance testing on every epoch, drives the SMS/email alarm layer, and serves the live web dashboard. Power and comms were solar with 4G backhaul, sized for wet-season cloud cover.
Atmospheric refraction — not instrument resolution — is the limiting factor over the longest sightlines across a heating tailings beach, so met-corrected scheduling and read averaging were designed in from the outset to suppress false alarms.
The result
The baseline confirmed a system repeatability of ±1.2 mm horizontal (3D) on the prism array and ±0.3 mm vertical on the levelled studs, comfortably inside the trigger levels set for the facility. Crucially, the programme replaced four data points a year with roughly 4,380 epochs a year per point.
The value showed within the first wet season. Where the previous quarterly regime saw only a slow, ambiguous crest trend, the automated system isolated a localised acceleration on the downstream toe — movement that climbed to 4.8 mm/week against a near-static surrounding array. Read against the tied-in piezometers, the signal correlated with a rise in pore pressure consistent with an underperforming toe drain. Because velocity itself was an alarm condition in the TARP, the system flagged the trend while cumulative displacement was still in the green band — well before the amber trigger that a cumulative-only programme would have waited for.
The geotechnical engineer mobilised toe-drain investigation and remediation on the strength of defensible, time-dense data rather than a single suspicious quarterly reading. Subsequent epochs confirmed the toe velocity returned to background after the drain works, and the live record provided an auditable demonstration that the response had worked.
The outcome
The operation now holds a continuous, alarmed deformation record on a high-consequence TSF that maps directly onto its dam safety management system and satisfies ANCOLD surveillance expectations and the Global Industry Standard on Tailings Management. The installed cost of approximately AUD $96,000 plus an ongoing monitoring fee was never weighed against production throughput — on a high-consequence dam the comparison is simply the cost of monitoring against the consequence of an undetected breach, and on that basis the programme was straightforward to justify.
Three outcomes mattered most to the client. First, decisions are now made on rate of change, not just cumulative magnitude, so emerging problems surface weeks earlier. Second, every value carries a stated measurement uncertainty, so the engineer of record can defend the data under regulator and insurer audit. Third, the same instrument and methodology return every epoch automatically, which is what makes a single millimetre meaningful over years of record. ISS continues to operate the system under a monitoring agreement, issuing formal monthly summary reports alongside the live dashboard and immediate exceedance alarms.
Frequently asked questions
Why was an automated system chosen over continuing periodic campaigns?
Periodic GNSS and levelling told the operator the embankment had moved but not whether the movement was accelerating — and on a High C consequence dam, a three-month wet-season blind spot was unacceptable to the engineer of record. An automated robotic total station reading every two hours resolves velocity and acceleration, which are the real early-warning signals, and alarms the moment a trigger is breached rather than at the next scheduled visit.
What accuracy did the dam monitoring system achieve?
The commissioned system achieved ±1.2 mm horizontal (3D) repeatability on the prism array and an independent ±0.3 mm vertical check via precise invar levelling. For deformation work, repeatability is what counts: movement is only reported as real when it exceeds roughly twice the measurement uncertainty, and every report states that uncertainty so the engineer can assess statistical significance rather than read figures at face value.
Which standards governed this structural monitoring dam case study?
There is no single Australian standard titled "dam monitoring". The programme was scoped against ANCOLD Guidelines on Dam Safety Management, the Global Industry Standard on Tailings Management, and the Queensland mining regulator's geotechnical management requirements, with the facility's site-specific TARP defining the trigger levels. All measurements were tied to ISO 17025 calibration and georeferenced to GDA2020/MGA2020 Zone 55 with AHD heights.
Could the system be installed without taking the dam out of service?
Yes. Monitoring is non-contact — prisms are read remotely — so the facility stayed fully operational throughout. The only on-site access required was brief, for installing the reference pillars, fixing the prisms, and commissioning the enclosure, none of which interrupted tailings deposition.
How long did installation and commissioning take?
Roughly two weeks on site, covering reference-network observation and adjustment, prism installation, a baseline measured twice on separate days to quantify the noise floor, and a verification period before the automated system went live. The baseline campaign element is comparable to a one-to-three-day periodic survey; the additional time is the automation, power, comms, and the alarm/dashboard layer.
Request a structural monitoring quote
Dam and tailings movement is gradual, measurable, and — caught early on rate of change rather than magnitude alone — manageable. Whether you need a defensible baseline before a raise, a periodic campaign on a low-consequence facility, or a fully automated, alarmed deformation system on a high-consequence TSF, ISS will design a programme around your consequence category and your trigger levels and deliver data your geotechnical engineers can defend under audit. To scope structural monitoring for your dam or tailings facility, contact Industrial Spatial Solutions on 0407 057 015 or request a fixed-price quote and we will respond within one business day.