Hunter Water is a State Owned Corporation (SOC) providing drinking water, wastewater, and recycled water services to a population approaching 600,000 people in homes and businesses across the Lower Hunter in New South Wales, Australia.
With over $2.9 billion in assets, Hunter Water is actively looking to adopt new digital technologies to improve resilience, provide exceptional value to their customers, and protect the environment.
Grahamstown Dam is Hunter Water’s largest drinking water supply dam, providing 40% of water to the region and meeting up to 75% of the company’s daily supply requirements. Constructed between 1955 and 1965, it impounds a reservoir of around 182,305 million litres behind a 5-km-long embankment.
Daniel Turnbull is Hunter Water’s Dam Safety Engineer. His field operations team are responsible for inspecting their entire dam portfolio, assessing whether they meet modern standards, and ensuring risks are effectively managed. Their objective is to establish trends that indicate emerging problems.
However, the need for comprehensive monitoring of Grahamstown Dam presents several challenges. “Our biggest challenge is to ensure monitoring is targeted and effective,” explains Daniel. “It is not physically practical to check every aspect of the dam, especially with a dam like Grahamstown where we’ve got over 5km of embankment.”
Grahamstown Dam presents many additional challenges, not least because the top of the dam is home to a major road with an 80kph speed limit. In addition, the shoulders of the dam are constructed of sand, which moves during normal operations. The stable portion of the dam is the solid clay core, which is directly underneath the road.
Monitoring under these circumstances was costly and inconvenient for road users, and
a health and safety concern for Hunter Water’s staff.
“We could only perform visual inspections of the embankments, which meant no measurable data,” says Daniel.
Hunter Water began to explore alternative approaches to traditional measurement surveys and visual inspections. Daniel and his team investigated the use of drone flights which could provide coverage of the embankment several times a year, relatively cost-effectively.
“Drones just can’t achieve the millimetric accuracy of satellite data. Our comparisons showed drones pick up 20 to 30 millimetres; whereas satellite data detects movement of 1 to 2 millimetres. That’s hugely significant.” Daniel Tunrbull, Hunter Water.
Reliant on photography and their engineers’ opinions on the changes that take place year-on-year, Hunter Water was looking for data-led solutions that provide accurate measurements and analyse trends over time.
Having explored other methods, including aerial surveys, Hunter Water signed a three-year subscription for Rezatec’s Dam Monitoring product, leveraging geospatial analytics to track structural and environmental changes at Grahamstown Dam. This was offered in collaboration with local partners Detection Services, who specialize in supporting multi-discipline technologies for the water and wastewater industries.
Rezatec’s Dam Monitoring solution uses Synthetic Aperture Radar (SAR) to track ground motion and multispectral data to study vegetation vigour and moisture, both of which are indicators for potential dam failure.
One of the unique features of this technology is the ability to call upon a library of historic data to ‘look back in time’ and establish a baseline for what is normal behaviour for any given dam.
In the case of Hunter Water, Rezatec analysed three years of historic SAR and two years of multispectral data to provide a retrospective analysis. This first step is critical to identifying anomalies, or observations outside ‘normal’ thresholds, by establishing a baseline trend of what is ‘normal’ and what isn’t.
The retrospective analysis on Grahamstown Dam spanned a period of time where, unbeknown to Rezatec, several pieces of major work had been completed. Firstly, pipes had been installed through the crest of the dam where, for a short time, there
was an open excavation. Secondly, top soil had been applied to the embankment to generate vegetation growth.
Using this baseline trend data, Rezatec and Hunter Water agreed the normal behaviour threshold. This enables the analytics to identify anomalous behaviour and notify Hunter Water’s team where to direct their resources.
As part of their three-year subscription, Hunter Water receives monthly data updates which pinpoint the location of potential issues. Data is presented on Rezatec’s online portal, providing both a dashboard view, as well as detailed reports on anomalous ground movement, vegetation vigour and moisture.
Additional datasets from the client, such as water level information, can be incorporated into the model to further improve accuracy.
Camilla Braithwaite, Product Manager at Rezatec explains: “Water level and seasonality can affect vegetation. Adding these into the analytics means we can identify observations that are truly anomalous. Vegetation is a good indicator for seepage, which is a worry for dam owners. It’s not something they can easily spot immediately. We provide dam operators with an understanding of what’s normal, benchmarking and then identifying anomalies.”
Dam Safety Engineer, Hunter Water
During the retrospective analysis, Rezatec identified exactly where the work had been undertaken and when. That gave me confidence it was accurately picking up movement, as well as vegetation vigour.
General Manager, Detection Services
Rezatec’s Dam Monitoring product is a game-changer for dam owners and operators. The ability to remotely assess infrastructure accurately and frequently, without sending teams into remote or unsafe locations, is a huge benefit.
The introduction of Rezatec’s Dam Monitoring solution means that Hunter Water is no longer reliant solely on routine visual inspections to monitor the integrity of Grahamstown Dam. The team still conduct inspections, but these are now less frequent and more targeted.
With monthly updates and millimetric accuracy, Hunter Water now has complete confidence they are doing everything in their power to exceed regulatory requirements and safeguard the dam and the surrounding population.
Geospatial AI allows dam owners to:
• Improve consistency and accuracy of monitoring
• Reduce inspection costs and risk to personnel
• Demonstrate a greater Duty of Care
“Regulations for an extreme consequence category dam require a movement survey once a year. For a high consequence category dam, it’s every two years. We wanted to have more scrutiny than the bare minimum that we’re required to have,” adds Daniel. “The cost of Rezatec’s product, which provides accurate monthly updates, is less than the cost of a single survey using traditional techniques.”
Perhaps more importantly, Rezatec’s Dam Monitoring provides Hunter Water with a repeatable, testable, accurate and higher frequency solution compared with sending a team to a site. Now convinced about the value that geospatial AI is delivering, the team at Hunter Water is looking to apply Rezatec’s dam monitoring solution to other dams within its portfolio.
Take a look at geospatial AI in action. Explore how satellite data can provide a view of your entire asset base across vast areas. Discover how analytics drill down to a level of detail you never knew existed.
