Over the past few years Australia has witnessed extreme climate events from widespread drought to bushfires and severe flood events. Our Water Resources team strive to stay at the forefront of supporting a more climate resilient future for Australia. Through our research and techniques, we are looking to drive industry change to deliver future-proofed, climate resilient designs that improve flood risk management.
A focus of our water specialists has been revaluating industry procedures for rainfall runoff modelling. Rainfall runoff modelling addresses how rainfall is translated into flooding, through hydrology; calculating the volume of water, and hydraulics; movement of water across the surface of the ground. In short, a rainfall runoff model is a computer simulation that is used to generate volume estimates. The volume estimates are used as the basis for defining water flows for the design of infrastructure ranging from drainage pipes to dam spillways.
Improving the accuracy of rainfall runoff modelling
The industry guidelines document ‘Australian Rainfall and Runoff 2016’ dramatically changed the approach we take to conduct rainfall modelling, from a simple algebraic calculation to far more sophisticated approaches.
“While research into statistical techniques has progressed, it hasn’t improved our understanding of how the water physically translates from the sky to flooding. These new techniques involve substantial increases in computational effort but uncertain improvements in accuracy.”
– Tim Rhodes, Water Resources Technical Principal.
We saw an opportunity to improve accuracy by making use of readily available data and improving analysis through new software tools. We embarked upon an ambitious program to re-define model set up, calibration and validation procedures.
“Our approach was to understand the details within our models to improve their reliability and accuracy. We have developed a program which includes a range of sub projects, each with the goal of exploring and characterising some aspect of the hydrology modelling process,” says Tim.
In seeking to understand the details within our models we identified four key considerations:
Consistent model setup
Consistency in modelling enables us to achieve an accurate outcome. “We have sought to understand model deficiencies and modify the set-up process for models accordingly. This means any modeller can apply the model in a consistent manner” says Hugh Nguyen-Mallen, Water Resources Senior Engineer.
Data driven model parameters
We created scripts that download big data sets from publicly available records. This enables us to download as much realistic data as we can about the nature conditions of catchments. We have downloaded over 10,000 historic records and analysed the data to develop localised and regional relationships. “Real data allows us to define the characteristics of the catchments and establish accurate model parameters” says Hugh. “A new model calibration method has been developed, which we are presenting to clients”.
Extreme flood method revision
When designing dam spillways, it is important to identify realistic extreme flood estimates. “We have partnered with Griffith University, Queensland and are working with students on research collaborations to characterise the largest floods and associated floods in Queensland.” says Tim.
This research collaboration extends work published in the Australian National Committee On Large Dams 2018 Conference that provided benchmarking techniques for the largest floods in Australia.
Validation procedures for model outcomes
Importantly, we benchmark the outcomes from our analyses against real world records, allowing us to assess whether the flow estimates are realistic.
Seeking progressive, industry wide change
We identified that some industry procedures lead to significant overestimation of flood volumes. When applied to infrastructure sizing, this can often lead to overspending. Our procedures seek to increase certainty, accuracy and improve efficiency in infrastructure spending.
These techniques improve accuracy in our modelling and improve our understanding of current flood challenges. It also gives greater certainty when we apply our modelling to manage climate change resilience, to develop future-proof, climate resilience designs that improve flood risk management, or to make efficient use of materials in infrastructure design.
As we are proposing different techniques to the industry standards, there are barriers to overcome to achieve progressive, industry wide changes. “Communication for this process is key. Communicating the problem and the benefits of our solutions. To support our new techniques, we have been publishing our research in conference papers for peer review” says Hugh. “Our Queensland and Victorian water resources teams have been in collaboration to expand the research and apply these new techniques across different jurisdictions.”
Tim is a water resources technical principal with over 20 years’ experience. Tim is responsible for high level scoping of project requirements, review and team direction. In addition, he is responsible for project technical direction to ensure that the outcomes are consistent with project and client requirements. Tim has an excellent understanding of the design, function and operation of water resource infrastructure. He has had lead roles on a range of major investigation and infrastructure projects located both within and outside Australia.
Hugh is a senior water resources engineer with 10 years’ experience. Hugh has experience in the design, function and operation of a wide range of infrastructure including highways and railways, dams, wetlands and waterways. Key skills include hydrologic and hydraulic modelling, drainage and terrain design, and project management. He is experienced in applying the latest modelling techniques to urban and regional flood modelling including ARR2019 guidelines and rain-on-grid.