We use numerical groundwater models to support our allocation planning and licensing decisions. We use these highly complex and resource-intensive tools when:
- groundwater resources have or are reaching their sustainable yield
- there is a risk to the resource or its dependent values
- there are multiple competing demands on the resource
- simpler analytical solutions are not suitable.
This project provided workflows, tools and modelling capacity to enable more efficient modelling practices and reduce run times for regional-scale groundwater models. It is a partnership between our State Groundwater Investigation Project, the Water Corporation and the University of Western Australia.
What we did during the groundwater investigation
The project developed a series of ArcGIS tools based on the Perth Regional Aquifer Modelling System (PRAMS) v3.5.2. The tools allow model input files for MODFLOW to be generated and output files to be processed and visualised within the ArcGIS platform.
Key findings and how we are using the information
This project developed a method to generate a robust surrogate model for PRAMS with significantly reduced model run times. Our team has used the surrogate model for detailed model calibration and predictive uncertainty. The research was published in the Water Resources Research journal.
This project has improved:
- how we develop and use numerical groundwater models
- modelling efficiency
- confidence in model outputs through uncertainty analysis.
It has also built our capacity, and that of our project partners, to undertake advanced modelling techniques.
Where to get more details
Siade AJ, Cui T, Karelse RN & Hampton C 2020, ‘Reduced-dimensional Gaussian process machine learning for groundwater allocation planning using swarm theory’, Water Resources Research, vol. 56, issue 3: https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2019WR026061
You can request more information on this project by emailing email@example.com.
Find out more about our groundwater investigations across Western Australia.