A collaboration between eWater Group, CSIRO and the Murray–Darling Basin Authority is delivering a step change in how Australia models river behaviour during dry periods, with implications for water sharing, environmental flows and drought planning across the Basin.
The work centres on a new modelling approach known as Alluvium as River Storage, or AARS, which has now been integrated into eWater Source, the national river system modelling platform used by Basin governments, consultants and researchers.
The research was presented at the Engineers Australia Hydrology and Water Resources Symposium in Canberra in November 2025 and forms part of the Australian Government–funded Murray–Darling Water and Environment Research Program.
- Want more information on what Inside Water is up to? Sign up for our weekly email, landing in inboxes with the latest news.
- Do you want the magazine delivered to your letterbox? Sign up here to subscribe.v
Low-flow conditions are becoming more frequent and prolonged across large parts of the Murray–Darling Basin as the climate becomes hotter and more variable. During these periods, accurately estimating how much water remains in rivers and how much is lost to evaporation or seepage into surrounding alluvial systems becomes critical for decision-making.
Traditional river models typically apply fixed loss rates, which can oversimplify how rivers behave after extended dry or wet conditions. This can lead to overestimation of downstream flows, particularly during drought, increasing uncertainty for water managers tasked with balancing consumptive use, environmental needs and critical human water requirements.
AARS addresses this limitation by explicitly representing the interaction between rivers and surrounding groundwater systems. It treats alluvial sediments and connected groundwater as a dynamic storage that can absorb or release water depending on antecedent conditions, effectively giving the model a memory of past wetness or dryness.
Testing by CSIRO demonstrated that AARS significantly improves the simulation of observed low-flow events in several northern Basin catchments, including the Border Rivers, Gwydir and Barwon–Darling systems, compared with conventional loss methods.
To move the approach from research into operational use, eWater Group worked with CSIRO and the MDBA to implement AARS as a community plugin within eWater Source. This allows users to apply the new method within existing models without changing established workflows or software architecture.
Testing on the Macintyre River confirmed that the plugin performs consistently with earlier research implementations, while remaining stable in complex modelling environments. The integration also marks the first time a groundwater-based process has been added to Source as a community plugin, expanding the platform’s flexibility.
For water agencies and planners, the improved representation of low-flow dynamics has practical implications. More accurate modelling supports fairer water sharing decisions, better timing and targeting of environmental water, and stronger confidence when planning for drought and extreme climate conditions.
The work also reflects the value of collaboration across science, policy and technology. CSIRO’s hydrological research, MDBA’s system-scale water management expertise and eWater Group’s software development capability have combined to translate complex science into a tool that can be applied consistently across jurisdictions.
Further work is planned to explore whether AARS concepts could be embedded even more deeply into Source over time, strengthening representation of surface water and groundwater interactions as climate pressures intensify.
For now, the AARS plugin provides a practical and immediately usable advance in low-flow river modelling, offering water managers a clearer view of how rivers behave when every drop counts.