Ecological systems are complex and dynamic. Freshwater ecological systems even more so, driven by climatic cycles that can span across annual and decadal scales. Over these time scales, perturbations can come in many forms, with shifting water conditions impacting both water availability and water quality. The past 20 years has seen a growth in predictive modelling as an approach to advance our understanding of the complex interactions and outcomes in freshwater systems and to predict potential change in systems under shifts, such as climate shifts. Predictive models have also been used to explore alternative pathways to understand how we can further develop systems or rehabilitate them. Increasing predictive models have been used to bring together multiple disciplines and diverse knowledge types to interface across science and policy to achieve impact.
Overtime, predictive models have evolved from simple stressor-response models, to causal models that capture systems, evolving to data-based dynamic models representing complex interactions. This presentation will focus on the key lessons on the pros and cons of using predictive models, using case studies drawn from the Murray-Darling Basin, Northern Australia and international examples. Case studies will also be used to describe a future of predictive modelling for working across disciplines, using new technologies that span data analytics to digital twins, and developing of predictive tools in data-poor environments. A few lessons on bridging science and policy will also be shared.