Modelling approach

The Australian continent has approximately 1000 estuaries of moderate to large size, each with its own unique shape, and each subject to unique combinations of human and natural pressures. Due to constraints of time and data availability, we have not tried to model the estuaries identified by the Audit in specific detail. Rather, we have divided estuaries into three hydrodynamic Estuarine types, and developed simple coarse resolution circulation models for each type. We have further identified a relatively small number of estuarine parameters which characterise the shape and the physical and chemical forcing over time. We have then conducted simulations for a small but representative set of values of each of these parameters. This has required some tens of thousands of model simulations (the "curse of dimensionality").

For each parameter combination, the model was run for 10 years, by which time the model state variables had reached a repeating seasonal cycle (the season cycle is a result of the seasonal forcing specified by the climate zone). The initial conditions, which were the same for all simulations, represented a mesotrophic estuary with a viable benthic community. The 10th year of the simulations represents the ecological state that the model predicts after 10 years of the forcing specified by the estuarine parameters. It should be noted that using an initial state with a viable benthic community may not be appropriate in highly degraded systems, without viable seagrass and/or macroalgae benthic communities. Model results were sampled every 5 days over the last year of simulation, and these time series were condensed into a standard set of indicator statistics which are stored in a database, and can be viewed at the SERM Interface.

For the purposes of the Audit, one can think of the simulation models as defining a multi-dimensional map from Estuarine Parameter space (which one can also think of as Pressure Indicator space) to State Indicator space. The SERM Interface provided here offers the user a versatile, interactive environment for visualising and understanding this relationship. For brief instructions on how to use the SERM Interface, see our quick guide.

The user should be careful in interpreting the SERM results for a particular estuary. Firstly, the user is only able to specify relatively coarse ranges for estuarine parameters. For example, the SERM interface represents point source loads between 20 and 100 mg N m^-2 d^-1, a five-fold change in load, by a single intermediate value. Secondly, in specifying the estuary geometry and circulation, the user is limited  to a choice of 6 estuarine types. For example, the SERM interface provides only a choice of two shapes (thin or wide) for tidal estuaries . Limitations in scientific understanding and knowledge create uncertainties even in carefully calibrated models of specific systems. The coarse resolution sampling of the estuarine parameter space provided by SERM may further compound this uncertainty.