Townley, L.R., Brooker, A.M.H, and Reid, L.B. (2013), 40 Years of Groundwater Modelling in the Perth Basin, Western Australia, IAH Congress, Perth, September.

Groundwater modelling is not new. Modelling can be criticised for its weaknesses, yet modelling remains one of the only ways to predict the future behaviour of groundwater systems, in response to changes. While commercial software is readily available and models can be set up and run with only a few days of training, far more experience is required to develop robust models that take into account physical processes, historical measurements and estimates of future stresses, in such a way that the level of confidence in model predictions can be understood by those who need to understand. The purpose of this paper is to summarise the history of groundwater modelling in the Perth Basin, in the context of the evolution of software capabilities during the past 40 years and a similar evolution in the questions being asked. While our focus is on the Perth Basin, the history of modelling near Perth is an analogue for modelling in many parts of the world. Early modelling by Burton (1976) and Pollett and Wiese (1977) using PLASM was followed by the Perth Urban Water Balance Study (1987) and the Perth Urban Water Balance Model, developed using the Golder Package. In the 1990s, the Golder Package was coupled to AQUIFEM-N, to represent deeper aquifers. This model was replaced by the Perth Regional Aquifer Modelling System (PRAMS), developed using MODFLOW. In the last 10 years, the South West Aquifer Modelling System (SWAMS) and the Peel-Harvey Regional Aquifer Modelling System (PHRAMS) have been developed for parts of the Perth Basin to the south of metropolitan Perth. Other models have been developed to the north of Perth. While the fundamentals of groundwater modelling have not changed, the complexity of models has been continuously increasing. The resolution of models has increased by four orders of magnitude, from hundreds to millions of nodes or cells. Representation of the land surface boundary condition (recharge and discharge) remains difficult, and there is no consensus on the best way to represent the coastal boundary condition, along or beneath the Indian Ocean. Recent interest in geothermal energy in the Perth Basin has led to investigations of the effects of temperature and density on regional flow, and there is also increasing interest in depositional processes that led to formation of the Perth Basin, and processes by which the salinity structure of the Basin has evolved. Confidence in model predictions will improve if we learn lessons from the past, and when we truly integrate our knowledge of processes and the numerical algorithms used to simulate those processes.