RECENT ADVANCES IN THE INTEGRATED SIMULATION OF SURFACE AND SUBSURFACE FLOW AND MASS TRANSPORT

Models that simulate coupled (integrated) surface water and groundwater flow are becoming increasingly popular to address issues related to water resources, such as the impact of climate change, increased water consumption or degradation of water quality. Several coupled models have been recently presented in the literature and there is also an increasing number of published case studies demonstrating the application of such models. The current development of integrated models follows the blueprint presented by Freeze and Harlan (Journal of Hydrology, 1969), who proposed that hydrologic systems be represented by a distributed-system model in which parameters are spatially and temporally distributed. The advantage of this distributed-system approach is that a representation of the controlling physical parameters does not require simplifying assumptions with respect to the flow processes, which makes the model applicable to a potentially large range of hydrologic conditions. These integrated models should therefore be applicable to any type of river-groundwater system. This presentation will review recent developments in the simulation of coupled surface water and groundwater flow, with an emphasis on those approaches and that are based on the 3D variably-saturated groundwater flow equation. The assumptions, governing equations and numerical formulations of various models will be compared and current capabilities for simulating mass, heat and sediment transport in hydrologic systems will also be reviewed. Although models of increasing complexity have been recently developed, challenges remain with respect to the further development of integrated hydrologic system simulations. An example is the representation of water cycling in northern climates where annual freezing and thawing modify the infiltration properties of surficial soils. Another challenge is that distributed (integrated) models can become too computationally intensive for some applications, which will require faster numerical methods, such as parallelization, or will force the use of simplified models. More work is therefore needed to identify hydrologic characteristics for which more simple and less computationally demanding approaches are adequate.