High resolution ¯ow modelling in hydrology and geomorphology
This Issue brings together a collection of papers dealing with the adoption by hydrologists and geomorphologists of sophisticated Computational Fluid Dynamics (CFD) techniques for simulating interactions between catchments and river channels and ¯ow and sediment transport processes at a variety of scales. Until relatively recently, such schemes were predominantly employed by engineers and applied mathematicians, and used to tackle industrial ¯uid ¯ow problems for design and development studies. Whilst models appropriate to environmental ¯uid dynamics have been available since the early 1970's, the complexity of even the simplest applications, in terms of topography, friction and turbulence characteristics, prevented their widespread adoption. However, continued improvements in process representation within these models have progressively improved their basic appeal. Since 1990 in particular, rapid numerical and computing advances have increasingly allowed these models to be applied to problems associated with much more complex boundary conditions typically of interest to the geomorphologist and hydrologist. The space, time and dimensional resolution at which such models may be applied has reached a stage where typical simulations can involve many thousands of grid cells, simulated for hundreds of thousands of time steps in anything up to three dimensions. The process representation and the feasibility of application are both such that CFD approaches are being enthusiastically adopted by some environmental scientists and used to examine a wide range of hydrological and geomorphological problems.
The papers contained in this Issue re¯ect much of current thinking and give some indication of the future potential of this area of research. In terms of model development, the physical processes which are beginning to be addressed constitute fundamental scienti®c problems. CFD models have to be informed by both theoretical and empirical research. Improvements in process representation often require innovative development of modelling methods, a classic example of which is the problem of dynamic moving boundaries (Tchamen and Kawahita). With improvements in process representation these models themselves can provide a further and novel means of gaining process insights into, for example, the nature of ¯ow structures that form at river channel con¯uences (e.g. Bradbrook et al.) and sediment transport (e.g. Moulin and Ben Slama; NinÄ o and Garcia).
In hydrology, CFD models are replacing simple one dimensional representations of surface ¯ow processes in catchment models (El-Hames and Richards; Steward et al.). These add to the physical realism of existing hydrological research, whilst at the same time rede®ning the requisite ®eld data collection strategies to generate necessary validation, calibration and parameterisation data. The latter don't go away, but manifest themselves in new ways. In geomorphology, these models have the potential to connect form and process in a way that hitherto proved impossible (Darby; Mosselman) and the increasing realism of high resolution approaches enables highly complex problems to be tackled (Lane and Richards; Hodskinson and Ferguson; Nicholas and Walling). Nevertheless, a certain degree of caution is perhaps necessary. Both the simulation problems and modelling techniques themselves are relatively sophisticated, and such applications may not be straightforward (cf. Bates et al.). It follows that because some of these research developments were previously in the domain of applied mathematicians and engineers, it is critical that hydrologists are familiar with the basic principles surrounding their use, and some of the fundamental constraints upon their adoption, if only to allow the `high ground' to be claimed. Further, whilst perceiving a veritable panacea of potential model applications, a modeller must give careful consideration to the assumptions that are inherent in the model being used to avoid mis-application of models to particular problems.