Review of Computational River Dynamics by Weiming Wu: Taylor and Francis; 2007; ISBN 9780415449618. Price: $189.95.

 The first time I discussed this book with my Ph.D. students, the immediate question was how many additional copies we should buy. The book contains the most extensive coverage of formulas for numerical modeling of sediment transport that I have seen. It is well written with readable English and informative figures. The topics cover most areas of numerical modeling of rivers, from the equations describing the water flow to discretization in 1, 2, and 3 dimensions. Of the twelve chapters, one discusses mathematical description of water and sediment flow, giving the basic formulas used. Another chapter covers fundamentals of numerical methods, with discretization principles and schemes, the finite difference and volume methods, coordinate transformation, pressure-algorithms and solvers. Then three chapters cover 1D, 2D, and 3D modeling, respectively, with special formulas used in each of the three approaches.

In addition to the numerical methods, some necessary physics are described in separate chapters on classical sediment transport, cohesive sediments, vegetation effects and contaminant transport. These chapters cover special formulas for the physical processes and their inclusion into the numerical model. Experience has shown that the accuracy of the numerical models is often limited by the correct empirical formulas for the physical processes, so the inclusion of these chapters is therefore necessary. Sediment transport modeling has been given a particularly detailed coverage in the book. A separate chapter has also been given for dam-break modeling, covering special formulas and numerical algorithms for rapidly varying flows. Dr. Wu has also included a few examples in the book, where results of the numerical models are given.

Computational river dynamics covers both numerical modeling and river physics. A complete coverage of the field including derivations of all formulas would produce a book with several thousand pages. The current book has only 494 pages, so Dr. Wu has had to make choices in what to present. He has included a fairly large number of formulas, in total 1,253, while most derivations have been omitted. Priority has been given to formulas and algorithms especially used in river engineering. The number of formulas and their complexity is still extensive and can easily scare novices. Beginners in this field should be advised that it is not necessary to be familiar with all the formulas in the book to be able to use numerical models in river engineering.

Although the book covers a wide range of subjects, it does not contain the state of the art in all fields. This would not be possible in the huge scientific area of computational fluid dynamics (CFD). The 3D CFD modeling chapters omits recent work on multigrid solvers, non-isotropic turbulence modeling, wetting/drying algorithms, and complex nonorthogonal unstructured grid generation. A large part of the book is devoted to algorithms for 1D and 2D depth-averaged flow. Algorithms for 3D modeling assuming a hydrostatic pressure distribution in the vertical direction are given. Solving the fully 3D Navier-Stokes equations without the hydrostatic pressure assumption is described briefly, and Dr. Wu has included an example where he has used such algorithms in computing a local scour hole. The results compare very well with experiments, both with regards to the scour depth and the geometrical shape. The example shows the usefulness of numerical models and their potential in solving practical engineering problems. Dr. Wu has not, however, included references to the many earlier articles on the same topic. Readers should be aware that a large number of studies where 1, 2, and 3D numerical models have been used successfully to compute river and sedimentation engineering problems are not referenced in the book.

One inconsistency in the book is that tensor notation is used for the formulas in some chapters, for example, on 3D numerical models, but not in others, for example on 2D modeling and contaminant transport. Many of the equations are large and look fairly complicated, and could be simplified by using tensor notation. On the other hand, some readers might think tensor notation is complicated in itself.

Dr. Wu’s book includes a considerable number of empirical formulas for many of the processes in the river. Many formulas contain parameters that must be calibrated for each case. An example is the adaptation length, $L_s$ , used in the nonequilibrium formula for sediment transport. The book contains an interesting explanation of the parameter and the process on page 187: clear water flowing into a straight flume where the bed is covered with sediments. It seems the nonequilibrium effects are related to the depth-averaging in 1D and 2D models. The case on page 187 is similar to a study by Apmann and Rumer (1970), who computed the case without the use of an adaptation length parameter. The results compared well with experimental studies. The case was also computed by a number of other researchers, for example, Hjelmfeldt and Lenau (1970) and Van Rijn (1987) without the use of an adaptation length. The nonequilibrium effects were computed directly by resolving the variation of the sediment concentration in the vertical direction in the grid. I think that in theory, it should be possible to make a 3D CFD model with sediment transport without site-specific calibration parameters. Empirical formulas with universal constants could be used to compute the bed form height and roughness. In my opinion, such a model is a feasible goal for our research in this field of science.

Who would I recommend this book to? With the large number of equations presented in the book, a user might ask which formulas and approach to be used for a given situation. Some of the formulas can have a varying degree of documentation, and it can be difficult to understand the assumptions a few of the formulas are made under. The reader must therefore have some experience in this field of science to be able to use the formulas correctly. I think the main audience is researchers working in the field of computational river hydraulics. But engineering professionals will also benefit from the comprehensive detail of the book. 3D CFD models have been used as a standard tool in mechanical engineering for more than 20 years now. The use of numerical models in river engineering is currently increasing, and in the future, these tools will also be essential for consultancies working in water hydraulics. Then literature on the topic will be needed, including a thorough documentation of the formulas used by the computer programs. I therefore believe Dr. Wu’s book is valuable contribution to the field of computational river dynamics.