The papers in this issue of Water Management focus mainly on mathematical and numerical modelling with one exception looking at the issues of delivering clean water in developing countries. However, these issues are not as separate as they might first appear. As engineers we must consider what Professor Michael Abbott called the ‘socio-technical' arena (Abbott, 1996). The range of papers presented reflects well the research that Abbott has focused on, with his early ground-breaking work on computational hydraulics (Abbott, 1979; Abbott and Ionescu, 1967) and his later work on the social significance of water engineering and the impact that societal and informatics developments have had on water engineering. This led to the development of hydroinformatics (Abbott, 1991; Abbott, 1993; Abbott, 1999) and in fact, all of the papers in this issue could be considered as hydroinformatics if we take a broad definition of that much used and abused word.
In the first paper (Horritt et al., 2010) present a comprehensive dataset for creating and testing flood inundation models along with an example of one such model. This dataset is for the 2005 inundation in Carlisle and covers not only data for creating the model, but also validation of the model. The former combines data from Lidar, Mastermap and ground surveyed cross-sections whilst the validation is done with a post-flood survey of debris, wrack lines and water marks using differential GPS (dGPS). In addition to the dataset the paper also presents a model based on the full shallow water equations, an interesting comparison to earlier work based on a diffusion wave model (Neal et al., 2009). In the context of calibration of inundation models it is interesting to note that Manning's n values for the main channel and the floodplain are calibrated against different datasets which may appear unusual, but which is justified well in the paper and which warrants careful consideration by those involved in model calibration. The authors note some interesting differences in the range of Manning's n needed for calibration between each model. Overall the paper brings new knowledge to the important area of urban flood modelling which is attracting increased interest (Hunter et al., 2008; Lin et al., 2006; Schubert et al., 2008).
The following two papers use similar techniques for two different problems: a complex spillway and a gated tunnel (Jacobsen and Olsen, 2010; Najafi and Zarrati, 2010). In each case the rapid changes in the free surface mean that the shallow water equations are not applicable and that the fully three-dimensional Navier-Stokes equations must be used. Until the past decade such solutions were rarely used due to the high computer performance and specialist knowledge required. However, they are now becoming more commonplace. In such applications the depth of the flow is not calculated directly and a volume fraction is calculated throughout the domain. From this the position of the free surface can be established by examining where the volume fraction goes from zero to one, a technique pioneered in the code Flow-3D (Hirt and Nichols, 1981) that is used by Jacobsen and Olsen. Such techniques need careful application, particularly when constructing a mesh and this can be seen in the second paper (Jacobsen and Olsen, 2010) where the geometry is complex and the authors have taken care to consider the sensitivity of the results to this. The good agreement seen here should increase confidence in the use of these techniques. All such studies benefit from validation against high quality experimental data as shown in the third paper (Najafi and Zarrati, 2010). The authors compare the computed and measured data for several variables; they then compute the percentage errors and consider how these changed across the domain. Finally they show how the numerical results can be used to establish a rating curve for practical use – an important point as fully three-dimensional models are not appropriate for use in practice due to the long run times.
The fourth and fifth papers form a pair and go beyond numerical modelling to consider in detail the mathematical and physical models that underlie the computer simulations. The analysis in Part 1 is elegant and informative (Leal et al., 2010a). There is first a detailed consideration of the physical situation, this is then used to postulate a set of equations before implementing these in a numerical form in Part 2 (Leal et al., 2010b). All models have these stages implicit within them, but we often forget the first two and focus on the latter. The flows under consideration in these two papers are debris-laden or ‘muddy' flows that are the result of the movement of steep waves over a mobile bed. Given that these are the sort of flows seen below dam failures, their consideration should be more widespread given the increased interest in dam safety in the UK and the increased likelihood of glacial lake outbursts (Carrivick, 2010; Carrivick and Rushmer, 2006).
The final paper does not involve any numerical modelling, but tackles an equally difficult issue (Peters and Mohamed, 2010). Increasing access to water in developing countries is a key objective of the UN's Millennium Development Goals (UN, 2005). This requires not only funding to establish the system, but a sustainable business model to ensure maintenance and continued operation. The business can be private, public or a combination of both, but it must be sustainable. This paper considers the question of whether users are prepared to pay more for improved services and what variation there is in attitudes across different groups. Their conclusions will be useful to anybody involved in advising on or operating water supply in developing countries. Further their conclusions on non-payers may be of interest in the UK given the variation in debt levels across the UK water companies (Clarke et al., 2009).
In conclusion, I would highlight two main messages to take away from these papers: that our work as civil engineers takes place within a wider society and that this wider context must always be considered rather than taking only a technical view; and that all modelling is based on physical assumptions that may or may not apply in the situation we are considering – we must always remember what lies behind the computer screen.
