Once again we publish an issue that contains a variety of papers dealing with the widely different aspects of flood risk management. From the titles one could conclude that, other than in the general area of flooding, these are unrelated topics. Yet they share an important aspect: that is the extent to which basic science underpins their argument. In particular I am encouraged by the papers that rely on experimentation, whether these be in the physical sciences, or in the social sciences that are of fundamental importance in flood risk management. New hypotheses need to build not just on the current state of the knowledge but on a sound evidence base. Experimentation as a means of collecting the objective evidence on which conclusions can be drawn can be difficult, time consuming and therefore expensive. Yet robust conclusions can only be built on that basis. The temptation to rely on anecdotes must be resisted.

This principle is in general well understood in academic circles. Many of the excellent manuscripts we receive arrive from academic teams at the forefront of their research area. However, we also encourage practitioners to write for our journal and the needs and constraints of the practitioner are somewhat different. Pressures on time and cost can lead to an over reliance on anecdotal information from a project area that is not verified through independent objective evidence. Even where experimentation is included in the analysis, it can be compromised by preconceived ideas about the results. This is understandable but not excusable. Sound science is as important in the area of professional practice as it is in research. This is especially true in some of the emerging areas. Recently the Journal has received a number of manuscripts concerned with the active management of flood risk. Here the role of community engagement in implementing pre-emptive measures and mitigating the consequences of flooding can be an important element in managing flood risk. For effectiveness of these functions to be fully evaluated, sound experimentation is needed to provide the evidence on which valuable conclusions can be made. This is particularly true if those conclusions are to be useful in areas beyond the study area. In both industrialised and developing countries we need to be able to take the many lessons learnt from such studies and translate the findings into strategies for practical implementation. If the process is flawed then the potential benefits will not be realised.

I would, therefore, encourage all those embarking on quantifying the effectiveness of non-structural measures in managing flood risk to ensure that their experimental approach to collecting evidence is objective and scientifically sound. This will differentiate their work from less profound studies, and ultimately make it more usable to the wider scientific community and in flood risk management practice.