Author(s): John Ewen; Greg O'Donnell; Josie Geris; Will Mayes; Mark Wilkinson; Paul Quinn; Enda O'Connell
Keywords: Flooding; Land use change; Land management change; Vulnerability mapping; Rainfall-runoff modelling
Abstract: Over the past fifty years, significant changes in land use and management practices have occurred in many countries, driven by agricultural intensification policies. In the UK, there is substantial evidence that modern land use and management practices have enhanced surface runoff generation at the local scale; such local impacts can be avoided or mitigated through the adoption of better land management practices and/or small scale surface runoff control measures. There is little evidence that local scale changes in runoff generation propagate downstream to create flood impacts at the larger catchment scale. This does not imply that impacts do not exist, but the very few studies in which evidence has been sought have not produced any conclusive evidence, largely as a consequence of a lack of suitable data and the dominating influence of climatic variability. There is also the question as to whether local scale runoff control measures could form part of a broad catchment flood risk management strategy. An integrated programme of multiscale catchment experimentation and modelling is being undertaken in the UK to gain a better understanding of how small scale changes in runoff generation propagate to larger catchment scales, and to support decision-making on flood risk mitigation. New information tracking and adjoint modelling techniques are being developed within a distributed, catchment scale, Source-Pathway-Receptor (SPR) model. Predictions in the form of vulnerability maps are being produced that can inform catchment managers as to where in the catchment landscape mitigating interventions might best be made to mitigate downstream flooding. Despite the urgent need for prediction for decision-support, sight has not been lost of the fact that data on downstream large-scale impacts are very scarce. The nature and role of complexity and simplicity in modelling are discussed.