An Evaluation of a Hydrological Model Used to Predict the Impact of Flow Attenuation on Downstream Flood Flows

Abstract

A number of recent high magnitude events have shown that fluvial flooding has increasingly detrimental impacts on people and properties. There is a perception that flood risk is increasing due to several factors, including urbanisation of floodplains and climate change. This has created a need to assess the potential of using the natural function of floodplains, to slow and temporarily store water, as a means of mitigating downstream flooding.

This project applied and assessed a reduced complexity hydrological model called Overflow, which was used to predict the impact of flow attenuation on downstream flood flows. The model behaviour was theoretically consistent with physical processes, allowing it to be calibrated for the River Seven, North Yorkshire for a flood event in June 2007.

At the catchment scale, multiple, spatially distributed flow attenuation measures were shown to reduce downstream peak flood flows. However, any individual intervention could have a positive, neutral or negative impact on downstream flooding by affecting the timing and synchronicity of hydrographs from different tributaries. This demonstrated the importance of the specific geographic location of interventions.

The results were compared with a hydraulic model, HEC-RAS, which is a standard tool for flood hazard prediction in the UK. HEC-RAS is not without its own uncertainties however, so the aim of the model comparison was to benchmark Overflow against an established model, rather than validating the model with respect to reality. At the reach scale, Overflow was less sensitive to an increase in flow resistance than HEC-RAS, so it will under-predict the effects of flow attenuation. However, Overflow predicted higher magnitude hydrographs than HEC-RAS. This was due to the inclusion of lateral overland flow routing in Overflow, which could not be simulated in HEC-RAS. Further model development is required to integrate Overflow’s predictions of overland flow with boundary conditions for hydraulic models.

This thesis therefore provides a framework for the assessment of flow attenuation as a flood risk management tool, in the case of an extreme event, on the River Seven, North Yorkshire.