Exploring the optimal placement of Natural Flood Management (NFM) in different catchment configurations

Fluvial flooding causes significant economic, environmental and social losses, which are predicted to increase in the future due to climate change, land use change and increased urbanisation. To mitigate flooding, focus has moved from hard engineering and structural defences to risk reduction utilising natural processes, particularly tree planting to slow the flow. Whilst Natural Flood Management (NFM) has received a lot of attention, most research has focused on modelling small, upland catchments in the UK, with limited evidence demonstrating where and when NFM interventions are likely to be most effective. Therefore, I aimed to use a coupled hydraulic-hydrological model to explore how the morphology of catchments, intensity of rainfall, and the spatial configuration and coverage of NFM (tree cover), effects peak flows and lag times. The ultimate goal was to ascertain where in a catchment tree cover could be optimally placed to maximise reductions in peak flows. I achieved this in abstract space, looking at hydrographs generated in 20 abstract catchments of differing size, shape and morphology, with three rainfall intensities (30, 60, 120 mm h-1). Results show that establishing 25% NFM in catchments reduced peak flow on average by 23%, whereas covering three-quarters of the catchment led to 57% reductions. However, further coverage beyond 75% of the catchment had negligible additional reductions on peak flow. In contrast, it was found that up to 75% catchment vegetation cover was needed to have a significant impact on lag times. Similar relative patterns were found between catchments and NFM coverages for different rainfall intensities, although absolute values of percent reduction increased with rainfall intensity. Substantial differences in peak reduction were found between catchments, with catchment slope and drainage density being most significant in controlling hydrograph response to NFM. Situating NFM in areas of flat slopes was found to be effective in reducing peak flows, with coverages as low as 0.7% leading to peak flow reductions equivalent to 17.1%. To optimise NFM placement, a catchment travel time model was employed to focus NFM into areas with similar travel times in an attempt to equalise travel times across the catchment, reducing peak flows. This was found to significantly impact lag times but not to have had as large of an impact on peak flows as hypothesised. The reasons for this are discussed, along with future work needed to assess where and when NFM is likely to be most significant in reducing flood risk