Natural flood management potential of leaky dams in upland catchments

Natural Flood Management (NFM) is an increasingly popular approach to flood risk management which aims to slow and store flood water in the landscape by restoring natural hydrological and geomorphological processes. Due to difficulties in obtaining precise hydrological data, a lack of robust, empirical evidence describing the effectiveness of NFM measures at reducing downstream flood risk currently undermines confidence in its efficacy and limits its adoption. Furthermore, uncertainty about the risks associated with NFM features can present a barrier to implementation. This research quantifies the benefits and risks of installing engineered leaky dams for the purpose of NFM in an upland, headwater catchment in North Yorkshire, England. To overcome difficulties associated with the empirical quantification of leaky dam impacts on flood peak magnitude, hydrological data from a three year Before-After-Control-Impact style monitoring campaign was combined with a data-based time series modelling approach. The results quantify, for the first time, the impact of a series of eight channel-spanning leaky dams on flood peak magnitude for a range of events. The leaky dams reduced flood peak magnitude of events with a return period up to one year by 10% on average, but their impacts were highly variable. In order for the benefits provided by leaky dams to be evaluated against their potential hazards the study included assessment of the failure probability of leaky dams based on observations of leaky dam failures from five UK sites. The empirical fragility analysis showed that leaky dam failure rates were lower than naturally occurring large wood mobility rates reported in literature. The novel application of both the data-based timeseries modelling approach and empirical fragility analysis to assess the potential of leaky dams for NFM demonstrates the opportunities that these techniques offer to quantify leaky dam impacts on downstream flood risk.