Nature-based solutions for flood mitigation:Monitoring and modelling leaky barriers. A case study from the South East of England.

Tonkin, B.R^a., Marti-Cardona, B^a., Hughes, S.J^a., Ru, T^a., Philpott, N^b.

^a Civil and Environmental Engineering Department, University of Surrey, UK.

^b Environment Agency, England.

b.tonkin@surrey.ac.uk

Nature-based solutions (NBS) are increasingly being recognised as a tool for flood mitigation, particularly relevant in the face of increased storm severity due to climate change. The simulation of NBS functioning is of great interest for their effective location and design, and there are substantial ongoing efforts in developing strategies to underpin their catchment scale modelling.  Leaky Barriers (LB) are a type of NBS interventions which consist of placing logs across a channel to hold back runoff during storm events and to slow down its travel. Despite their common adoption in the UK, there are relatively few studies that have addressed the best hydraulic representations of LB’s through calibrated and validated measurements during flood events. 

This study is based within the Thames basin in the Southeast of England and encompasses the 21km² headwater catchment of the Pipp Brook. In 2019, the Environment Agency of England installed over thirty LB’s in the Pipp Brook as a trail study. Monitoring data has been continuously acquired by sensors installed at the site (water level gauges, ultrasonic flow gauge, fixed-point infrared cameras) and during periodic inspections (structural monitoring) since 2019. This enhanced monitoring programme, one of the most comprehensive in the UK, provides rigorous evidence to understand and assess the effectiveness of the NFM measures installed in the catchment. To date, this has included the capture of multiple high flow events, up to a peak magnitude of 1 in 20 years.

This research seeks to address a gap in the strategy to simulate individual leaky barriers using 1-dimensional hydraulic models. To this aim, one LB in the Pipp Brook was simulated with an industry leading hydraulic modelling software package (Flood Modeller, Version 6.10 by Jacobs) using six different 1D modelling strategies reported in the literature: i) Orifice, ii) Bridge, iii) Weir, iv) Increased roughness (Manning’s n), v) Bernoulli loss, vi) Blockage. High-flow records from a double-peak event in October 2021 were used to calibrate and assess these hydraulic representations. Upstream boundary conditions were produced by ReFH2 rainfall-runoff modelling, using the precipitation records from a nearby meteorological station.

The comparison of calibrated models to the gauged data revealed a maximum difference of circa 0.20m to the measured upstream water elevation, for a maximum water depth of 0.75m. Our results showed that the best approximation was achieved by using the bridge unit. A common approach in the literature is to represent LB’s with a high roughness coefficient (Manning’s n), which in our case resulted in the poorest performance. The results of this ongoing research will improve the ability of flood practitioners to predict the effectiveness of leaky barrier configurations in a catchment, hence informing their optimal design.