The effect of soil bunds as Natural Flood Management features on soil water chemistry and hydraulic conductivity

Abstract

Nature-based solutions to flooding have drawn growing attention in recent years as climate change models predict a significant increase to flood-risk in the UK and around the world. The lack of systematic empirical evidence to support Natural Flood Management (NFM) initiatives still presents a key barrier to the widespread implementation of NFM techniques.
Both rural and urban areas in the North Pennine Hills in Northeast England have historically been substantially affected by flooding. This study focused on a small-scale pilot installation of five earth bunds designed as temporary storage units for flood water in the upper catchment of the River Wear in Weardale. An upscaled installation of such features on a larger spatial scale may have a significant impact on reducing flood-risk by slowing down flood flows. To assess the effectiveness of the bunds, water sensors were installed in each of the five bunds to record the frequency of flood storage. Changes in soil hydraulic conductivity were measured throughout the study period (December 2021 to June 2022) to assess changes in infiltration capacity as a measure of bund construction quality. Finally, the chemistry of soil water was analysed through Inductively Coupled Plasma Mass Spectrometry (ICP-MS), to establish whether the bunds can act as filters for pollutants in a catchment suffering from heavy metal contamination with possible implications to water quality improvements.
Due to instrument malfunction, no data was recorded on the frequency with which the bunds acted as active storage features. No evidence was found for the bunds’ ability to act as sinks for heavy metals, although the data are not conclusive. Hydraulic conductivity measurements found no variation across the five bunds, nor across different areas of each bund, suggesting the bund structure remains stable through time and does not vary in quality. Further research is required to corroborate these results, with more extensive sampling and over a longer time scale.