Engineered leaky barriers are increasingly used as natural flood management methods providing ecosystem and water quality benefits in addition to flood attenuation, complementing hard engineering flood defences. Field-based monitoring of a natural flood management site, Wilde Brook in the Corvedale catchment, England (UK) studied the rainfall-runoff relationship for a 5.36 km reach with 105 leaky barriers over two years. Paired pressure transducers were placed upstream and downstream of three channel spanning leaky barriers, allowing evaluation of upstream backwater rise relative to rainfall intensity, storm magnitude, and frequency. By increasing backwater rise, the leaky barriers caused overbank flows, resulting in a reduction in the cross-sectional area velocity after the event. The incidence of overbank flow depended on the local stream cross-sectional profile, barrier properties, location in the reach, and storm magnitude. Barrier operational flow conditions were classified into five modes according to relative bank inundation and barrier submergence extent. The backwater rise magnitude depended on barrier physical properties and evolution over time through cycles of accretion and build-up of brash and leafy material in the barrier, in addition to local bedload sediment transport dynamics, where instances of scour around the barriers were observed. Backwater rise and net volume hydrographs showed rapid filling up behind the barriers on the rising limb and slower water release on the falling limb. For a ∼4 yr return period storm event, results indicated that one leaky barrier increased storage volume by up to 102 m³, which translates to an overall net volume increase of ∼10,700 m³ for the full reach. These new findings provide quantitative evidence of leaky barrier backwater and storage performance, and leaky barrier design recommendations for storms up to a 4 yr return period. This evidence can be used to develop and validate flood modelling generalised approaches for smaller, more frequent storms, and work towards the development of an approach for modelling leaky barriers for larger storm magnitudes.

工程渗漏屏障越来越多地用作自然洪水管理方法，除了洪水衰减外，还提供生态系统和水质效益，补充硬工程防洪设施。英格兰（英国）Corvedale 集水区的 Wilde Brook 自然洪水管理地点的实地监测研究了 5.36 公里范围内 105 个渗漏屏障两年内的降雨-径流关系。成对的压力传感器被放置在三个跨越漏水屏障的通道的上游和下游，从而可以评估上游回水上升与降雨强度、风暴强度和频率的关系。通过增加回水水位，渗漏屏障导致越岸水流，导致事后横截面积速度降低。越岸水流的发生率取决于当地河流的横截面剖面、屏障特性、河段位置和风暴强度。根据相对堤岸淹没和屏障淹没程度，将屏障操作流量条件分为五种模式。回水上升的幅度取决于屏障的物理特性和随着时间的推移通过屏障中泥沙和多叶物质的增生和堆积循环而演变，此外还取决于局部床载沉积物输送动力学，在屏障周围观察到冲刷实例。回水上升和净体积水文过程线显示上升肢的障碍物后面快速填充，下降肢的水释放较慢。对于 ~4 年重现期的风暴事件，结果表明一个漏水屏障增加了高达 102 m 的蓄水量³，这意味着整个航程的总净体积增加了 ∼10,700 m ^{3 。}这些新发现提供了渗漏屏障回水和存储性能的定量证据，以及针对长达 4 年重现期的风暴的渗漏屏障设计建议。这些证据可用于开发和验证针对更小、更频繁的风暴的洪水建模通用方法，并致力于开发一种方法来为更大的风暴量级的漏水屏障建模。