Understanding geomorphic response to hydrological events: filling the data gaps

Lead Research Organisation: Loughborough University
Department Name: Geography


We will improve the management of river sediment and modelling of flood risk by developing and applying a 'sediments toolkit' for the characterisation of river bed grain size at local to catchment scales.
Flood intensity is increasing in the UK and many rivers are becoming more mobile, particularly Scotland's flashy and sediment-laden systems. Our understanding of how geomorphic systems respond and adjust to hydrological events comes principally from traditional morphological measurements, obtained from aerial photographs, LiDAR and maps. We know that the nature of the constituent sediments-especially size-is a critical element of geomorphic systems, controlling, for example, entrainment and mobility of sediment, channel conveyance during floods, and the ability to support ecologically and economically important fauna & flora. However, data on sediment properties is relatively sparse, principally because it is expensive and time-consuming to collect.
SEPA have identified the lack of grain-size data at bar to catchment scales as a critical data gap, limiting their ability to predict future patterns of channel erosion/deposition and to model flood risk. Obtaining such data is therefore a key priority.
Despite the development of several image-based grain-sizing methods in recent years (including by Graham/Rice), uptake of these methods by practitioners has been partial owing to limited guidance on their selection and use, and a reluctance to adopt new/unproven technologies. If these obstacles are overcome, image-based sediment characterisation could significantly improve understanding and management of geomorphic systems.
In this project, we will develop an integrated 'sediments toolkit' consisting of software, documentation and guidance for image-based sediment characterisation at multiple scales. It will be targeted at practitioners without specialist knowledge of image processing methods and released under an Open Source licence.
Utilising the ability of small unmanned aircraft to obtain data over a wide range of spatial scales/resolutions, we will collect a heterogeneous dataset for a variety of river environments. Published and original (e.g. using structure-from-motion derived point clouds) algorithms will be tested to define a matrix of recommended methods for different scenarios, including: the nature of the environment being studied; the spatial and temporal resolution of the study (e.g. bar scale vs catchment scale); the nature of the data required (e.g. proportion of sand/gravel vs complete grain-size distribution); the accessibility of the site; and the resources (financial/human/technological) available.
The toolkit will have wide applicability to rivers in Scotland, the UK and beyond. We will use the River Dee, Aberdeenshire-where the 2015/16 floods caused significant erosion, destroying productive land and threatening infrastructure (e.g. Abergeldie Castle)-as a case study of how the incorporation of distributed grain-size data can improve river management. We will estimate sediment entrainment thresholds and identify potentialzones of sediment supply, erosion and deposition. Management strategies will be designed to reduce flood risk, mitigate threats to infrastructure and maintain hydromorphological integrity (as required by Water Framework Directive, WFD). We will then explore applications in other environments where grain size is of fundamental importance for understanding geomorphic adjustment, such as beaches.


10 25 50