Improved prediction of cohesive sediment erosion based on inter-particle forces

Lead Research Organisation: Cranfield University
Department Name: School of Water, Energy and Environment


Cohesive sediment (more commonly known as mud) forms the bed of many waterways and coastal environments and is thus the foundation on which critically important engineering structures are built. These structures alter local water flows that can induce the erosion (or scour) of sediment, resulting in the undermining of bridge piers, abutments and revetments; compromising the foundations of off-shore wind turbines; and causing the self-burial of submarine pipelines. Scientists have long known that the composition of the sediment will affect its resistance to erosion, but have not been able to develop universal models to predict erosion thresholds, rates or depths of scour. Research in this topic has been hindered by (i) inconsistencies in how erosion is defined and measured and (ii) the large number of sediment properties that control cohesion and adhesion in cohesive sediment. Better predictions of erosion thresholds and rates are urgently required to improve the assessment of scour risk to protect essential transportation and energy infrastructure, particularly as climate change is predicted to increase the frequency and severity of storms that drive the hydrodynamic forces responsible for sediment erosion.

This innovative study will advance the development of a physically-based predictive model of cohesive sediment erosion by focusing on the particle-particle and particle-fluid interactions that underlie cohesive and adhesive forces within cohesive sediment. It will combine (i) new field and laboratory research on erosion thresholds and rates of cohesive sediment and (ii) a novel computational dynamics model to simulate cohesive sediment mechanics and erosion dynamics. The research is composed of three work packages (WPs). In WP 1, erosion threshold and rates of natural and artificial sediments covering a diversity of sediment and water properties will be measured in the new EPSRC-funded UKCRIC facilities at Cranfield University. It will be the first time that the effects of sediment and water properties on cohesive sediment erosion will be investigated systematically in a single study using facilities capable of high-precision monitoring of water flow, the sediment bed surface, and the erosion of surface particles and aggregates. In WP 2, a Discrete Element Model (DEM) that represents cohesive sediment as a mixture of rigid, non-cohesive elements and smaller 'soft' cohesive elements will be combined with a Computational Fluid Dynamics model (CFD). This new coupled model incorporates several recent advances in the modelling of materials (a mixture of particle sizes with varying physical, chemical and mechanical properties) and permits the simulation of erosion at similar spatial and temporal scales as the laboratory analyses in WP 1. Finally, WP 3 will analyse, evaluate and compare the results of the empirical study (WP 1) and coupled numerical model (WP 2), employing statistical and probabilistic approaches to infer significant relationships between erodibility and sediment properties and evaluate the performance of the modelling.

The study will produce a step-change in the scientific understanding of cohesive sediment mechanics and the prediction of erosion thresholds, rates and depth of scour. The greatly improved predictions of cohesive sediment erosion will have wide ranging applications that will help to protect critically important aquatic environments and water resources from contaminated sediment and engineering infrastructure from scour-induced failure.

Planned Impact

Cohesive sediment is a natural part of waterways and coasts, but alterations to its quality and transport dynamics make it a significant management challenge. Better predictions of erosion thresholds and rates will have wide ranging applications.
Application of research outputs to address sediment scour will benefit (i) public and private organisations that build or maintain infrastructure in aquatic environments that are exposed to the risk of scour-induced undermining and failure and (ii) private companies that are developing tools to assess scour risk (e.g. bathymetric sonar manufacturers). Sediment erosion is one of the main causes of failure for hydraulic structures. Sediment erosion undermines bridge piers, abutments and revetments; compromises the foundations of off-shore wind turbines; and causes self-burial of submarine pipelines that affects their lateral stability and buckling. The principal output of this research is a new physically-based predictive model of erodibility (i.e. erosion thresholds and rates) of cohesive and mixed fine sediment beds based on their sediment and water chemistry properties. The modelling will be readily transferable to a wide variety of environments and applications, be they in rivers, lakes, estuaries, coasts or open ocean, because the research is based on the inter-particle and inter-molecular forces that determine cohesion and adhesion. The new predictive models can be incorporated into existing hydraulic models of sediment transport and scour, replacing the current empirically-derived or estimated erosion thresholds, to generate more accurate predictions of sediment scour. Project partner ITER Systems (a small-medium size enterprise with a UK office) supplies bathymetric sonar systems to clients in a range of sectors (e.g. shipping, exploration, flood risk management), and they are responding to market pressure to develop better data analysis tools. By integrating our predictive models of erosion thresholds with their sonar-derived estimates of sediment properties, they can provide clients with critical information on the likelihood of erosion over large swathes of river or sea bed so clients can better predict the longevity of hydraulic infrastructure, inform maintenance regimes, or critically appraise scour risk for alternative designs and mitigations.
Application of research outputs to address contaminated sediment will benefit (i) the government to inform the management of contaminated sediment, and (ii) communities currently impacted by poor water quality due to contaminated sediment that threatens water supplies and local economies based on tourism. Many contaminants bind to cohesive sediment, and thus the beds of waterways and coastal areas represent a significant store of contaminants from a legacy of industrial and municipal pollution. Contaminated sediment is a worldwide problem, and the assessment of its risk and identification of solutions is a significant priority for environmental regulators (e.g. US DoD estimates $2 billion USD in liabilities from contaminated sediment). Whilst damaging to ecological communities in situ, these contaminates pose an even greater threat to water quality, drinking water supplies, human health and aquatic ecological communities if the sediment is resuspended. In the UK, heavily contaminated sediment is found in rivers, canals and estuaries near former mining and heavy industry and in agricultural areas. Project partner Natural England and society will benefit from the application of the research to the management of the Broads National Park where sediment resuspension is impacting local, economically-important fisheries, tourism and recreation. The research proposed in this study will develop improved models of erosion thresholds and rates that can be applied to any hydrodynamic model to better predict and assess contaminant risk and flux in any aquatic environment to protect human health, environment, and locally-important economies.


10 25 50
Description Collaboration with NanJing University 
Organisation Nanjing University (NJU)
Country China 
Sector Academic/University 
PI Contribution None
Collaborator Contribution The researcher is providing guidance on the forcefield to use for clays in the molecular dynamics modelling. He suggested an alternative software to set up the simulation. The researcher kindly provided a demonstration on how to prepare clay particles with Materials studio and the method to convert Materials studio data file to lammps input data file.
Impact Non yet - work is still in progress
Start Year 2020
Description Collaboration with Penn State University 
Organisation Penn State University
Country United States 
Sector Academic/University 
PI Contribution None
Collaborator Contribution The researcher provided guidance on setting up a molecular dynamics simulation for clay/water interactions using the reaxff force field.
Impact None -
Start Year 2020
Description Collaboration with Sandia National Laboratories, USA 
Organisation Sandia Laboratories
Country United States 
Sector Private 
PI Contribution None
Collaborator Contribution The researcher is providing advice on the representation of clay particles in our molecular dynamics simulations and suggestions on analytical approaches post-simulation. This collaboration has been very helpful in the first stage of our research related to the force balance of clay particles as it swells.
Impact No outputs or outcomes have been generated yet.
Start Year 2020
Description Project blog - Sand castles 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact A blog post on Cranfield University's official blog site describing the aim of the project and early research
Year(s) Of Engagement Activity 2021