The final report of this project can be accessed by clicking the thumbnail below:
To request a copy of the software and the associated user guide, please click this link to create a request: create request
On this wiki webpage, we summarize the activities and findings of Phase I of the project “Development of a numerical rapid assessment tool to simulate fate and environmental impact of fluidized sediment layers”. This project is a collaborative Topconsortium Kennis en Innovatie (TKI) Deltatechnologie project between Deltares and Boskalis. The project is divided in two main phases:
- Phase I: development and verification of a numerical tool; and
- Phase II: validation and application of this tool on specific dredging operations
Here, we only discuss the results of Phase I. The results of Phase II are produced by Boskalis and are not published as they may contain data and information confidential to Boskalis. Below, the word “project” refers specifically to “Phase I”.
Within the dredging industry, it is common practice to generate fluidized sediment layers when executing dredging projects in, for example, port areas. The fluidized sediment layer that is generated can flow by gravity to deeper sections where it can be trapped. During flow, the fluidized layer will interact with the ambient water column under local hydrodynamic forcing (i.e. tidal currents or waves). Interaction of the fluidized layer with the water column can lead to an increase in turbidity of the water. Before a project is executed, this turbidity plume must be evaluated carefully, so the environmental impact of the dredging activities can be estimated and mitigated.
In this project, we develop a rapid assessment numerical tool to assess the environmental impacts of fluidized sediment layers induced by a moving turbidity source. To facilitate rapid computations, the tool is based on the Deltares 1DV model (e.g. Winterwerp & Uittenbogaard, 1997). The tool should be able to model the following two aspects of the fluidized sediment layer flow and its interaction with the ambient water:
- The transport path and stability (i.e. fate) of the fluidized sediment layer (location, extent, thickness and density of the layer); and complementary
- The entrainment of sediment from the fluidized layer into the water column (which can be used as turbidity source for environmental evaluation).
The fate of the fluidized layer, and therefore the effectivity and impact of these dredging operations, depends on several factors. The most important factors are: bed slope, ambient flow conditions, initial density, and thickness and momentum of the generated fluidized layer. The model should be able to handle these factors adequately.
Goals and deliverables
The main goal of this project is to develop and verify the numerical rapid assessment tool. This tool should have the following capabilities:
- Compute the thickness and density of the fluidized layer as function of the distance from the source in relation to the hydrodynamics in the water column and bed slope;
- Be capable to calculate a turbidity source term (in kg/s), which can be used in far-field modelling of turbidity plumes (in analogy with the method described in Becker et al (2014)); and
- Be (offline) coupled with a near-field production model (provided by Boskalis). This model provides initial conditions for the 1DV model computations. The near-field production model is not developed further in this project.
The project deliverables are:
- A report describing model verification and model development;
- A beta research executable of the numerical tool; and
- A user guide on how to use the numerical tool.
The report is directly available by clicking the link at the top of this web page. The executable of the numerical tool and the associated user guide are available upon request. This request can be files by clicking the other link at the top of this web page.
In this project, we have successfully developed a tool with the capabilities described above. This tool is based on the Deltares 1DV model. We have adapted the 1DV model to make it suitable for a Lagrangian 1DV modelling approach. The Lagrangian 1DV approach entails that we follow the development of the fluidized layer flow along a user-defined trajectory using a moving frame of reference.
The main model developments that were carried out in this project are:
- Validation of the Lagrangian 1DV approach;
- Specification of fluidized layer initial conditions in the 1DV model;
- Testing sensitivity of model outcome to fluidized layer initial conditions; and
- Implementation of mud dynamics formulations.
The main conclusions of these 4 model developments are discussed below:
- Validation of the Lagrangian 1DV approach
The Lagrangian 1DV model shows a good qualitative agreement with selected results from literature. It also shows the expected behaviour for test cases when the fluidized layer flows over a downsloping or upsloping bed. Therefore, the Lagrangian 1DV model seems to be fit for purpose.
2. Specify initial conditions for Lagrangian 1DV model
To make sure the model can be easily set up, we specify three parameters in the model input to characterize the initial condition of the fluidized layer flow. These parameters are: the fluidized layer height (hfl), average fluidized layer velocity (ufl), and average fluidized layer concentration (cfl). Using these parameters, vertical profiles are constructed of: velocity, concentration and turbulence. In our approach, we make sure that sediment mass and sediment mass flux, specified by the three parameters listed above, are conserved.
3. Test the sensitivity of model outcome on initial conditions
When setting up the Lagrangian 1DV model, it is advised to give most attention to the setup of the initial concentration profile, since its potential influence is (much) larger than the influence of the velocity profile. Model outcome is very sensitive to the initial concentration profile, when only conservation of sediment and sediment mass flux apply. The main sensitivity of model outcome lies in ‘centre of gravity’ of the fluidized layer. If this remains constant for different initial profiles, the shape of the initial concentration profile does not have a large effect.
4. Implement mud dynamics formulations
We have implemented formulations for hindered settling and erosion/deposition of muddy sediments implemented. After implementation, we have also verified and tested these new formulations. This makes the model suitable for modelling fluidized sediment layers that consist of sand or mud, when they are turbulent and behave as a Newtonian fluid.
Outlook and further developments
With this new version of the 1DV model as a basis, possible follow-up developments include:
1. Modelling settling and consolidation of muddy sediments
Consolidation, intended as dewatering, changes the thickness, the strength and erosion potential of the newly deposited bed. This can have applications on the thickness of the deposit or the tendency to resuspension from variable hydrodynamic conditions during or after operations. The same features and the same model can be utilized in settling estimates in soft sediment deposits, such as land reclamation or construction of natural islands where requirements are demanded on final topography. Consolidation processes are already embedded in other research versions of the 1DV model, therefore they can be easily included in this new (and official) version.
A step further would be to include the effects of vegetation and ripening in subaerial compartments. These processes and their addition to this model are in line with Deltares’ development ambitions, as they are important for various applications world-wide. However, their addition would still require a significant effort.
2. Incorporate the influence of rheological properties on fluidized layer flow
When the sediment concentration (especially the cohesive fraction) exceeds about 100 g/l (depending on initial flow conditions and rheological properties of mud), sediment flow will likely behave as non-Newtonian and laminar. In this case, specific non-Newtonian rheology models should be applied. These models were recently developed and embedded in a specific version of the 1DV model (Hanssen, 2016) and then transferred to Delft3D (Delft3D-Slurry, Sittoni et al., 2017). Including non-Newtonian rheology to the version of the 1DV model developed in this report is therefore a logical and relatively simple step that allows improving the accuracy (in fact the correctness) of the prediction to higher sediment concentration.
In this project, a rapid assessment numerical tool was successfully developed. This tool can be used to assess the environmental impacts of fluidized sediment layers induced by a moving turbidity source. Because model setup is relatively easy and computational effort is small, this tool enables users to make rapid calculations. This makes it particularly useful in the engineering and design phases of dredging projects, where engineers must be able to act quickly upon receiving information about the project site and changes in project execution. Furthermore, this approach enables users to test the sensitivity of model outcome to different parameter settings. However, to correctly apply the model, the assumptions underlying the model must be carefully considered.