 |
|
Cohesive Sediment Mechanics
|
<back>
Cohesive sediments are
a mixture of clay, silt, sand and organic matter. Mixed in water
they form what commonly is known as mud. This sediment is predominantly
fine-grained and the physico-chemical characteristics of the clay
minerals cause aggregation of these particles into flocs. Flocculation
is furthermore enhanced by the presence of natural organic matter
found in the aquatic environment, often of biological origin (e.g.
slimes and mucus). Hence, cohesive sediment particles in general
are not solid particles but flexible flocs which can aggregate or
break-down, depending on various external factors (such as turbulent
shear, concentration and salinity of the ambient water). Their size,
structure (i.e., their density) and strength are thus variable,
much in contrast to non-cohesive particles like sand and gravel.
Therefore, the sediment mechanics of cohesive particles is different
and much more complex than that of sand.
As these sediments are so fine, they are often transported by the
river all the way down to their mouth, where they ultimately deposit
often massively forming typical mud banks and intertidal flats.
Cohesive sediment research at the K.U.Leuven started in the early
1980s with a feasibility study of mud pumping of deposited sediments
in the entrance of a harbour lock (Berlamont 1989). The work was
continued by developing numerical models, using the Finite Element
Method, for fluid mud flow and settling and consolidation of mud
(Toorman 1992). Further research then extended the research to the
modelling of all the sediment processes in the water column, which
leaded to the general purpose 2D code FENST.
Several aspects of the cohesive sediment research have been carried
out with E.U. funding through the Marine Science and Technology
projects G6-M (MAST I), G8-M (MAST II) and
COSINUS (MAST III). The latter project was coordinated by the
Hydraulics Laboratory of the K.U.Leuven.
Some major publications:
- Toorman, E.A. (2001). Cohesive sediment transport modeling:
European perspective. In: Proceedings in Marine Science,
Vol.3: Coastal and Estuarine Fine Sediment Processes (W.H.
McAnally & A.J. Mehta, eds.; Proc. INTERCOH'98, Seoul, May 1998),
pp.1-18, Elsevier Science, Amsterdam.
- Berlamont, J., Ockenden, M., Toorman, E. & Winterwerp, J. (1993).
The characterisation of cohesive sediment properties. Coastal
Engineering, 21:105-128.
- Toorman, E.A. (1992). Modelling of fluid mud flow and consolidation.
PhD thesis, K.U.Leuven.
- Berlamont, J. (1989). Pumping of fluid mud: theoretical and
experimental considerations. J. Coastal Research, Special
Issue No.5:195-205.
- COSINUS publications cf. COSINUS
website.
Cohesive sediment research at the Hydraulics Laboratory focuses on
the following topics:
 |
Sedimentation,
consolidation and liquefaction of mud (Erik Toorman) |
 |
|
The laboratory is equipped with a temperature controlled,
dark room with six transparent settling/consolidation
columns (2 m high, 0.1 m diameter). Measurements are
taken of pore pressures at various heights (piezometric
tubes), bulk density (non-destructive measurement by
gamma-ray absorption) and interface height as a function
of time. This set-up has been extensively used for research
and consulting on the settling and consolidation behaviour
of artifical and natural muds, as well as of sand/mud
mixtures.
A general theory has been developed to unify the classical
theories of sedimentation and consolidation (Toorman
1994). A computer model has been developed to solve
the sedimentation/consolidation problem (Toorman & Berlamont
1991, Toorman 1996). Experimental data have been analysed
in order to develop new closure relationships (Toorman
& Huysentruyt, Toorman 1996, Toorman & Leurer 2000).
The erosion of a sediment bed is governed by the structure
of its surface. Its erosion resistance strength changes
due to consolidation (strengthening) and/or fluidisation
and liquefaction (weakening). During the COSINUS project
a first attempt was made to develop a numerical model
that could simulate both processes, i.e. a bed dynamics
model for a soft soils with extremely large deformations
(Toorman et al. 2000).
Major publications:
- Toorman, E.A. & Berlamont, J.E. (1991). A hindered
settling model for the prediction of settling and
consolidation of cohesive sediment. Geo-Marine
Letters, 11(3-4):179-183.
- Toorman, E.A. (1996). Sedimentation and self-weight
consolidation: general unifying theory. Géotechnique,
46(1):103-113.
- Toorman, E.A. (1999). Sedimentation and self-weight
consolidation: constitutive equations and numerical
modelling. Géotechnique, 49 (6):709-726.
- Toorman, E.A. & H. Huysentruyt (1997). Towards a
new constitutive equation for effective stress in
self-weight consolidation. In: Cohesive Sediments
(N. Burt, W.R. Parker & J. Watts, eds.; Proc. INTERCOH'94,
Wallingford, July 1994), pp.121-132, J. Wiley & Sons,
Chichester.
- Torfs, H., H. Mitchener, H. Huysentruyt & E. Toorman
(1996). Settling and consolidation of mud/sand mixtures.
Coastal Engineering, 29 (1-2):27-45.
- Toorman, E.A., I. Brenon & K.C. Leurer (2000). Bed
dynamics modelling based on the generalized Biot theory.
Report no. HYD/ET/00/COSINUS6, Hydraulics Laboratory,
K.U.Leuven.
- Toorman, E.A. & K.C. Leurer (2000). An improved
data-processing method for consolidation column experiments.
Report HYD/ET/00/COSINUS8, Hydraulics Laboratory,
K.U.Leuven.
|
|
|
Mud rheology
(Erik Toorman) |
|
|
Rheology studies the deformation behaviour of materials.
In relationship to fluid mud flow, rheology is used to
determine a closure relationship for the bulk viscosity
of the concentrated mud suspension. As results of the
flocculated nature of the particles, the viscosity of
mud varies with shear rate (mud is a shear thinning fluid)
and with time (thixotropic behaviour). This makes that
rheological measurements of mud are cumbersome, as the
interpretation of the results even more. A methodology
has been developed to analyse vane rheometry experiments
(Toorman, 1994). A new, more general thixotropy model
has been developed (Toorman, 1997).
Major publications:
- Toorman, E.A. (1997). Modelling the thixotropic behaviour
of dense cohesive sediment suspensions. Rheologica
Acta, 36 (1):56-65.
- Toorman, E.A. (1994). An analytical solution for the
velocity and shear rate distribution of non-ideal Bingham
fluids in a concentric cylinder viscometer. Rheologica
Acta, 33:193-202.
|
|
|
Erosion of sand-mud
mixtures (Erik Toorman) |
|
|
The Hydraulics Laboratory is equipped with a straight,
recirculating erosion flume (9 m long, with a test section
of 2 m, 0.4 m wide, 0.5 m high). Concentration and velocity
profiles can be measured, as well as the water surface
slope and bed load transport.
The composition of the sediment bed in many estuaries
or along the coast are mixtures of sand and mud. Flume
experiments have been carried out to study the influence
of clay content on the erodibility of mixed sand-clay
bottoms. Both homogeneous and layered beds (obtained by
sedimentation in a removable tank) have been investigated.
Major publications:
- Mitchener, H. & H. Torfs (1996). Erosion of mud/sand
mixtures. Coastal Engineering, 29:1-25.
- Torfs, H. (1995). Erosion of mud/sand mixtures. PhD
thesis, Civil Eng. Dept., K.U.Leuven, 223 pp.
|
|
Sediment-Turbulence Interaction ( Erik Toorman) |
|
|
Sediment particles are held in suspension by
turbulence. But their presence alters the characteristics
of the turbulence, affecting the velocity profile
and the bed shear stress. Subsequently, sediment
transport models should couple the sediment mass
balance with the hydrodynamics. The effect of stratification
can easily be dealt with through the buoyancy term
in the k-epsilon turbulence closure, used
to determine the vertical turbulent diffusion in
currently used state-of-the-art 3D sediment transport
models for estuarine and coastal scales. However,
consistent implementation of this effect requires
modification of the traditional hydrodynamic near-bottom
boundary conditions. Subsequently, it has been demonstrated
that drag reduction can be simulated correctly (Toorman,
2002).
Furthermore, it has been found that the suspension
capacity is characterized by a constant flux Richardson
number. Subsequently, better insight is obtained
in the interpretation of many experimental concentration
profiles, where a subdivision can be made into a
saturated above a super-saturated layer near the
bed (Toorman, 2003).
Part of this research has been carried out as part
of the MAST III COSINUS project and was funded by
the Flemish Science Foundation (FWO) and the K.U.Leuven
Special Research Fund.
Current research (funded by the K.U.Leuven Special
Research Fund) is focusing on the modeling of this
super-saturated layer and its relationship to sheetflow
in the case of non-cohesive sediments and non-Newtonian
fluid mud in the case of cohesive sediment.
A major weakness in sediment transport models is
the lack of a generally valid closure for the turbulent
Schmidt number (the ratio between eddy viscosity
and eddy diffusivity). Experimental data sets, either
from lab experiments or generated by LES and/or
DNS models will be collected and investigated within
this context. LES data will be generated by the
Fluid Mechanics Section of the Vrije Universiteit
Brussel within the framework of the joint FWO project
"Development of simulation models for two-phase
flow on geophysical scale, with application to sediment
transport in estuaries and coastal zones" (funded
by the Flemish Science Foundation).
Major publications:
- Toorman, E.A. (2001). Cohesive sediment transport
modeling: European perspective. In: Proceedings
in Marine Science, Vol.3: Coastal and Estuarine
Fine Sediment Processes (W.H. McAnally &
A.J. Mehta, eds.; Proc. INTERCOH'98, Seoul, May
1998), pp.1-18, Elsevier Science, Amsterdam.
- Toorman, E.A. (2002). Modelling of turbulent
flow with cohesive sediment. In: Proceedings
in Marine Science, Vol.5: Fine Sediment
Dynamics in the Marine Environment (J.C. Winterwerp
& C. Kranenburg), pp.155-169, Elsevier Science,
Amsterdam.
- Toorman,
E.A. (2003). Validation of macroscopic modelling
of particle-laden turbulent flows. Proc. 6th
Belgian National Congress on Theoretical and Applied
Mechanics (Gent, 26-27 May 2003), CD Rom,
7 pp.
- COSINUS publications cf. COSINUS
website.
|
|
Sediment Transport Modelling |
|
|
Results from fundamental sediment transport
research are implemented in the FENST-2D
research code of the K.U.Leuven and 3D community
software. Implementation of Large Eddy Simulation
(LES) techniques for a better description
of horizontal mixing is under investigation
within the framework of PhD research with
a K.U.Leuven-IRO scholarship.
A major challenge for the application of sediment
transport models to real environmental problems
in estuaries and coastal areas is the overcoming
of the scale deficiencies, because computational
grids are still very coarse (vertical resolution
of the order of 1 m, horizontal resolution
of the order of 100-1000 m), since they cover
a large area (with order 10^5-10^6 nodes)
and the computational cost should remain acceptable.
Many processes occur at subgrid scales and
cannot be taken into account with the same
degree of accuracy as in a research code,
such as FENST-2D. One of these aspects is
the description of the processes near the
bed and the determination of the effective
roughness (accounting not only for grain roughness,
but also form and topographic roughness and
roughness modification due to high sediment
concentration effects). This topic forms the
focus of PhD research (funded by a VLIR scholarship)
and the FWO-project "Development of
simulation models for two-phase flow on geophysical
scale, with application to sediment transport
in estuaries and coastal zones" (funded
by the Flemish Science Foundation) in collaboration
with the Fluid
Mechanics Section of the Vrije Universiteit
Brussel. The VUB will contribute in the generation
of detailed
fine-scale LES data for sediment-laden turbulent
flow under specified conditions (e.g. specified
bottom profiles).
A model for the Yzer Mouth in Nieuwpoort
(Belgium) is under development.
Major publications:
|
|
|
|
|
