STUDY AND ANALYSIS OF SILTATION AT NAHRE GHASR FISHERY PORT

Mehrdad Orooghi Amidi 1, Mohsen Soltanpour 1, Zahra Ranji 1

The discharge of Arvand River and tidal currents affect the large siltation at Nahre Ghasr Fishery Port, located at the
Nahre Ghasr channel, the Persian Gulf. Using 2DH and 3D numerical models, the flow pattern at the channel was
investigated to study the details of siltation problem. The local model was calibrated by the measured data of water
surface and suspended sediment rate. The main flow mechanisms that can affect the rate of siltation including the
horizontal exchange, the vertical exchange and tidal filling, were analyzed. The study showed that tidal filling plays the
dominant role on the sedimentation at Nahre Ghasr channel. Different scenarios were also examined for reducing the
siltation rate at the port.

Keywords: Siltation analysis, Flow pattern, Sill, Current Deflecting Wall

INTRODUCTION
Siltation is defined as the transportation of the silt by streams or currents which can contribute in
water turbidity. Siltation in river ports can cause many difficulties in marine transportation resulting
to the necessity of costly dredging operations. Using periodic dredging, the port can be preserved in its
draft depth ensuring the desirable navigation. However, this can highly affect the port productivity by
reducing its economic and trading efficiency, which might even force the authorities to abandon the
port after few years. Therefore, sediment control practices are necessary to cut down dredging costs at
ports in order to reduce the undesirable siltation effects.
Siltation of muds and silts in port basins is a known problem that goes back to the history of
building the ports and it is related to their basic function, i.e. providing shelter by creating quiescent
conditions (Van Rijn, 2004). It is also important to know that port siltation in fresh water is much less
than in salt and brackish water conditions (Nasner 1992). As much as rivers’ water become salty, they
can play significant role in siltation. The siltation at a river port consists of the following major
factors (Winterwerp 2005 and Van Rijn 2004):
1. Tidal filling (filling and emptying of the tide)
2. Horizontal eddy circulation generated at the entrance by the main river flow
3. Vertical circulation at the entrance, generated by density differences, inside and outside of the
basin
These mechanisms can affect the siltation rate depending on the geometry of port entrance. Thus,
understanding the effective mechanisms and their contributions should be considered at the first step
to evaluate suitability of different methods for the reduction of siltation rate. Geometry and conditions
of river stream as well as tidal regime needs to be carefully studied to control the imposed
sedimentation. Different methods have been purposed for the reduction of siltation at a port:
 Improvement of the port geometry
 Current Deflecting Wall (CDW)
 Sill in the entrance
 Upstream pile screen
 Mechanical gate in the entrance
 Silt curtain
 Re-suspending systems
 Chemical methods
 Biological methods
CDWs are curved vertical walls in horizontal plane that are extended to the top of water.
Construction of a CDW is an effective method to control the horizontal exchange and flow separation.
On the other hand, sill is used to control the vertical exchange and tidal filling at the entrance. The
main purpose of the CDW construction can be classified as (Winterwerp 1994; Schewarze 1995;
Christiansen and Kirby 1991):
 Prevent flow separation at the upstream corner of the entrance
 Reduce the exchange volume related to horizontal circulation
 Reduce the retention time of the sediments in the basin
 Suppress the generation of eddies in the entrance region
Parkhafen basins of the Port of Hamburg is an example of the design of the CDWs (Delft Hydraulics
2001). A curved wall with a length of 150 m was also constructed near the entrance of the Köhlfleet
harbor basin, located seawards of the Parkhafen, with the purpose of minimizing sediment input to
1
K. N. Toosi University of Technology, Tehran, Iran
1
2 COASTAL ENGINEERING 2016

the basin. Its design was based on extensive hydraulic scale model tests at Franzius Institute in
Germany. This CDW proved its worthiness over the years, reducing the dredged quantities in the
Köhlfleet by approximately 40% (Winterwerp et al. 2005), resulting in a 30% reduction of the
deposition volume at the entrance of that basin. The effect of a CDW critically depends on the local
hydraulic conditions and the geometry of the entrance. The proper design of a CDW should be based
on experiments in a laboratory scale model (Delft Hydraulics 1992 and 2001; Van Leeuwen and
Hofland 1999; Crowder 2001 and 2002). An incorrect design may even cause deterioration of the flow
patterns by generating additional eddies and more siltation.
STUDY AREA
Arvandkenar is part of Abadan County in Khuzestan Province, Iran, where the natural tidal inlet
of Nahre Ghasr, branches Arvand River near the Persian Gulf, see Fig. 1. Nahre Ghasr Fishery Port is
located at Nahre Ghasr channel, about 10 km upstream of Arvand River mouth. The port, opened after
dredging the branch in 2009, has been encountered by a siltation, resulting to the reduction of the
navigability of fishery boats. Fig. 2 shows the large siltation at the port pier in 2014.

Figure 1. Nahre Ghasr Fishery Port in a branch of Arvand River, North-West of the Persian Gulf and the natural
tidal inlet of Nahre Ghasr (left), branches of Arvand River near the Persian Gulf (right)

Figure 2. Nahre Ghasr pier (left: 2009 after dredging, right: 2014)

FIELD MEASUREMENTS
Three series of bathymetric surveys in 2001 (natural channel condition), 2009 (after dredging)
and 2014 have been conducted at the site (Fig. 3). 3-days measurements of the water level and
suspended sediment concentration during neap and spring tides have been performed at the beginning
(ST1) and the end (ST2) of Nahre Ghasr channel, respectively (SPI 2003). Sediment samples were
also taken from three locations along the channel. Fig. 4 shows the locations of sediment samples and
two measuring stations ST1 and ST2. The distributions of grain sizes of sediment samples are shown
in Fig. 5. It is observed that more than %95 of the particles consist of fine silt and clay particles. The
figure also reveals the higher percentage of finer materials in front of the berth (sample 1) and along
the channel (sample 2) in comparisons to the collected sample at the entrance of the Nahre Ghasr
channel (sample 3), as expected.
 COASTAL ENGINEERING 2016 3

Figure 3. Hydrographic surveys: a) 2001 (before the port construction), b) 2009 (after dredging the channel),
and c) 2014

Figure 3. Locations of sediment samples and stations of measured water levels and suspended sediment
concentrations (ST1 and ST2) along Nahre Ghasr channel

Figure 4. Grain size distributions of sediment samples

Fig. 6 shows the continuous increase of the salinity of Arvand River during past 40 years based on
the data in literature. Constructions of dams, which have resulted in the decrease of upstream fresh
water, are blamed for this rapid change. The measurements of surface salinity of Arvand River at the
4 COASTAL ENGINEERING 2016

entrance of Nahre Ghasr in 2007 and 2014 also confirm a clear increase of the water salinity, which
can result in the significant vertical exchange.

Figure 5. Changes of Arvand River salinity with time

NUMERICAL MODELING
Regional and Local Models
Regional modeling of the Persian Gulf and local modeling of Nahre Ghasr have been performed
by DHI MIKE software (DHI 2011). The water surface elevation at the open boundary of regional
model at Strait of Hormuz was adopted from TPXO 8 global tide model (Egbert and Erofeeva 2002)
and the upper boundary was located at the Arvand River mouth with discharge of 1500 m 3/s (Fig. 7).
Various bed roughness coefficients at different locations of the domain have been examined to
minimize water level differences between the model outputs and observations along the north and
south coastlines of the Persian Gulf.

Figure 6. Regional and local numerical models

The lateral and south boundaries of the local Nahre Ghasr model were adopted from the regional
model. The upper boundary of local model is set at upstream of Arvand River. The average discharge
at upstream, water level at downstream (output of regional model), which was linearly applied to the
boundary, and zero discharge at the middle of River's cross section, were applied as the boundary
conditions of local model (Fig. 7). Annual and monthly average discharges of Arvand River in
October and November of 1980 were estimated based on the discharges of three main incoming rivers
of Karoon, Tigris, and Euphrates. It should be noted that the average discharge at the intermediate
longitudinal section of the river is not zero, considering the slight curvature along the river, but the
small effect of assuming no cross river discharge on the modeling results was neglected. Moreover, 20
layers were considered in the three-dimensional local modeling using sigma method. Smagorinsky
and k-ε models were also adopted for horizontal and vertical simulations of turbulence, respectively.
 COASTAL ENGINEERING 2016 5

Model Calibration
Different values of roughness and mesh size have been evaluated in regional model to verify the
measured data near Arvand River mouth. Unstructured triangular grids with a size of 5×107 m2 for
the regional model and 105 m2 for the river were finally selected. Model results with constant and
varying bed roughness were evaluated and a varying bed resistance, with a roughness of 50 m1/3/s at
the boundary and 85 m1/3/s near Arvand River, was chosen (Fig. 9).

Figure 7. Regional model computational grid (left) and selected varying bed resistance (right)

shows the comparisons at ST3 station (see Fig. 9) using a variable roughness in regional
modeling, with the correlation and RMS coefficients of 0.95 and 0.17, respectively.

Figure 8: Verification of regional model at ST3 station with varying bed resistance

The local modeling before port construction was performed applying Hydrodynamic (HD) and
Mud Transport (MT) Modules of MIKE (DHI 2011). The outputs were verified using observed water
levels and suspended sediment concentration data, respectively, in November and October 1980
during neap and spring tides. The input data at ST1 station was introduced to the model and the
outputs at ST2 station were compared with the observations. An unstructured grid with a size of 300m
was selected for Nahre Ghasr channel. Fig. 11 shows the comparisons between 3-days outputs of neap
and spring tides with observed data at ST2.
The local model was calibrated assuming 2 layers (Layer1: soft mud and Layer2: hard mud) with
a time step of 600 s, upstream discharge of 650 m 3/s. The boundary conditions were selected from the
outputs of the regional model of the Persian Gulf. Table 1 shows the parameters of the numerical
simulation.

Table 1. Selected parameters for local model simulation (HD and MT)
Parameters of MT module Parameters of HD module
Parameters Value Parameters Value
Density of sediment 2637 kg/m3 No. time step (Oct & Nov) 2000
6 COASTAL ENGINEERING 2016

Settling velocity 0.018 m/s Bed resistance 85 (m1/3/s)

Critical shear stress 0.05 N/m2 Salinity (1980 model) 0.5 ppt
Layer1:180 kg/m3
Density of bed layer Salinity (2009 model) 5 ppt
Layer2:450 kg/m3
Bed roughness 0.001 m Upstream average discharge 650 (m3/s)
Horizontal dispersion 1 (Constant) Horizontal Eddy viscosity (Smagorinsky formulation) 0.28

Figure 9. The comparisons of water levels at ST2 with the data during spring and neap tides

In order to calibrate the hydrodynamic and mud transport models, the sensitivity analysis was
conducted on the grid size and roughness for HD and fall velocities of sediment particles for MT,
respectively. A roughness value of 85 m 1/3/s and a grid size of 100 m 2 for the Nahre Ghasr channel and
a grid size of 1800 m 2 for Arvand River were finally selected (Fig. 12) . The fall velocity of 0.018 m/s
was also chosen. Fig 13 presents the comparisons of model results of HD and MT modules with
measured data. Table 2 offers the correlation and RMS coefficients of the simulated water surface
elevation and suspended sediment concentration at two stations.

Figure 10. Computational grid of the local model

 COASTAL ENGINEERING 2016 7

Figure 11.Comparisons of water surface elevations and suspended sediment concentrations at ST1 (left) and ST2
(right) with measured data
8 COASTAL ENGINEERING 2016

Table 2. Statistics indices of local model simulation results

Suspended sediment concentration Water surface elevation
Correlation RMSE (kg/m3 ) Correlation RMSE (m)
Tide
ST2 ST1 ST2 ST1 ST2 ST1 ST2 ST1
Spring 0.66 0.15 0.29 0.35 0.86 0.86 0.42 0.42
Neap 0.71 0.64 0.24 0.59 0.83 0.77 0.34 0.42

Figure 12. Comparison between model prediction and measured hydrography in 2014

After calibration, MT module was run to calculate the total siltation volume from 2009
(immediately after port dredging) to 2014. Figure 12 shows the comparison between hydrography data
and calculated bathymetry at 2014, with an error of %9.2.
SILTATION REDUCTION
The rate of siltation at a river harbor is governed by the suspended sediment concentration and
factors that affect the flow pattern (PIANC working group 2008). Two main strategies of “keeping the
sediment out” or “keeping the sediment moving” can be considered to minimize the siltation (Krone
1987). Different mechanisms can cause the formation of vortices and flow exchange between the
river/channel and port including horizontal exchange, vertical exchange, tidal filling, tidal flood and
river flow interactions (Figure 13).
Numerical modeling of these mechanisms shows that the horizontal exchange between Arvand
River and channel causes a flow separation zone near the entrance at some time intervals. The
conflict of river flow and flood tide also leads to a calm zone and the increase of the siltation at the
entrance in some time steps. Three-dimensional modeling by MT module MIKE3 (DHI 2011) reveals
that the differences of flow densities are also effective, resulting to a vertical circulation which brings
high concentration sediment into the port.

de
Horizontal exchange Vertical exchange Tidal filling d ti tide
Conflict of river current andoflood
l o
F
Figure 13. Four mechanisms of flow pattern and siltation at a river port
 COASTAL ENGINEERING 2016 9

Different scenarios were examined for the reduction of siltation rate at Nahre Ghasr channel
including active measures such as constructing CDWs and submerged sills to change the flow patterns
and horizontal eddies at the port entrance.
16 and 17 show two modeling scenarios of CDW applications and their effects on velocity
vectors at the entrance of Nahre Ghasr channel, respectively. The first one is a long 200 m CDW. The
second scenario was performed with a standard CDW, where its total length should be about 30% to
40% of entrance width and need to be located at a distance of 10-15% of main river channel from the
riverbank. Modeling of these two CDWs resulted in 3.2 % and 2% reductions of siltation, respectively.

Figure 16. Modeling of standard CDW and long CDW at the entrance of the channel

Figure 14. The effect of standard and long CDW on the flow pattern and siltation

Figure1818 presents the modeled results of a 1 m height submerged sill, i.e. about 30% of the local
water depth, showing a 7.3% reduction of siltation. The higher reduction, in comparisons to
construction of CDWs, can be related to the high sediment concentration near the bed. This solution is
also effective on the vertical exchange and tidal filling of the harbor. However, construction of a sill
has a major negative impact on the navigation.
10 COASTAL ENGINEERING 2016

Figure18. The effect of constructing a sill at the entrance of Nahre Ghasr channel

SUMMERY AND CONCLUSION

Nahre Ghasr Fishery Port is affected by the suspended sediments in Arvand River, which mostly
consists of silt and clay. Modeling results by the flow model and mud transport revealed that different
phenomena, such as tidal filling, vertical circulation as density differentiation on the sea surface and
seabed, horizontal circulation as flow separation, and the interaction of tides and river flow, are
affecting the high siltation at the Nahre Ghasr channel.
In addition to tidal filling, the discharge of Arvand River causes the horizontal entrainment
(through formation of horizontal eddies) at the entrance of channel. The vertical exchange due to the
sharp density gradient, which results in vertical circulation and carrying the high concentration mud
bed into the port is also effective in the siltation rate. Additionally, interaction between river flow and
tidal flow at some times leads to high deposition at stagnation points.
Modeling of different scenarios proved that although constructions of CDWs and sills have
positive impacts on horizontal/vertical exchanges, their impacts on the reduction of siltation at the
channel are both limited. This is mainly due to the high importance of the tidal filling at Nahre Ghasr
channel, because of the vicinity of the port entrance to the coastline. It can thus be concluded that
none of these two active countermeasures are feasible solutions to reduce the siltation rate at the port.

ACKNOWLEDGMENTS
The authors are grateful to the secretary of the Technical Committee, Mr. Bahirayy, and Head of
Design and Construction Department, Mr. Beheshtian, of Iranian Fisheries Company for their support
of the study and providing parts of the field data.

REFERENCES
Christiansen, H. and Kirby, R., 1991. Fluid mud intrusion and evaluation of a passive device to reduce
mud deposition, Proceedings of CEDA-PIANC Conference “Accessible Harbours”, E1-E14.
Crowder, R.A. 2001. Study of the design and operation of current deflecting walls to reduce siltation
in harbours. Delft Hydraulics, The Netherlands, LIP III report, 8 pp.
Crowder, R.A. 2002. Study of current deflecting walls to reduce siltation in coastal basins, University
of Bradford, Great Britain, Doctoral thesis, 201 pp.
Delft Hydraulics. 2001. Design of a current deflection wall for the Parkhafen in Hamburg, Delft,
Report Z2666, 47 pp.
DHI (Danish Hydraulic Institute). 2011. MIKE 21 & MIKE 3 Flow model FM, Hydrodynamic and
transport module, Hørsholm, 58 pp.
Egbert G.D. and S. Y. Erofeeva. 2002. Efficient inverse modeling of barotropic ocean tides, Journal of
Atmospheric and Oceanic Technology, 19, 183-204.
Krone R.B. 1987. Reducing sedimentation rates in harbor facilities: Sedimentation control to reduce
maintenance dredging of navigational facilities in estuaries, National Academy Press, 128-140.
Nasner, H. 1992. Siltation in tidal harbours, part I (in German), Die Küste, 127-170.
 COASTAL ENGINEERING 2016 11

PIANC working group. 2008. Minimizing harbor siltation, PIANC report No. 102, 75 pp.
SPI (Sazeh Pardazi Iran) Consulting Engineers Company. 2003. Study report of Nahre Ghasr Fishery
Port-phase 1, Iranian Fisheries Company, 274 pp.
Van Rijn, L.C. 2004. Harbour siltation and control measures, online resource, URL: www.leovanrijn-
sediment.com.
Van Leeuwen, S. and B. Hofland. 1999. The current deflecting wall in a tidal harbor with density
influences. Technical University of Delft, Master thesis, 161 pp.
Winterwerp, J.C., W.D. Eysink, F.E. Van Kruiningen, H. Christiansen, R. Kirby and T.J. Smith. 1994.
The current deflection wall: a device to minimise harbour siltation. Dock and Harbour Authority,
74, 243-246.
Winterwerp, J.C. 2005. Reducing harbor siltation. I: Methodology, Journal of Waterway, Port,
Coastal, and Ocean Engineering, 131, 258-266.
Winterwerp. J.C., C. Kuijper, H. Christiansen and J.M. Cornelisse. 2005. Reducing harbor siltation.
II: Case study of Parkhafen in Hamburg, Journal of Waterway, Port, Coastal, and Ocean
Engineering, 131, 267-276.