Obaidullah reduced local scour around structure. An experimental


Obaidullah safie

Department of
Civil Engineering, Nagoya Institute of Technology, Gokiso-cho, Showa-ku

Nagoya, 466-8555,


Akihiro tominaga

Department of
Civil Engineering, Nagoya Institute of Technology, Gokiso-cho, Showa-ku

Nagoya, 466-8555,


Impermeable and permeable spur dikes are used to
protect banks of an open channel. Impermeable spur dike suffers from structural
instability due to local scour; whereas, high permeable spur dike cannot protect
the bank sufficiently. In this study, pile-group dikes are investigated in
order to reduce the velocity behind structure for the purpose of bank
protection and improving aquatic habitat with expected reduced local scour
around structure. An experimental study on the flow characteristics around pile-group
dike structures is conducted using particle image velocimetry (PIV) method. Pile-groups were arranged with in-line and
staggered arrays. Quantitative analysis on flow field around impermeable spur
dike and various permeability pile-group dikes are considered. The results
indicate that staggered arrangement pile-group dike demonstrated efficient significance
than impermeable and in-line arrangement dikes.



For protecting the river bank from erosion, several common measures such as
use of spur dikes (or groins), revetments, riprap structures, concrete armor,
gabion mattress, etc., are presented as a result of years of practice and research
1. Spur dikes (or groins) are hydraulic structures constructed projecting
from a bank of the channel into the current. They redirect the flow in a
desired direction, reduce the flow velocity along the riverbanks, and create
recirculation zones downstream of the structure providing a favorable aquatic
habitat 2.

Spur dikes are classified into impermeable and
permeable types according to the flow penetration through it. Various spur dike
types produce various flow characteristics and patterns around structure and in
the mainstream. Selection of a specific type of spur dike depends on the
purpose of its construction. If a dike is mainly applied for navigation
purposes, the mainstream velocity to secure appropriate depth is considered.
However, if a dike is aimed at protecting a bank or improving the stream environment,
flow velocity reduction behind the structure and the recirculation zones
affected by flow separation become important. Therefore, their analysis is
necessary to select the appropriate type of spur dike for a specific purpose.

Kang et al. 2 studied flow
patterns and characteristics of impermeable, permeable pile of single row, and
triangular shape groins. It is stated for impermeable groin, an increased
velocity at the groin tip and in the mainstream; and a broad spectrum of
recirculation zones were observed which have high effect on the local scour at
the groin tip and the bed changes of the main channel. While, for permeable
pile groins, decreased velocity at the groin tip and strength of the vortex is
reported which has the advantages of excellent stability and relatively easy maintenance

In addition, pile-groups have been also
studied to protect other structures along the bank of a river, e.g. the bridge abutments,
irrigation intakes, or an impermeable spur dike 1. Installation
of a pile group at the upstream adjacent of the spur dike can reduce the local
scour and volume of scouring 3. Additionally, due to the penetration of flow through
a pile dike, it is also useful to prevent changes such as a sudden increase in
the levels at river bends 2. Nevertheless, flow characteristics around pile-group
dikes are not treated sufficiently. It needs for further investigations.

Herein, quantitative analysis on flow characteristics around pile-group
dike is considered. A series of experiments were conducted using particle image
velocimetry (PIV) method. Pile-groups are arranged in two patterns of in-line
and staggered arrays with different permeability as indicated in Figure
1. The main purposes were to analyze the velocity fields around structure, in
the main stream, and behind the structure near protected bank to evaluate the
performance of different pile-group dikes for bank protection and creation of
diverse ecological habitat.

Experimental procedures

The experiments were conducted in a 7.5m long,
0.3m wide, and 0.4m high rectangular flume. The slope of the flume S was set to
0.001. The pile-groups were made of acrylic cylinders of 0.5cm diameter and height
hd of 10cm. The
experimental conditions and the schematic view of the flume with structure are
shown in Table 1 and Figure 1 respectively. The dike was installed
perpendicular to the flume axis 3.0m downstream from the channel entrance. Three
types of dikes were applied; the impermeable (permeability P=0%), the in-line,
and the staggered arrays permeable pile-groups (P of 46.7, 60, and 73.3%). Permeability
rate P is defined by Equation 1.




Length Ld and the width Wd
of dikes were 0.075m in all cases. Diameter of pile was d=0.5cm and n
is the number of piles per row (Figure 1). Permeability rate was controlled by
changing the number of piles, and then the spacing between piles was calculated
according to it. For each case, the same number of piles per row and column (n=m)
hence the same center to center spacing of the piles in the x and y directions
(Sx=Sy) was kept. The discharge Q
was 0.003m3/s, then the mean velocity U0 was 0.251m/s
and the water depth h was set to 0.04m.

Velocity vectors were measured by PIV method in horizontal planes. For
visualization of the flow, nylon resin particles with 80 micron in diameter and
1.02 in specific weight were used. A green laser light sheet was projected on
horizontal (x-y) planes. For each case, the height of the laser projection in
horizontal planes was from 5 to 35mm with 5mm interval. The visual images were
taken by a high speed video camera with 200 frames in a second and they were recorded
as AVI files with 1024 x 1024 pixels. The velocity vectors were measured with a
cross-correlation method by using the commercial PIV software (FlowExpert by
Katokoken). Time averaged velocity vectors were obtained by processing 3200
successive images in 16 seconds.


Table 1. Experimental conditions

Discharge Q



Permeability P





Water Depth h


Number of
piles (n × m)
(per row x per column)

4 × 4

6 × 6

8 × 8


Mean velocity
U0 (m/s)


Pile spacing Sx
= Sy (mm)




Channel Slope


Pile dimeter d(mm)


Channel width
B (m)


Dike length Ld


Froude number


Dike width Wd


        (a)                                                                               (b)

           (c)                                            (d)                                            (e)
Figure 1. Flume and dikes layout: (a) Plan view of dike placement on the
flume bed, (b) side veiw of the flume, (c) in-line array, (d) staggered
array, and (e) impermeable dikes. All dimention are in meters (m).

Results and discussion

The experimental results were analyzed for the velocity in the mainstream,
the recirculating vortices, and the local velocity in the vicinity of protected
bank downstream of structure. Velocity measurements were conducted at a range
of -1Ld to 4Ld in the x direction
and for the entire width of the channel (0 to B) in the y
direction. Figure 2(a)-(g) show the contours of longitudinal velocity U/U0
in the plane of z=2.0cm. Longitudinal velocity U is normalized by
the mean velocity U0 and the y axis by the width of
the channel B. Figure 2(h) summarizes the maximum longitudinal velocity values
Umax in the mainstream that was generated due to installation of
the dike. The values of the longitudinal velocity in the vicinity of the
protected bank Ub downstream of the structure are presented
by Figure 2(i). The values of Umax and Ub
are calculated as regional averages. Umax was obtained by
averaging the values in an area of (2Ld ? x ? 3Ld)
and (0.6B ? y ? 0.8B), while the values of Ub was
obtained by averaging the longitudinal velocity in the section of y=2cm for (1Ld ? x ? 4Ld)
for each case.

(a) P=73.3%                                         (b) P=60%                                             (c)

    (d) P=73.3% staggered                        (e) P=60% staggered                          (f) P=46.7%
                   (g) P=0, Impermeable                                      (h) Umax/U0
                                              (i) Ub/U0

Figure 2. Longitudinal velocity: (a)
– (g) Contours of longitudinal velocity for different permeability rate P,

(h) maximum longitudinal velocity in the mainstream, and (i) longitudinal velocity
in the vicinity of protected bank. All values are for z = 20mm.


3.1      Velocity in the mainstream

Velocity contours of Figure 2 indicate that the increase of velocity in
mainstream is inversely proportional to the permeability rate. In addition, considering
the same P rate, staggered pile-group dike enhanced the mainstream more
than in-line type. While for high permeability rate due to the large opening
between piles, the behavior of both tend to become identical. On the other hand, impermeable dike strongly enhanced
the velocity in the mainstream. The maximum velocity in the mainstream was
about 1.9U0 for impermeable dike while it was lower for pile-group
dikes of any arrangement.


Recirculating vortices and velocity in the
vicinity of protected bank behind the dike

For the impermeable dike, a large vortex was
generated downstream of the structure. Two additional smaller vortices rotating
in opposite direction occurred in front of and behind the dike. For pile-group
dikes this large recirculating vortex was not generated, rather small vortices
behind the piles appeared. The change of velocity due to the presence of pile-group
dike occurred in both the regions, in the mainstream and behind the dike. The
change in these two regions is inversely related to each other. Figure 2(i)
indicates that staggered arrangement has significant influence on velocity
reduction near the bank while the mainstream does not affected strongly.

Furthermore, for any pair of pile-group dike with different arrangement,
but having the same permeability rate, regardless of the magnitudes of
velocities, the contour shapes express similar pattern in the mainstream but
different behind the structure (Figure 2). Behind the structure, for in-line
arrangement pile-group dikes the velocity becomes higher near the bank and then
decreases toward the mainstream up to the width of dike, while for staggered
pile-group; the velocity is minimized near the bank and increases regularly to
the mainstream. However, the impermeable dike has a returned flow near the



Three types of dikes were investigated
experimentally, namely; the impermeable, in-line, and staggered pile-group
dikes. Flow characteristics around the above mentioned dikes were studied. The
results stated that, impermeable dike enhanced strongly the mainstream. An
increased velocity around the structure and a large recirculation zone were
observed. These have a high effect on the local scour around dike and the bed
changes of the main channel. In contrast, it may provide favorable aquatic
habitat in the recirculation zone.

On the other hand, for the pile-group dikes, the velocity behind the
structure was decreased while the mainstream was not enhanced strongly. The flow
penetrated thorough pile-group dikes and did not deviate suddenly to the
mainstream. In addition, the penetrated flow discharged from the pile-group
dike with reduced velocity. This function of pile-group dikes can reduce the
flow velocity for downstream bank protection purpose while reducing the local
scour around structure. Furthermore, by varying the openings between the piles,
the flow gradient along the bank can be adjusted. In addition, slow velocity
zone behind the dike can encourage vegetation growth, providing further stabilization
of the banks as well as improved habitat for aquatic species. Application of
these dikes can be most appropriate for narrow sections where installation of
an impermeable dike results in extreme acceleration in the mainstream. Pile-group
dikes are expected not to enhance the main stream strongly and to reduce local
scour around structure while protecting the bank.

Among the two arrangements of pile dikes, the staggered arrays demonstrated
better significance regarding flow pattern. It reduced the velocity near the
bank and then it is increased gradually to the mainstream, while for the
in-line arrays it was in opposite. This phenomenon needs for further
investigations to clarify its mechanism.




Sayed Hashmat
SADAT and Akihiro TOMINAGA, “Optimal distance between
pile-group and spur-dike to reduce local scour”, Journal of Japan Society of Civil Engineers, Ser. B1 (Hydraulic
Engineering), Vol. 71, No. 4, (2015), pp I_187-I_192.

JOONGU KANG, HONGKOO YEO, SUNGJUNG KIM and UN JI, “Permeability effects of single groin on flow characteristics”, Journal of Hydraulic Research, Vol. 49, No. 6, (2011), pp 728-735.

Sayed Hashmat
SADAT and Akihiro TOMINAGA, “Influence of pile group
density on minimizing local scour of a double spur dike group”, Journal of Japan Society of Civil Engineers, Ser. B1
(Hydraulic Engineering), Vol. 70, No. 4, (2014), pp I_85-I_90.