Environmental Fluid Mechanics

https://doi.org/10.1007/s10652-023-09932-2

ORIGINAL ARTICLE

Numerical simulation of local scour around the pier

with and without airfoil collar (AFC) using FLOW‑3D

Lav Kumar Gupta1 · Manish Pandey1 · P. Anand Raj1

Received: 3 February 2023 / Accepted: 11 June 2023

© The Author(s), under exclusive licence to Springer Nature B.V. 2023

Abstract
This paper studies the pier with Air-Foil Collar (AFC) and its effect on local scour. Sim-
ulation is carried out with two collars having diameters of 2 b and 3 b, where b is pier
diameter, designated as dc1 and dc2 respectively using FLOW-3D. A total of six simulations
are carried out; pier without AFC, pier with collar dc1, dc2 and dc1 in reverse to the flow
direction on the sediment bed and pier with dc1 and dc2 at y∕2 above the sediment bed.
The turbulence model used in the study is Large Eddy Simulation (LES), and the bed-load
transport model utilized is van Rijn. A nested mesh configuration is used with 12.234 mil-
lion mesh cells. The error between the experimental and simulated results is 7%, indicating
a good correlation between them. Temporal variation of scour depth and the percentage
reduction of scour depth using the AFC are explored. The percentage reduction of scour
depth (𝜂) for various case studies and different orientations of the collar ranges between
11 to 100%. Scour depth contours, longitudinal scour hole profiles, transverse scour hole
profiles, and streamlines are developed and analyzed for salient features.

Keywords Air-foil collar · Local scour · Scour depth · Large eddy simulation · FLOW-3D

1 Introduction

Deposition and erosion of sediment bed is a natural and complex phenomenon in the allu-
vial channel, which causes the change in the direction of river flow [1–3]. When the direc-
tion of river flow changes, it carries dangerous and unwanted debris which can destroy
any society’s sustenance [4]. Nowadays, flooding of the river is more common due to
climate change which causes damage to hydraulic structures like bridge elements, abut-
ments, caissons etc.[4–9]. Therefore, flooding of the river is a natural hazard [9–12]. The

* Lav Kumar Gupta

gk701902@student.nitw.ac.in; lkguptace@gmail.com
Manish Pandey
mpandey@nitw.ac.in; manishpandey3aug@gmail.com
P. Anand Raj
anand@nitw.ac.in
1
Department of Civil Engineering, National Institute of Technology Warangal, Warangal 506004,
India

13
Vol.:(0123456789)
 Environmental Fluid Mechanics

characteristics of natural hazards and sediment transport depend upon the change in the
river flow direction [11]. The river flow capacity is affected by the change in bed elevation,
i.e., erosion and deposition [13, 14]. The erosion and deposition of bed materials occur
due to the action of running water, known as scouring [5, 15, 16]. Scouring around the
hydraulic structure is a global problem. It is of three types, i.e., natural, contraction and
local scour. Local scour occurs by destruction in the flow, resulting from downflow and
horseshoe vortex [8, 16–21].
The construction of bridges has always been a complex and challenging process, espe-
cially when spanning over water bodies [22]. In some instances, it becomes necessary to
place piers within the water to span long distances, which can be problematic since the
foundations can be undermined, jeopardizing the structural safety of the bridge. As a
result, it poses a significant concern for public safety and infrastructure, given that scouring
has been responsible for 60% of bridge failures in the United States [22, 23]. For instance,
in 1994, tropical storm Alberto made landfall in Georgia, USA, and caused damage to over
500 bridges due to scour. Out of these, 31 were considered unsafe and needed reconstruc-
tion because scouring reached depths between 4 and 6 m [23]. The widespread nature of
scouring is evident, and it has caused significant damage through various incidents [24].
With the advancement in computer science, Computational Fluid Dynamics (CFD)
became an essential tool for analyzing and visualizing the scouring process [25]. It gives
significant and reliable results compared to experimental results. Nowadays, CFD is more
accessible than experimental work because of difficulties like data collection, scale issues,
time restriction, human error and human resources in experimental work [26]. Experi-
mental results are obtained only where the gauges and sensors are fixed. Also, during the
pandemic and health emergency, it is almost inaccessible to conduct experiments, whereas
CFD becomes useful [27]. The three main turbulence models used in CFD are Reynolds-
Averaged Navier–Stokes (RANS), Large Eddy Simulation (LES), and Direct Numerical
Simulation (DNS) [28]. RANS solves steady-state Navier–Stokes equations, which include
time-averaged velocity. It does not consider the fluctuating components, unsteadiness and
anisotropic turbulence on sediment transport [29].
In turbulent flow, energy is transferred from larger scales of motion to smaller scales of
motion through a process called energy cascade. The large-scale flow structures, which are
characterized by their long time and length scales, dominate the momentum and energy trans-
port. However, at smaller scales, the flow becomes more chaotic and turbulent, with small-
scale eddies and vortices appearing and dissipating rapidly. In Large Eddy Simulation (LES),
the goal is to simulate the large-scale flow structures accurately while modelling the smaller-
scale eddies [29–32]. This is achieved by filtering out the small-scale turbulent fluctuations
using a filter function and then applying a turbulence model to the remaining large-scale flow
structures. The subgrid-scale turbulent fluctuations are modelled using various approaches,
such as eddy viscosity models or dynamic models [25]. Chen et al. [33] investigated the
effects of collar width and hook height on scour reduction through laboratory experiments and
numerical simulations using FLOW-3D. They used a single hook for the laboratory experi-
ments and a double hook for the simulations, both with collar widths of 1.25b and a hook
height of 0.25b, where b is the pier diameter. The results showed that the pier with a sin-
gle hooked collar placed on the bed reduced scouring by 42%, while the pier with a double
hooked collar placed on the bed reduced scouring by 50%. Qi et al. [25] investigated the effect
on local scour around an isolated pier under clear water conditions due to the tilted anticlock-
wise circular collar. FLOW-3D was utilized to simulate the six different cases, i.e., collar tilted
on the bed at 0°, 5°,10°, 15° and 20°. It was concluded that a circular collar on the bed with
0° tilt is best suited for reducing local scour. Valela et al. [27] developed a new collar over

13
Environmental Fluid Mechanics

the traditional flat collar. The best collar design was achieved using OpenFOAM software.
Furthermore, an iteration hybrid process was utilized to get the improved collar, and it was
tested physically and observed that it performed better than the traditional collar by 11%. This
paper investigates scour depth reduction around the bridge pier with and without AFC using
FLOW-3D.

2 Governing equation

2.1 Large Eddy Simulation (LES)

For simulating the turbulent flow, LES is considered as the most popular technique. Accord-
ing to Kolmogorov’s theory of self-similarity from 1941, the geometry of the flow affects the
large eddies, while the smaller scales tend to be more universal [34–37]. This characteristic
allows for explicit solving of the large eddies during calculations, with the smaller eddies
being implicitly accounted for by incorporating a Sub Grid-Scale (SGS) model. For incom-
pressible flow, the filtered Navier–Stokes equation is employed as LES. This is given by the
following eqs:
( )
𝜕𝜌 𝜕 𝜌ui
+ =0 (1)
𝜕t 𝜕xi

( )
𝜕ui 𝜕ui 1 𝜕p 𝜕 𝜕ui 1 𝜕𝜏ij
+ uj =− + 𝜈 + (2)
𝜕t 𝜕xj 𝜌 𝜕xi 𝜕xj 𝜕xj 𝜌 𝜕xj

where 𝜌 is the fluid density, t is time, ui are filtered velocities in ith direction, xi
(i = 1, 2and3) are three axes in the cartesian coordinate system, p is filtered pressure, 𝜈 is
kinematic viscosity and the term 𝜕xij brings the nonlinearity in the system because of
𝜕𝜏
j

uj 𝜕xi ≠ uj 𝜕xi hence,

𝜕u 𝜕u
j j

𝜏ij = ui uj − ui uj (3)
where 𝜏ij is SGS stress and similar equations can be derived for the SGS field (i.e., the
residual field) and it is calculated using Boussinesq’s hypothesis.
1
𝜏ij − 𝜏kk 𝛿ij = −2𝜇t Sij (4)
3
where, 𝜏kk is the SGS stress turbulent kinetic energy, 𝛿ij is filtered scale length, 𝜇t is SGS
turbulent viscosity and Sij is the rate of strain tensor for the resolved scale defined by
( )
1 𝜕ui 𝜕uj
Sij = + (5)
2 𝜕xj 𝜕xi

Then filtered Navier–Stokes equation can be the following equation

( )
𝜕ui 𝜕u 1 𝜕p 𝜕 [ ] 𝜕ui
+ uj i = − + 𝜇 + 𝜇t (6)
𝜕t 𝜕xj 𝜌 𝜕xi 𝜕xj 𝜕xj

13
 Environmental Fluid Mechanics

2.2 Sediment transport model

The volumetric sediment transport rate per bed width is given by van Rijn equation and is
given as [38–40].
( )2.1
𝜃
Φ = 𝛽d∗−0.3 − 1.0 Cb (7)
𝜃cr

where 𝜙 is dimensionless rate of bed-load transport, 𝛽 is bed load coefficient, d∗ is dimen-

sionless particle size, 𝜃 and 𝜃cr are local and critical Shields parameters, respectively which
are given by the Eqs. (8) and (9) [26, 41–43]. Cb is the amount of sediment in a sample,
expressed as a percentage of the total volume of the sample.
(( ) )1∕3
Gs − 1 g
d∗ = d50 (8)
𝜈2

0.3 ( )
𝜃cr = + 0.055 1 − e−0.2d∗ (9)
1 + 1.2d∗

where Gs is the relative density of sediment particles, g is the gravitational acceleration

constant.

3 Experimental and numerical setup

Experiments are conducted in a rectangular flume of length 10.3 m, width of 0.8 m and
0.5 m depth located in the Hydraulic lab in the department of civil engineering at NIT
Warangal, as shown in Fig. 1. The working section of the flume is situated at a distance
of 4 m from the inlet and has dimensions of 2.3 m in length, 0.8 m in width, and 0.3 m
in depth. The circular pier of 5 cm diameter (b) is considered in this study and is kept at
the centre of the working section. The sediment bed is prepared with uniform √ sand with
median size diameter (d50) of 0.32 mm with standard deviation (𝜎 = d84 ∕d16 ) 1.31
where the median particle size d84 and d16 are diameters for which 84% and 16% of the sed-
iment particles are finer respectively [44]. The thickness of the sediment bed is 30 cm. Pier
diameter (b) is selected such that effect of sidewall and contraction effect are negligible. In
channel, discharge is measured by an ultrasonic flowmeter with an accuracy of ± 0.0001
­m3/s installed at the pipe inlet’s exit. All the experiments are conducted with constant
water depth (y = 10 cm) and under clear condition, i.e., V∕Vc < 1 to ensure no transport of

Fig. 1  Layout of the experimental setup (plan view)

13
Environmental Fluid Mechanics

sediment into the scour hole from upstream, where V and Vc are approach and critical flow
velocities respectively. The study is carried out with the flow intensity (V∕Vc ) of 0.95.
Figure 2a shows the simulation setup of the working section in FLOW-3D. The pier
is placed at center of the working section with AFC attached to it. The AFC geometry
is built in AUTOCAD and is imported to FLOW-3D. The two different collars of 4 mm
thickness are used in the study with a diameter of dc1( 2b) and dc2 (3b) and a length of 4b
and 6b. Figure 2b shows the description of the AFC dc1. The input properties of sediment
and hydraulic conditions are shown in Table 1. Simulations are carried out for six cases, as
shown in Table 2.

3.1 Mesh sensitivity analysis

Sensitivity analysis of meshing in CFD is an important consideration to estimate scour

depth. Four different mesh cell sizes are used to validate the numerical results with experi-
mental results. A nested mesh configuration in which coarse mesh cells cover the entire
geometry and finer meshing is used near the pier with AFC. The total number of mesh cells
is 12,233,789 (12.234 million), with coarse cells of 2,225,664 and fine cells of 10,008,125.
Table 3 shows the mesh sensitivity analysis and it is observed that decreasing the mesh cell
size increases the accuracy of the results. So, the study uses coarse and fine cell sizes of
0.8 and 0.4 cm, respectively. Meshing is done using the Fractional Area-Volume Obstacle
Representation (FAVOR) method for complex geometry [45]. A computer with 128 GB
RAM and i-7 processor is used to simulate the scour process with the simulation running
for eight to twelve days.

3.2 Boundary and initial conditions

The input boundary conditions used in the simulation is shown in Fig. 3b. The upstream
and downstream boundary conditions are velocity (V) and continuative (C), respectively.
Floor, lateral and free surface boundary conditions are wall (W), wall (W) and symmetry
(S), respectively. The initial water depth and pressure distribution used in the simulation
are 0.5 m and hydrostatic pressure, respectively.

3.3 Validation of LES model from previous research

The experimental data of Melville and Raudkivi [16] is utilized to verify the accuracy of
the LES. This same data was also utilized by Qi et al. [25] in their numerical simulation
validation process. The experiments were carried out in a rectangular flume of 19 m in
length, 0.456 m in width and 0.44 m in depth. A cylindrical pier of diameter 5.08 cm was
used as bridge pier with uniform sediment of median size (d50) of 0.385 mm. The approach
velocity and flow depth are 0.25 m/s and 15 cm, respectively. The experiment continued
for several hours, and during the first 30 min, the depth of the scour hole increased rapidly,
but subsequently progressed at a slower rate. The simulation has same conditions as in
experimental setup. The velocity, continuity, wall, wall, and symmetry boundary condi-
tions are applied to the upstream, downstream, floor, lateral side, and free surface, respec-
tively. To validate the observed data of Melville and Raudkivi [16] in a simulation model,
the parameters such as critical Shields number, entrainment coefficient, angle of repose,
bed roughness, and bed load coefficient of 0.033, 0.018, 32°, 2.5, and 0.050 respectively

13
13
Fig. 2  a Simulation setup and b description of collar dc1
Environmental Fluid Mechanics
Environmental Fluid Mechanics

Table 1  Sediment and hydraulic conditions in the simulation setup

Median Specific Critical Angle Bed load Flow Critical Froude number,Fr
particle size, gravity of Shields of coefficient depth, y flow veloc-
d50 (mm) sediment,Gs param- repose (cm) ity, Vc (m/s)
eter,𝜃cr

0.32 2.65 0.033 29° 0.045 10 0.26 0.26

are considered. The study uses nested mesh configuration of coarse and fine cell sizes of
0.8 and 0.4 cm, respectively.

4 Results and discussion

4.1 Results of LES model from previous research

The maximum scour depth around the pier is 4.073 cm and 3.89 cm from Melville and
Raudkivi [16] and this simulation, respectively. The error between simulated and observed
results is below 6%. Melville and Raudkivi’s [16] findings indicate that the point of maxi-
mum scour depth is located at the upstream face (0°) of the pier. However, simulations
suggest that it is situated within the range of 0° to 45°. The comparison between Melville
and Raudkivi [16]and the simulated are shown in Fig. 4 for scour profiles. Figure 4a shows
the longitudinal scour profile in which maximum scour occurs at the upstream face of the
pier along the centre line of the pier, and deposition occurs behind the downstream face
of the pier. As shown in Fig. 4 (b), the traverse scour hole profile is symmetrical about
the pier’s centre line. The findings of this study align more closely with previous research,
as evidenced by the comparable maximum depth of local scour and the similar shape of
the scour hole observed. From the Melville and Raudkivi [16] and simulated results, it is
observed that both profiles are significantly similar. It can be concluded that LES model
utilized is both reasonable and accurate, indicating its suitability for computing the utmost
scouring depth surrounding the bridge pier.

4.2 Validation of numerical result with the present experimental result

Experiment is conducted to validate the maximum scour depth around the pier without collar,
which is found to be 6.8 cm and numerical simulation for the similar condition is 6.33 cm.
Thus, an error of 7% is observed, with simulation underestimating the scour depth as shown
in Fig. 5. To check the accuracy of the numerical results, various statistical indices are evalu-
ated by comparing the numerical and experimental results. The coefficient of correlation
(R2) is calculated to assess the strength of the linear relationship between the numerical and
experimental results. Additionally, the mean absolute error (MAE) is computed to determine
the average magnitude of the errors in the numerical simulations [46]. The relative error per-
centage (REP) is utilized to quantify the discrepancy between the simulated and experimental
results as a percentage of the experimental results. The root mean square error (RMSE) is
also employed to measure the magnitude of the errors in the simulated results. The model’s
performance is evaluated using the Nash–Sutcliffe efficiency (NSE) coefficient, which com-
pares the model’s performance to the experimental results based on the mean squared error of

13
 Environmental Fluid Mechanics

Table 2  Description of simulation for six cases

Case 1 2 3 4 5 6

Location of Pier Pier with Pier with Pier with Pier with Pier with AFC having
AFC with without AFC having AFC having AFC having AFC having diameter 3b at y/2
respect to AFC diameter 2b diameter diameter 2b diameter above the sediment bed
sediment on sediment 2b on sedi- at y/2 above 3b on the (dc2 )@y∕2
bed bed (dc1) ment bed the sedi- sediment
in reverse ment bed bed dc2)
to the flow (dc1 )@y∕2
direction
(dc1R )

the residuals. The value of R2, MAE, RMSE, and NSE are evaluated using the following Eqs.
(10–13), which are found to be 0.981, 3.65, 6.30 and 0.924, respectively. Overall, the results
of the study indicated that the simulation has a high level of accuracy, as demonstrated by the
low values of MAE, RMSE, and REP and the high value of the R2 and NSE.
2
⎡ ∑ ∑ ∑ ⎤
⎢ (N Mexp Msim ) − ( Mexp )( Msim ) ⎥
2
R = ⎢� � (10)
� � � � ⎥
⎢ N(∑ M 2 ) − ∑ M 2 N(∑ M 2 ) − ∑ M 2 ⎥
⎣ exp exp sim sim ⎦

∑� �
�Mexp − Msim �
MAE = � � (11)
N
√
√
√1 ∑ N
( )2
RMSE = √ Mexp − Msim (12)
N i=1

∑N � �2
i=1
Mexp − Msim
NSE = 1 −
∑N � �2 (13)
i=1 M exp − M exp

where Mexp and Msim are experimental and simulated values, N and Mexp are total no of
data and average experimental values, respectively.
From the numerical simulation, 67 and 83% of equilibrium scour depths are obtained
at 10 and 20% of equilibrium time. From the experimental results, the maximum scour
depth location is at the pier’s upstream face (0°). But, in simulation, it is between 0° and
45°. Simulation did not estimate the location of maximum scour depth accurately because,
it does not measure the velocity of turbulent pulsation. Also, the smallest mesh cell size is
not capable of capturing the actual scour hole location. The second reason might be that
it does not consider the sliding, rolling and jumping of sediment particles as in real-world
scenarios.

13
Environmental Fluid Mechanics

4.3 Temporal variation of scour depth with and without AFC

The equilibrium scour depth around the pier without AFC is 6.33 cm, and with collars
dc1, dc1R and dc2, it is 1.4, 3.5 and 0 cm, respectively when they are on the sediment bed
as shown in Fig. 6. When dc1 and dc2 are at y∕2 above the sediment bed then, equilibrium
scour depth around the pier is 5.6 and 3.6 cm, respectively. The percentage of scour reduc-
tion (η) is defined as
( )
d − dc
𝜂= ∗ 100 (14)
d

where d and dc are the maximum scour depth around the pier without and with AFC,
respectively. The percentage of scour reduction using dc1, dc1R, dc2 on sediment bed is
77.78, 46 and 100% and for dc1 and dc2 at y/2 above sediment bed is 11.12 and 42.86%. It is
observed that collar dc2 on sediment bed is the best suited for reducing local scour around
the bridge pier. The increasing order of percentages of scour reduction for collars are dc1 at
y/2 above the sediment bed, dc2 at y/2 above the sediment bed, dc1 in the reverse direction
to flow on the sediment bed, dc1 and dc2 on sediment bed, respectively. It is observed that
the location of AFC also affects the scour depth around the pier. The collar dc1 is 31.78%
more efficient than dc1R in reducing the scour around the pier. The streamlined geometry
of the AFC reduces the turbulent intensity of the flow, leading to a decrease in the strength
of horseshoe and wake vortices and a reduction in the transportation of sediment particles
downstream of the pier.

4.4 Morphological changes and scour depth contours

Figure 7 shows the bed morphological changes around the pier with different collar sizes
placed at different elevations. These figures show equilibrium scour hole shapes, reflecting
the scouring and deposition area. Using the collar, minimum scour is observed at dc2 on
sediment bed as shown in Fig. 7 (e) and maximum scour depth at dc1 at y/2 above the sedi-
ment bed as shown in Fig. 7d. The maximum depositions of 1.54, 1.6, 1.64, 1.55, 2.02 and
0.6 cm occurred at the distances of 33.5, 22, 23.4, 32.6, 19.2 and 31.9 cm from the centre
of the pier respectively.
Scour depth contours are developed in six cases as shown in Fig. 8. Figure 8b and e
shows that in the longitudinal direction, no scouring occurred around the pier. From
Fig. 8c, there is no scouring in the downstream face of the pier. Also, no scour is developed
around the pier with collar bc2 on bed level as shown in Fig. 8e. Similar pattern in contour

Table 3  Mesh sensitivity analysis

S. N Meshing
Coarse mesh Fine mesh Experimental FLOW-3D Error (%) Estimated Total no of
cell size (cm) cell size results (cm) results (cm) computational mesh cells
(cm) time (days) (million)

1 2 1 6.8 3.3 51.47 2.5 0.783

2 1 0.5 6.8 4.7 30.88 8 6.533
3 0.8 0.5 6.8 6.3 7.35 10 7.374
4 0.8 0.4 6.8 6.33 6.91 18 12.234

13
13
Fig. 3  a Meshing around the geometry and b boundary conditions annotated
Environmental Fluid Mechanics
Environmental Fluid Mechanics

(a) Scour hole profile in the longitudinal (b) Scour hole profile in the transverse
direc on direc on

Fig. 4  Scour hole profile from Melville and Raudkivi [16] and simulated results

Fig. 5  Dimensionless scour vs

time for experimental and simu-
lation results

profiles is observed when the pier is without the collar and when the collar is placed above
the sediment bed. Collars bc1 and bc2 on the sediment bed have similar profile and differ
when the collar bc1 is placed in reverse to the flow direction.

4.5 Scour hole profiles

Longitudinal profiles of six cases are shown in Fig. 9. Scour occurs at the upstream face
of pier and deposition at the downstream of the pier. Length of the longitudinal scour hole
in six cases is 30.7, 2.17, 8.67, 26.5, 0 and 16 cm respectively. Length of transverse scour
hole in six cases is 22.23, 4.1, 25.5, 22, 0 and 18 cm, respectively. Transverse profiles are
symmetrical because of the circular shape of the collar in downstream face of the pier as
shown in Fig. 10.

13
 Environmental Fluid Mechanics

4.6 Plotting of the streamlines

Streamlines pattern around the pier with AFC in six cases is shown in Fig. 11. The wake
regions behind the pier exhibit low flow velocity, whereas the maximum flow velocity
is observed at the pier’s side. When pier is attached with AFC, results in a gradual flow
separation, thereby avoiding sudden flow separation. Reduction of intensity of vortices is
directly related to scour reduction. Furthermore, the collar increases the number of zero-
velocity streamlines around the pier, which signifies a reduction in the intensity of horse-
shoe and wake vortices. AFC is similar to streamline body and greatly influences the vor-
tex system i.e., horseshoe and wake vortices. When flow bypassed the pier attached with
collar, vortex system is divided the into two parts; vortex above the collar and below the
collar. As the local scour hole around the pier increases, the number of streamlines also
increases. As a result, the vortex below the collar also increases. Number of streamlines are
lesser below the collar because of the smaller scour hole.

5 Conclusions
This study uses numerical simulation to explore the effect of the airfoil collar on local
scour around the bridge pier. A total of six simulations are carried out with and without
AFC. The error between the experimental and simulation result is 7% which is a good
performance.
The percentage of scour reduction using dc1, dc1R, dc2 on sediment bed is 77.78, 46
and 100%, respectively. For dc1 and dc2 at y∕2 from sediment bed, it is 11.12 and 42.86%,
respectively. The increasing order of percentages of scour reduction for collars are dc1 at
y∕2 above the sediment bed, dc2 at y∕2 above the sediment bed, dc1 in reverse to flow direc-
tion on the sediment bed, dc1 and dc2 on sediment bed. The collar dc1 is 31.78% efficient
than dc1R in reducing the scour around the pier. When the collar dc1 and dc2 are placed on
the sediment bed, it is observed that no scouring occurred around the upstream face of the
pier. Also, with collar dc1R no scouring occurred in the downstream face of the pier.
Length of the longitudinal scour hole in six cases is 30.7, 2.17, 8.67, 26.5, 0 and 16 cm,
respectively. Length of the transverse scour hole in six cases is 22.23, 4.1, 25.5, 22, 0 and
18 cm, respectively. The presence of AFC significantly reduces horseshoe and wake vorti-
ces, as evidenced by the higher number of zero velocity streamlines around the pier.
Fig. 6  Temporal variation of
scour depth around the pier using
AFC

13
Environmental Fluid Mechanics

(a) Circular pier without AFC (b) Circular pier with AFC having diameter 2b
on sediment bed ( )

(c) Circular pier with AFC having diameter 2b (d) Circular pier with AFC having diameter 2b
on sediment bed in reverse to the flow at y/2 above the sediment bed ( )@
direcon )

(e) Circular pier with AFC having diameter 3b (f) Circular pier with AFC having diameter 3b
on the sediment bed ( ) at y/2 above the sediment bed ( )@

Fig. 7  Morphological changes in six different cases

13
 Environmental Fluid Mechanics

(a) Circular pier without AFC (b) Circular pier with AFC having diameter
2b on sediment bed ( )

(c) Circular pier with AFC having diameter (d) Circular pier with AFC having diameter
2b on sediment bed in reverse to the flow 2b at y/2 above the sediment bed
direcon ) ( )@

(e) Circular pier with AFC having diameter (f) Circular pier with AFC having diameter 3b
3b on the sediment bed ( ) at y/2 above the sediment bed ( )@

Fig. 8  Scour depth contours developed around the pier in six cases

13
Environmental Fluid Mechanics

(a) Circular pier without AFC (b) Circular pier with AFC having diameter 2b
on sediment bed ( )

(c) Circular pier with AFC having diameter 2b (d) Circular pier with AFC having diameter 2b
on sediment bed in reverse to the flow at y/2 above the sediment bed ( )@
direcon )

(e) Circular pier with AFC having diameter 3b (f) Circular pier with AFC having diameter 3b
on the sediment bed ( ) at y/2 above the sediment bed ( )@

Fig. 9  Longitudinal scour hole profiles for six cases

13
 Environmental Fluid Mechanics

(a) Circular pier without AFC (b) Circular pier with AFC having diameter
2b on sediment bed ( )

(c) Circular pier with AFC having diameter 2b (d) Circular pier with AFC having diameter
on sediment bed in reverse to the flow 2b at y/2 above the sediment bed
direcon ) ( )@

(e) Circular pier with AFC having diameter 3b (f) Circular pier with AFC having diameter 3b
on the sediment bed ( ) at y/2 above the sediment bed ( )@

Fig. 10  Transverse scour hole profiles for six cases

13
Environmental Fluid Mechanics

(b) Circular pier with AFC having diameter

(a) Circular pier without AFC
2b on sediment bed ( )

(c) Circular pier with AFC having diameter 2b on (d) Circular pier with AFC having diameter
sediment bed in reverse to the flow direcon 2b at y/2 above the sediment bed
) ( )@

(e) Circular pier with AFC having diameter 3b on (f) Circular pier with AFC having diameter 3b
the sediment bed ( ) at y/2 above the sediment bed ( )@

Fig. 11  Streamlines pattern around the pier with collar in six cases

Author contributions Conceptualization, M.P. and P.A.R.; methodology, L.K.G. and M.P.; software,
L.K.G..; validation, L.K.G.; investigation, M.P. and L.K.G.; resources, M.P. and L.K.G.; data analysis,
L.K.G.; writing—original draft preparation, L.K.G.; writing—review and editing, M.P., L.K.G. and P.A.R.;
visualization, M.P. and P.A.R..;. All authors have read and agreed to the published version of the manuscript.

Funding This research received no external funding.

Declarations
Competing interests The authors declare no competing interests.

Conflict of interest The authors declare no conflict of interest.

References
1. Karna N, Hari Prasad KS, Giri S, Lodhi AS (2015) Intrusion of fine sediments into river bed and its
effect on river environment - A research review. ISH J Hydraul Eng 21:142–150. https://​doi.​org/​10.​
1080/​09715​010.​2014.​982000

13
 Environmental Fluid Mechanics

2. Gupta LK, Pandey M, Raj PA, Shukla AK (2023) Fine sediment intrusion and its consequences for
river ecosystems : a review. J Hazardous, Toxic, and Radioactive Waste 27:1–12. https://​doi.​org/​10.​
1061/​(ASCE)​HZ.​2153-​5515.​00007​29
3. Mathers KL, Rice SP, Wood PJ (2019) Discharge and suspended sediment time series as controls on
fine sediment ingress into gravel river beds. CATENA 173:253–263. https://​doi.​org/​10.​1016/j.​catena.​
2018.​10.​001
4. Deng L, Wang W, Yu Y (2016) State-of-the-art review on the causes and mechanisms of bridge col-
lapse. J Perform Constr Facil 30:04015005. https://​doi.​org/​10.​1061/​(ASCE)​cf.​1943-​5509.​00007​31
5. Melville BW (1997) Pier and abutment scour: integrated approach. J Hydraul Eng 123:125–136.
https://​doi.​org/​10.​1061/​(ASCE)​0733-​9429(1997)​123:​2(125)
6. Kothyari UC, Kumar A (2012) Temporal variation of scour around circular compound piers. J Hydraul
Eng 138:945–957. https://​doi.​org/​10.​1061/​(ASCE)​hy.​1943-​7900.​00005​93
7. Hong J-H, Chiew Y-M, Yeh P-H, Chan H-C (2017) Evolution of local pier-scour depth with dune
migration in subcritical flow conditions. J Hydraul Eng 143:04016098. https://​doi.​org/​10.​1061/​(ASCE)​
hy.​1943-​7900.​00012​61
8. Shahriar AR, Ortiz AC, Montoya BM, Gabr MA (2021) Bridge Pier Scour: an overview of factors
affecting the phenomenon and comparative evaluation of selected models. Transp Geotech 28:100549.
https://​doi.​org/​10.​1016/j.​trgeo.​2021.​100549
9. Tafarojnoruz A, Gaudio R, Calomino F (2012) Evaluation of flow-altering countermeasures against
bridge pier scour. J Hydraul Eng 138:297–305. https://​doi.​org/​10.​1061/​(ASCE)​hy.​1943-​7900.​00005​12
10. Pandey M, Pu JH, Pourshahbaz H, Khan MA (2022) Reduction of scour around circular piers using
collars. J Flood Risk Manag 15:1–16. https://​doi.​org/​10.​1111/​jfr3.​12812
11. Bussi G, Francés F, Horel E et al (2014) Modelling the impact of climate change on sediment yield in
a highly erodible Mediterranean catchment. J Soils Sediments 14:1921–1937. https://​doi.​org/​10.​1007/​
s11368-​014-​0956-7
12. Aly AM, Dougherty E (2021) Bridge pier geometry effects on local scour potential: a comparative
study. Ocean Eng. https://​doi.​org/​10.​1016/j.​ocean​eng.​2021.​109326
13. Harrison E, Norris R, Wilkinson S (2007) The impact of fine sediment accumulation on benthic mac-
roinvertebrates: implications for river management. Proceedings of the 5th Australian Stream Man-
agement Conference. Australian rivers: making a difference. Proc 5th Aust Stream Manag Conf Aust
rivers Mak a Differ 139–144
14. Khullar NK, Kothyari UC, Ranga Raju KG (2013) Study of deposition of fine sediment within the
pores of a coarse sediment bed stream. Int J Sediment Res 28:210–219. https://​doi.​org/​10.​1016/​S1001-​
6279(13)​60032-3
15. Chiew YM, Melville BW (1987) Local scour around bridge piers Local scour around bridge piers. J
Hydraul Res 25:15–26
16. Melville BW, Raudkivi AJ (1977) FLOW CHARACTERISTICS IN LOCAL SCOUR AT BRIDGE
PIERS. J Hydraul Res 15:373–380. https://​doi.​org/​10.​1080/​00221​68770​94996​41
17. Breusers HNC, Nicollet G, Shen HW (1977) Local Scour Around Cylindrical Piers. J Hydraul Res
15:211–252. https://​doi.​org/​10.​1080/​00221​68770​94996​45
18. Chiew Y (1992) Scour Protection at Bridge Piers. J Hydraul Eng 118:1260–1269. https://​doi.​org/​10.​
1061/​(ASCE)​0733-​9429(1992)​118:​9(1260)
19. Camporeale C, Perucca E, Ridolfi L, Gurnell AM (2013) Modeling the interactions between river mor-
phodynamics and riparian vegetation. Rev Geophys 51:379–414. https://​doi.​org/​10.​1002/​rog.​20014
20. Ahmed QA, Atiq M, Rehman U (2022) Effects of hooked-collar on the local scour around a lenticular
bridge. Int J Sediment Res. https://​doi.​org/​10.​1016/j.​ijsrc.​2022.​07.​002
21. Mming L, Cheng WH, Qiang TG et al (2022) Investigation of local scour around two vertical piles by
using numerical method. Ocean Eng 244:110405. https://​doi.​org/​10.​1016/j.​ocean​eng.​2021.​110405
22. Lagasse, P. F., Clopper, P. E., Zevenbergen, L. W., & Girard LG (2007) Countermeasures to Protect
Bridge Piers from Scour. Transportation Research Board
23. Lagasse PF, Clopper PE, Pagan-Ortiz JE et al (2009) Bridge scour and stream instability countermeas-
ures: experience, selection, and design guidance-third edition. Hydraul Eng Circ 2:376
24. Hong J-H, Chiew Y-M, Lu J-Y et al (2012) Houfeng bridge failure in Taiwan. J Hydraul Eng 138:186–
198. https://​doi.​org/​10.​1061/​(ASCE)​hy.​1943-​7900.​00004​30
25. Qi H, Tian W, Zhang H (2021) Modeling local scour around a cylindrical pier with circular collar with
tilt angles (Counterclockwise around the direction of the channel cross-section) in clear-water. Water
(Switzerland). https://​doi.​org/​10.​3390/​w1322​3281
26. Pourshahbaz H, Abbasi S, Pandey M et al (2020) Morphology and hydrodynamics numerical simula-
tion around groynes. ISH J Hydraul Eng 00:1–9. https://​doi.​org/​10.​1080/​09715​010.​2020.​18300​00

13
Environmental Fluid Mechanics

27. Valela C, Nistor I, Rennie CD et al (2021) Hybrid modeling for design of a novel bridge pier collar for
reducing scour. J Hydraul Eng. https://​doi.​org/​10.​1061/​(ASCE)​hy.​1943-​7900.​00018​75
28. Bhide K, Abdallah S (2022) High-order accurate numerical simulation of supersonic flow using RANS
and LES guided by turbulence anisotropy. Fluids. https://​doi.​org/​10.​3390/​fluid​s7120​385
29. Kirkil G, Constantinescu G, Ettema R (2009) Detached Eddy Simulation Investigation of Turbulence
at a Circular Pier with Scour Hole. J Hydraul Eng 135:888–901.
30. Chamani R, Rajaratnam N, Beirami MK, Sakhalkar S V (2011) Discussions and Closures Discussion
of “ Turbulent Jet Energy Dissipation at Vertical Drops ” by Mohammad Closure to “ Turbulent Jet
Energy Dissipation at Vertical Drops ” by Mohammad R . Chamani , Discussion of “ Energy Dissipa-
tion and Turbulent Producti. 134: 1532–1535.
31. Koken M, Constantinescu G (2008) An investigation of the flow and scour mechanisms around iso-
lated spur dikes in a shallow open channel: 1. Conditions corresponding to the initiation of the erosion
and deposition process. Water Resour Res 44:1–19. https://​doi.​org/​10.​1029/​2007W​R0064​89
32. Gupta LK, Pandey M, Raj PA, Pu JH (2023) Scour reduction around bridge pier using the airfoil-
shaped collar. Hydrology 10:77.
33. Chen SC, Tfwala S, Wu TY et al (2018) A hooked-collar for bridge piers protection: Flow fields and
scour. Water (Switzerland) 10:1–12. https://​doi.​org/​10.​3390/​w1009​1251
34. Hinterberger C, Rodi W (2007) Three-dimensional and depth-averaged large-eddy simulations of some
shallow water flows. J Hydraul Eng 133:857–872
35. Armenio V (2017) Large eddy simulation in hydraulic engineering: examples of laboratory-scale
numerical experiments. J Hydraul Eng 143:1–19. https://​doi.​org/​10.​1061/​(ASCE)​hy.​1943-​7900.​00013​
57
36. Dhamankar NS, Blaisdell GA, Lyrintzis AS (2018) Overview of turbulent inflow boundary conditions
for large-eddy simulations. AIAA J 56:1317–1334. https://​doi.​org/​10.​2514/1.​J0555​28
37. Keating A, Piomelli U, Balaras E, Kaltenbach HJ (2004) A priori and a posteriori tests of inflow condi-
tions for large-eddy simulation. Phys Fluids 16:4696–4712. https://​doi.​org/​10.​1063/1.​18116​72
38. van Rijn LC (1984) Sediment transport, part I: bed load transport. J Hydraul Eng 110:1431–1456
39. van Rijn LC (1984) Sediment transport, part II: suspended load transport. J Hydraul Eng
110:1613–1641
40. van Rijn LC (1984) Sediment transport, part III: bed forms and alluvial roughness. J Hydraul Eng
110:1733–1754
41. Ghaderi A, Abbasi S (2019) CFD simulation of local scouring around airfoil-shaped bridge piers with
and without collar. Sadhana - Acad Proc Eng Sci. https://​doi.​org/​10.​1007/​s12046-​019-​1196-8
42. Li GLL (2016) 3D numerical simulation of flow field and local scour around spur dikes. Shuidong-
lixue Yanjiu yu Jinzhan Chinese J Hydrodyn Ser A 31:372–378
43. Kuhnle RA, Alonso CV, Shields FD (2002) Local scour associated with angled spur dikes. J Hydraul
Eng 128:1087–1093. https://​doi.​org/​10.​1061/​(ASCE)​0733-​9429(2002)​128:​12(1087)
44. Dey S, Bose SK, Sastry GLN (1995) Clear water scour at circular piers: a model. J Hydraul Eng
121:869–876. https://​doi.​org/​10.​1061/​(ASCE)​0733-​9429(1995)​121:​12(869)
45. Flow Science I (2016) Flow-3d User Manual: V11.2 ; Flow Science: Santa Fe, NM, USA
46. Guguloth S, Pandey M (2023) Accuracy evaluation of scour depth equations under the submerged ver-
tical jet. AQUA-Water Infrastructure, Ecosyst Soc 00:1–19. https://​doi.​org/​10.​2166/​aqua.​2023.​015

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under
a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted
manuscript version of this article is solely governed by the terms of such publishing agreement and applicable
law.

13