INITIALLY CALCULATED THE WAVE ATTENUATION

IN THE MANGROVE FOREST USING

THE DELFT – 3D MODEL
Nguyen Hoang Phong1

1
VNUHCM–University of Science

Email: nhphong@hcmus.edu.vn

Abstract: Delft3D model is used to calculate wave propagation in mangrove forests. The
research method is based on the changes of Collins bottom coefficient (1972) and the
Chezy roughness coefficient (Baptis, 2005). They depend on the characteristics of the
mangrove forest as density, height of tree, trunk diameter and hydrodynamic parameters.
This study was applied to estimate wave height dissipation in the first 500 m forest from
shore at Can Gio mangrove forest. The results show that the wave height reduction
depends on initial wave height, cross-shore distances, and mangrove forest structures. In
particular, with 1.0 m initial wave height, 1.0 tree/ m2 forest density and 1.0 m water
level, significant wave height is 0.92 m (8% reduction), 0.43 m (57% reduction), 0.29 m
(71% reduction), 0.22 m (78% reduction), 0.18 m (82% reduction) and 0.15 m (85%
reduction) away from the edge forest 10m, 100m, 200m, 300m, 400m and 500m
respectively.

Keywords: Delft3D, Collins bottom coefficient, Chezy roughness coefficient, mangrove

forest.

I. Introduction

Mangrove forest are widely distributed in the coastal areas of Vietnam which have the
effect of preventing erosion, protecting the coastal line, reducing the influence of waves,
preventing storms and tsunamis. This research uses open source Delft3D model to
calculate wave attenuation when entering Can Gio mangrove forest (Ho Chi Minh City).
Two Delft3D modules are used: Delft-3D Flow (hydrodynamic module) and Delft-3D
Wave (wave module). The research method is based on the calculation of Chezy
coefficient and Collins coefficient to suit the research area before being included in the
calculation model. These two coefficients depend on the characteristics of the mangrove
1
forest: density of mangrove trees, trunk height, trunk diameter, Von Karman coefficient,
roughness coefficient without mangroves, Cb, drag coefficient , Cd.

II. Research area and methods

II.1 Research area

Can Gio mangrove forest has a basin topography in the central area, the topography has
many variations in a small area but the difference is not large, most of the terrain is 0.0 -
1.5m high, except Giong Chua mountain is the highest point in the forest with an altitude
of 10.1m. The area has a network of sister services, intertwined. Government area
accounts for 31.76% of the total natural area of Can Gio district. Can Gio mangrove
forest is located in an irregular semi-tidal regime (twice high water and low water twice a
day). Tidal amplitude is about 2m during mean tide and 4m during high tide. The
maximum amplitude in mangroves is from 4.0 to 4.2m, the highest observed in Vietnam.
The time of maximum tidal amplitude is most frequent from September to January with
an amplitude of 3.6 - 4.1m in the southern region and from 2.8 - 3.3m in the northern part
of Can Gio (fig.1).

Figure 1. Research area.

II.2 Research methods

In this research, the influence of mangrove characteristics on wave attenuation is

included in the calculation model through two coefficients: Chezy roughness coefficient,
C and Collins bottom friction coefficient, Cf.
9
In the Delft-3D Flow module, the influence of mangroves is expressed through the Chezy
roughness coefficient (Baptist, 2005).

C=
1
+
(h )
√ g ln h

√ 1 C D mD h v κ v (1)
+
C b2 2g

With:

C b: roughness coefficient.

m: density of mangrove trees.

D : diameter of mangrove trees.

h : water depth.

h v: height of mangrove trees.

κ : Von Karman constant (~0.4).

C D : Coefficient of drag (0.9-1.0).

In the Delft3D - Wave module, the Collins (1972) coefficient is used to describe the
effects of mangroves. Wave energy dissipation, S:

σ2
Sdx ,b ( σ , θ )=−Cbottom E (σ , θ) (2)
g2 sinh kd

With:

S: wave energy dissipated,

ω: wave frequency ¿ ( 2Tπ ),

θ : wave direction

C bottom: coefficient of friction

k : number of waves

h : water depth

E : total energy

9
Collins has shown the coefficient of bottom friction, C bot with the wave velocity
trajectory, Uorb and the Collins coefficient, Cf as follows:

C bot =c f g U orb (3)

Substituting into the above formula, we get the energy dissipation of the S wave
expressed as follows:

−1 3
S= c ρU orb (4)
2 f

Formula (4) is used for normal conditions without mangroves. De Vries and Roelvink
(2004) show that the Collins coefficient of bottom friction, C f is equal to the mangrove
coefficient, Cv. When assessing the impact of mangroves, the coefficient Cv instead of Cf:

C v =f w Dn d Z (5)

With:

f w : bottom friction component

D : trunk diameter

n : number of trees per square meter

d Z : trunk height

The research uses formulas (1) and (5) to calculate the change of coefficients of Chezy
roughness coefficient and Collins bottom friction coefficient depending on the
characteristics of mangroves. The model ignores wind resistance, the influence of tree
roots on wave attenuation.

III. Results and Discussion

III.1 Calculation of Chezy coefficient and Collins coefficient for the research area

In the model, using formulas (1) and (5) to calculate Chezy coefficient and Collins
coefficient with the following values: Chezy roughness coefficient in the absence of
mangroves Cb=65, trunk height 3m, Von Karman coefficient 0.4, depth 1m, drag
coefficient Cd = 0.04 (Cd value is used uniformly for the entire mangrove area, trunk
diameter 0.1m and friction component there fw = 0.9 The calculation area is assumed to
9
be 1 m water level at the edge of the forest with a negligible slope (~1/600). Table 1
show the results of calculating the coefficients in the model with the change of Tree
density The greater the density, the larger the Collins coefficient and the opposite of the
Chezy coefficient.

Table 1: Changes of Chezy and Collins coefficients by density.

Tree density (trees/m2) Chezy coefficient (m1/2/s) Collins coefficient

0.5 34.33 0.135
1 25.73 0.27

1.5 20.83 0.405

III.2 Wave attenuation when entering the mangroves

The calculation area has a total length of 500 m, starting from the edge of the alluvial
ground (position 0m) extending into the forest (position 500 m), including: Alluvial
beach about 70 m long and 430 m long forest range. The first 92 m, extending from the
edge of the mudflat into the forest are topographic data measured by the Department of
Oceanography, Meteorology and Hydrology (Fig. 2).

9
 Figure 2. topography muddy flat of the calculated area.

From the position of 92m extending deep into the forest, to the position of 500m, the
terrain is assumed to be unchanged. As shown in Figure 3.5, at position 55m (position A)
the wave intrusment is placed on the mudflat; At position 92m (position B) set the wave
intrusment in the forest, the distance between them is 37m. The results from these 2 wave
intrusment are compared with the calculated results from the model.

The water level at the location of the wave intrusment in the forest changes depending on
the rise and fall of the tide, can reach the highest value of 0.8m. In this calculation, the
selected water level value is 0.7 m, if the water level is lower than 0.7 m, the significant
wave height at the location of the wave meter in the mudflat (position A) is quite small
(<0.25 m). Corresponding to a water level greater than 0.7 m, the significant wave height
at position A is in the range of 0.25m - 0.35m, and the significant wave height at the
estuary is in the range of 0.35 m - 0.5 m. Therefore, the boundary condition (at the edge
of the mudflat) uses a significant wave height value of 0.4 m. The model is used to
calculate for 2 cases: With regard to the effects of mangroves and without considering the
effects of mangroves. In addition, 8 real wave height data measured at positions A and B
satisfy the requirements: the water level at location B is greater than 0.7m and the wave
9
height at position A is greater than 0.3m are used to compare with calculation results.
(Fig. 3)

0.45

0.4

0.35
Độ cao sóng có nghĩa (m)

0.3

0.25

0.2

0.15

0.1

0.05

0
0 50 100 150 200 250 300 350 400 450 500
Khoảng cách so với vách rừng (m)

Figure 3: Wave attenuation in mangroves in the absence and presence of mangroves

in the calculated area.

In the first 70 m, high speed means in offset clearance in the same 2 cases. When entering
the forest, in the case of mangroves, high wave speed means a faster decline than in the
case of no mangroves. At the 500 m position, the mean high transmission rate is 0.16 m
and 0.23 m respectively for the case of mangroves and no mangroves. This shows the
obvious wave energy-reducing effect of mangroves.

In the case of mangroves, the high wave speed means that at position 50 m (close to
position A in the clearing) and 90 m (close to position B in the forest) have a value of
0.34 respectively. m and 0.29 m, high wave speed means a decrease of about 14.7%.
Actual data measured at 2 locations A and B, mean wave height decreased from 13% -
33%, averaged about 17.9%.

IV. Conclusion

The Delft-3D model can be used to calculate wave attenuation in mangroves through the
Collins coefficient and the Chezzy coefficient.

The rate of wave height decline in mangroves is directly proportional to the initial wave
height, tree density in mangroves, and inversely proportional to the water level.

9
Calculation results in Can Gio mangrove forest shows a significant reduction in wave
height of mangroves. The difference between the calculated results and the actual
measurement results is not too big

Limitations of the research:

- The characteristics of mangroves are collected in a limited way, not fully showing the
specific characteristics of mangroves.

- Do not consider the influence of mangrove roots, mangrove species.

Future development direction:

- The characteristics of mangroves need to be further studied to find the wave resistance
coefficient suitable for each different tree species in the forest.

- Research using computational Delft3D models for some other mangrove areas in
Vietnam.

References

[1]. Blackmar, P. J. (2013). Experimantal and Numerical Modeling of Wave Height

Attenuation by Emergent Coastal Vegetation, Oregon State University, USA.

[2]. Burger, B., 2005, Wave attenuation in mangrove forests. Numerical modelling
of wave attenuation by implementation of a physical description of vegetation in SWAN,
MSc thesis, Delft University of technology, Netherlands.

[3]. Dekker, F. (2006). Hydrodynamics and Morphodynamics in and around Mangrove

Forests, University of Twente, Netherlands.

[4]. Hendriks, J. M. (2014). Wave Dissipation in Mangroves, University of Twente,

Netherlands.

[5]. Le, T. L. (2008). Effect of hydrology on the structure and function of mangrove
ecosystems in the Can Gio Mangrove Biosphere Reserve, Vietnam, Louisiana State
University, USA.

[6]. Massel S. R., Furukawa K., Brinkman R. M. (1999). Surface wave propagation in
mangrove forests, Fluid Dyn. Res., 24 (4), 219–249.
9
[7]. Mazda, Y., Magi, M., Ikeda, Y., Kurokawa, T., Asano, T., 2006. Wave reduc- tion in
a mangrove forest dominated by Sonneratia sp, Wetlands Ecology and Management, 14,
365-378.

[8]. Mullarney, J. C., Henderson, S. M., Heyns, J. A. H., Norris, B. K., Bryan, K. R.,
(2017), Spatially varying drag within a wave-exposed mangrove forest and on the
adjacent tidal flat, Continental Shelf Research, 147, 102 – 113.

[9]. Tran, Q. B., (2011), Effect of mangrove forest structures on wave attenuation in
coastal Vietnam, Institute of Oceanology, 53, 3, pp. 807 – 818.

[10]. Vo Luong, H.P. (2006). Surface waves propagation in mangrove forest and induced
suspended sediment concentration, Polish Academy of Sciences, Poland.

[11]. Vo Luong, H.P., (2008), Energy dissipation in non-uniform mangrove forests of

arbitrary depth, Journal of Marine Systems, 74, 603 – 622.

[12]. Vo Luong, H.P., (2013), Preliminary research on turbulence in mangrove forests,

Tuyển tập Nghiên cứu biển, 19, 51 – 60.

[13]. Vo Luong, H.P., Massel, S.R. (2006). Experiments on wave motion and suspended
sediment concentration at Nang Hai, Can Gio mangrove forest, Southern Vietnam,
Institute of Oceanologia PSA, Vol 48, 1, pp. 23-40.