CONCENTRATION OF SUSPENDED SEDIMENTS IN MANGROVE FORESTS

1VÕ LƯƠNG HỒNG PHƯỚC,  2STANISLAW MASSEL,
1ĐẶNG TRƯỜNG AN, 1NGUYỄN CÔNG THÀNH

1University of Natural Sciences, Việt Nam National University, Hồ Chí Minh City,
227 Nguyễn Văn Cừ, Distr. 5, HCMC. Email: vlhphuoc@phys.hcmuns.edu.vn;
2 Institute of Oceanology, Polish Academy of Sciences,
Powstancow Warsarzawy 55, PL-81-712 Sopot, Poland

Abstract: Hydrodynamic study in mangrove forests shows that rainfall, riverine flows, tidal currents and waves are the main causes of suspended sediment transport. In this study, a 1-D model of vertical profile of sediment suspension in mangrove forest is considered, in which the role of tidal currents and waves are taken into account. Some formulae of suspended settling velocity and diffusion coefficient in mangrove forest are given. Based on measured data of suspended sediment concentration (SSC) in 2004 and 2005 at Nàng Hai area, Cần Giờ Mangrove Biosphere Reserve, Hồ Chí Minh City, Việt Nam, some parameters of suspended sediment matters, such as suspended sediment settling velocity, manning friction coefficient and diffusion coefficient can be estimated and their values depend strongly on wave action and tidal currents. From the model, the vertical distribution of suspended sediments can be determined and the total SSC during a tidal cycle can be calculated. The calculated results of SSCs during one tidal cycle at station ST1 before the mangrove forest and at station ST3 in mangrove forest were considered under three different cases of weather conditions: influence of tidal currents, under high waves and influence of tropical low pressure. As results of the study, the erosion / accumulation probability at a given area can be estimated. It is believed that wave action is one of the main factors inducing sediment transport and erosion processes at Nàng Hai, even the wave field in the Đồng Tranh estuary is not so strong. The results also prove that suspended sediment concentrations depend on wave intensity and tidal current velocity and that mangrove vegetation can encourage the deposition of sediment and protect the coastland from high waves and storms.


I. INTRODUCTION

Mangrove forests are densely vegetated mudflats that exist at the boundary of marine and terrestrial environments. Erosion processes on cohesive shores are distinctly different from those on sandy shores. The eroded fine sediments (silt and clay) are winnowed, carried offshore, and deposited in deep water in contrast to the sand fraction, which usually remains in the littoral zone. Furthermore, cohesive sediments are not transported as bed load, except in the form of fluid mud. They almost always are transported in suspension form. Therefore in this study, we focus on the suspended sediment concentration, rather than on the sediment transport.

Hydrodynamic studies in mangrove forests show that rainfall, riverine flows, tidal currents and waves are the main causes of suspended sediment transport [1-3, 5, 9, 12]. In general, the suspended sediment concentration (SSC) near the bottom is much higher than SSC on the surface due to higher turbulence at the bottom. Therefore, a 1-D model of vertical profile of sediment suspension is considered, in which the role of tidal currents and waves are taken into account. The measured data of SSC in 2004 and 2005 at Nàng Hai area, Cần Giờ Mangrove Biosphere Reserve, Hồ Chí Minh City were analysed and some parameters of suspended sediment matters can be estimated. From the model and SSC data, the calculations of total SSC in one tidal cycle under tidal current, strong waves and low pressure are presented. As a result, the erosion / accumulation probability at a given area can be estimated.

II. SUSPENDED SEDIMENT PROFILES AND LOAD TRANSPORT IN MANGROVE FORESTS

In order to determine the SSC profile when it is not homologous from the bottom to the surface, we use the simple 1-D model of vertical suspended sediment profile [4, 6]. Within a water column of depth h, the vertical sediment transport is governed by upward mass diffusion due to turbulence and particulate settling. The change of concentration C with time at any elevation, z (measured positively upward from the mean water level), is determined by the magnitude and direction of the net sediment flux due to diffusion and settling. The vertical settling-diffusion equation can be expressed as a particular case of the general mass conservative equation:

                  (1)

The equation (1) can be solved for some initial condition, C(z,0), and surface (z = 0) and bottom boundary (z = -h) conditions, provided Ws and Kz are specified as shown in the next section.

The boundary condition at the water surface is that the net flux of sediment at z=0 is nil, i.e.

          (2)

and at the water-fluid mud interface (z = -h) the net flux of sediment is determined by sediment entrainment, E, and deposition, S, i.e.

          (3)

Specification of the fluxes, E and S, is crucial to an accurate simulation of the suspended sediment profile.

In the governing equation (1), the unknown concentration, C, is a function of the independent variable z and time t. The simplest approach is to represent function C as a product of functions Cz(z) and T(t), each depending on variable x and t, respectively, i.e.

                            (4)

After some calculations, we can get an expression for vertical concentration Cz(z) in the following form:

(5)

Equation (5) is the linear second-order differential equation with the boundary conditions , in which Cs and Cb are the concentrations at the surface and at the bottom. The differential equation (5) can be approximated by a finite difference equation and becomes a system of matrix equations. To solve this equation, the Cholesky’s method for a tridiagonal band type matrix is well suited.

III. SETTLING VELOCITY OF SUSPENDED SEDIMENTS AND DIFFUSION COEFFICIENTS DUE TO WAVES AND CURRENTS

1. Settling velocity

In mangrove forests, the sediment particles carried in suspension during tidal inundation are cohesive, mainly clay and fine silt, and form large flocs [2]. The observed exponential decrease in sedimentation rate with distance from the creek enables the estimation of the settling velocity Ws of the suspended sediment in mangroves [3]. Assuming zero re-suspension, the settling velocity Ws can be calculated from the equation for conservation of sediment mass as follows:

                                     (6)

Where C is the suspended-sediment concentration, x is the distance across the mangrove forrest, h is the water depth and U is the current velocity.

2. Diffusion coefficients due to waves and currents

Various researchers have tried to model the suspension process by introducing an effective diffusion coefficient according to specific scenarios such as: suspended sediment induced by currents for steady flow or for non-steady (tidal) flow, suspended sediment induced by non-breaking waves, breaking waves or ripples, and suspended sediment induced by wave and current combination.

Most expressions for the diffusion coefficient are empirical or semi-empirical. The corresponding expressions are presented in more detail in Van Rijn [7, 8]. In this study, for modelling of the suspended sediment concentration, the following diffusion coefficient is used [4, 6]:

                                            (7)

Where  is the Monin-Obukov correction for stratification; K0  is the diffusivity under non-stratified or neutral flow conditions.       

For simulating the influence of stratification under wave motion and for sediment-induced stratification under tidal current,  takes the form:

                                 (8)

In which  and  are coefficients depending on the effect of suspended sediment on the turbulent mixing length; Ri(z) is the gradient Richardson number:

                                       (9)

Where ρ is the fluid density; g - gravitational acceleration; and u(z) - horizontal velocity.

To simulate enhanced diffusion for the combined effect of waves and current, Li and Parchure [4] introduced the following linear form of the diffusion coefficients:

                        (10)

In which the wave and current diffusions Kow and Koc respectively must be specified, as well as the corresponding weighting coefficients αw and αc. In particular, the diffusion coefficient for wave motion only is:

     (11)

In which h is water depth from the surface; z - vertical coordinate, directed upward; ω - angular wave frequency; A - wave amplitude; k - wave number; and α0w - diffusion scaling coefficient, which for a given sediment depends on the flow field.

On the other hand, the diffusion coefficient for the current-induced boundary layer is given by:

    (12)

Where κ is the von Karman constant = 0.4;  - mean current velocity; and n - Manning friction coefficient. The Manning friction coefficient n is an important parameter and is the focus of extensive studies on rivers and channels. This coefficient can be expressed in MKS units, as follows [2, 13]:

                                   (13)

Where I is the gradient of water surface; and h - water depth.

When sediment concentration profiles are known, the mixing coefficient can be computed by a simple relationship [7]:

                            (14)

However, this equation cannot be applied when concentrations in the water layer are well mixed due to high turbulence, especially for strong wave action, in very shallow water in mangroves.

IV. SSC DATA AND MODEL APPLICATION IN CẦN GIỜ MANGROVE FOREST

1. Location of the studied site

Nàng Hai area is chosen as the studied site. It is about 2.0 km far from the estuary of the Đồng Tranh river, which with a length of 67.50 km is the longest of the main rivers in Cần Giờ Mangrove Biosphere Reserve. The Đồng Tranh area is less affected by strong wind-induced waves as it is sheltered by Capes Lý Nhơn and Long Hòa, either side of the estuary. The selected chosen site has a more direct influence on wave motion and the instrumentation transect (10023.427 N; 106052.761 E – 10023.442 N, 106052.793 E) was chosen parallel to the direction of wave propagation towards the forest. Nàng Hai is a complicated site with many surrounding creeks and mixed-type mangroves. Especially, the topography of this site changes remarkably between muddy flat and mangrove area and being eroded seriously.

Two field trips in April-May 2004 and January-February 2005 were made at Nàng Hai area. The aim of the field measurements was to study the influence of wave motion in mangrove forests and induced sediment concentrations. Data collection methodology and analysis results were described by [10, 11].

2. Parameters of suspended sediments

From the SSC analysis and calculations, the final parameters of the suspended sediments at the Nàng Hai site can be estimated as follows:

- Suspended sediment settling velocity W: - For tidal currents and weak waves, Ws ≈2.0 - 4.5×10-4 m/s; - For strong waves, Ws 2.0 x10-3 m/s.

-  Manning friction coefficient n: - The mean values, n ≈ 0.016 - 0.021 in MKS units; - For highest spring tides n ≈ 0.030 in MKS units.

- Diffusion coefficient Kz: The mean value, Kz ≈ 2.46×10-4 m2/s; - For tidal currents, Kz ≈ 3.35×10-4 m2/s; - For strong waves, Kz ≈ 1.18×10-3 m2/s.

It is noticeable that these parameters are calculated to estimate the general characteristics of the Nàng Hai studied site and to be applied in the 1D model of vertical suspended sediment profile.

3. Calculated results of vertical suspended sediment concentrations

When the SSC is not homogeneous, the 1-D model of vertical suspended sediment transport can be applied to determine the SSC profile. When the suspended sediment profile C(z) at a given point is determined, the total SSC Ctol during a tidal cycle can be calculated. The calculated results of SSCs during one tidal cycle were considered under three different cases of weather conditions: influence of tidal current, under high waves and influence of tropical low pressure.

The field measurement of SSC on 19-20 April 2004 was carried out in calm weather with weak wind and calm sea. Figs. 1a, b show the changes of SSC at stations ST1 and ST3, respectively. The station ST1 is deployed in the muddy flat 20 m before the mangrove front and the station ST3 is 45 m within the mangroves. It is obvious that the changes of SSCs depend mainly on tidal level; the SSCs are higher during the ebb and the flow tide, but decrease at the slack tide when the tidal current is almost zero. Fig. 1a shows SSC at ST1 at three different levels, namely: near-bottom, middle and top levels. It is obvious that SSC near the bottom is much higher and fluctuates more than SSC in the middle and near-surface levels. Inside the mangroves at ST3, the concentration changes also due to tidal level and has lower values than at the muddy flat station ST1, namely 50 mg/l at slack tide and about 100 mg/l at flow and ebb (Fig. 1b). The lower concentration at ST3 in the mangroves suggests that mangrove vegetation can encourage the retention of sediment.

It is known that flow currents flow toward the coast during flow phase and ebb currents go away from the coast. Therefore, the concentrations can be considered as positive for towards mangrove flow and as values for seaward flow.

The total concentration is the difference between concentrations for flow tide and for ebb tide. From the calculation, it can be seen that the concentrations at ebb tide Cebb are higher than those at flow tide Cflow at two stations. Especially at ST1, concentration Cebb is much higher than concentration Cflow. As a result, the total concentrations Ctol at theses stations have negative values. The negative value of SSC suggests that more sediment tends to move seaward. This can encourage erosion at these locations, particularly of the mangrove front at ST1.

 


Figure 1. Suspended sediment concentration at ST1 (a) and ST3 (b) on 19-20 April, 2004.


Fig. 2 shows suspended sediment concentration at 2 different stations during the high wave influences on 06-09 May, 2004. Obviously, suspended sediments at both stations also fluctuate due to tidal variations with higher concentrations due to wave impact. At ST1, the SSCs at 1.5 m above the mudbed reach the value of about 200 mg/l, while the SSCs at 0.2 m above the mud bottom reach the value of about 400 mg/l  at lack time, about 600 mg/l at flow tide and 700 mg/l at ebb tide. The calculated total concentration in one tidal day at ST1 Ctol is about -20,000 mg/l. The sediment at ST1 exhibits some loss. The negative value shows that SSC before the mangrove front ST1 also is directed seaward during strong wave action and the station ST1 represents the erosion place. It can be explained that waves,  especially breaking waves, can “pump up” the suspended matters from the mudbed


 

Figure 2. Suspended sediment concentration at ST1: a) and ST3; and b) on 06-09 May, 2004


 

 

  

Figure 3. Suspended sediment concentration at ST1 (a) and  ST3 (b) in 16-18 May, 2004.


and cause higher turbulence; however, the suspended sediment movement along the transect depends not only on wave action but on tidal currents as well. In contrast to ST1, at station ST3 in the mangroves, the SSC at flow tide was higher than SSC at ebb tide: namely about 400 mg/l for flow tide and about 300 mg/l and 200 mg/l at ebb and slack tide. The resulting total concentration Ctol in one tidal day is about 20 000 mg/l with positive direction. Sediment at ST3 was in the accumulation stage. About half of suspended sediment was deposited in the mangroves. This means that mangroves can “receive” sediment from the sea as flow is incoming. Mangroves do not want to “return all the sediment” when the water is flowing to the sea. These results prove the significant role of mangroves in sediment retention and coastal protection from wave impact.

On 16-18 May, 2004, there was a tropical low pressure in the south of the East Sea, along with heavy rain. On 17 May, 2004, the low pressure system was nearest to the coastlines of Vũng Tàu - Cần Giờ. As a result, it can be easily seen that measured SSC had highest values from 13h16 May, 2004 to 13h17 May, 2004 as shown in Figs. 3a, b. Fig. 3a shows that the fluctuation of SSCs at the ST1 was quite different to the  previous  cases. As the tropical low pressure dominated, SSC at the level 0.85 m above the mud bed did not clearly change due to tidal level except at flow. Differently from the previous cases at ST1, the calculated concentration was much higher at flow tide than that at ebb tide and the resulting total concentration had a positive value, more than 50,000 mg/l. This fact suggests that a considerable amount of sediment was brought to the mangroves from the open sea due to the tropical low pressure season.

At station ST3, the chaotic fluctuation of SSC due to rain during the low pressure did not exhibit so clearly as SSC fluctuation at ST1. During the low pressure, SSC was much higher but showed variations due to tides. The concentration at flow tide was equal to about 78000 mg/l, while SSC at ebb flow was about 26800 mg/l. Mangroves kept more than 65 % of suspended sediment coming from the sea during the low pressure weather. This confirms the conclusion that mangrove forests are effective barriers to protect coastlines from waves and storms [11]. 

V. CONCLUSIONS

The experiments of suspended sediment concentrations clearly prove that suspended sediment concentrations depend on wave intensity and tidal current velocity. Wave action can make a significant contribution to the  increase in SSC in  mangroves and to  accelerate the movement of suspended sediments. From the SSC analysis and calculations, it proves that mangrove vegetation can encourage the deposition of sediment and protect the coastland from high waves and storms. However, the sediment flux before the mangrove front is directed seaward due to tidal currents and especially due to wave action. Therefore, it is strongly believed that wave action is one of the main factors inducing sediment transport and erosion processes at Nàng Hai site. Moreover, the high gradients of sediment flux due to waves erode mud from around the root system, as well as destroy the individual trees. These reasons can contribute to the erosion at Nàng Hai edge [11].

The parameters of suspended sediment settling velocity, manning friction coefficient and diffusion coefficient are calculated to estimate the general characteristics of the Nàng Hai study site and to be applied in the 1D model of vertical suspended sediment profile. However, to achieve higher accuracy, more experimental data and more calibrations at the field site are required.

Acknowledgements

This study is supported partially by the project No. 322 (Code 171/2001/2/TS) of the Ministry of Education and Training (MOET), Việt Nam and by the Centre of Excellence for Shelf Seas Science (CeSSS) at the Institute of Oceanology, Poland. Sincere thanks are due to the SEDYMAN project, which provided some logistical support. The authors would like to express their gratitude to Prof. La T. Cang, Drs. K. Schwarzer and K. Ricklefs for invaluable advice and assistance with field experiments.

REFERENCES

1. Bryce S., Larcombe P., Ridd P.V., 2003. Hydrodynamic and geomorphological controls on suspended sediment transport in mangrove creek systems, a case study: Cocoa creek, Townsville, Australia. Estuarine, Coastal and Shelf Science, 56, 515-431.

2. Furukawa K., Wolanski E., 1996. Sedimentation in mangrove forests.  Mangroves and Salt Marshes, 1/1 : 3-10.

3. Furukawa K., Wolanski E., Mueller H., 1997. Currents and sediment transport in mangrove forests. Estuarine, Coastal and Shelf Science, 44 : 301-310.

4. Li Y., Parchure T.M., 1998. Mudbanks of the Southwest coast of India. VI: Suspended sediment profiles. J.of Coastal Res., 14/4 : 1363-1372.

5. Mazda Y., Magi M., Nanao H., Kogo M., Miyagi T., Kanazawa N., Kobashi D., 2002. Coastal erosion due to long-term human impact on mangrove forests. Wetlands Ecol. and Man., 10 : 1-9.

6. Mehta A. and Li Y., 2003. Principles and process: Modelling of cohesive sediment transport. Univ.of Florida.

7. Van Rijn L.C., 1989. Sediment transport by currents and waves. Delft Hydraulics, I, II .

8. Van Rijn L.C., 2005. Principles of sedimentation and erosion engineering in rivers, estuaries and coastal seas. Aqua Publ.

9. Victor S., Neth L., Golbuu Y., Wolanski E., Richmond R.H., 2006. Sedimentation in mangroves and coral reefs in a wet tropical island, Pohnei, Micronesia. Estuarine, Coastal and Shelf Sci., 66 : 409-416.

10. Vo Luong Hong Phuoc, 2006.  Surface waves propagation in mangrove forest and induced suspended sediment concentration. PhD Thesis. Inst. of Oceanology. Sopot, Poland.

11. Vo Luong H.P., Massel S.R., 2006. Experiments on wave motion and suspended sediment concentration at Nang Hai, Cần Giờ mangrove forest, Southern Vietnam. Oceanologia, 48/1 : 23-40.

12. Wolanski E., Spagnol S., 2000. Environmental degradation by mud in tropical estuaries. Reg. Envir. Change, 1 : 152-162.

13. Wolanski E., Mazda Y., Ridd P., 1992. Mangrove hydrodynamics. In Robertson, A.I., Alongi, D.M. (Eds). Tropical mangrove ecosystems. Am. Geoph. Union, pp. 43-62. Washington.