international journal of civil and structural … · countries for more than a century as...

INTERNATIONAL JOURNAL OF CIVIL AND STRUCTURAL ENGINEERING Volume 1, No 4, 2011

© Copyright 2010 All rights reserved Integrated Publishing services Research article ISSN 0976 – 4399

852

Controlling of overtopping flow of embankment with vegetative barrier: A Flume study

H. M. Rasel 1 ,Md. Abu Sayeed 2 , Md. Aftabur Rahman 3 , M. A. Sobhan 2 1 Master’s Student, Saitama University, Japan and Member, Faculty of Civil Engineering,

Rajshahi University of Engineering and Technology (RUET), Rajshahi6204, Bangladesh.

2 Member, Faculty of Civil Engineering, Rajshahi University of Engineering and Technology (RUET), Rajshahi6204, Bangladesh.

3 Master’s Student, Saitama University, Japan and Member, Faculty of Engineering, Chittagong University of Engineering and Technology (CUET), Chittagong4349,

Bangladesh. [email protected]

ABSTRACT

For earth coastal dikes or river embankments, overtopping is one of the most damaging factors for the downstream (d/s) slope. Very little is known about the stability of d/s slope covered with grass during overflow. For an understanding the processes and properties of the overflow on the d/s slope of the embankment with vegetal cover, a smallscale laboratory flume experiment with an embankment using emerged vegetal cover (polypropylene Hechimaron with 80% porosity) on the d/s slope have been performed with three different unit discharges (0.018, 0.013 and 0.010 m 3 /s). The results of the smallscale flume experiment indicate that the vegetation markedly reduces the overtopping velocity (6070%) and effective shear stress (3175%) at the surface of the d/s slope due to water compared with the case without vegetation. This demonstrates that a significant reduction of damage from overtopping flows can be achieved by providing vegetation cover on the d/s slope of the embankment.

Keywords: Vetiver grass, hechimaron, vegetation, overtopping, shear stress, drag force, erosion.

1. Introduction

Bangladesh, with its repeated cycle of floods, cyclones, and storm surges, has proved to be one of the most disasterprone areas of the world. River bank and coastal erosion, and embankment failures happen continuously throughout Bangladesh. These are endemic and recurrent natural hazards, causing loss of lands and livelihoods along major rivers and coastlines. From a strictly economic point of view, the cost for mitigating these problems is high. In addition, the State budget for such works is never sufficient which confines rigid structural protection measures to the most acute sections, never to the full length of the river bank or coastline and embankment. This bandage approach compounds the problem. Not only that, hard engineering structure makes the scenic environment unpleasant and helps only to transfer the problem to another place, to the opposite site or downstream, which aggravates the problem rather than reducing (Grimshaw, 2008). Establishment of vegetation as a soft bioengineering technique to rigid or hard structures accepted all over the world due to its low cost and longevity.



853

Bangladesh Water Development Board (BWDB) has been constructed nearly 13000 km of embankments (over 4000 km of coastal embankments along the coastline surrounding the Bay of Bengal and offshore islands, nearly 4600 km of embankments along the bank of big rivers and nearly 4500 km of lowlying embankments along the small rivers, haors and canals). A large number of sea dike or river embankment failures have been initiated from damages induced by wave or flow overtopping (Islam, 2000). To minimize the impact of natural disasters as well as to achieve the aim of agricultural production, sustainable and cost effective maintenance of those embankments is a sine qua non for Bangladesh. For earth coastal dikes or river embankments, overtopping is one of the most damaging factors for the downstream slope. Eventually a failure of the downstream slope may lead a failure of the dike. Water infiltrates into dike crest, downstream slope and reduces the shear resistance of the soil (Hanson and Temple, 2001). Vegetation planted in the downstream slope may have a considerable effect in reducing overtopping erosion. Vegetative barriers increase the surface roughness and slow the flow rate. It dissipates energy and protects the slope from erosion.

An ecofriendly vetiver grass (Vetiveria zizanioides) has been used successfully over 120 countries for more than a century as traditional technology for riverbank stabilization and embankment erosion control (Truong and Loch, 2004). Recent research and case studies have shown that vetiver grass can be used as a cheap method to protect shorelines or embankments. Mature vetiver is extremely resistant to washouts from high velocity flow due to its extraordinary root depth and strength. The stiff shoots and strong roots can keep the plant stand steadily in water with 0.60.8 m deep and 3.5 m/s velocity of water flow (Ke et al., 2003). On this subject, Bangladesh is in advantageous position as it has been abundant supply of vetiver grass. But vetiver grass technology (VGT) is very new in Bangladesh and using vetiver as a living barrier against erosion remains untouched excepting as few small scale trials in some foreign aided projects. Moreover, its application has been based on experience rather than hydraulic principles (Das and Tanaka, 2009). Basically, the upstream side of earthen sea dikes or embankments are armored with stones or concrete blocks where there any exist of important infrastructures while the downstream slope often covered with grass. Very little is known about the strength and stability of downstream slope of sea dikes or river embankments covered with grass during overflow. Although it is difficult to scale down the properties of natural resources like vetiver grass, the main objective of this study is to introduce the use of vegetation to protect the downstream slope of embankments or sea dikes during overflow and act as a solution of embankments/sea dikes stabilization and to encourage research and development with vetiver grass in Bangladesh.

2. Materials and Method

2.1 Experimental setup and procedure

For an understanding the processes and properties of the overflow on the downstream slope of the sea dike or embankment with the vetiver hedges, physical modeling is indispensable. Use of numerical model in this stage would be too much influenced by the assumptions on the governing processes between vetiver grasses and overflow. A small scale physical model tests were carried out to protect the downstream (d/s) slope of embankments or sea dikes during overflow. Tests were conducted in a water recirculating flume of hydraulic and environmental engineering laboratory at Saitama University in Japan. The flume is 0.50 m in depth, 0.70 m in width, and 5.0 m in length with head box (intake) and tail box as shown in



854

(a)

(b)

figure 1 below. The intake is 1.0 m long and flow straighteners generate undisturbed inflow. Model–scale wooden embankment was constructed and placed at the middle of the flume which separated the flume along its length, forming main stream on one side (called upstream) and the floodplain on the other side (called downstream). The size of the embankment is 0.25 m in height, 0.25 m in crest width and 1.5 m in length, with slopes 3H:1V and 2H:1V in the upstream and downstream, respectively (fig.1).

The discharges were measured with an electromagnetic flow meter (model: MK 515/ 8510 XX, paddle flow sensor, Georg Fischer Signet LLC, USA). An electromagnetic velocity meter (type of main amplifier: VM2000; type of sensor: VMT2–200 04P, Kenek Company, Ltd.) measured the flow velocities at the centerline of the channel and the model. The water surface elevation was measured at the same locations as the velocity profiles by the point gauges (with accuracy up to 0.1 mm), fixed and mounted on a movable sliding carriage.

Vetiver when planted close together and properly managed it forms rigid hedge and provides outstanding protection against erosion. Keeping this in mind the experiments were conducted using vegetation model (commercially available polypropylene Hechimaron with 80% porosity) fixed on the downstream slope of the wooden model. The height of the vegetation model was kept constant, 0.10 m and width and length was kept same as the d/s side of the embankment model as shown in fig. 2 considering emergent flow conditions. The steady flow with three different unit discharges as 0.018, 0.013 and 0.010 m 3 /s (later we called it discharge 1, discharge 2 and discharge 3) was taken in our experiments. The longitudinal flow velocities and water levels were measured along the centre line of the flume at 27 points with an interval of 0.10 m up to upstream (u/s) crest of the model and later 0.05 m interval from u/s crest to the d/s end of the flume until steady flow condition was established. The scale factor (ratio of a variable in the model to the corresponding variable in its prototype) 0.0625 was kept constant throughout the tests. In the tests the velocities and the water depths were measured for situations with and without vegetation in the flume.

Figure 1: Profile of testing facility (a) front view and (b) top view



855

Figure 2: Experimental flume with vegetation model

2.2 Drag force measurement

A twoaxis load cell (streamwise (X) and transverse (Y) directions, type LB60, SSK Co., Ltd.) with a resolution of 1/1000 that can measure a maximum load of 10.0 N was used to measure the drag force on the vegetative roughness model. The measurement technique was almost the same as the method in the earlier study (Takemura and Tanaka, 2007).

3. Results and discussions

3.1 Analysis of flow parameter

The resistance to flow is a strong function of the flow parameters with vegetation (Baptist et al., 2007). The flow parameters like overtopping depth, overtopping velocity, Froude number (Fr), Reynolds number (Re) etc. (Table 1) were measured with three different discharges as mentioned in the earlier section. The threshold values of overtopping height, velocity and discharges were checked by Froude law of similitude between model and prototype for comparison and the results found within the permissible limit based on the previous researches and case studies (Powledge et al., 1989a&b, Nadal and Hughes, 2009).

Table 1: Hydraulic Parameters of the experiment (WOV: without vegetation and WV: with vegetation).

Test 1 Test 2 Test 3 WOV WV WOV WV WOV WV

Discharge (m3/s/m) 0.018 0.013 0.010 Average overtopping depth (cm) 3.27 8.14 2.83 6.95 2.41 5.46

Average overtopping velocity(cm/s) 55.77 28.30 50.50 26.62 48.05 24.93

Froude No along the slope 2.424.90 0.370.57 2.444.45 0.390.55 2.084.48 0.420.58

Reynolds No along the slope

17000 18000

27000 39500

13000 14000

18600 30500

10000 11000

15200 22400



856

The hydraulic characteristics of overtopping flow in without vegetation condition shows that the flow was static at the beginning upstream, accelerating subcritical flow state over a portion of the embankment crest; through critical flow on the crest and supercritical flow over the remainder of the crest; and supercritical flow on the downstream slope and extend to the further downward same as observed in the previous study (Powledge et al., 1989b). On the other hand, the hydraulic characteristics of overtopping flow in WV condition shows the flow was static at the beginning of upstream, accelerating subcritical flow state both on the embankment crest and the downstream slope. Froude number varies from 2.0 to 5.0 along the slope in the case of without vegetation (WOV) and 0.4 to 0.6 in the case of with vegetation (WV). The flow was turbulence in nature for both of the cases but it was more than 2 times higher in case of vegetative model than without vegetation due to combined effect of steep slope and dense vegetation cover (Table 1, Fig. 3).

Figure 4 and Figure 5 show the variation of water depth and velocity along the flume with different discharges, respectively. In the case of WOV water depth decreased along the downstream slope and further slightly increased from the toe to the downstream of the flume which consequently increased the velocity as water runs down the slope for all discharges. But this trend was not consistent and variation of water depth and velocity found more complex especially inside the vegetation in the case of WV. This might be the complex interactions between vegetation and flow depth. Moreover, we used commercially available polypropylene Hechimaron as a vegetation model which are not uniformly dense. Another point is that we made some hole at the centerline of the vegetation model to measure water depth and velocity inside the model. This might be the cause of disturbance of the homogeneity of the model (porosity) and affected the results.

Figure 3: Hydraulic characteristics of overtopping flow (a) without vegetation (b) with vegetation



857

0.0 2.0 4.0 6.0 8.0 10.0

265.0 285.0 305.0 325.0 345.0 365.0 Distance (cm)

Water depth (cm)

Q1=0.018_WV Q2=0.013_WV Q3=0.010_WV Q1=0.018_WOV Q2=0.013_WOV Q3=0.010_WOV

Vegetation may protect embankment in many ways but the most important one is the reduction of velocity of water at the soil surface to a value that required causing erosion. The fig. 6 shows the percentage variation of average velocity and water depth along the embankment slope with different discharges. Vegetation reduced the velocity of water significantly (68 to 70% for three discharges) although water depth increased more higher than velocity reduction (more than 400%) due to blocking and resistance by vegetation.

Figure 4: Variation of water depth along the flume with different discharges

Figure 5: Variation of velocity along the flume with different discharges

0.00 0.40 0.80 1.20 1.60 2.00

2.65 2.85 3.05 3.25 3.45 3.65 Distance (m)

Velocity (m

/s)

Q1=0.018_WV Q2=0.013_WV Q3=0.010_WV Q1=0.018_WOV Q2=0.013_WOV Q3=0.010_WOV



858

68

69

69

70

70

71

0.0090 0.0140 0.0190

Discharge (m 3 /s/m)

% decrease (velocity)

410 420 430 440 450 460 470

% increase (w

ater depth)

Velocity

Water depth

3.2 Calculation of effective shear stress

During the experiments the steady uniform overflow condition was established at the upstream side of the model and during overtopping flow became unsteady and nonuniform. But for simplicity we consider steady nonuniform flow for calculation of effective shear stress. For a control volume of unit area along the slope the balance of horizontal momentum in this case can be expressed as total shear stress which is equal to the sum of the bed shear stress and the equivalent shear stress due to the vegetation drag:

= + w b v τ τ τ (1) where, τb= bed shear stress transfer to the soil; τw= stream wise component of the weight of water mass and τv=resistance due to the drag around the vegetation. The drag force was measured directly by a twoaxis load cell apparatus as mention above. The streamwise weight component of the weight of water mass per unit bed area is expressed as,

( ) = e w ρghi λ τ 1 (2) where, λ = area concentration due to vegetation.

Based on the Eq.(2) the effective shear stress which directly transfers to the soil was calculated by deducting the drag forces induced by the vegetation for different discharges as shown in Table 2 and Fig. 7. The total shear stress with vegetation found significantly high compared with the case without vegetation due to high water depth inside the vegetation but a significant portion of the total shear stress acted on the vegetation. The remaining portion i.e. the effective shear stress which directly transfers to the soil was well below the shear stress developed in the case of WOV because of the induced drag force due to vegetation. Vegetation markedly reduced the effective shear stress (75%) especially when the discharge was low (0.010 m 3 /s/m) but in the case of high discharge (0.018 m 3 /s/m) the reduction was only 31%.

Figure 6: Percentage variation of velocity and water depth along the embankment slope with different discharges



859

Table 2: Drag forces and effective shear stresses with different discharges

Water level near Up stream flume

(m)

Stream wise component of water

mass (N/m 2 )

SL No.

Discharge (m 3 /s/m)

WOV WV WOV WV

Drag force by vegetation (N/m 2 )

Effective shear stress transfer to soil (N/m 2 )

1 0.018 0.305 0.333 33.08 260.74 237.94 22.80

2 0.013 0.298 0.323 32.77 216.31 207.57 8.74

3 0.010 0.291 0.306 27.64 162.09 155.25 6.84

The above calculation was done based on same discharges as in the cases of WOV and WV. But in this smallscale model test (flume length is only 5.0 m) model was placed at the center of the flume and backwater effect was high at the up stream point of the flume (Table 2) especially when the discharge was high. The might be influenced the results of effective shear stress calculations. In reality this backwater effect may be not so high. Fig. 8 shows the comparisons of effective shear stress considering same discharges and same water levels in the case of WOV and WV. Results revealed a significant reduction of shear stress (6889%) by WV compared with the case of WOV by considering same water levels as without vegetation case. Overtopping of earthen embankment produces fast, turbulent flow velocities on the downstream side slope and initial erosion may occur within the supercritical flow region especially at a point of slope discontinuity such as d/s crest or at the toe of the d/s embankment. But smallscale laboratory flume experiments with an embankment using emerged vegetal cover on the d/s slope revealed no supercritical condition. Results also indicate that the vegetation markedly reduces the overtopping velocity and effective shear stress on the surface which are responsible for erosion.

0.0 50.0 100.0 150.0 200.0 250.0 300.0

0.009 0.011 0.013 0.015 0.017 0.019

Discharge (m3/s/m)

Shear stress (N

/m2)

Streamwise component of water (WV) Streamwise component of water (WOV) Effective shear stress tranfer to soil (WV)

Figure 7: Variation of shear stress with different discharges



860

4. Conclusions

A new methodology for stabilization of embankment with soil bioengineering is presented in this paper. To understand the processes and properties of hydraulics of overtopping flow on the downstream slope of the embankment with the vetiver hedges, a smallscale laboratory flume experiments with an embankment using emerged vegetal cover were carried out with three different unit discharges (0.018, 0.013 and 0.010 m3/s). The study explored the effectiveness of vegetated condition on the d/s of embankment compared with nonvegetated condition during overflow maintaining subcritical condition alone the flume. Vegetation markedly reduced the overtopping velocity and shear stress on the surface which in turn limits the extent of flooding/erosion.

However, the reduction capacity found very high in the case of low discharge compared with the high discharge. The results demonstrates that a significant reduction of damage and risk of failure from over topping flows can be achieved by providing vegetation cover in the downstream slope of the embankment. So it could a better solution against erosion induced by hydraulic action. We conducted limited experiments considering constant downstream slope of embankment and constant porosity. But the value depends on d/s slopes and density/porosity of vegetation cover. So further study is needed considering varying slopes and densities.

0

10

20

30

40

Discharge 1 Discharge 2 Discharge 3

Shear stress (N

/m2)

WOV WV considering same discharge as WOV

0

10

20

30

40

Water Level 1 Water Level 2 Water Level 3

Shear stress (N

/m2)

WOV WV considering same waterlevel as WOV

(a)

(b)

Figure 8: Variation of shear stress with (a) different discharges (b) water levels



861

5. References

1) Grimshaw, D. (2008), “Vetiver System Applications”. Technical Reference Manual (second edition) published by The Vetiver Network International, US, pp. 1101.

2) Islam, M. N. (2000), “Embankment erosion control: Towards cheapand simple practical solutions for Bangladesh”, Proc. Second International Conference on Vetiver. Office of the Royal Development Projects Board, Bangkok, pp. 307321.

3) Hanson, J. G and Temple, M. D. (2001), “Evaluation mechanics of embankments erosion during overtopping”, Proc. Seven Federal Interagency sedimentation conference, March 25 to 29, 2001, Reno, Nevada, USA, pp. 2430.

4) Truong, P. N. V. and Loch, R. (2004), “Vetiver system for erosion and sediment control”, In Paper Presented at 13 th International Soil Conservation Organization Conference, Brisbane 47 July (Paper no. 247), pp. 16.

5) Ke, C. C., Feng, Z. Y., Wu, X. J. and Tu, F. G. (2003), “Design principles and engineering samples of applying Vetiver ecoengineering technology for landslide control and slope stabilization of riverbank”, Proc. Third Int. Conference on Vetiver, Guangzhou, China, October, pp. 349357.

6) Das, S. C. and Tanaka, N. (2009), “The effectiveness of vetiver grass in controlling water borne erosion in Bangladesh”, Proc. 11th International Summer Symposium, JSCE, 11 September, Tokyo, Japan, pp. 6972.

7) Takemura, T. and Tanaka, N. (2007), “Flow structures and drag characteristics of a colonytype emergent roughness model mounted on a flat plate in uniform flow”, Fluid Dyn. Res. 39, pp. 694 – 710.

8) Baptist, M. J., Babovic, V., Uthurburu, J. R., Keijer, M., Uittenbogaard, R. D., Mynett, A., Verwey, A. (2007), “On inducing equations for vegetation resistance”, J. Hydr. Res. 43(3), pp. 435450.

9) Powledge, G. R., Ralston, D. C., Miller, P., Chen, Y. H., Clopper, P. E. and Temple, D. M. (1989), “Mechanics of Overflow Erosion on Embankments, I: Research Activities”, Journal of Hydraulic engineering, ASCE, 115(8), pp. 10401055.

10) Powledge, G. R., Ralston, D. C., Miller, P., Chen, Y. H., Clopper, P. E. and Temple, D. M. (1989), “Mechanics of Overflow Erosion on Embankments, II: Hydraulic and Design Considerations”, Journal of Hydraulic engineering, ASCE, 115(8), pp. 10561075.

11)Nadal, N. C. and S. A. Hughes. (2009), “Shear stress estimates for combined wave and surge overtopping at earthen levees”, Coastal and Hydraulics Engineering Technical Note ERDC/CHL CHETNIII79, Vicksburg, MS: U.S. Army Engineer Research and Development Center, 2009 available at http://chl.erdc.usace.army.mil/chetn.

international journal of civil and structural … · countries for more than a century as...

Documents