Experimental study of sediment trapping by geotextile

Geosynthetics International, 2013, 20, No. 6

Experimental study of sediment trapping by geotextile mattress installed with sloping curtain L. Xie1 , W. Huang2 and Y. Yu3 1

Professor, Department of Hydraulic Engineering, Tongji University, 1239 Siping Road, Shanghai, China, Telephone: 0826 21 65981543, Telefax: 0826 21 65989220, E-mail: [email protected] 2 Adjunct Professor, Department of Hydraulic Engineering, Tongji University, 1239 Siping Road, Shanghai, China and Professor, Department of Civil and Environmental Engineering, Florida State University, FL, USA, Telephone: +1 850 410 6199, Telefax: +1 850 410 6236, E-mail: [email protected] 3 Professor, State Key Laboratory of Hydroscience and Engineering, Tsinghua University, Beijing, China, Telephone: 0826 10 62797200, Telefax: 0826 10 62785593, E-mail: [email protected] Received 3 December 2012, revised 14 August 2013, accepted 9 September 2013 ABSTRACT: Geosynthetic structures created for channel erosion protection offer environmental friendly benefits and have demonstrably lower construction and lifetime costs than similar hard structures. A geotextile mattress with a sloping curtain (GMSC) offers an alternative countermeasure against channel erosion. In the present study, experiments were conducted to investigate the working mechanism and effectiveness of GMSCs which were installed on movable beds in a rectangular flume. The bathymetry of the plastic sand beds was measured before and after the tests. The results showed that the presence of the GMSC led to sediment deposition and dune formation at both upstream and downstream edges of the GMSC structure. This will prevent bottom erosion near the structure and increase its stability against flow-induced sediment scour, so that the erodible beds will be protected. KEYWORDS: Geosynthetics, Geotextile, Mattress, Sloping curtain, Erosion control, Coverage ratio, Sediment trapping REFERENCE: Xie, L., Huang, W. & Yu, Y. (2013). Experimental study of sediment trapping by geotextile mattress installed with sloping curtain. Geosynthetics International, 20, No. 6, 389–395. [http://dx.doi.org/10.1680/gein.13.00026]

1. INTRODUCTION Fluvial erosion and shoreline retreat are causing significant economic loss, risks to public safety and are degrading waterways not only in China but also elsewhere in the world. During the past 50 years about 50 000 dams have been constructed throughout the Yangtze River watershed, and this has resulted in downstream channel erosion and coarsening of the bottom sediment, and in erosion of the Yangtze’s sub-aqueous delta, especially after the closure of the Three Gorges Dam in 2003 (Yang et al. 2011). In the coming decades the Yangtze’s sediment load will probably continue to decrease, and its middle–lower river channel and delta will continue to erode as more dams are built. Riprap is a most common form of riverbank-toe protection and the US Army Corps of Engineers have presented riprap design guidance for flood channels, which takes account of the effects of bends, blanket thickness, sideslope angle, particle shape, and gradation on riprap stability (Maynord 1993). In China, the protection of a kilometre length of the riverbank in the middle and

downstream reaches of the Changjiang (Yangtze) River requires an average of about 100 000–150 000 m3 of stone (Chen and Sun 2000) and the necessary mining for stone has caused substantial environmental damage. Moreover, in recent years there have been problems involving a lack of suitable-size rocks, difficulties in quarrying, transporting, and placing the stone, and in providing the large amounts of material needed for deeper rivers or sea, and so it has become very expensive to build and maintain the traditional forms of river and coastal structures. The materials used to control erosion in hydraulic and coastal structures are therefore changing from the traditional rubble and concrete systems to cheaper materials and more environmentally friendly countermeasures. In this respect, geosynthetics have attracted much attention worldwide and are becoming the most effective solution of choice against scour, having been used in a broad range of coastal protection, erosion protection, seismic aspects, sinkhole bridging and hazardous waste. Innovative geosynthetic engineering practices have been developed to satisfy

1072-6349 # 2013 Thomas Telford Ltd

Downloaded by [ Tongji University] on [11/10/18]. Copyright © ICE Publishing, all rights reserved.

389

390

Xie, Huang and Yu

a wide range of applications in damage prevention, natural disaster mitigation, rehabilitation measures and environmental protection (Brandl 2011). Geosynthetic casings can be constructed easily and cost-effectively and can be filled using cheap filling material such as sludge or slurry that is dredged on site by hydraulic pumping. Sandbags were the first geosynthetic containers and new applications of geotextile bags are being continuously developed in response to the lack of traditional material for erosion protection, riverbank stabilisation and shoreline protection (Krahn et al. 2007; Hornsey et al. 2011; Oberhagemann and Hussain 2011). The geosynthetic containers can be manufactured according to engineering requirements and there are two main types, namely tubes and mattresses, which are defined by their shape and structural characteristics (Heibaum 2002). Geotextile tubes are similar in concept to geotextile bags and containers, and have been widely used in structural and dewatering applications. They are typically engineered to resist short- and long-term forces. Geotextile tubes can be used in the construction of dykes or breakwaters for the prevention of beach erosion and the protection of coastal infrastructure (Shin and Oh 2003). They have also been used to protect underwater pipelines, to prevent scour under bridge piers and other structures, and for slope protection. By acting as a confining mechanism, geotextile tubes have been used to dewater dredged materials and contain contaminated materials (Cantre and Saathoff 2011; Yee and Lawson 2012). The tubes are factory manufactured by sewing multiple sheets of woven geotextiles together to create an enclosed tube. The use of geosynthetics offers the chance to minimise the impact of structures on the environment. Figure 1 shows an application of geotextiles in erosion control at the Yangtze estuary in China, in which some sand-filled tubes were installed on a geotextile at a constant spacing. Lateral erosion and bed degradation have the effect of increasing the bank heights and angles, which decreases the stability of the river banks (Osman and Thorne 1988) and eventually may cause them to collapse (Darby and Thorne 1996). Figure 2 shows a classical cross-section of revetments against slope erosion, in which the slope is

fully covered by armouring countermeasures acting as a resistant layer to the hydraulic shear stresses and providing protection to the soils underneath. The armouring countermeasures include riprap and sand-filled geotextile bags (Oberhagemann and Hussain 2011). It should be noted that the stability of these structures is reduced when scour undermines the toe support or when pore water pressures in the underlying soil builds because of clogging of the revetment filter fabric. Moreover, the adverse effects of revetments such as river habitat destruction may occur due to complete coverage of the bank slope (Shields et al. 2000). It is therefore important to reduce the mattress coverage ratio in erosion control. The properties of the soil and bottom shearing stress induced by flow or waves are important factors for bottom erosion in rivers and coastal waters (Liu and Huang 2009; Zhang 2010; Le et al. 2011; Zhang et al. 2011). In the present study, a new alternative technique to control erosion in the form of a geotextile mattress with sloping curtain (GMSC) is presented. The principles of the technique are described and the results of physical model tests to verify the feasibility of the geotextile mattress with a sloping curtain and to summarise the mechanism of sediment harvest are reported. The structural efficiency of the GMSC is discussed in detail and potential future applications in erosion control and sedimentation improvement are discussed.

2. GEOTEXTILE MATTRESS WITH SLOPING CURTAIN 2.1. Definition A geotextile mattress with sloping curtain, as shown in Figure 3, is composed of two sheets of geotextiles sewn together to form a string of tubes which can be filled with granular materials such as sand, gravel, concrete, or cement. The seam can be of any suitably durable conventional stitching or other suitable fastener for fabric. The bottom edge of the curtain is fixed to the middle of the mattress and the top edge is connected to a pipe to make the sloping curtain. When the pipe at the top edge is floating, the sloping curtain becomes a floating curtain. Some openings are cut at the bottom of the curtain to allow the bottom flow to pass through, with the aim of harvesting

Armoring countermeasures

High (flood) water level (HWL)

Low water level (LWL) Soil slope

Figure 1. Photograph of geotextile revetments with sandfilled geotextile tubes for deployment to the riverbed for erosion control

Figure 2. Schematic diagram of revetments against bottom erosion in riverbed or beach



Experimental study of sediment trapping by geotextile mattress installed with sloping curtain

downstream, which makes some of the bottom current flow upward, and the rest pass through the sand-pass openings with high sediment concentration. The currents passing over the curtain will collide with the discharge of the current from the bottom openings, forming two circulation zones: one at the back of the topside curtain, and the other at the bottom area close to the bed surface, just downstream of the geotextile mattress. The top circulation is not a key consideration in the present research, as it only affects the flow of the upside current. The bottom circulation provides a long safe zone with a lower velocity of flow, because of the strike of the bottom current from the sand-pass openings. Therefore, the bed in this zone can be protected from scouring. According to the vertical distribution characteristics of the sediment concentration, the highest sediment concentration appears mainly in the bottom area and this is the reason why the sand-pass openings are cut in the bottom part. After the GMSC structure has been in place for a long time a lot of sediment will have been deposited in the bottom circulation area, causes a sand dune to form and stabilise the bed landform in that area.

3 5

2

1

4

6

391

(a)

Flow

3. LABORATORY MODEL TEST 3.1. Test and apparatus conditions The laboratory experiments were conducted at the State Key Laboratory of Hydroscience and Engineering at Tsinghua University, Beijing. The main pieces of equipment used were the flume, plastic sand, an acoustic profiler and the velocimeter. The flume was of rectangular cross-section with transparent glass walls, 22 m long (the X direction), 0.5 m wide (the Y direction) and 0.6 m deep (the Z direction). The flume was supported in the centre and at two points on the ends by screw jacks, which allowed the channel bed to be adjusted. The water in the flume was recirculated using a pump, and the power was adjusted according to the requirements of the water flow velocity. The water entered the flume through a honeycomb structure whose purpose was to provide rectilinear flow. The sand bed in the flume was formed by plastic sand and the GMSC were placed on the sand bed.

(b)

Figure 3. Example of geotextile mattress with sloping curtain (GMSC) and a model installed in running flow: 1, geotextile mattress; 2, sloping curtain; 3, pipe (either floating or fixed); 4, sand-pass openings; 5, belts; 6, mattress tubes

the high concentration of sediment load in the bottom flow. Belts are sewn on the mattress and curtain to strengthen the integrity of the geotextile mattress and curtain. Figure 4 shows the working principles of a GMSC structure. A sloping curtain in operation is always inclined

Flow Upward flow

Top circulation Hc Flow Erodible zone

Low velocity zone Lt

Bottom circulation LW

ΔL Wd

Wu

Figure 4. Schematic diagram to show operating principles of a geotextile mattress with sloping curtain Geosynthetics International, 2013, 20, No. 6


392

Xie, Huang and Yu

3.2. Materials used in model tests

Percentage finer than a given size (%)

Before the ests, the plastic sand was placed uniformly on the flume base to form a movable bed, 40 mm thick and 10 m long. The grading curve of the plastic sand is shown in Figure 5. The sand had a median grain size of 0.24 mm, with a dry unit weight (ªd ) of 6.56 kN/m3 and a specific gravity of 1.045–1.056. The velocity of the incipient motion was nearly 81 mm/s, corresponding to the flow depth controlled in the model tests. The sand was recirculated via a slurry pump. The model of the GMSC used in the model tests is shown in Figure 6, with the curtain height Hc ¼ 43 mm, curtain angle 458and length 500 mm, which was the same as the width of the flume. The widths of the upstream geotextile mattress and the downstream geotextile mattress were both 75 mm. The bottom openings were rectangular

100 80 60 40 20 0

1

0.1 Grain size (mm)

0.01

and three different sizes were included in the various tests, namely 30 mm 3 10 mm with opening ratio ¼ 10%, 40 mm 3 10 mm with ¼ 13.3% and 35 mm 3 15 mm with ¼ 17.5%. The geotextile used for the GMSC should meet the filter criteria as well as the requirement of sufficient tensile strength. The apparent pore opening size should be small enough to retain the soil inside the mattress and but sufficient to provide permeability. Heerten (2007) recommended that geotextiles used in drainage ditches or other hydraulic engineering structures should be larger than 300 g/m2 in order to avoid puncturing. In applications for erosion control, the slope angle of the curtain changes with variation of stream velocities, and hence the effective opening size of the curtain openings for sediment passing was changed accordingly as this decides the effectiveness of sediment trapping by the GMSC. The main purpose of this experimental study was to investigate the sediment harvesting behaviour and performance of the GMSC structure. The model system was supported by a rigid frame. A woven polypropylene geotextile with a mass per unit area of 130 g/m2 was chosen to meet the model requirements. It had a tensile strength of 29 kN/m in the longitudinal and 27 kN/m in the transverse direction and a maximum elongation of 20% in the longitudinal and 17% in the transverse direction, respectively. The apparent pore opening size (O95 ) of the geotextile was 0.14 mm and the thickness was 0.5 mm. The permeability of the geotextile was 0.3 mm/s. 3.3. Test programme

Figure 5. Grading curve of the plastic sand

45° (a)

To investigate its performance a GMSC was installed in the flume and the basic sedimentation phenomena were checked. The experiments assessed the advantages or disadvantages and highlighted the structural properties of the GMSC used for erosion control. In the tests, the flow was controlled at a depth of 370 mm and the flow velocities were varied with time as 7.2cm/s for 1 h, 10.5cm/s for 23 h, 14.1cm/s for 14 h. The test model was designed to a geometric scale of model:prototype ¼ 1: 36, and the corresponding velocity scale was model: prototype ¼ 1: 6. For this scale model, the incipient motion velocity for the prototype was about 0.48 m/s, flow depth was about 13 m and the curtain height was about 1.5 m. Figure 7 shows the installation of the GMSC, and the phenomena and characteristics of sediment trapping were verified by the collected data. Velocity and sandy bed transformation were measured in the model tests. At the end of the tests, the flow velocity was decreased slowly to avoid any flow fluctuation that might destroy the sand bed, and an acoustic profiler was used to measure the bed transformation.

4. RESULTS AND DISCUSSION (b)

Figure 6. GMSC model: (a) frame of a GMSC, (b) GMSC in flume

Tests were conducted with and without the GMSC. The resultant transformation of the sand beds is shown in Figures 8–13.



Experimental study of sediment trapping by geotextile mattress installed with sloping curtain

393

Flow GMSC

Paved sandy bed

Flume bed

Figure 7. Installation of GMSC in the tests

Figure 8. Sand bed before test (a)

(b)

Figure 9. Bottom scour condition after 3 h with no GMSC matt protection. Scour depth will increase as flow continues

Figure 11. Test 2: Transformation of the sand bed around GMSC (ä 13.3%): (a) T 16 h; (b) T 38 h

(a)

(b)

(c)

Figure 10. Test 1: Transformation of the sand bed around GMSC (ä

17.5%): (a) T



16 h; (b) T

23 h; (c) T

38 h

394

Xie, Huang and Yu

(a)

(b)

(c)

10%): (a) T

Figure 12. Test 3: Transformation of the sand bed around GMSC (ä 40

16 h; (b) T

23 h; (c) T

38 h

δ ⫽ 17.5%

Erosion/deposition (mm)

δ ⫽ 13.3% 20

0

δ ⫽ 10.0%

0

0.2

0.4

0.6 0.8 Distance to GMSC (m)

1.0

1.2

⫺20

⫺40

Figure 13. Transformation of the sand beds downstream of the GMSC

4.1. Phenomena of bed erosion without installations of GMSC To demonstrate the difference after the installation of the GMSC, a control test without the GMSC was conducted with the same bed material and flow conditions. Figure 8 shows the sand bed before the test and Figure 9 shows the transformation of the sand bed after 3 h of water flow during which the sand bed was subjected to obvious scour to 15 mm depth. 4.2. Phenomena of sediment trapping To verify the sediment trapping mechanism and performance of the GMSC, three tests were performed using only one GMSC with three different bottom opening sizes. Figures 10–13 show the transformation of the sand beds after different periods of channel flow in the three tests. Before the tests the original bed was set at the same level by careful measurement, so the initial bed level was regarded as zero in Figure 13. From the transformation of the sand beds, the effectiveness of the GMSC can be seen as sand dunes formed on the upstream and downstream sides of the structure. The upstream dune was smaller than the downstream dune and the distance between the downstream dune and the curtain was about the same as the height of the curtain. A much longer zone of the area downstream of the GMSC was protected from scouring as shown in Figure 13. From the photos of the bed transformation, it can be concluded that the GMSC was ineffec-

tive when the percentage of bottom opening area was too large, (i.e. 17.5%) and the geotextile mattress should be designed with sufficient width in accordance with the sand-pass opening ratio. The erosive length ˜L increased with the opening ratio, and therefore the width of mattress should be increased accordingly. 4.3. Potential applications of GMSC As shown in Figure 2, the bottom of the riverbank can be protected by geotextile revetments isolating the slope soil from erosive flow using a GMSC with its easier installation, instead of rock or concrete cubes. If the negative effects of geotextile revetments on the ecological environment of a river due to full coverage of bank slope and a large enough area of riverbed can be kept low enough, the range of applications geotextile would be increased. In this respect a GMSC may be a good choice as the mattress coverage ratio is much less, leaving most of the protected bed surface undisturbed. A GMSC structure can work well for sediment trapping when submerged, so for erosion protection it should be installed below the lowest water level. However, protection of the slope toe of the riverbank is of great importance for its stability (Shields et al. 2000). Erosion of the slope toe will cause riverbank collapse, so it is essential to trap sediment by installing many elements of GMSC structure and build up the sand base aiming to prevent bottom erosion in order to improve riverbank stability. The GMSC can be installed at variable



Experimental study of sediment trapping by geotextile mattress installed with sloping curtain spacing in the bottom area of the riverbank slope and the riverbed, and this provides protection for the erodible bed at a lower coverage ratio. As it needs only partial coverage of the erodible site, this technique is an effective and ecological alternative to protect beds from serious erosionand furthermore, the cost of erosion control by GMSC will be lower because of the lesser mattress coverage. In view of the sediment trapping mechanism of sand build up at the upstream and downstream sides of the GMSC structure, GMSC can be used in more complex hydrodynamic conditions such as reciprocating flow. If the main current direction of the most threatening flow in a marine environment is not changing obviously during the year, GMSC can be used to protect many kinds of offshore structures such as pipelines, platforms, submarine cables, etc.

5. CONCLUSIONS The GMSC offers an alternative technique which compensates for the shortcomings of a mattress covering the entire erodible bed and forms a material structure that effectively controls erosion of the riverbank slope toe or sea bed. The sloping curtain with bottom openings intercepts the approaching current and improves the local flow structure, which makes it possible to reduce the approaching flow velocity around the GMSC structure. After installation of a GMSC on the bed, dunes gradually form around it and its stability increases with the increasing weight of dunes. Some other important results from these experiments are listed here.

(1)

(2)

(3)

Under erosive flow conditions, dunes form at both upstream and downstream sides of the GMSC structures. The volumes of the dunes become steady as the deposition and erosion balance and finally the maximal dune height is nearly half the curtain height. The sand-pass openings provide a passageway for the river load to pass through normally and a material source for the deposited dunes. This greatly reduces the ecological effect of installations of GMSC on the sediment discharge of rivers. The bottom opening ratio needs to be controlled within a reasonable range, because the ability of sand to pass through cannot meet the requirement of sediment discharge for a low opening ratio and the area around the GMSC may be subjected to serious erosion. For the flow conditions in the reported tests, the ideal opening ratio was about 14%.

395

ACKNOWLEDGEMENTS This work was supported by the National Natural Science Foundation of China (Grant No. 50979071, No. 51279134, and No. 91225301).

REFERENCES Brandl, H. (2011). Geosynthetics applications for the mitigation of natural disasters and for environmental protection. Geosynthetics International, 18, No. 6, 340–390. Cantre, S. & Saathoff, F. (2011). Design parameters for geosynthetic dewatering tubes derived from pressure filtration tests. Geosynthetics International, 18, No. 3, 90–103. Chen, Z. Y. & Sun, Y. S. (2000). Probing the cause of process of Yangtze dyke collapse and corresponding structural measures. China Water Resources, No. 2, 28–29 (in Chinese). Darby, S. E. & Thorne, C. R. (1996). Development and testing of riverbank-stability analysis. Journal of Hydraulic Engineering, 122, No. 8, 443–445. Heerten, G. (2007). Basics of filtration with soils and geosynthetics for geotechnical constructions. Proceedings of the 10th Australia New Zealand Conference on Geomechanics, Brisbane, Australia, vol. 2, pp. 381–323. Heibaum, M. H. (2002). Special issue on geosynthetic containers. Geotextiles and Geomembranes, 20, No. 5, 279–280. Hornsey, W. P., Carley, J. T., Coghlan, I. R. & Cox, R. J. (2011). Geotextile sand container shoreline protection systems: Design and application. Geotextiles and Geomembranes, 29, No. 4, 425–439. Krahn, T., Blatz, J., Alfaro, M. & Bathurst, R. J. (2007). Large-scale interface shear testing of sandbag dyke materials. Geosynthetics International, 14, No. 2, 119–126. Le, T. T., Cui, Y. J., Mun˜oz, J. J., Delage, P., Tang, A. M. & Li, X. L. (2011). Studying the stress-suction coupling in soils using an oedometer equipped with a high capacity tensiometer. Frontiers of Architecture and Civil Engineering in China, 5, No. 2, 160–170. Liu, X. & Huang, W. (2009). Modeling sediment resuspension and transport induced by storm wind in Apalachicola Bay, USA. Journal of Environmental Modelling & Software, 24, No. 11, 1302–1313. Maynord, S. (1993). US Army Corps of Engineers riprap design for flood channels. Transportation Research Record, 1420, 14–22. Oberhagemann, K. & Hussain, M. M. (2011). Geotextile bag revetments for large rivers in Bangladesh. Geotextiles and Geomembranes, 29, No. 4, 402–414. Osman, A. M. & Thorne, C. R. (1988). Riverbank stability analysis. I: theory. Journal of Hydraulic Engineering, 114, No. 2, 134–150. Shields, F. D. J., Knight, S. S. & Cooper, C. M. (2000). Warmwater stream bank protection and fish habitat: A comparative study. Environmental Management, 26, No. 3, 317–328. Shin, E. C. & Oh, Y. I. (2003). Analysis of geotextile tube behaviour by large-scale field model tests. Geosynthetics International, 10, No. 4, 134–141. Yang, S. L., Milliman, J. D., Li, P. & Xu, K. (2011). 50000 dams later: Erosion of the Yangtze River and its delta. Global and Planetary Change, 75, No. 1–2, 14–20. Yee, T. W. & Lawson, C. R. (2012). Modelling the geotextile tube dewatering process. Geosynthetics International, 19, No. 5, 339– 353. Zhang, F., Ye, B. & Ye, G. L. (2011). Unified description of sand behavior. Frontiers of Architecture and Civil Engineering in China, 5, No. 2, 121–150. Zhang, L. Y. (2010). A simple method for evaluating liquefaction potential from shear wave velocity. Frontiers of Architecture and Civil Engineering in China, 4, No. 2, 178–195.

The Editor welcomes discussion on all papers published in Geosynthetics International. Please email your contribution to [email protected] by 15 June 2014.

