Applications of two costal pretection structure construction methods.

Yan, Shuwang ; Chu, Jian

Two methods that can be used for the construction of costal protection structures are introduced in this paper. The first is the geotextile bag method which uses either sand or clay to fill geotextile tubes or bags to form dikes or breakwaters. The second is to a method for using prefabricated semicircular reinforced concrete caissons to construct offshore dikes or structures on soft soil. Two case studies are presented to illustrate the applications of the two methods.

INTRODUCTION

The importance of costal protection has been highlighted during the devastating tsunami in December 2004. However, a large amount of resources are required to construct costal defence or protection facilities. This is particularly true when the coast to be protected is long. Therefore, the selection of the most cost-effective costal protection structures and the construction techniques becomes important in reducing the cost of the project. When the dikes or other types of costal protection structures are long, a small improvement in the design could result in a significant amount of saving. Therefore, it is beneficial to review as many methods as possible so the most cost-effective methods can be identified or to develop new methods that suit the local conditions the best. In this paper, two methods that can be used for the construction of costal protection structures or costal disaster rehabilitation works are introduced. The first is the geotextile bag method which uses either sand or clay to fill geotextile tubes or bags to form dikes or breakwaters. The second is the use of prefabricated semicircular reinforced concrete caissons to construct offshore structures or dikes on soft soil. Both methods have been used successfully in costal protection projects in China. Two case studies are presented to illustrate the applications of the two methods.

GEOTEXTILE BAG METHOD

The traditional method of constructing shoreline structures is to use rock or prefabricated concrete units. In recent years, several methods have been developed to use geotextile materials for the construction of coastal structures such as breakwaters and dikes.

One of the methods is to use geotextiles acting as formwork for cement mortar units cast in situ (Silvester and Hsu, 1993). The mortar mix need be only of sufficient compressive strength to support the weight above, plus the moment from the side force of the waves. Since the flexible membrane is required to hold the mixture in place until it sets, any subsequent deterioration due to UV rays or other conditions is of little concern. Thus, the method tends to be cheaper than the conventional methods. Applications of the mortar filled geotextile tubes are illustrated in Figs. 1 and 2. Details are referred to Silvester and Hsu (1993).

[FIGURES 1-2 OMITTED]

Similar methods, but using sand or dehydrated soil as the fill material have also been used for dike construction (Kazimierowicz, 1994; Miki et al., 1996). Sand or sandy soil is the most ideal fill material for this purpose. For near shore or offshore project, a suction dredger can used to pump sand from the seabed or a sand pit directly into the geotextile tubes. In case sand is not readily available, silty clay or soft clay may also be used. In this case, the clayey fill would have to be in a slurry state in order to be pumped and flow in the tube. The slurry would have to be dewatered in the geotextile bags or tubes under an ambient pressure. Then the selection of the geotextile used for the bags or tubes becomes important. The geotextile has to be chosen to meet both the strength and filter design criteria. Some analytical methods have been developed to estimate the required tensile strength for the geotextile (Kazimierowicz, 1994; Miki et al., 1996). The apparent opening size (AOS) of the geotextile needs to be selected to allow the pore pressure to dissipate freely and yet retain the soil particles in the bags.

One technique of using clay slurry fill geotextile bags for dike construction was developed in Tianjin, China, and used for one land reclamation project along the coast of Tianjin. The cross-section of the dike is illustrated in Fig. 3 and a picture showing the alignment of the bags is given in Fig. 4. It can be seen that flat geotextile bags, instead of tubes, are adopted in this method.

[FIGURES 3-4 OMITTED]

As shown in Fig. 3, the designed height of the dike was 4.8 m with base and top elevations at 0.7 m and 5.5 m respectively. The top width of the dike was 2.43 m. The water levels were at 4.7 m elevation during high tide and at nearly 0.7 m elevation during low tide. The outer and inner slopes of the dike were chosen to be 2L:1H and 1.5L:1H, respectively. For the bottom bag, the dimension used was 30 m in circumference. Clay slurry was dredged from the seabed of a selected area and pumped directly into the bags through an injection hole. The height of the bag after consolidation was around 0.5 m. Nine layers of geotextile bags were used. The surfaces of the slopes formed by the geotextile bags were to be covered with a cast-in-situ concrete layer of 25 cm. The concrete was cast-in-situ using moulds formed by geotextile bags, a technique which is commonly used in China (Chi, 1991). As shown in Fig. 3, berms were used to enhance the stability of the dike and to protect the toes of the slopes. The berms were made of crushed stones of 50 to 80 kg. The slopes of the berms were 2L:1H for the inner side and 3L:1H for the outer side. A 4 m thick of hydraulically filled slurry was to be placed behind the dike after the dike was constructed. An instrumentation scheme was also suggested as shown in Fig. 3. A woven polypropylene geotextile with a mass density of 120 g/[m.sup.3] was chosen for the bags. It had a tensile strength of 28 kN/m in the longitudinal and 26 kN/m in the transverse direction respectively. The AOS ([O.sub.95]) of the geotextile was 0.145 mm. The bags are sewn together using sewing machines on site. The soil used to fill the bags was classified as SC-CL according to the Unified Soil Classification System (USCS), that is, a borderline case between sandy clay and low plasticity clay. The liquid limit and plastic limit of the soil were 20.4% and 8.9% respectively and the plasticity index was 8.9%.

PREFABRICATED CAISSON METHOD

The geotextile bag method may only be feasible when dikes are to be constructed in relative shallow or quiet water. When water is too deep or the wave is rough, gravity retaining structures using prefabricated reinforced concrete segments or caissons may be a better option. The use of caisson or concrete segments is not new. However, the most cost-effective design methods are still yet to be established. These concrete segments or caissons have to be tall enough to match the water depth and heavy enough to provide stability against the waves. However, when the concrete segments or caissons are too heavy, they cause settlement or bearing capacity problems. This is particularly the case when the foundation soil is soft. The weak foundation soil can be improved. However, it is difficult and costly to treat soil offshore and over a large area or distance. Therefore, it has become a challenge on how to construct large size gravity structures on soft soil.

In one of costal protection projects along Yangtze Estuary, some dikes for navigation purposes needed to be constructed. The method of using prefabricated reinforced concrete caissons was adopted. The dike was to be constructed at 40 km away from the coast. The water depth ranged from 5.0 to 8.5 m. The design wave height was 3.32 to 5.90 m with a return period of 25 years. The total length of the dike was about 17 km.

The design of the dike is schematically shown in Fig. 5. The caisson used was prefabricated reinforced concrete hollow segment. It was semicircle shaped, as shown in Fig. 5. The radius of the semicircle was 5.7 m. The advantage of using a semicircular cross-section is that the direction of the resultant wave force on the semicircular shaped structure will always pass through the center of the circle, which will greatly improve the loading condition of the structure. The hollow caisson would be filled with sand after installation through a 600 mm diameter hole on top of the caisson. In order to prevent the foundation soil from scouring, a geotextile sheet was used to cover the seabed. A cushion which acted as the foundation bed was placed on top of the geotextile. The cushion was 1 to 2 m high. It was made of crushed stones of 1 ~ 100 kg for the centre and 200 ~ 400 kg for the edge. After the caisson was placed, berms were placed on two sides of the caisson. The berms were made of 400 ~ 600 kg crushed stones.

[FIGURE 5 OMITTED]

The soil profile below the dike consisted of a 1.5 to 3.5 m thick silt sand followed by 2 to 4 m thick clay mud and a roughly 30 m thick soft clay underlying the mud. The basic properties of the soil are given in Table 1. Although the soil below the dike was weak, with the use of geotextile and the cushion, the bearing capacity was estimated to be sufficient. However, the shear strength of the soil could be weakened by the wave action, which in turns would affect the stability of the dike. Nevertheless, due to time and other constraints, one section of the dike was constructed without improving the soil first. As the dike was a strip load, the load would only be distributed to a certain depth. It was hoped that under the surcharge of the crushed stone and the caissons, the geotechnical properties of both the upper silty sand layer and the top few meters of lower soft clay layer would be improved with time, and thus the stability of the dike would have been enhanced with time.

However, this proved to be a mistake. Only 2 months after the construction of the dike, an unusual strong storm took place in December 2002. The dike failed during the storm. Large settlements incurred suddenly and some caissons large lateral movements. A picture of the dike after the storm is shown in Fig. 6. It can be seen that the caissons had undergone either large lateral or vertical movements. The settlement versus time curve is shown in Fig. 7. The wave height versus time curve is also plotted in Fig. 7. It can be seen that a sudden settlement took place at the time when there was a surge in the wave height. The dike had become unstable under the wave action. Apparently, the soil needed to be improved. How to improve the seabed soil which was 5 to 8.5 m below the sea level with a wave height of 3 to 6 m in a cost-effective way became the key issue.

[FIGURES 6-7 OMITTED]

Several soil improvement methods were considered. The soil replacement method was not feasible as the amount of fill required and the amount of excavated soil to be disposed were too excessive. The method to accelerate the consolidation process of the soft clay using prefabricated vertical drain (PVD) was considered the most economical. This method was also relatively easy to implement. The influence zone of the strip load was estimated to be 7.0 m below the seabed. Therefore, it was only necessary to install the PVD to a depth of 10 m deep. The weight of the crushed stone cushion was considered sufficient in providing surcharge. A special vertical drain installation barge was used to install the PVD offshore. The procedure for improving the soft soil was as follows: (1) PVDs was installed from the PVD installation barge at a spacing of 1.0 m to a depth of 10 m below the seabed; (2) The crushed stone cushion was laid on the seabed as a surcharge to consolidate the soil below; (3) The caisson segments were only placed after an average degree of consolidation of 80% was achieved, which took about 90 days after the placement of the crushed stone.

In addition to soil improvement, anti-slippery rubber pads were also used for the caissons. The rubber pads were embedded into the base of the caisson during the casting stage. Pins were also precast into the base of the caisson to enhance the anchoring effect. Large scale laboratory tests have shown that the use of rubber pads and pins can enhance significantly the resistance of the caisson against sliding.

To assess the effectiveness of soil improvement, the undrained shear strength profiles obtained from vane shear tests conducted before and after the preloading are compared in Fig. 8. The duration of consolidation was 90 days, It can be seen that the vane shear strength increased by almost 2 folds as a result of consolidation using PVDs.

It can also be seen from Fig. 8 that there were little improvement in the soil where PVDs were not reached. However, soils at these depths were not affected by the cyclic wave loads.

[FIGURE 8 OMITTED]

The construction of the dike was completed in Dec 2003. The settlements of the dike were monitored after the construction. The settlements measured at six different sections with time are shown in Fig. 9. The wave height versus time curve is also plotted in Fig. 9. It can be seen from Fig. 9 that the dike had experienced several strong storms caused by typhoons. Some of them were even stronger than the one that caused the failure as shown in Figs. 6 and 7. Even through, the dike was stable and there was no additional settlement caused by the storms. The settlement also stabilized in 5 to 6 months. The total settlements were within the expected range. Therefore, the use of PVDs has proven to be a successful method for this project. The dike after construction is shown in Fig. 10.

[FIGURES 9-10 OMITTED]

CONCLUSIONS

Two methods that can be used for the construction of coastal protection structures or for costal disaster rehabilitation works are introduced in this paper. The first is the geotextile bag method which uses either sand or clay to fill geotextile tubes or bags to form dikes or breakwaters. The second is to use prefabricated reinforced concrete caissons to construct offshore dikes or structures on soft soil. The applications of these methods are illustrated using case studies. The following conclusions can be made on the two methods:

(1) The geotextile bag or tube method is suitable for the construction of breakwaters or dikes in shallow water. The bags can be fabricated using geotextile on site into any desirable dimensions. Although sand or lean concrete mortar are normally used as the fill materials, the presented case study has shown that clay slurry dredged from the seabed can also be use as a fill material to fill the bags. This method is cost-effective, particularly when sand or rocks are not readily available as fill material.

(2) A method to use prefabricated reinforced concrete caissons to construct dikes of offshore structures on soft clay is presented using a case study. The study shows that it is necessary to improve the soft seabed soil before constructing the dike even though a thick cushion made of crushed rocks is used. The dike built on soil without improvement failed during a heavy storm. For the case presented, the soil can be improved by simply installing PVDs to accelerate the consolidation process of the seabed clay. With the use of PVDs, the soft soil would consolidate much faster and gain sufficient strength quickly to maintain the stability of the caissons.

REFERENCES

Chi, J. K. (1991). Technical Report on the Technique of Pumping Concrete into Geotextile Mould. Shanghai Geotechnical Research Institute (in Chinese).

Kazimierowicz, K. (1994). "Simple analysis of deformation of sand-sausages." Proc. 5th Int. Conf. Geotextiles, Geomembranes and Related Products, Singapore, 5-9 Sept., Vol. 2, 775-778.

Miki, H., Yamada, T., Takahashi, I., Shinsha, H., and Kushima, M. (1996). "Application of geotextile tube dehydrated soil to form embankments." Proc. 2nd Int. Conf. on Environmental Geotechnics, Osaka, 5-8 Nov, 385-390.

Silvester, R. and Hsu, J. R. C. (1993). "Costal Stabilization--Innovative Concepts", Prentice-Hall Inc.

SHUWANG YAN

Geotechnical Research Institute, Tianjin University, Tianjin, 300072, China

JIAN CHU

School of Civil and Environmental Engineering, Nanyang Technological University, Singapore 639798

Table l. Soil properties

Soil Water Unit Void Liquid
Stratum content weight ratio limit
 (%) (kN/[m.sup.3])

Silt sand 29.3 19.0 0.827 -
Mud 57.5 16.4 1.672 45.2
Soft clay 51.5 16.8 1.470 45.8

Soil Plastic Undrained Compression
Stratum limit shear Index (2)
 strength (1) [Mpa.sup.-1]
 (kPa)

Silt sand - - -
Mud 23.6 13.4 1.59
Soft clay 23.8 22.0 1.31

(1.) The undrained shear strength was measured by Unconsolidated
undrained tests.

(2.) The compression index was measured by oedometer tests within
the stress range of 100 to 200 kPa.