Sedimentary Changes of a Sand Layer in Liquefied Silts

REN Yupeng, ZENG Yu, XU Xingbei, and XU Guohui, *

Sedimentary Changes of a Sand Layer in Liquefied Silts

REN Yupeng1), 2), ZENG Yu3), XU Xingbei1), 2), and XU Guohui1), 2), *

1) Shandong Provincial Key Laboratory of Marine Environment and Geological Engineering, Qingdao 266100, China 2) Laboratory for Marine Geology, Qingdao National Laboratory for Marine Science and Technology, Qingdao 266061, China 3) Zhejiang Huadong Construction Engineering Co., Ltd., Hangzhou 310000, China

A flume experiment was conducted to investigate the restratification of liquefied sediment strata under a wave load with the focus on the interbedded strata of coarse and fine sediments formed in estuarine and coastal areas. The aim of this research was to study the characteristics and processes of liquefied sediment strata in terms of wave-induced liquefaction. In the experiment, the bottom bed liquefied under the wave action and the liquefied soil moved in the same period with the overlying waves, and the track of the soil particles in the liquefied soil was an ellipse. The sand layer consisting of coarse particles in the upper part, settled into the lower silt layer. The sinking of coarse particles and upward migration of the fine particles of the lower layer induced by liquefied sediment fluctuations are the likely reasons for sedimentation of the sand layer in liquefied silt.

storm-liquefied sediment strata; liquefaction; sand layer settlement; silt interlayer; particle sorting

1 Introduction

In marine geology engineering, indoor experimentation, theoretical analysis, and on-site observation have been used to demonstrate that coastal area seabed sediments can be liquefied under strong waves. A series of flume wave experiments (Sleath, 1970; Tsui and Helfrich, 1983; Tzang, 1998; Sumer, 1999; Tzang and Ou, 2006; Zhang, 2009; Kirca, 2013), centrifuge tests (Sassa and Sekiguchi, 2001; Sekiguchi, 2001, 2004), shaking table tests (Xu, 2019; Ko and Chen, 2020; Wang, 2020; Zhang, 2020), and one-dimensional (1-D) cylinder compressive tests (Chowdhury, 2006; Zen and Yamazaki, 2008a, b; Liu and Jeng, 2015; Liu, 2015) have revealed wave-induced liquefaction in sandy/silty seabeds under laboratory conditions. Chang(2004) analyzed the liquefaction potential of the Yi-Lan nearshore sandy seabed area using the Ishihara and Yamazaki (1984) method, which showed a maximum liquefaction of 6.1m at a water depth of 8.6m. Field pore pressure monitoring on the coast also demonstrated that sandy seabeds can be liquefied under strong wave conditions (Sassa,2006; Chang, 2007). For silty/silty-clay seabeds, using geographical interpretation of the shallow formation profile (Coleman and Prior, 1981), a collapsed stratum was found in the Mississippi River Delta and the disturbance of the strata was discovered in the Yellow River Delta (Xu, 2009). Further research revealed that this phenome-non was the result of seafloor liquefaction under the action of waves (Xu, 2012; Wang, 2013).monitoring demonstrated that liquefaction depths in the subaqueous delta of the Yellow River exceeded 3.3m (Xu, 2018).

Liquefied sediments oscillate under the effect of water fluctuations (Sumer, 2006b; Xu, 2012). During the fluctuation of sediments, objects (initially settled on/buried in the sediments) heavier than the sediments will sink. Sumer(1999) investigated the sinking of pipelines, spherical objects, and cubic objects in liquefied marine sediments under waves and suggested a model with empirical relations for the drag coefficient as a function of the ultimate sinking depth of the objects. Sumer(2006a) and Teh(2006) experimentally studied the stability and flotation/sinking of pipelines under wave-induced liquefaction. Sawicki and Mierczyński (2009) established an empirical approximation of the sinking depth of a large object in liquefied soil according to their results from a cyclic triaxial testing apparatus, which was also related to the viscosity of liquefied soil. The fluctuation also caused a change in sediment. Wang(2013) and Xu(2016) found that wave-induced liquefaction resulted in the differentiation of variable size deposits using flume wave tests. Liquefied sediments fluctuated with wave activity and moved elliptically. With continuous cycling between the peaks and troughs of waves, coarse and fine particles of the liquefied sediments gradually differentiatedto form restratified strata. This process is called storm liquefaction deposition. Recent research has verified storm liquefaction deposits in the subaqueous delta of the Yellow River (under submission).

Estuarine delta seabed sediments can be liquefied under wave action and the liquefaction depth reaches 3–6 m. The sedimentary sequence of the estuarine and coastal areas under strong waves will be reconstructed because of liquefaction of the sediments. With seasonal changes in river water volume or flooding events, sediments are carried by rivers into the sea to form interbedded layers of coarse-grained and fine-grained sediments in estuarine andcoastal areas, where coarse-particle sediments overlie fine-particle sediments. Sediments with this type of structure are easily affected by storm waves at certain depths along coastlines. Researchers still do not fully understand what happens to sedimentary strata when coarse-particle sediments overlie fine-particle sediments that are liquefied by the influence of strong storm waves. In this study, a wave flume test was conducted by constructing a silt–sand–silt layered bed. The sedimentation of the sand layer between the silt layers was investigated and the cause of sedimentation of the sand layer under liquefaction conditions was analyzed.

2 Flume Experiments

2.1 Instruments and Equipment

The experimental wave flume consisted of three areas: a wave generation area, a bed area, and a wave elimination area (see Fig.1). The bed bottom was 0.6m lower than the wave generation and wave elimination areas. The size of the bed was 2.6m(L)×0.5m(W)×0.6m(H) and was used to hold the sediment. Wave heights and wave periods were measured using a WG-55 wave-height instrument (RBR Ltd., Canada). Samples of the initial sediment bed were obtained using a piston-type PVC sampler (Fig.2A). Particle size analysis of the sediments was performed us- ing sieving and hydrometer methods. The strength of the soil was tested using a miniature penetration instrument (Fig.2B).Glass tracer beads were uniformly dispersed using four sizes of glass beads, and the particle sizes were 2.6, 5.16, 7.64, and 9.3mm, and 300, 300, 150, and 100 particles, respectively, were used for 850 total glass particles.

Fig.1 Experimental flume.

Fig.2 Sampler and penetration instrument. A, Piston-type PVC sampler; B, Miniature penetration instrument.

2.2 Preparation of the Test Bed

The silt used in this experiment was soil from the Yellow River Delta (containing approximately 9% clay). The sand was obtained from the Qingdao coast.The silt was air dried, crushed, and then mixed with water in a mixer to form a uniform slurry containing 35% water. The sand used for the experiment consisted of coarse sand (1.0–2.0 mm), fine sand (0.125–0.25 mm), and Yellow River Delta silt mixed in a mass ratio of 3:3:4. Water was then added to obtain a 35% water content (see Table 1 for the physical indices of the initial bed sediments).

First, a supporting plate for sampling (with a size 0.4m×0.4m, the location is shown in Fig.3) was placed in the flume. The mixed silt slurry was then transferred slowly along a sloping panel into the sediment section of the flume to form a bottom silt layer with dimensions of 2.6m(L)×0.5m(W)×0.35m(H). The prepared sand was then gradually spread on top of the silt layer to establish a sand layer with a thickness of 0.15m. Finally, silt was placed on the sand layer to form the whole sediment bed with a height of 0.66m from the bottom. After this process was complete, the surface of the sediment bed was 0.06m higher than the wave generation and wave elimination areas,which was the reserved height necessary to allow the sedi- ment bed to subside. After the test bed was complete, water was added to the flume to a depth of 0.40m.

Table 1 Physical indices of the initial bed sediment

Fig.3 Test bed.

2.3 Experimental Procedures

After the test bed had settled in the flume for seven days, the test bed thickness was reduced to 0.63m. During the test, waves with heights of 8, 11, and 14.5cm were continuously applied for 43, 118, and 60min, respectively. In this process, the bottom of the bed did not liquefy, and the bed height decreased nearly 0.03m because of the vibration of the waves. At this time, the bed height was the same as the experimental design height. The wave load was then suspended, and a piston-type PVC tube was inserted into the bottom bed layer to obtain a sample, which was analyzed for particle size. The mechanical pro- perties of the sediment bed were tested using the penetration instrument (the test points are shown in Fig.3 with red dots). These results were used as the initial values of the bottom bed layer prior to the liquefaction test.

Subsequently, a wave with a height of 15.4cm (=1.02s) was applied to the bottom of the bed for the liquefaction test. However, this wave condition was not sufficient to liquefy the bed because of the long consolidation time of the bed. Therefore, at the periphery of the bed, the load on the bed was artificially increased by hammering of the bed surface, which resulted in bed liquefaction. Observation of the bed soil body fluctuations from the sidewall of the flume confirmed that the bed had been liquefied, and the liquefaction depth continued to develop downwards under the effect of the continuous wave. The glass tracer beads were evenly scattered across the glass bead placement area (Fig.3).

At the onset of the wave load, the clay on the surface of the sediment bed entered the water, and the water became cloudy. The sediments began to liquefy, beginning at the surface layer, and the liquefaction depth gradually increased. The upper body of water progressively became more turbid. As observed from the sidewall of the flume, the liquefied sediments in the liquefied bed oscillated at the same cycle frequency as the waves. The trajectory of the liquefied deposits was elliptical. The horizontal and vertical minor axes of the trajectory ellipse gradually decreased from top to bottom, and the liquefaction bottom only exhibited reciprocal movement in the horizontal direction. The sand layer also displayed a fluctuating movement during liquefaction; however, the sand layer appeared to sink during the liquefaction process. Thefluctuation boundary of the liquefied sediments observedthe sidewall of the flume was determined to be the visible liquefaction depth. During the liquefaction test, the soil bed reached a maximum liquefaction depth of 41cm in 435min.The visible liquefaction depth was shallower than the actual liquefaction depth because of the effect of the wall.The final scope of the observed liquefaction zone is shown in Fig.4 (the area above the orange line indicates the range). After reaching the maximum liquefaction depth, the liquefaction area began to gradually deposit sediment from bottom to top, although there was a continuous wave load. The wave was stopped at 768min. During the entire wave-enhancement process, the bottom bed was subjected to several mechanical property tests using the miniature penetration instrument as sediment liquefaction progressed (see Table 2 for the test time and Fig.3 for the test area).

Table 2 Time arrangement and test position of the penetration resistance test

Fig.4 Data marking of the sediment bed liquefaction test.

Two days after ending the wave load, the water in the flume was drained to air dry the bed surface. Two weeks later, the sediments in the flume were excavated, and a cubic sample (0.4m×0.4m×0.56m) was removed from above the supporting plate. Vertical profiles were cut in this cubic sample along the wave direction, the wave front, and the behind-wave surface. Stratum profile observation and layered sediment particle size analysis were performed. The distribution of the glass tracer beads was also examined.

3 Results

In this study, the penetration resistance of the sediment bed and particle size were analyzed using the cubic samples. During the wave flume test, the penetration resistance at each layer depth of the sediment bed was measured to indicate the position change of the sand layer. Particle size analysis was performed at each layer depth of the cubic sediment sample to elucidate the particle size changes in the silt and sand layers.

3.1 Penetration Resistance of the Test Sediment Bed

Before, after, and during the test, a sediment penetration resistance test (Fig.5, the shaded part of the figure is the sand layer) was performed in areas A and B of the flume (see Fig.3) from the bottom bed surface to a depth of 0.52m. The strength of a sand layer is generally greater than that of a silt layer; therefore, we can use this condition to distinguish the position of the sand layer. The penetration resistance tests of the two areas showed the following results.1) At the beginning of liquefaction, the surface sediment strength became zero. From the surface layer, the sediment strength gradually increased from zero and slowly increased with the depth. The sediment strength then decreased sharply after reaching its peak. Based on the different layers, the surface zero-strength layer was the upper silt layer and the middle high-strength layer was the sand layer. 2) In the liquefaction development stage (0–368min), the high-strength zone was a gradual downward trend, and the peak intensity increased at the same time. Therefore, the sand layer gradually subsided with the downward development of the liquefaction depth during the liquefaction process. 3) In the deposition return stage (386–768min), the liquefaction bottom began to move upwards after liquefaction reached the maximum depth, and the high-intensity area remained in the same layer. In addition, the sedimentation intensity continued to increase with time. At this point, the sand had settled to the lowest horizon, and the bottom particles stopped moving and redepositing.

3.2 Granularity Characteristics and Glass Tracer Bead at Each Layer Depth

When the cubic sediment samples (Figs.6A and B) were removed from the flume, the boundary between the sand and silt layers was clearly identified. In addition, a silt interlayer (two types of interlayers: horizontal and inclined) and distributed glass beads were observed in the sand layer.

Particle size analysis was performed every 20 mm from the top to the bottom of the cubic sample (see Fig.6C for the sampling approach). The particle size analysis results of the samples taken before liquefaction of the sediment bed and the post-test cubic samples were plotted against the depth (Figs.7A and B, respectively). The sediment bed after reconstitution exhibited notable changes as a result of the liquefaction process.

The overall height of the sediment bed was reduced from 0.6m to 0.56m. The depth of the uppermost silt layer changed from 47–60cm (from the bottom) to 32.9–56cm, and the thickness increased from 13cm to 23cm. The content of fine particles (silt and clay) decreased slightly from 88%–90% to 78%–88%, with the content in the surface layer decreasing by up to 12%. The fine sand content increased from 5%–7% to 9%–14%. The medium and coarse sand contents slightly decreased. The thickness of the sand layer in the middle varied little. The sand layer moved from a depth of 35–50cm from the bottom to 16.7–32.9cm and then moved downwards approximately 17–18cm. The content of fine particles (silt and clay) increased in the middle, corresponding to the observed fine-grained interlayer. In the middle of the sample, the fines content was high, lower in the upper and lower layers. Additionally, the fine sand content increased, and the medium sand content decreased, while the coarse sand content changed only slightly. The content of fine particles(powder-sized particles and clay) decreased in the silt layer at the bottom, and the fine sand content increased. The sediment bed particle analysis data collected after the test are summarized in Table 3.

The glass tracer beads were concentrated in the middle and lower parts of the sand layer (a total of 252, the spe-cific distribution is shown in Table 4). The lower parts of the sand layer and the larger glass beads reached a deeper depth.

Fig.5 Penetration resistance during wave loading.

Fig.6 Sediment sample. A, Side view; B, Front view; C, Sampling for particle size analysis.

Table 3 Physical indices of the bed sediment (after the test)

Fig.7 Particle analysis (A. Initial sediment bed; B. Reconstructed sediment bed).

Table 4 Glass bead distribution (9.5cm×10cm)

4 Analysis and Discussion

In this study, the settlement of sand particles from the upper layer (coarse-particle layer) in the silt layer (fine-particle layer) was investigated under the wave load during sediment liquefaction. Glass tracer beads, which have the same density as sand grains, helped to distinguish the sedimentation behavior of the coarse particles in the fine-particle layer.Before the test and after a week of consolidation of the static silty sand bed, the scattered glass beads were only slightly submerged and were not completely covered with silt. Additionally, when the wave load was applied, the deposited glass beads only settled to a depth of 1–4cm without sedimentation, and the beads did not sink any further.

Observations through the sidewall of the flume during the test indicated that the liquefied sediments fluctuated with the waves (see Xu, 2016, Fig.6). The settlement of sand particles and glass beads was the result of wave sorting of the liquefied sediments. Finally, the cause for the settlement of sand particles in the silt layer under liquefaction conditions were analyzed.

4.1 Causes of Sedimentation

Previous researchers examined the fluctuation of high-density liquefied soil using a wave flume test on a sediment bed with a thickness of 17.5cm (Sumer, 2006b).A double fluid fluctuation effect was formed by the liquefied sediment and upper water body, and a phenomenon of sand pattern formation on the surface of the liquefied sediment was reported.However, the fluctuation phase of the liquefied sediment (Sumer, 2006b)was completely the opposite of that of the upper water body, perhaps because of the smaller bed thickness and shorter test duration. In this study, the liquefied sediments exhibited the same wave motions under the effect of the upper wave, and the trajectory was circular or elliptical. The amplitude gradually decreased from the top to the bottom.

During the fluctuation, the cohesive forces between the sediment particles were weakened. Under the combined effects of self-gravity and other external forces (the inertia force caused by wave motion, the buoyancy caused by the upward movement of pore water, the pore water pressure, the shear force generated because of particle friction during motion,), the soil particles responded differently according to their own physical state (size, shape, specific gravity,) and produced different trajectories. The liquefaction sediment fluctuations caused the particles to exhibit an approximately elliptical cyclic motion and generate inertial forces. According to Stokes’ settling theory, there are different settlement resistances (differ-ences in gravity and buoyancy) depending on the particle size. The gravity of relatively fine particles is not sufficient to counterbalance the buoyancy caused by the upward drainage of pore water. As the pore water moves upward, the fine particles at the near surface of soil beds will directly enter the fluctuating water body. After losing the support of other particles relatively coarse particles, the gravity is greater than the buoyancy and the particles move downward.

Liquefaction sedimentation occurs because of fluctuations in the subsidence of coarse grains and rise of fine grains, resulting in sorting that leads to restratification. The particle size of the sand layer is large, while the particle size of the silt layer is small. The sand layer will sink in the fluctuating process of liquefaction. This phenomenon is the reason the test sand layer settled 17–18cm. The glass tracer beads settled on the surface of the sediment bed and reached the lower part of the sand layer through the upper silt layer. The sedimentation distance reached 44cm. Additionally, the coarse particles settled because of undulation of the liquefied sediment.

4.2 Two Speculative Sorting Patterns

The results of this experiment confirmed that in the fluctuation process of liquefied deposits, the sand layer dropped deeper, and the upper silt layer was thickened. In the cubic sample, a fine-grained interlayer appeared withinthe sand layer. This observation was in contrast to the storm liquefaction deposition results of Xu(2016). As a result, this study analyzed the two liquefaction deposition processes during storms. This fluctuation in deposition of the liquefied sediments may be because of the two sorting patterns that can occur at different locations.

4.2.1 Dispersive sorting pattern

This pattern describes the behavior of the coarse and fine particles in sediment liquefaction, where these particlesfluctuate independently and are sorted during the fluctuations.

Initially, during the development of liquefaction, sediment liquefaction begins in the upper silt layer and gradually develops downwards. A liquefaction interface develops within the sand layer, and the sand particles start to exhibit a cyclical elliptical motion because of the upper liquefied silt particles. Concurrently, the coarse and fine particles begin to differentiate. When the bottom of the liquefaction interface developed downward into the silt layer below the sand layer, the development speed of liquefaction of the silt layer was greater than that of the sand layer. The amplitude of the elliptical oscillation increased at the interface between the sand layer and underlying silt layer because of the deeper liquefaction boundary, making the relative vertical motion between the upper coarse particles and fine particles more intense. The sand particles individually overcame buoyancy and continued their downward movement along the trajectory of an elliptical spiral (Fig.8). In the entire sand layer, the sinking of numerous sand grains together with the upward movement of the fine particles of the lower silt layer resulted in the replacement of the sand layer and lower silt layer (Fig.9).

Then, during the restratification process, the fluctuating sediments began to be redeposited from the liquefaction bottom, and the liquefaction bottom gradually moved upwards again. After restratification of the liquefied lower silt layer, the liquefaction interface will rise, which will cause the fluctuation of the liquefied sediment in the upper part to weaken. When the liquefaction bottom reaches the depth of the settled sand layer, the pore water will dissipate quickly because the sand layer is more porous than the silt layer, and the sand layer undergoes rapid deposition. In this case, the fine particles from the lower silt layer moved upward through the sand layer and settled in the sand layer; as a result, the fine-particle content in the sand layer did not show a significant reduction (Figs.7 and 9).

4.2.2 Channel sorting pattern

In the sediment samples examined in this study, the silt interlayer in the sand layer was observed at different positions, always horizontal and inclined. The inclined silt interlayer was connected to the upper silt layer because of the channel sorting pattern process.

First, when a liquefaction interface develops in the silt layer below the sand layer, the sand layer has a relatively downward settling force, so that the lower liquefied silt layer bears a relatively high upper layer pressure.Since silt has a higher fluidity than sand after liquefaction, the lower liquefied silt is squeezed into the weak parts of the sand layer in the aggregated state because of the pressure, which opens an upward migration channel. As the liquefaction bottom continues to develop downward after channels are formed, the lower silt particles pass through the channel into the upper layer. As the sand layer moves down, the fluctuating silt layer under the sand layer be comes thinner, and the fluctuation capacity of the sediment decreases near the liquefied bottom interface. Therefore, the movement of the silt in the lower part of the sand layer into the sand layer will also be reduced.

Fig.8 Settlement pattern of coarse particles in the fluctuation of the liquefied sediment.

In the restratification process, the upward movement channel through the sand layer becomes filled with silt because of the rapid deposition of the whole sand layer, and the movement of the silt layer is stopped. The silt in the channel becomes locked in the sand layer, forming a silt interlayer (Fig.10).

Fig.9 Dispersive sorting pattern.

Fig.10 Channel sorting pattern.

5 Conclusion

Liquefaction of a coarse-grained sedimentary layerabove a fine-grained sedimentary layer occurs in coastal environments under the influence of storm waves.In this study, a wave flume test was used to investigate the sedimentation characteristics of a sand layer in a liquefied silt layer. The results of the sedimentation process of the sand layer were experimentally revealed. The causes of the sedimentation of the sand layer were also analyzed, and the fluctuation sorting patterns of the liquefied sediments based on the distributions of the coarse and fine layers was examined.

1) The liquefied sediment fluctuated under the wave load, and the sand layer that consisted of coarse particles in the upper part, settled into the lower silt layer. Additionally, the sand layer settled near the maximum liquefaction depth of the sediment.

2) Sedimentation of the sand layer in the liquefied silt layer was mainly caused by separation of the coarse and fine particles during the liquefaction process of the sediments, which exhibited fluctuations.

3) The presence of silt interlayers were dependent on the different depths of the settled sand layer. The dispersive and channel sorting patterns describe the fluctuating deposition process as a result of liquefied sediment sorting.

Acknowledgements

The funding for this project was provided by the National Natural Science Foundation of China (No. 41976049), and the Fundamental Research Funds for the Central Universities (No. 202061028). We thank engineer Lin Lin at the College of Geoscience and Engineering, Ocean University of China, for conducting the particle size analysis. We would also like to thank graduate students Mingchao Cao, Wenhao Chen and Xinzhi Wang for their help in sampling and sample processing.

Chang, C. H., Chien, L. K., and Chang, Y. H., 2004. 3-D liquefaction potential analysis of seabed at nearshore area., 12: 141-151.

Chang, W. J., Rathje, E. M., Stokoe, K. H., and Hazirbaba, K., 2007.pore-pressure generation behavior of liquefiable sand., 133: 921-931.

Chowdhury, B., Dasari, G. R., and Nogami, T., 2006. Laboratory study of liquefaction due to wave-seabed interaction., 132: 842-851.

Coleman, J. M., and Prior, D. B., 1981. Subaqueous sediment instabilities in the offshore Mississippi River Delta: Section 5. AAPG Special Volumes, 1-53.

Ishihara, K., and Yamazaki, A., 1984. Wave-induced liquefaction in seabed deposits of sand., 24: 85-100.

Kirca, V. S. O., 2013. Sinking of irregular shape blocks into marine seabed under wave-induced liquefaction., 75: 40-51.

Ko, Y. Y., and Chen, C. H., 2020. On the variation of mechanical properties of saturated sand during liquefaction observed in shaking table tests., 129: 105946.

Liu, B., and Jeng, D. S., 2015. Laboratory study for influence of clay content (CC) on wave-induced liquefaction in marine sediments., 34: 280-292.

Liu, B., Jeng, D. S., Ye, G. L., and Yang, B., 2015. Laboratory study for pore pressures in sandy deposit under wave loading., 106: 207-219.

Sassa, S., and Sekiguchi, H., 2001. Analysis of wave-induced liquefaction of sand beds., 51: 115-126.

Sassa, S., Takayama, T., Mizutani, M., and Tsujio, D., 2006. Field observations of the build-up and dissipation of residual pore water pressures in seabed sands under the passage of storm waves., 39: 410-414.

Sawicki, A., and Mierczyński, J., 2009. On the behaviour of liquefied soil., 36: 531-536.

Sekiguchi, H., Miyamoto, J., and Sassa, S., 2004. Progressive solidification of a liquefied sand layer during continued wave loading., 54: 617-629.

Sekiguchi, H., Sassa, S., and Miyamoto, J., 2001. Analysis of progressive liquefaction as a moving-boundary problem., 51: 847-857.

Sleath, J. F. A., 1970. Wave-induced pressures in beds of sand., 96: 367-378.

Sumer, B. M., Freds?e, J., Christensen, S., and Lind, M. T., 1999. Sinking/floatation of pipelines and other objects in liquefied soil under waves., 38: 53-90.

Sumer, B. M., Hatipoglu, F., and Hansen, N. E. O., 2006a. Critical flotation density of pipelines in soils liquefied by waves anddensityofliquefiedsoils., 132: 252-265.

Sumer, B. M., Hatipoglu, F., Freds?e, J., and Sumer, S. K., 2006b. The sequence of sediment behaviour during wave-induced liquefaction., 53: 611-629.

Teh, T. C., Palmer, A. C., Bolton, M. D., and Damgaard, J. S., 2006. Stability of submarine pipelines on liquefied seabeds., 132: 244-251.

Tsui, Y., and Helfrich, S. C., 1983. Wave-induced pore pressures in submerged sand layer., 109: 603-618.

Tzang, S. Y., 1998. Unfluidized soil responses of a silty seabed to monochromatic waves., 35: 283-301.

Tzang, S. Y., and Ou, S. H., 2006. Laboratory flume studies on monochromatic wave-fine sandy bed interactions: Part 1. Soil fluidization., 53: 965-982.

Wang, J. T., Salam, S., and Xiao, M., 2020. Evaluation of the effects of shaking history on liquefaction and cone penetration resistance using shake table tests., 131: 106025.1-106025.14.

Wang, X., Xu, G., Sun, Y., Hu, G., Yu, Y., and Lin, L., 2013. Storm-waves seabed sediment liquefaction and re-stratificationon the Yellow River subaqueous delta., 33: 29-40 (in Chinese with English abstract).

Xu, G., Liu, Z., Sun, Y., Wang, X., Lin, L., and Ren, Y., 2016. Experimental characterization of storm liquefaction deposits sequences., 382: 191-199.

Xu, G., Sun, Y., Wang, X., Hu, G., and Song, Y., 2009. Wave-induced shallow slides and their features on the subaqueous Yellow River Delta., 46: 1406-1417.

Xu, G., Sun, Y., Yu, Y., Lin, L., Guang-Hai, H. U., Zhao, Q. P.,, 2012. Discussion on storm-induced liquefaction of the superficial stratum in the Yellow River subaqueous delta., 14: 80-89.

Xu, Z., Xu, G., Chen, C., Xu, X., Ren, Y., Ma, C.,, 2018. The rate characteristics of wave-induced liquefaction process of the silty soil in the seabed., 48: 115-122 (in Chinese with English abstract).

Zen, K., and Yamazaki, H., 2008a. Mechanism of wave-induced liquefaction and densification in seabed., 30: 90-104.

Zen, K., and Yamazaki, H., 2008b. Oscillatory pore pressure and liquefaction in seabed induced by ocean waves.,30: 147-161.

Zhang, J. M., Zhang, J. H., Wang, G., and Chen, Y., 2009. Safety evaluation of breakwaters based on physical and numerical modelling., 36: 852-862.

Zhang, W. G., Han, L., Feng, L., Ding, X. M., Wang, L., Chen, Z. X.,, 2020. Study on seismic behaviors of a double box utility tunnel with joint connections using shaking table model tests., 136: 106118.

. E-mail: xuguohui@ouc.edu.cn

June 9, 2020;

December 8, 2020;

January 4, 2021

? Ocean University of China, Science Press and Springer-Verlag GmbH Germany 2021

(Edited by Xie Jun)