Sedimentary stratigraphy of natural intertidal flats with various characteristics

Soils and Foundations 2012;52(3):411–429 The Japanese Geotechnical Society

Soils and Foundations www.sciencedirect.com journal homepage: www.elsevier.com/locate/sandf

Sedimentary stratigraphy of natural intertidal flats with various characteristics Yoichi Wataben, Shinji Sassa Port and Airport Research Institute, 3-1-1, Nagase, Yokosuka, Japan Available online 4 July 2012

Abstract Intertidal flats are key elements in marine environments; they provide regions of rich bioactivity and also have the function of water purification. In the present study, the shear wave velocity structures of intertidal flats were explored by a surface wave method called the multichannel analysis of surface waves (MASW), which is a very useful method for surveying the sedimentary stratigraphy of various intertidal flats, in order to evaluate the cross-shore and the along-shore sedimentary stratigraphy. Field surveys were carried out in sandflats, mudflats, and subtropical tidal flats during spring tides in summer, when these flats were exposed. In the sandflats, the sedimentary stratigraphy evaluated by MASW (as distributions of the shear wave velocity/stiffness) and by physical soil tests (as distributions of the void ratio) can be explained as being a consequence of the cyclic elastoplastic contraction of the soils that are subjected to a variety of suction dynamics under the tide-induced groundwater table fluctuations. In the mudflats, the soil stiffness was extremely homogeneous, although a variation in the void structure was observed in association with the grain-size distributions, because suction did not develop in the soil. In the subtropical intertidal flats, lime rock was clearly identified beneath the soil sediments. A spectrum analysis in the cross-shore direction was carried out in order to quantitatively characterize the dominant wavelength for the variations in both shear wave velocity and morphological ground surface profiles of intertidal flats with various scales and soil types. In the intertidal flats with a multibar-trough structure, dominant wavelengths in the range 40–90 m were obtained corresponding to the unit length of the bar-trough structure, while in the mudflat or the huge-scale homogeneous sandflat, no dominant wavelength was obtained. In the subtropical tidal flats, the dominant wavelength was strongly influenced by protruding lime rock rather than by the morphological ground surface profile. & 2012 The Japanese Geotechnical Society. Production and hosting by Elsevier B.V. All rights reserved. Keywords: Tidal flat; MASW; Stiffness; Sedimentary stratigraphy; Suction; Groundwater level

1. Introduction Intertidal flats are key elements in marine environments; they provide zones of rich bioactivity and also serve as n

Corresponding author. E-mail addresses: [email protected] (Y. Watabe), [email protected] (S. Sassa). Peer review under responsibility of The Japanese Geotechnical Society.

water purifiers. Most of the previous studies that have challenged issues associated with intertidal flats have done so on the basis of the following: a biological approach (e.g., Kuipers et al., 1981; Ellison, 1984), in which the major concern was the food chain of the inhabitant organisms, a hydrological approach (e.g., Jarvis and Riley, 1987; Stevenson et al., 1988; Perillo and Sequeira, 1989; Le Hir et al., 2000), in which the major concern was the sediment transportation during the period of submergence, and a chemical approach (e.g. Widdows et al., 2000), in which the major concern was the chemical mass balance in the seawater and/or the pore water. Although

0038-0806 & 2012 The Japanese Geotechnical Society. Production and hosting by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.sandf.2012.05.003

412

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

the most important characteristic of intertidal flats is the period of exposure, the authors of the above studies have mainly focused their attention on the period of submergence. Thus, such studies have mainly treated intertidal flats as being equivalent to the sea bottom, not the exposed ground. The present study approaches these issues by using a geotechnical and geoenvironmental methodology, which enables a direct investigation of the intertidal flat soils that are the primary habitats for benthic life/activity. In surveys of marine soil stratigraphy, some useful techniques, such as core sampling (e.g., Hori et al., 2001; Hasiotis et al., 2005; Heyvaert and Baeteman, 2007), seismic reflection (e.g., Terrinha et al., 2003; Hasiotis et al., 2005; Mienis et al., 2006; Schlu¨ter and Uenzelmann-Neben, 2007), acoustic reflection (e.g., Mienis et al., 2006; Hutri and Kotilainen, 2007), and ground-penetrating radar (e.g., Neal et al., 2002) have been applied. Some of these techniques require a sea vehicle (e.g., a survey ship and sidescan sonar) and are applicable to the investigation of the submerged sediment stratigraphy with a sufficient water depth. Many other techniques require a high level of trafficability for the equipment. The important points in the identification and the description of tidal flat stratigraphy are the portability of the equipment, the mobility of the survey, the spatial continuity and the high accuracy of the data set, and the low impact on marine life. For example, core sampling is the most accurate technique; however, the obtained data is present as stratigraphic succession at discrete points. On the other hand, acoustic reflection and seismic reflection are available for obtaining spatially continuous data sets; however, the equipment used for these techniques is not portable. With this in mind, the authors have explored the capability of a surface wave method, called the multichannel analysis of surface waves (MASW) (Park et al., 1999; Hayashi and Suzuki, 2004), proving that MASW is an efficient and versatile method for investigating intertidal sedimentary stratigraphy (Watabe and Sassa, 2008). The purpose of this study is threefold, namely, to evaluate the sedimentary stratigraphy of natural intertidal flats with various characteristics, to gain insights into the formation mechanisms of such sedimentary stratigraphy, in light of the relevant geoenvironmental dynamics, and to characterize the morphological features of cross-shore sedimentary stratigraphy. Indeed, the authors were motivated by recent field observations (Sassa and Watabe, 2005, 2007) in an intertidal sandy flat. The continuous measurements closely captured the dynamics of suction in the shallow soils under the tide-induced groundwater table fluctuations. Suction is defined by s ¼ ua 2uw

ð1Þ

where ua is the atmospheric air pressure and uw is the pore water pressure. Sassa and Watabe (2007) have demonstrated that such suction dynamics plays a substantial role in the temporal and the spatial evolution of voids, stiffness, and

surface shear strength in cyclically exposed and submerged soil. Furthermore, the state of suction in association with the groundwater level is found to be closely linked to the performance of benthic activity (Sassa and Watabe, 2008). The outputs from the MASW exploration were the spatial distributions of the shear wave velocities vs, which were used to assess the stiffness of the intertidal flat soils. Field surveys were firstly conducted in a typical sandflat and a typical mudflat during spring tides in summer, when these flats were exposed, and the formation mechanisms of the sedimentary stratigraphy that were detected by MASW are discussed with regard to the relevant geoenvironmental dynamics. Then, MASW was applied to intertidal flats with various scales and soil types. A spectrum analysis in the cross-shore direction was carried out in order to quantitatively characterize the dominant wavelength for the variations in both shear wave velocity structures and morphological ground surface profiles of intertidal flats with various scales and soil types. 2. Geoenvironmental dynamics and geomorphology in intertidal flats The authors have performed field observations (Sassa and Watabe, 2005) in the Banzu intertidal sandflat, Kisarazu City, Chiba Prefecture, Japan and have demonstrated the nature of the geoenvironmental dynamics in the soil surface layer, which involve coupled processes including evaporation, suction development, upward moisture movement, and salinity accumulation. The temporal variations in the groundwater level are shown in Fig. 1(a). The astronomical tide level is plotted for the purpose of comparison. During periods of submergence, the measured fluctuations in the water level essentially coincide with the fluctuations in the astronomical tide level. By contrast, during periods of exposure, there were apparent decays and phase lags in the measured variations in the groundwater level in comparison with those of the astronomical tide level. This observation implies that the fluctuations in the groundwater table significantly attenuate in the offshore–onshore direction, which is attributed to both the hydraulic conductivity of the soil and the dimensions of the intertidal flat. The associated variations in suction at three different soil depths – 10, 100, and 200 mm below the ground surface – are shown in Fig. 1(b). The shapes of these three curves are essentially the same, but are vertically shifted to correspond to the soil depths. The ratio of the increase in suction to the decrease in groundwater level was found to be essentially equal to the unit weight of water gw, indicating that the soil in an intertidal flat always remains saturated. In fact, during the period of the field observations, the measured volumetric water content was constant. The saturated state is important to understanding the characteristics of suction development in intertidal flat soils. This is the main difference between intertidal sandflats, which have rich bioactivity,

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

Depth of G.W.L. (m)

and sandy beaches, which have poor bioactivity in their unsaturated state. Such internal geodynamics under ebbing and flooding tides have been shown to play important roles in geomorphology, as described in Sassa and Watabe (2007). Namely, the cyclic

0.0 0.1 0.2 0.3 0.4

Observed

Astronomical tide

0.5 0.6

Submergence

Exposure

413

variations in suction, with observed magnitudes of about 2.5 kPa at the soil surface (2.0–3.0 kPa in many cases), eventually caused marked contrasts in the soil densification of the zones above the groundwater level in the cross-shore direction, causing characteristic changes in the morphological profile. Here, the groundwater level is defined as the depth corresponding to zero suction. The elastoplastic theoretical model proposed by the authors (Sassa and Watabe, 2007) explained well the temporal variation in the void structures observed in a model experiment; typically, severe suction fluctuations caused soil with an initial void ratio (e) of 1.1 to eventually become densified to 0.9. This magnitude of soil densification is large enough to bring about substantial changes in soil stiffness (Lohani et al., 2001). In fact, based on the vane shear tests conducted in the laboratory on specimens of reconstituted Banzu sand with different void ratios (0.73–1.01), a decrease in void ratio of 0.2 corresponds to a threefold increase in the vane shear strength under the same suction conditions (Sassa and Watabe, 2007).

Suction s (kPa)

3 G.L.–10mm

2

3. MASW—geophysical exploration

G.L.–100mm

1 0 -1 G.L.–200mm

-2 -3 0

60

120

180 240 Time t (min)

300

360

420

Fig. 1. Temporal variations in (a) water/groundwater level and (b) suction at three different depths (10, 100 and 200 mm below the ground surface) at a distance of approximately 400 m from the shoreline in the Banzu intertidal flat (approximately 700 m south of the MASW survey array to be described). The observation was performed on May 25, 2005. The dotted line in (a) represents the astronomical tide level.

MASW is a seismic method for geophysical site investigations and is applied here to explore the internal sedimentary stratigraphy of the soil deposits in intertidal flats. The measurement principle of MASW and the instruments (seismograph: Model McSEIS-SXW, OYO Corporation, Tokyo, Japan and geophone: Model GS11D, OYO Geospace Corporation, Houston, Texas, USA) used for the purpose of the present study are schematically illustrated in Fig. 2. In the field, a land streamer was used (Inazaki, 1999) comprising 24 geophones aligned in a series with an interval length of one meter on the ground surface. Before each measurement, the land streamer was moved by

Triggering point (wooden hummer) Array of geophones at 1-m intervals connected to the “land streamer”.

pull

short wavelength (high frequency)

Rayleigh waves

long wavelength (low frequency)

Fig. 2. Measurement principle of multichannel analysis of surface waves (MASW) and the apparatus for investigating the stratigraphy of the intertidal flats. The geophones are connected to a land streamer at intervals of 1 m. Rayleigh waves with various frequencies generated by a triggering shock from a wooden hammer were received by the geophones.

414

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

pulling it for 2, 4, 5, or 10 m, depending on the conditions. A large portable wooden hammer was used to impose a vertical impact on the soil surface, thereby triggering Rayleigh waves in the soil. In the tidal flat soils, the high-frequency, short-wavelength waves propagated in the shallow soil region, while the low-frequency, long-wavelength waves propagated in a soil region that included both shallow and deep parts, following the characteristics of Rayleigh waves. The common mid-point (CMP) cross-correlation analysis (Hayashi and Suzuki, 2004) was applied to shot gathers and CMP cross-correlation (CMPCC) gathers were calculated in order to improve the lateral resolution of shear wave velocity profiles. The MASW (Park et al., 1999) was applied to each CMPCC gathers, and phase velocity images were obtained in the frequency domain. The MASW enables the direct calculation of the multi-modal dispersion curves from multi-traces or cross-correlations. The phase velocities were determined as the maximum amplitude in each frequency so that the dispersion curve (relationship between the phase velocity and the frequency) could be constructed. Low and high frequencies corresponded to the deep and the shallow regions of the soil layer, respectively, as mentioned above. A non-linear least squares method was applied to each dispersion curve in order to reconstruct the one-dimensional shear wave velocity profile. An initial model was generated by a simple wavelength-depth conversion (Xia et al., 1999). The number of layers was fixed at 15 and only shear wave velocities were varied throughout the reconstruction. Primary wave velocities and densities automatically changed based on the empirical relationship (Kitsunezaki et al., 1990; Ludwig et al., 1970). The theoretical dispersion curves were calculated by a matrix method (Saito and Kabasawa, 1993). The two-dimensional shear wave velocity structure was obtained by aligning the one-dimensional depth profiles of the shear wave velocity. This resulted in the shear wave velocity structure along each survey array. Typical data sets obtained at a stiff part of the sandflats, a soft part of the sandflats, a soft part of the mudflats, and a soft part of the sand-mud layered flats are shown as wave patterns, dispersion curves, and shear wave velocity structures in Watabe and Sassa (2008). Shear wave velocity vs can be converted to initial (maximum) shear modulus G0 by using G0 ¼ rt  v2s

ð2Þ

where rt is the bulk density. Since the shear modulus is proportional to the square of the shear wave velocity, the shear wave velocity structure represents the soil stiffness structure. Note that, in this regard, high surface wave velocities correspond to high soil stiffness. The outputs from the MASW survey are the spatial distributions of the shear wave velocities in the intertidal flat soils, which are used to assess their stiffness structures.

4. Typical sedimentary stratigraphy of tidal flats and its formation mechanisms 4.1. Typical intertidal flats investigated Typical sedimentary stratigraphy of a sandflat and a mudflat was characterized using MASW. The locations of the flats investigated are illustrated in Fig. 3. The Banzu intertidal flat is located on the east coast of Tokyo Bay in Kisarazu City, Chiba Prefecture, Japan. It is a sandflat located in the estuary of the Obitsu River. The hydrological, morphological, and ecological characteristics are well-known from prior studies; e.g., Uchiyama (2007), Kuwae et al. (2003), and Sassa and Watabe (2005, 2007, 2008). MASW was performed in the cross-shore direction along array A and in the along-shore direction along array B during ebbing tide from August 16 to 18, 2004 (Watabe and Sassa, 2008). The diurnal difference between high tide and low tide was about 1.6 m during the day. The Shiranui intertidal flat is located in the closed-off section of Yatsushiro Bay, Uki City, Kumamoto Prefecture, Japan. It is a mudflat in the estuary of the Ohno River. Since the Shiranui Polder was reclaimed in the 1960s, the muddy surface layer has gradually thickened. It is now impossible to walk on the flat because of the softness of the muddy soil; therefore, a special paddle boat designed for mudflats (Fig. 3(d)) was used. MASW was performed in the crossshore direction along arrays A and B and in the along-shore direction along array C during ebbing tide from August 18– 20, 2005 (Watabe and Sassa, 2008). The diurnal difference between high tide and low tide was about 4.0 m during the day. 4.2. Sedimentary stratigraphy of the typical intertidal flats The shear wave velocity structures were explored for the cross-shore and the along-shore intertidal flat soils using MASW. The phase velocity profiles were calculated so as to accurately reproduce each dispersion curve. This showed the shear wave velocity structure along each line of the survey. Since the shear modulus is proportional to the square of the shear wave velocity, the shear wave velocity structure represents the above-mentioned soil stiffness structure. 4.2.1. Shear wave velocity structures of the Banzu intertidal flat (sandflat) The shear wave velocity structure in the cross-shore direction along array A is shown in Fig. 4. The indicated distances correspond to the distance from the shoreline; the region from  5 to  30 m represents the reed bed in the backshore, and the region from 0 to 1100 m represents the intertidal zone. Surface layers with shear wave velocities of less than 60 m/s correspond to very soft regions, where it would be difficult to walk and where each step would cause a depression of about 50 mm in the sand. In this region, there are many burrows of benthic fauna on the soil surface. Offshore, in the region from 800 to 1100 m, distinct contrasts are visible between stiff bars, corresponding to high

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

415

Fig. 3. Locations of the intertidal flats investigated as a typical sandflat and a typical mudflat: (a) the Banzu intertidal flat (sandflat) in Kisarazu City, Japan, (b) scene at a region of soft sand around a distance of 300 m along array A, (c) the Shiranui intertidal flat (mudflat) in Uki City, Japan with contours of the soil surface elevation obtained by bathymetry and (d) the special paddle boat used in the field work in the mudflat.

Fig. 4. Shear wave velocity structures obtained by the MASW survey in the cross-shore direction along array A and in the along-shore direction along array B in the Banzu intertidal flat.

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

416

Fig. 5. Typical scene of multibar-trough structure on the offshore side at the Banzu intertidal flat.

400

Soil surface 1m

350

Ground water level 200m

Phase-velocity (m/s)

300 Fig. 7. Profiles of the morphological ground surface and the groundwater level in the cross-shore direction in the Banzu intertidal flat. The groundwater level was observed just at the time of the lowest tide on that day.

250 200 150 Bar 100 Trough

50 0 0

10

20 30 Frequency (Hz)

40

50

Fig. 6. Dispersion curves for the offshore bar-trough regions at the Banzu intertidal flat. The dispersion curves are clearly divided into two groups corresponding to stiff bars and soft troughs.

velocities, and soft troughs, corresponding to low velocities. These contrasts between bars and troughs compare well with a typical scene of the bar-trough region at the Banzu intertidal flat, as shown in Fig. 5. Typical examples of the dispersion curves for an offshore bar-trough region in the Banzu intertidal flat are shown in Fig. 6. It is interesting to note that the dispersion curves are clearly divided into two groups that represent stiff bars and soft troughs, which correspond to shear wave velocities greater than 110 m/s and less than 80 m/s, respectively, at the ground surface. Based on an empirical approximation of the surface waves, the wavelength, which is calculated from the ratio of the wave velocity to the frequency, was found to coincide with triple the corresponding soil depth. The point of convergence of the two groups (25 Hz and 130 m/s) corresponds to a depth of 1.7 m. Thus, the marked difference in stiffness in the bar-trough region was limited to the shallow layer above this depth. In other words, the soils were essentially homogeneous below this depth (see Fig. 4). Also, there is field evidence that the bar regions are generally stationary, as theoretically explained by Sassa and Watabe (2009). Since this site is subjected to the net deposition, as observed by Uchiyama (2007), the stiff region up to a depth of 1.7 m is evidence of a stationary densified zone.

The shear wave velocity structure in the along-shore direction along array B is also shown in Fig. 4. Soil surfaces with shear wave velocities of approximately 50 m/s correspond to regions where the fine-particle content (particles smaller than 0.075 mm in diameter) is greater than 20%. The soil structure is very horizontal, despite the presence of the creek. Since the bulk density rt of the soil is 1900 kg/m3 in the stiff regions and 1800 kg/m3 in the soft regions, shear wave velocities vs of 120 m/s in a bar, of 60 m/s in a trough, and of 50 m/s in the soft regions near the shoreline, can be converted into shear moduli G0 of 27.4, 6.5, and 4.5 MN/m2, respectively, using Eq. (2). An example of the offshore soil surface profile in the crossshore direction is shown in Fig. 7 with the associated groundwater level profile. The section corresponds to a line approximately 100 m south of array A. The groundwater level was just beneath the morphological ground surface, except in the trough regions where the water level was just above the soil surface. The photograph in Fig. 3(b) confirms this, showing the groundwater on the soil surface during the period of exposure when we walked on the exposed soil at a distance of around 300 m along array A. The water levels of 0.1–0.3 m beneath the soil surface in the bar regions generate suctions of 1–3 kPa, causing changes in the level of the ground surface. Indeed, semidiurnal fluctuations in the tidal water level bring about cyclic elastoplastic contraction and elastic expansion of the surface soil. This results in an accumulated contraction in the soil, as indicated in Sassa and Watabe (2007), since the elastoplastic contraction is significantly larger than the elastic expansion in the soil. The magnitude of such contraction behavior depends largely on the state of the suction in the bar and the trough regions, giving rise to marked contrasts in the void ratios of the soil. In fact, although the grain-size distributions were essentially the same, as shown in Fig. 8, the values for the void ratio were measured in soils at the surface (to the depth of 300 mm) and the values obtained were about 0.9 at the

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

4.2.2. Shear wave velocity structures of the Shiranui intertidal flat (mudflat) The shear wave velocity structure in the cross-shore direction along arrays A and B, and that in the along-shore direction along array C, are shown in Fig. 10. The structure is very homogeneous, indicating a uniform distribution in the horizontal direction and a mild increase with depth. This trend contrasts sharply with that in the sandflats. The shear wave velocity was less than 50 m/s on the 2 m surface soil and was greater than 100 m/s below depths of 5–6 m. Since the bulk density rt of the soil is 1400 kg/m3, a shear wave velocity vs of 40 m/s corresponds to a shear modulus G0 of 2.2 MN/m2. The temporal variation in the suction observed just beneath the ground surface in the Shiranui intertidal mudflat, during the period of exposure in summer, is shown in Fig. 11. The suction values are in the range 0–0.1 kPa, indicating that the moisture tension on the ground surface is slightly increased under the strong sunlight of summer, even in the mudflat. Since the suction value of 0.1 kPa is equivalent to a hydraulic head of only 0.01 m, it can be said that during periods of exposure, the groundwater level in the natural mudflat remains essentially at the ground surface. In Fig. 3(c), the contours of the soil surface are represented on the map. Due to the very flat ground surface with a slope of about 1/700, the very low hydraulic conductivity in the order of 10–8 m/s, and the very flat horizontal soil surface profile, the groundwater level coincided with the soil surface during low tide, as is visible in Fig. 11. This fact can also be confirmed from Fig. 3(d), which is a scene of the MASW survey at the Shiranui intertidal flat. The influence of the fluctuation in the tidal

Percentage finer in weight (%)

bars and about 1.1 at the troughs, as shown in Fig. 9. The profile of relative soil density Dr, defined as (emax–e)/(emax– emin), shows this aspect of the soil behavior more clearly. The observed difference in the void ratios of the bars and the troughs was found to be consistent with the results of the theoretical, experimental, and field investigations described in Sassa and Watabe (2007). The magnitude of the stiffness/shear wave velocity distribution obtained by the MASW survey (Fig. 4) is consistent with the magnitude of the above-mentioned strength/stiffness variation corresponding to void ratios of 0.9 and 1.1. Therefore, such morphology in the bar-trough region can be consistently explained by the above-mentioned results in which suction plays an important role.

100 Bar Trough

80 60 40 20 0 0.01

0.1 1 Soil particle diameter (mm)

417

10

Fig. 8. Grain size distributions in the bar-trough regions at the Banzu intertidal flat. The 300 mm surface soil samples were tested in the laboratory.

B: Bar, T: Trough G.L. & G.W.L.

B T 100m

B T

B T

W.L./ G.W.L

B T

1m

T

Void ratio e

1.4 1.2

emax

1.0

en

0.8

Relative Density Dr (%)

0.6

emin

80 60 40

Fig. 9. Profiles of (a) the morphological ground surface, (b) the soil void ratio and (c) the soil relative density in the cross-shore direction at the Banzu intertidal flat. The 300 mm surface soil samples were tested in the laboratory to obtain natural void ratio en and the maximum and minimum void ratios (emax and emin).

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

418

Depth z (m)

0

Array A (Shore Offshore)

2

Soft mud

4 6

vs(m/s) 200 500 400

Very soft mud

300

100

Sand

50 0

100 200 Distance x (m)

200

300

Array B (Shore Offshore) Depth z (m)

0 Very soft mud 2 Soft mud 4 Sand 6 0

Depth z (m)

0

100

200

300 400 Distance x (m)

500

600

700

600

700

Array C (North South) Very soft mud

2

Soft mud

4 Sand 6 0

100

200

400

300

500

Distance x (m) Fig. 10. Shear wave velocity structures obtained by the MASW survey in the cross-shore direction along arrays A and B and in the along-shore direction along array C in the Shiranui intertidal flat.

Suction s (kPa)

0.4 0.2 0.0 -0.2 -0.4 0

50

100 150 Time t (min)

200

Fig. 11. Temporal variation in the suction observation just beneath the ground surface at the Shiranui intertidal flat during the period of exposure in summer.

water level is limited to a very thin layer that is less than 0.01 m from the soil surface; consequently, suction does not develop in the soil, resulting in a very soft and homogeneous soil structure, as indicated above. Fig. 12 shows the distribution of the (a) water content, (b) void ratio, and (c) grain-size fractions for the 50 mm surface soil. The void ratio decreased significantly from about 5.0 to about 2.5 in the offshore direction due to the associated increase in the sand fraction. This increase in the offshore direction indicates that the sediment transportation is mainly governed by the waves and the current rather than by the river flow, because large and fine particles generally deposit upstream and downstream, respectively. As shown in Fig. 12(a), the consistency at the soil surface indicates that

normalized water content w/wL was in the range 2.5–3.0 over the whole area. Based on the fact that the vane shear strength of the clay in the slurry state is a unique function of w/wL, this value corresponds to a vane shear strength in the order of 0.01 kPa (Leroueil et al., 1983). Mud in this condition is normally in the slurry state, which cannot support marine life (gobies and crabs) on the surface. In the field, however, the soil had enough bearing capacity to support such life, suggesting the involvement of additional strength generated by chemical reactions. In fact, the temperature during summer was greater than 40 1C; this could have possibly accelerated the chemical reaction (bonding) between the soil particles. The temperature profile observed in the field is shown in Fig. 13; it indicates a temperature greater than 35 1C at the soil surface during periods of exposure. The vane shear strength of the surface layer was measured at depths of 0.04, 0.1, 0.2, and 0.4 m, and its profile was obtained at distances of 0, 570, and 1100 m along array B, as shown in Fig. 14(a). Here, the dimensions of each vane blade were 20 mm in width, 40 mm in height, and 0.8 mm in thickness. The rotational shearing rate of the vane in the sediment was maintained at 11/s. This vane apparatus is also used in the following sections. The strength at a depth of 0.04 m was about 0.4 kPa, and it increased linearly with depth with the ratio varying from 2.0 kPa/m near the shore to 8.0 kPa/m offshore, indicating that the ratio increases with the sand fraction. The vane shear strength measured in the

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

6

2

cross-shore

along-shore

4 3 cross-shore

2

1 1 0

0 0

Percentage finer in weight (%)

5

along-shore

3

Void ratio e

Normalized water content w/wL

4

419

100

100

200 Distance x (m) sand

80

sand

silt

300

0

100

200 Distance x (m)

300

silt cross-shore along-shore

60 clay

40 clay

20 0 0

100

200 Distance x (m)

300

Fig. 12. Planar distributions of (a) normalized water content w/wL (natural water content w and liquid limit wL), (b) void ratio e and (c) grain size fractions (sand, silt and clay fractions) in the cross-shore and along-shore directions in the Shiranui intertidal flat.

Temperature T (°C) 20 -0.1

25

30

35

40

20 mm above ground surface

0

Depth z (m)

0.1 0.2 0.3 0.4 0.5 0.6 Fig. 13. Depth profile of the soil temperature obtained at a distance of 0 m along array B in the Shiranui mudflat. The measurement was performed at the lowest tide level on August 24, 2006.

field (0.4 kPa) is more than 10 times larger than that expected from the w/wL value (0.01 kPa). A typical example of the G0 profile calculated with Eq. (2), from the shear wave velocity structure obtained at a distance of 570 m by MASW, is shown in Fig. 14(b). Here, bulk density rt was assumed to be 1400 kg/m3. We obtain the best fit increase ratio for G0 as 2.2 MPa/m. Since it is known that G0 increases linearly with undrained shear strength cu, G0 is

expressed as being approximately 300 (E2200/8¼ 275) times the value of cu from the comparison between the G0 profiles obtained by both the MASW and the vane shear strength measurements. The factor of 300 is consistent with the values indicated by D’Appolonia et al. (1971). 5. Morphological features of intertidal flats with various characteristics 5.1. Intertidal flats investigated Six intertidal flats, including the above-mentioned two flats, are studied here. The MASW survey arrays for the additional four flats are shown in Figs. 15 and 16, with typical scenes of the surveys. The Buzen Sea intertidal flat (Usa City, Oita Prefecture, Japan) and the Roberts Bank intertidal flat (Delta City, BC, Canada) are studied as representing the sandflats, while the Awase intertidal flat (Okinawa City, Okinawa Prefecture, Japan) and the Naha Airport intertidal flat (Naha City, Okinawa Prefecture, Japan) are studied as representing the subtropical flats. The field surveys were carried out during the spring tides in summer when these flats were exposed. The MASW survey was performed during the ebbing tide from August 9 to 11, 2006 in the Buzen Sea intertidal flat (the diurnal difference between high tide and low tide was approximately 3.3 m during the day), from July 27 to 29, 2007 in the Roberts Bank intertidal flat (the diurnal difference was approximately 3.5 m), from March 9 to 11, 2005 in the Awase intertidal flat (diurnal difference was approximately

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

420

Vane shear strength f (kPa) 0

1

2

3

4

5

G0 (kPa) 6

0

0.2

x=0m x=570m x=1100m x: distance along array B

10000

15000

20000

0.3

0.4

0.5

2 Depth z (m)

Depth z (m)

0.1

5000

0

0

4

6

8

Fig. 14. Depth profile of (a) the vane shear strength of the surface soil layer at 0.04, 0.1, 0.2 and 0.4 m depths at distances of 0, 570 and 1100 m and (b) a typical example of a G0 profile calculated with Eq. (2) from the shear wave velocity structure obtained by the MASW survey at a distance of 570 m, along array B in the Shiranui intertidal flat.

1.7 m), and from June 2 to 6, 2008 in the Naha Airport intertidal flat (diurnal difference was approximately 2.3 m). The Buzen Sea intertidal flat is a sandflat located in the estuary of the Yakkan River. A multibar-trough structure developed in both the cross-shore direction with a long interval and the along-shore direction with a short interval. There are some offshore regions with cobble stones. The MASW survey was carried out in the cross-shore direction along array A, with a length of approximately 1100 m, and along array C, corresponding to a bar-trough unit with a length of approximately 45 m on array A, and in the along-shore direction along array B with a length of approximately 500 m (Fig. 15(a)). The Roberts Bank intertidal flat is a sandflat located in an estuary of the Frazer River. The scale of the intertidal zone in the cross-shore direction is more than 4000 m. In an area near the shore, muddy deposits are dominant. The MASW survey was carried out in the cross-shore direction along array A with a length of approximately 3000 m (Fig. 16(a)). The end of array A was a large water path route. The Awase intertidal flat is a subtropical sandflat. The MASW survey was carried out in the cross-shore direction along array A with a length of approximately 850 m (Fig. 15(e)). The deposits are composed of sandy soil with coral gravel in the near-shore region, coral gravel with sandy soils in the middle region, and sandy soil in the offshore region. The main contents of the sand are crushed coral and shells. The Naha Airport intertidal flat is a subtropical sandflat with coral gravel. In some regions, lime rock is exposed at the ground surface. The MASW survey was carried out in the along-shore direction along array A with a length of approximately 1350 m and along array C with a length of approximately 23 m, and in the cross-shore direction along array B with a length of approximately 1400 m (Fig. 15(c)). Array C was approximately 110 m apart from array A. There is a submerged region between arrays A and C.

5.2. Shear wave velocity structure of the Buzen sea intertidal flat The shear wave velocity structures, obtained along the arrays, are shown in Fig. 17. The morphological ground surface profile data leveled along the arrays are reflected in the figure. As mentioned above, the shear wave velocity structure essentially represents the stiffness structure. Along the arrays, a multibar-trough structure was clearly captured. It is notable that array C captured the details of a unit of bar-trough structure. The interval of the multibar-trough structure in the cross-shore direction along array A is longer than that in the along-shore direction along array B. Along array A, the low shear wave velocity in the near shore region corresponds to the soft deposit, and the high shear wave velocity in the offshore region corresponds to the stiff deposit. This fact is consistent with the shear wave velocity structure obtained in the crossshore direction at the Banzu intertidal flat (Fig. 4). The distribution of surface shear wave velocities, calculated as the average at depths of 0–2 m, along arrays A, B, and C are shown in Figs. 18(a), 19(a), and 20(a), respectively. Along array A, the region at a distance from 0 to 220 m is soft with a shear wave velocity of 50–65 m/s, the region at a distance from 250 to 450 m shows a shear wave velocity of 70–80 m/s, and the region at a distance of 500 m is stiffer with a shear wave velocity of 80–100 m/s. The shear wave velocity in the bar regions is 10–15 m/s higher than that in the trough regions. Along array B, the region at a distance from  20 to 100 m distance is stiffer with a shear wave velocity of 80–90 m/s, and the region at a distance from 150 to 500 m is softer with a shear wave velocity of 60–70 m/s. This difference corresponds to the variation in soil properties, i.e., the softer deposits correspond to the finer soils. The shear wave velocity in the bar regions is 5–10 m/s higher than that in the trough regions. Array C focuses on the bar-trough unit at a distance of approximately 600 m along array A. The maximum shear

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

421

Fig. 15. Locations of (a) the Buzen Sea intertidal flat (sandflat) in Usa City, Japan, (b) a scene of the MASW survey at a region of the multibar-trough structure, (c) the Naha Airport intertidal flat (subtropical flat) in Naha City, Japan, (d) a scene of the MASW survey at a region of sand with coral gravels, (e) the Awase intertidal flat (subtropical flat) in Okinawa City, Japan and (f) a scene of the MASW survey at a region of sand with coral gravels.

wave velocity in the bar region was 102 m/s, while the minimum shear wave velocity in the trough region was 94 m/s. The shear wave velocity distribution shows a peak at a distance of around 12 m in the bar region, while it is almost constant at a distance from  17 to  6 m in the trough region. These results indicate that, in the trough region, soil contraction and hardening does not occur because suction does not develop under the always-submerged condition, while in the bar region, they occur due to developed suction during

the exposed periods. Moreover, the distribution of developed suction is not uniform. More details will be given below. The morphological ground surface profile data, leveled along the arrays, are shown in Figs. 18(b), 19(b), and 20(b), respectively, as well as the water level in the trough regions and the ground water level in the bar regions (except for array B). At some points representing bar and trough structures on arrays B and C, soil samples were collected to obtain mean particle diameter D50 and relative

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

422

Roberts bank intertidal flat (sandflat)

W123°05’09”

N49°04’39”

W123°14’21”

A part of the river mouth of the Frazer River

2000m

Delta Port Ferry Terminal

N49o00’13” Fig. 16. Locations of (a) the Roberts Bank intertidal flat (sandflat) in Delta City, BA, Canada and (b) a scene of the MASW survey at a region of soft sand with muddy coverage.

Array A (Shore Offshore) 4m

60

0

80 100 120

Array B

C

100

4m

A'

100 120

120 140

140 200

400

600

Array B (East West) Array A 80

Distance x (m)

800

B' vs (m/s)

100 120 140 0

C'

200 Distance x (m)

400

Array C C'

4m

100 120

-20

140 0 Distance x (m)

20

Fig. 17. Shear wave velocity structures obtained by the MASW survey in the cross-shore direction along arrays A and C and in the along-shore direction along array B in the Buzen Sea intertidal flat. Array C corresponds to the sequential bar-trough structure unit at a distance of around 600 m along array A.

density Dr. These results are indicated in Figs. 19(b) and 20(b). The multibar-trough structure in the cross-shore direction along array A shows a longer wavelength and a larger amplitude than that in the alongshore direction along

array B. In this tidal flat, the complex morphological formation with a multibar-trough structure in two directions is affected by the tidal current, the wave direction, the wave frequency, and the wave height. Then, the bar regions are hardened by suction dynamics.

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

C

Vane shear strength f (kPa)

C'

A'

100 80 60 40

G.L. W.L./G.W.L.

1m

G.L. and W.L./G.W.L.

Shear wave velocity vs (m/s)

A

423

8 4 0 0

200

400

600 Distance x (m)

800

1000

Fig. 18. Profiles of (a) the shear wave velocity (average for 0–2 m surface soil), (b) the morphological ground surface (G.L.) and the water level (W.L.)/ groundwater level (G.W.L.) and (c) the vane shear strength at a depth of 0.02 m in the cross-shore direction along array A in the Buzen Sea intertidal flat.

Fig. 19. Profiles of (a) the shear wave velocity (average for 0–2 m surface soil) and (b) the morphological ground surface (G.L.) in the along-shore direction along array B in the Buzen Sea intertidal flat.

Mean particle diameter D50 shows a large variation; however, those values obtained at a unit of sequential bar and trough, i.e., at distances of 140 m and 170 m along array B and at distances of 10 m and 25 m along array C, respectively, are quite similar. This is consistent with the results obtained at the Banzu intertidal flat (see Fig. 8). Even though mean particle diameter D50 was equivalent at each sequential bar and trough, relative density Dr in the bar region is higher than that in the trough region. This is consistent with the tendency of the accumulated contraction and hardening in the bar regions. Along array C, for the detailed survey for the sequential bar and trough unit, the point showing the maximum shear wave velocity did not coincide with the highest elevation point of the bar structure. This fact indicates that the point showing the maximum Dr value was not coincident to the

highest elevation point, because the soil with the slightly larger grain size was in an unsaturated condition at the highest elevation point. Sassa et al. (2007) showed that the relationship among air entry suction value saev, effective soil particle diameter D10, and void ratio e, for coastal sands can be expressed by saev ¼ cgw =ðeD10 Þ

ð3Þ

where gw is the unit weight of the water and c is a constant (c ¼ 20 mm2). At the point with Dr ¼ 69.4% on the bar of array C, with a void ratio e of 0.900 and an effective mean particle diameter D10 of 0.17 mm, the water head corresponding to the air entry suction value saev can be calculated as haev ¼  saev/gw ¼  131 mm. Since the ground level at the highest point of the bar was approximately  220 mm below the ground level (Fig. 20), the surface soil could become unsaturated. Therefore, the

424

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

Fig. 20. Profiles of (a) the shear wave velocity (average for 0–2 m surface soil) and (b) the morphological ground surface (G.L.) and the water level (W.L.)/groundwater level (G.W.L.) along array C in the Buzen Sea intertidal flat.

Fig. 21. Shear wave velocity structure obtained by the MASW survey in the cross-shore direction along array A in the Roberts Bank intertidal flat.

suction (i.e., effective stress) became smaller than the value calculated with the linear assumption between the suction and the ground water level as the saturated condition. In fact, the ground water level at the point where the maximum shear wave velocity was observed was approximately 150 mm below the ground level, corresponding to the place where the accumulated contraction and the densification had occurred under the saturated condition. Similarly, it was noticed that in the along-shore direction along array B (Fig. 19), in which each bar is 200–250 mm higher than the sequential trough, the two bars could be compared. The bar with a Dr of 59.3% at a distance of 110 m had a void ratio e of 0.913 and a D10 of 0.13 mm, resulting in a haev of  169 mm. The bar with a Dr of 79.0% at a distance of 170 m had a void ratio e of 0.877 and a D10 of 0.10 mm, resulting in a haev of  228 mm. Since the ground water level was below the level corresponding to haev at the former, the point showing the maximum shear wave velocity was different from the morphological ground surface peak point. Meanwhile, because the ground water level was above the level corresponding to have at the latter, the point showing the maximum shear wave velocity was coincident with the

morphological ground surface peak point. This is consistent with the above discussions for array C. Vane shear strengths tf at a depth of 0.02 m along array A are shown in Fig. 18(c). The shear strengths in the trough regions were smaller than 2 kPa, and those in the bar regions were in a range 5–7 kPa, indicating that the difference between the trough and the bar regions at extremely shallow depths was well captured as the surface shear strength rather than the shear wave velocity structure. 5.3. Shear wave velocity structure at the Roberts Bank intertidal flat The shear wave velocity structure obtained by the MASW survey is shown in Fig. 21, and the grain-size distribution curves are shown in Fig. 22. Moreover, the morphological ground surface profile data and the void ratio are shown in Fig. 23. The region from 0 to 700 m was homogeneous in the cross-shore direction, showing shear wave velocities of around 80 m/s and 160 m/s at depths of 1 m and 4 m, respectively. The region from 0 to 100 m shows a shear

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

wave velocity lower than 50 m/s at the ground surface. At a distance of around 100 m, the fine particle content was approximately 100%, indicating that the presence of muddy soil is detectable as ground softness. In the region at a distance from 700 to 900 m, the 2-m surface soil was similar to the surrounding soil in stratigraphy; however, the deeper portion was significantly softer with a shear wave velocity of, e.g., 150 m/s at a depth of 9 m. Although it has not yet been clarified why this portion was so soft, it is believed, based on a study of sequential sedimentation history investigated by MASW technology at a different flat (Watabe and Sassa, 2010), that this is the location of a previous water path. In the region at a distance from 1000 to 2200 m, the depth corresponding to a certain shear wave velocity gradually became deeper with distance. Shear wave velocities of 80 m/s and 160 m/s were observed at depths of 2 m and 6 m, respectively, at a distance of 2100 m. In the region at a distance of 2200–2300 m, there was a soft portion at larger depths; however, in the region at a distance from 2300 to 3000 m, the stratigraphy was very homogeneous in the cross-shore direction. The

Percentage finer in weight P (%)

100

80 100 m 103 m 275 m 512 m 812 m 945 m 1068 m 1273 m 1295 m 1788 m 2138 m 2390 m 4127 m

60

40

20

0 0.001

0.01 0.1 1 Soil particle diameter D (mm)

10

Void ratio e

MASW survey was conducted up to a distance of 3000 m corresponding to the relatively large water path (see Fig. 23(a)). The fine particle content of the surface soil decreases with distance in the cross-shore direction, e.g., it is only 20% at a distance of 2390 m. Despite the coarse particle content increasing with distance, the surface shear wave velocity was smaller than 50 m/s in the region from a distance of 2000 m. As mentioned above, in the sandflats with a distance of approximately 1000 m of intertidal zone in the cross-shore direction, the deposits at the offshore side were stiffer than those at the shore side. Suction dynamics corresponding to the ground water level fluctuations resulted in accumulated contraction and hardening in the offshore region, particularly in the bar regions. The accumulated contraction was not observed in the Roberts Bank intertidal flat of large dimensions or without the bartrough structure, because the ground water level was coincident with the ground surface in the whole region of the flat. In other words, there is no effect of suction dynamics on the sediments. This condition is very similar to the whole region of the Shiranui muddy intertidal flat as well as the near-shore region of the Banzu intertidal flat. In fact, in most of the regions, suction did not develop, except for a small clayey mound at a distance of 100 m (0.7 kPa) and a small sandy mound near a water path at a distance of 1000 m (1.0 kPa). The void ratio showed a high value of 1.5–1.7 near the shore, it gradually decreased toward the offshore to be 1.0 at a distance of 1000 m, and then it became constant at 0.9 in the region from 1000 m. The void ratio distribution corresponds to the variation in the grain-size distribution from the shore to the offshore; this can be explained by the absence of suction dynamics. 5.4. Shear wave velocity structure of the Awase intertidal flat The shear wave velocity structure obtained by the MASW survey is shown in Fig. 24. Unlike at the Banzu intertidal flat and the Buzen Sea intertidal flat, the surface

2m

G.L.

Fig. 22. Grain size distributions for the samples collected along array A in the Roberts Bank intertidal flat. The 50 mm surface soil samples were tested in the laboratory.

425

2.0 1.5 1.0 0.5 0

500

1000

1500

2000 2500 Distance x (m)

3000

3500

4000

4500

Fig. 23. Profiles of (a) the morphological ground surface (G.L.) and (b) the void ratio for 50 mm surface soil in the cross-shore direction along array A in the Roberts Bank intertidal flat.

426

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

Fig. 24. Shear wave velocity structure obtained by the MASW survey in the cross-shore direction along array A in the Awase intertidal flat.

Fig. 25. Shear wave velocity structures obtained by the MASW survey in the along-shore direction along arrays A and C and in the cross-shore direction along array B in the Naha Airport intertidal flat. Array C is an extension of array A corresponding to a distance of 420 m on array A.

soil is stiffer at the near-shore region and is softer in the offshore region. This tendency cannot be explained by soil contraction or by hardening caused by suction dynamics. Since sand with coral gravel and relatively homogeneous sand are dominant in the near-shore and the offshore regions, respectively, it can be said that shear wave velocity structures are strongly affected by the variation in the physical properties of the soil. The near-shore region, with a shear wave velocity of 130 m/s, was significantly stiffer than the regions at the Banzu intertidal flat and the Buzen Sea intertidal flat with shear wave velocities of 50–100 m/s. This is also stiffer than the stiffest bar regions of those intertidal flats with shear wave velocities of 110–130 m/s. The offshore region with shear wave velocities of 130 m/s in the bar regions and 90–110 m/s in the trough regions are equivalent to the typical shear wave velocity structures in the offshore region of the Banzu intertidal flat and the Buzen Sea intertidal flat. In the region at a distance from 0 to 550 m, except for around 100 m and 400 m, a stiff layer was found with a

shear wave velocity of greater than 300 m/s at depths of more than 8 m. The region at a distance from 550 m became slightly softer, but in the region at a distance from 750 to 850 m, the stiff layer was protruding. These uneven levels of the stiff layer are thought to be a lime rock layer formed from coral bodies, not a dense sandy soil. 5.5. Shear wave velocity structure of the Naha Airport intertidal flat The shear wave velocity structures obtained by the MASW survey are shown in Fig. 25. Firstly, we focus on the arrays in the along-shore direction. Array C is an extension of array A, corresponding to a distance of  420 m on array A. In the regions of  420 m, from 0 to 120 m, around 600 m, from 750 to 800 m, and from 880 to 920 m, lime rock with a shear wave velocity above 300 m/s was exposed on the ground surface. Lime rock protruding at the surface was also detected along array C. Sandy soils were deposited in the valleys

In this section, the shear wave velocity structures in the cross-shore direction obtained at the six tidal flats mentioned above, i.e., the Banzu intertidal flat, the Buzen Sea intertidal flat, and the Roberts Bank intertidal flat as sandflats, the Shiranui intertidal flat as mudflats, and the Awase intertidal flat and the Naha Airport intertidal flat as subtropical tidal flats, are considered. To examine a dominant wavelength for each shear wave velocity structure, the shear wave velocities at depths of 0.5 m and 2.5 m were read along the survey array, and then a spectrum analysis with fast Fourier transformation (FFT) was carried out for the data at each depth. For the three tidal flats leveled along the survey arrays, i.e., the Buzen Sea intertidal flat, the Roberts Bank intertidal flat, and the Naha Airport intertidal flat, a spectrum analysis with FFT was carried out for each ground surface profile data, to examine a dominant wavelength for the morphological ground surface profile. For the Banzu intertidal flat, the morphological ground surface profile data obtained approximately 100 m south of array A was examined. The morphological ground surface profiles analyzed here were subtracted from each trend component in elevation, which was assumed to be linearly lowered with distance from the shore to the offshore. The results of the spectrum analysis for the shear wave velocity structure and the morphological ground surface profile are shown in Figs. 26 and 27, respectively. Firstly, we focus on the shear wave velocity at a depth of 0.5 m (Fig. 26(a)). In the Banzu intertidal flat, there are clear peaks at wavelengths of approximately 40 m, 65 m, and 90 m. In the Buzen Sea intertidal flat, there are some small peaks in the range of wavelengths of 45–75 m. Meanwhile, in the Roberts Bank intertidal flat, there is no dominant wavelength. As seen in the shear wave velocity structure of the Banzu intertidal flat (Fig. 4) and the Buzen Sea intertidal flat (Fig. 17), the interval of the bar-trough structure was 40–60 m. In addition, the dominant wavelengths for the morphological ground surface profile (Fig. 27) are similar to those for the shear wave velocity

Amplitude of shear wave velocity distribution (m/s)

5.6. Spectrum analysis for morphological features of intertidal flats

3.5

427

: Banzu sandy flat : Buzenc Sea sandy flat

3.0

: Roberts bank sandy flat

: Shiranui muddy flat : Awase subtropical flat

2.5

: Naha Airport subtropical flat

2.0 1.5 1.0 0.5 0.0 0

10

20 30 40 50 60 70 80 90 100 110 120 Wavelength of shear wave velocity distribution (m)

3.5 : Banzu sandy flat : Buzen Sea sandy flat

3.0

: Roberts bank sandy flat : Shiranui muddy flat

2.5

: Awase subtropical flat : Naha Airport subtropical flat

2.0 1.5 1.0 0.5 0.0 0

10 20 30 40 50 60 70 80 90 100 110 120 Wavelength of shear wave velocity distribution (m)

Fig. 26. Spectrum for the shear wave velocity structure (a) at a depth of 0.5 m and (b) at a depth of 2.5 m.

0.011 Amplitude of ground level distribution (m)

between the protruding lime rock layers. The thickness of the sandy deposits was larger than 15 m in the region from  320 to  150 m, but less than 6 m in the other regions because of a shallow lime rock layer. The shear wave velocity at the surface of the sandy deposits was approximately 100 m/s, which is equivalent to that measured at sandflats, e.g., the Banzu intertidal flat and the Buzen Sea intertidal flat. Next, we focus on the array in the cross-shore direction. Along array B, in the regions around distances of 180 m, from 320 to 440 m, and around 1350 m, a lime rock layer with shear wave velocities higher than 300 m/s was exposed on the ground surface. In the regions at distances from 600 to 670 m, around 900 m, around 1000 m, and around 1100 m, the thickness of the sandy deposits was approximately 10 m; however, in the other regions, that was only 2–6 m. The geological stratigraphy was very complex corresponding to the lime rock protruding at the surface.

Amplitude of shear wave velocity distribution (m/s)

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

: Banzu sandy flat

0.010

: Buzen Sea sandy flat : Roberts bank sandy flat : Naha Airport subtropical flat

0.009 0.008 0.007 0.006 0.005 0.004 0.003 0.002 0.001 0.000 0

10

20 30 40 50 60 70 80 90 100 110 120 Wavelength of ground level distribution (m)

Fig. 27. Spectrum for the morphological ground surface profile.

428

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

structure (Fig. 26(a)). These facts indicate that the dominant wavelengths are mainly governed by the morphological ground surface profile corresponding to the multibar-trough structure. They are consistent with the above descriptions for the accumulated contraction and hardening caused by suction dynamics. It is very interesting that the shapes of the spectrum are very similar between the Shiranui intertidal flat (mudflat) and the Roberts Bank intertidal flat (huge scale sandflat) without dominant wavelengths or morphological variations. This fact indicates that morphological variations, such as bar-trough structures, strongly affect the shear wave velocity structure. In the subtropical intertidal flats, the spectrum of the Naha Airport intertidal flat was very different from that of the Awase intertidal flat. In the Awase intertidal flat, the dominant wavelength for the shear wave velocity structure was obtained at approximately 60 m, equivalent to that in the sandflats (the Banzu intertidal flat and the Buzen Sea intertidal flat). In the Naha Airport intertidal flat, the dominant wavelength for the shear wave velocity structure was obtained at approximately 100 m, as well as some other peaks. According to the additional spectrum analysis for longer wavelengths, a wavelength of 200 m was obtained as the dominant one. These dominant wavelengths are significantly longer than those of other tidal flats, indicating that these are not governed by the morphological formation affected by hydraulic conditions. The lime rock exposed on the ground surface strongly affects the dominant wavelength. Next, we focus on the shear wave velocity at a depth of 2.5 m (Fig. 26(b)). The spectrum is similar to that obtained at a depth of 0.5 m; however, there are no dominant wavelengths, except for the Naha Airport intertidal flat with a protruding lime rock. This is consistent with the fact that the effect of the bar-trough structure on the shear wave velocity structure was limited to the depth of less than around 1.7 m (Figs. 4 and 6). 6. Conclusions In this study, MASW technology was applied to investigate the sedimentary stratigraphy at a typical sandflat (Banzu intertidal flat) and a typical mudflat (Shiranui intertidal flat), and its formation mechanisms were discussed from the viewpoints of geoenvironmental dynamics in association with the semidiurnal tidal fluctuations. The main conclusions are as follows: (1) MASW is a very useful method for quantitatively measuring the eventual significant changes in the soil stiffness structure by using the representation of the shear wave velocity. (2) In the sandflat at the Banzu intertidal flat, the sedimentary structures evaluated by MASW (as the distribution of the shear wave velocity/stiffness) and by the physical soil tests (as the distribution of the void ratio) can be

explained as being a consequence of the elastoplastic contraction of soils that experience suction dynamics due to the tide-induced groundwater level fluctuations. Bar and trough regions are hydrologically formed by ocean waves and tidal currents; however, even though the sediments initially possess homogeneous void structures, only the bar regions gradually densify because of the suction-induced cyclic elastoplastic contraction. (3) In the mudflat at the Shiranui intertidal flat, although a variation in the void structure was observed in association with the grain-size distribution, the soil stiffness was extremely homogeneous because suction did not develop in the soil.

In addition, MASW technology was applied to the four intertidal flats with various characteristics, and the features of the sedimentary stratigraphy was described. The main conclusions are as follows: (4) In the Buzen Sea intertidal flat, multibar-trough structures developed in both the cross-shore direction, with a long interval, and in the along-shore direction, with a short interval. The morphological surface variation, the water level/ground water level, and the ground stiffness are closely related to each other. Although the multibar-trough structures in the cross-shore and along-shore directions are of a different scale, the eventual contraction and hardening at the bar regions can be explained by the formation mechanisms concluded for the Banzu intertidal flat. Namely, the peak of the shear wave velocity structure corresponds to the location where the sediments underwent the severest suction dynamics under the saturated condition in consideration of the air entry suction value of the soil. (5) In the Roberts Bank intertidal flat of large dimensions without morphological variations (without a bar-trough structure), no accumulated contraction was observed because the ground water level was coincident with the ground surface in all the regions of the flat. The void ratio structure corresponds to the variation in grain-size distribution from the shore to the offshore, which can be explained by the absence of suction dynamics. (6) In the subtropical intertidal flats with a deep lime rock layer at the Awase intertidal flat, the shear wave velocity structure is governed by the properties of the shallow soil layer. In the subtropical intertidal flats with a protruding lime rock layer at the Naha Airport intertidal flat, the shear wave velocity structure is strongly influenced by the morphological profile of the lime rock layer.

Finally, a spectrum analysis for the shear wave velocity structure and a morphological ground surface profile in the cross-shore direction were carried out for the six intertidal flats investigated in order to quantitatively characterize the dominant wavelengths. The main conclusions are as follows:

Y. Watabe, S. Sassa / Soils and Foundations 52 (2012) 411–429

(7) In the sandflats with a multibar-trough structure, the dominant wavelength for the shear wave velocity structure is in a range 40–90 m, corresponding to the interval of the bar-trough structure. Meanwhile, in the mudflats or the huge-scale sandflats, there is no dominant wavelength. In the subtropical flats with a protruding lime rock, the dominant wavelength is strongly affected by the morphological profile of the lime rock layer. Acknowledgments The authors would like to thank Dr. K. Hayashi of Geometrics, USA (formerly, Oyo Corporation, Japan) for his contribution to MASW. The field observations were carried out in collaboration with the Kyushu Regional Development Bureau of the Ministry of Land Infrastructure, Transport and Tourism, Japan, and Okinawa General Bureau, Cabinet Office, Government of Japan. This research work was supported by the Grant-in-Aid for Scientific Research (Nos. 23360208 and 20360216) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. References Ellison, R.L., 1984. Foraminifera and meiofauna on an intertidal mudflat, Cornwall, England: Populations; respiration and secondary production; and energy budget. Hydrobiologia 109 (2), 131–148. D’Appolonia, D.J., Poulos, H.G., Ladd, C.C., 1971. Initial settlement of structures on clay. Journal of Soil Mechanics and Foundations Division—American Society of Civil Engineers 97 (10), 1359–1377. Hasiotis, T., Papatheodorou, G., Ferentinos, G., 2005. A high resolution approach in the recent sedimentation processes at the head of Zakynthos Canyon, Western Greece. Marine Geology 214, 49–73. Hayashi, K., Suzuki, H., 2004. CMP cross-correlation analysis of multichannel surface-wave data. Exploration Geophysics 35, 7–13. Heyvaert, V.M.A., Baeteman, C., 2007. Holocene sedimentary evolution and palaeocoastlines of the Lower Khuzestan plain (Southwest Iran). Marine Geology 242, 83–108. Hori, K., Saito, Y., Zhao, Q., Cheng, X., Wang, P., Sato, Y., Li, C., 2001. Sedimentary facies of the tide-dominated paleo-Changjiang (Yangtze) estuary during the last transgression. Marine Geology 177, 331–351. Hutri, K.-L., Kotilainen, A., 2007. An acoustic view into Holocene palaeoseismicity offshore Southwestern Finland, Baltic Sea. Marine Geology 238, 45–59. Inazaki, T., 1999. Land Streamer: a new system for high-resolution Swave shallow reflection surveys. In: Proceedings of the Symposium on the Application of Geophysics to Engineering and Environmental Problems ’99, pp. 207–216. Jarvis, J., Riley, C., 1987. Sediment transport in the Mouth of the Eden estuary. Estuarine, Coastal and Shelf Science 24, 463–481. Kitsunezaki, C., Goto, N., Kobayashi, Y., Ikawa, T., Horike, M., Saito, T., Kurota, T., Yamane, K., Okuzumi, K., 1990. Estimation of P- and S-wave velocities in deep soil deposits for evaluating ground vibrations in earthquake. Shizen-Saigai-Kagaku 9 (3), 1–17. Kuipers, B.R., de Wilde, P.A.W.J., Creutzberg, F., 1981. Energy flow in a tidal flat ecosystem. Marine Ecology Progress Series 5, 215–221. Kuwae, T., Kibe, E., Nakamura, Y., 2003. Effect of emersion and immersion on the porewater nutrient dynamics of an intertidal sandflat in Tokyo Bay. Estuarine, Coastal and Shelf Science 47, 929–940. Le Hir, P., Roberts, W., Cazaillet, O., Christie, M., Bassoullet, P., Bacher, C., 2000. Characterization of intertidal flat hydrodynamics. Continental Shelf Research 20, 1433–1459.

429

Leroueil, S., Tavenas, F., Le Bihan, J.P., 1983. Proprie´te´s caracte´ristiques des argiles de l’est du Canada. Canadian Geotechnical Journal 20 (4), 681–705. Lohani, T., Imai, G., Tani, K., Shibuya, S., 2001. Gmax of fine-grained soils at wide void ratio range, focusing on time-dependent behavior. Soils and Foundations 41 (5), 87–102. Ludwig, W.J., Nafe, J.E., Drake, C.L., 1970. Seismic Refraction, in the Sea, vol. 4, Part 1. Wiley-Interscience, pp. 53–84. Mienis, F., van Weering, T., de Haas, H., de Stigter, H., Huvenne, V., Wheeler, A., 2006. Carbonate mound development at the SW Rockall Trough margin based on high resolution TOBI and seismic recording. Marine Geology 233, 1–19. Neal, A., Richards, J., Pye, K., 2002. Structure and development of shell cheniers in Essex, Southeast England, investigated using high-frequency ground-penetrating radar. Marine Geology 185, 435–469. Park, C.B., Miller, R.D., Xia, J., 1999. Multichannel analysis of surface waves. Geophysics 64 (3), 800–808. Perillo, G.M.E., Sequeira, M.E., 1989. Geomorphologic and sediment transport characteristics of the middle reach of the Bahia Blanca estuary (Argentina). Journal of Geophysical Research 94, 14,351–14,362. Saito, M., Kabasawa, H., 1993. Computation of reflectivity and surface wave dispersion curves for layered media, 2. Rayleigh wave calculations. Butsuri-Tansa 46, 283–298. Sassa, S., Watabe Y., 2005. The dynamics of tidal flat geo-environments in an estuary. Coastal Dynamics ’05 American Society of Civil Engineers. http://dx.doi.org/10.1061/40855(214)25. Sassa, S., Watabe, Y., 2007. Role of suction dynamics in evolution of intertidal sandy flats: field evidence, experiments, and theoretical model. Journal of Geophysical Research 112, F01003. http://dx.doi.org/10.1029/2006JF000575. Sassa, S., Watabe, Y., Ishii, H., 2007. Mechanisms of water retention dynamics in intertidal sandy flats and beaches. Annual Journal of Coastal Engineering, Japan Society of Civil Engineers 54, 1151–1155. Sassa, S., Watabe, Y., 2008. Threshold, optimum and critical geoenvironmental conditions for burrowing activity of sand bubbler crab, Scopimera globosa. Marine Ecology Progress Series 354, 191–199. Sassa, S., Watabe, Y., 2009. Persistent sand bars explained by geodynamic effects. Geophysical Research Letters 36, L01404. http://dx.doi.org/10.1029/2008GL036230. Schlu¨ter, P., Uenzelmann-Neben, G., 2007. Seismostratigraphic analysis of the Transkei Basin: a history of deep sea current controlled sedimentation. Marine Geology 240, 99–111. Stevenson, J.C., Ward, L.G., Kearney, M.S., 1988. Sediment transport and trapping in marsh systems: implications of tidal flux studies. Marine Geology 80, 37–59. Terrinha, P., Pinheiro, L.M., Henriet, J.-P., Matias, L., Ivanov, M.K., Monteiro, J.H., Akhmetzhanov, A., Volkonskaya, A., Cunha, T., Shaskin, P., Rovere, M., 2003. Tsunamigenic-seismogenic structures, neotectonics, sedimentary processes and slope instability on the southwest Portuguese Margin. Marine Geology 195, 55–73. Uchiyama, Y., 2007. Hydrodynamics and associated morphological variations on an estuarine intertidal sand flat. Journal of Coastal Research 23, 1015–1027. Watabe, Y., Sassa, S., 2008. Application of MASW technology to identification of tidal flat stratigraphy and its geoenvironmental interpretation. Marine Geology 252, 79–88. Watabe, Y., Sassa, S., 2010. Evaluation of sedimentation history of sandbars at entrance of Lake Tofutsu, Hokkaido, Japan, by MASW technology. Geotechnical Special Publication No. 210, American Society of Civil Engineers (Proceedings of the 5th International Conference on Scour and Erosion), pp. 1013–1022. Widdows, J., Brinsley, M.D., Salkeld, P.N., Lucas, C.H., 2000. Influence of biota on spatial and temporal variation in sediment erodability and material flux on a tidal flat (Westerschelde, The Netherlands). Marine Ecology Progress Series 194, 23–37. Xia, J., Miller, R.D., Park, C.B., 1999. Estimation of near-surface shearwave velocity by inversion of Rayleigh waves. Geophysics 64, 691–700.