Difference between revisions of "Wave-induced soil liquefaction"
Dronkers J (talk | contribs) |
Dronkers J (talk | contribs) |
||
(2 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
− | |||
− | |||
− | |||
{{ Definition| title = Soil liquefaction | {{ Definition| title = Soil liquefaction | ||
Line 7: | Line 4: | ||
− | + | Seabed soils which are not well consolidated are prone to liquefaction when subjected to pressure fluctuations. Soil liquefaction can be produced for example by earthquakes, but here we concentrate on submarine soils which are subjected to cyclic loading by waves. Structures built on freshly deposited submarine soils can collapse or sink when the seabed is liquified<ref name=S99>Sumer, B.M., Fredsøe, J., Christensen, S. and Lind, M.T. 1999. Sinking/floatation of pipelines and other objects in liquefied soil under waves. Coast. Eng. 38: 53–90</ref>. A simplified mathematical introduction to the theory of soil liquefaction is given in the Appendix. | |
Line 25: | Line 22: | ||
Soils prone to liquefaction are typically water-saturated, consisting of loosely packed fine sediments – fine sand, silt and some clay – with poor drainage capacity (low hydraulic conductivity). When such a soil is subjected to cyclic wave-induced loading and associated shear deformations, the soil grains tend to rearrange such that the soil skeleton is progressively compressed. (The reverse happens for a densely compacted soil.) The resulting reduction of the pore volume is associated with an increase of the pore water pressure. When the pore pressure exceeds the pressure exerted by the load of the overlying water and soil (pore pressure higher than the initial mean normal effective stress, taking soil cohesion into account), the contact friction points between soil grains are broken and the soil skeleton collapses. The soil grains become unbound and free to move, carried by the pore water (Fig. 2). The soil begins to act like a liquid, losing its stiffness and bearing capacity. Any structure built on this soil will sink or collapse<ref name=S99>Sumer, B.M., Fredsøe, J., Christensen, S. and Lind, M.T. 1999. Sinking or floatation of pipelines and other objects in liquefied soil under waves. Coastal Engineering 38: 53–90</ref>. | Soils prone to liquefaction are typically water-saturated, consisting of loosely packed fine sediments – fine sand, silt and some clay – with poor drainage capacity (low hydraulic conductivity). When such a soil is subjected to cyclic wave-induced loading and associated shear deformations, the soil grains tend to rearrange such that the soil skeleton is progressively compressed. (The reverse happens for a densely compacted soil.) The resulting reduction of the pore volume is associated with an increase of the pore water pressure. When the pore pressure exceeds the pressure exerted by the load of the overlying water and soil (pore pressure higher than the initial mean normal effective stress, taking soil cohesion into account), the contact friction points between soil grains are broken and the soil skeleton collapses. The soil grains become unbound and free to move, carried by the pore water (Fig. 2). The soil begins to act like a liquid, losing its stiffness and bearing capacity. Any structure built on this soil will sink or collapse<ref name=S99>Sumer, B.M., Fredsøe, J., Christensen, S. and Lind, M.T. 1999. Sinking or floatation of pipelines and other objects in liquefied soil under waves. Coastal Engineering 38: 53–90</ref>. | ||
− | + | Experiments in a wave flume have shown that the residual pore pressure generated by random waves can be much larger than that by regular waves with equivalent mean wave height and period. Low-frequency waves in particular can make a significant contribution to the residual pressure<ref>Niu, J.W., Xu, J.S., Dong, P. and Li, G.X. 2019. Pore water pressure responses in silty sediment bed under random wave action. Sci. Rep. 9, 11685</ref>. Other flume experiments have shown that soil liquefaction through excess pore pressure enhances sediment re-suspension<ref>Jia, Y., Zhang, L., Zheng, J., Liu, X., Jeng, D-S. and Shan, H. 2014. Effects of wave-induced seabed liquefaction on sediment re-suspension in the Yellow River Delta. Ocean Engineering 89: 146–156</ref>. Seabed liquefaction and seabed scour reinforce each other, as removal of top soil strengthens the upward pore pression gradient in the soil underneath<ref>Wang, L., Zhang, J., Jeng, D-S., Zhang, Q. and Chen, T. 2023. Experimental study on the dynamic response of a silty seabed under waves. Ocean Engineering 269, 113554</ref> | |
Line 31: | Line 28: | ||
− | The onset of liquefaction occurs first at the surface of the bed, and rapidly spreads out across the soil depth, causing the entire soil to behave like a liquid. When the liquefaction reaches the impermeable base, the soil begins to compact. The large pores between particles in the liquefied state allow pore water | + | ==Liquefaction-consolidation process== |
− | behaviour during wave-induced liquefaction. Sedimentology 53: 611–629</ref><ref>Sui, T., Sumer, B.M., Kirca, V.S.O., Carstensen, S., Zheng, J. and Fuhrman, D.R. 2023. Effect of history of wave exposure on seabed liquefaction. Coastal Engineering 183 (2023) 104307</ref>. After | + | The onset of liquefaction occurs first at the surface of the bed, and rapidly spreads out across the soil depth, causing the entire soil to behave like a liquid. When the liquefaction reaches the impermeable base, the soil begins to compact. The large pores between particles in the liquefied state allow the expel (drainage) of pore water and wave-induced compaction. Soil grains fall out of the liquid state, settle through the pore water until they come into contact with each other. The compaction gradually progresses from the impermeable base in the upward direction, and the entire sequence of the liquefaction/compaction process comes to an end when the compaction reaches the soil surface<ref>Sumer, B.M., Hatipoglu, F., Fredsoe, J. and Sumer, S.K. 2006. The sequence of sediment |
+ | behaviour during wave-induced liquefaction. Sedimentology 53: 611–629</ref><ref>Sui, T., Sumer, B.M., Kirca, V.S.O., Carstensen, S., Zheng, J. and Fuhrman, D.R. 2023. Effect of history of wave exposure on seabed liquefaction. Coastal Engineering 183 (2023) 104307</ref>. After liquefaction, a loosely packed soil will not recover its initial state. The soil is compacted and much less susceptible to renewed liquefaction. The liquefaction-compaction process is depicted schematically in Fig. 3. | ||
Line 39: | Line 37: | ||
==Influence of soil composition== | ==Influence of soil composition== | ||
− | Flume experiments show that the liquefaction potential strongly depends on the soil composition. Experiments by Gratchev et al. (2006<ref>Gratchev, I.B., Sassa, K., Osipov, V.I. and Sokolov, V.N. 2006. The liquefaction of clayey soils under cyclic loading. Eng. Geol. 86: 70–84</ref>) revealed the significant influence of plasticity on soil liquefaction resistance. Soil vulnerable to liquefaction had an open microfabric in which clay aggregates mainly gathered at the silt particle contact points, forming low-strength 'clay bridges' that were easily destroyed during cyclic loading (Fig. 2). The contact friction between particles that maintained the soil skeleton structure was reduced by the clay aggregates, so grains could easily slip and rotate once the seabed was subjected to wave loading. Clay | + | Flume experiments show that the liquefaction potential strongly depends on the soil composition. Experiments by Gratchev et al. (2006<ref>Gratchev, I.B., Sassa, K., Osipov, V.I. and Sokolov, V.N. 2006. The liquefaction of clayey soils under cyclic loading. Eng. Geol. 86: 70–84</ref>) revealed the significant influence of plasticity on soil liquefaction resistance. Soil vulnerable to liquefaction had an open microfabric in which clay aggregates mainly gathered at the silt particle contact points, forming low-strength 'clay bridges' that were easily destroyed during cyclic loading (Fig. 2). The contact friction between particles that maintained the soil skeleton structure was reduced by the clay aggregates, so grains could easily slip and rotate once the seabed was subjected to wave loading. Clay minerals thus enhance the plasticity and compressibility of the clay-silt seabed and increases the sensitivity of the seabed to liquefaction. |
Experiments with kaolinite-fine sand and illite-fine sand showed that the susceptibility of a soil to liquefaction increases with increasing clay content if the clay content is low (0.5-5%), but decreases with a higher clay content<ref>Zang, J., Jiang, Q., Jeng, D., Zhang, C., Chen, X. and Wang, L. 2020. Experimental Study on Mechanism of Wave-Induced Liquefaction of Sand-Clay Seabed. J. Mar. Sci. Eng. 8, 66</ref>. At a clay content of more than 10-30%, depending on the type of clay, the silt-clay mixture was not susceptible to liquefaction. This can be explained by the fact that in case of high clay content the silt grains are completely encapsulated in the clay matrix and therefore cannot rearrange under cyclic shear stresses, causing resistance to liquefaction. Experiments by Kirca et al. (2014<ref>Kirca, V., Sumer, B.M. and Fredsøe, J. 2014. Influence of clay content on wave-induced liquefaction, J. Waterw. Port Coast. Ocean Eng. ASCE 140, 04014024</ref>) showed that the sensitivity of a sand-clay mixture to wave-induced liquefaction not only depends on the clay content but also on the sand grainsize. Liquefaction occurred at a much larger clay content for medium sand than for fine sand or silt. The presence of shell fragments also decreases the susceptibility of silt to liquefaction<ref name=S20>Sui, T., Kirca, V.S.O, Sumer, B.M., Carstensen, S. and Fuhrman, D.R. 2022. Wave-induced liquefaction in a silt and seashell mixture. Coastal Engineering 178, 104215</ref>. Experiments for a medium-sand-clay (kaolinite) seabed supporting a submerged breakwater showed that wave-induced liquefaction required a clay content greater than 40%<ref>Chavez, V., Mendoza, E., Silva, R., Silva, A. and Losada, M.A. 2017. An experimental method to verify the failure of coastal structures by wave induced liquefaction of clayey soils. Coastal Engineering 123: 1–10</ref>. These experiment also showed that the initial water content in the soil is relevant to the liquefaction potential as it affects the pressure build-up process. Higher pore water fractions allow the residual pore pressure to accumulate faster. | Experiments with kaolinite-fine sand and illite-fine sand showed that the susceptibility of a soil to liquefaction increases with increasing clay content if the clay content is low (0.5-5%), but decreases with a higher clay content<ref>Zang, J., Jiang, Q., Jeng, D., Zhang, C., Chen, X. and Wang, L. 2020. Experimental Study on Mechanism of Wave-Induced Liquefaction of Sand-Clay Seabed. J. Mar. Sci. Eng. 8, 66</ref>. At a clay content of more than 10-30%, depending on the type of clay, the silt-clay mixture was not susceptible to liquefaction. This can be explained by the fact that in case of high clay content the silt grains are completely encapsulated in the clay matrix and therefore cannot rearrange under cyclic shear stresses, causing resistance to liquefaction. Experiments by Kirca et al. (2014<ref>Kirca, V., Sumer, B.M. and Fredsøe, J. 2014. Influence of clay content on wave-induced liquefaction, J. Waterw. Port Coast. Ocean Eng. ASCE 140, 04014024</ref>) showed that the sensitivity of a sand-clay mixture to wave-induced liquefaction not only depends on the clay content but also on the sand grainsize. Liquefaction occurred at a much larger clay content for medium sand than for fine sand or silt. The presence of shell fragments also decreases the susceptibility of silt to liquefaction<ref name=S20>Sui, T., Kirca, V.S.O, Sumer, B.M., Carstensen, S. and Fuhrman, D.R. 2022. Wave-induced liquefaction in a silt and seashell mixture. Coastal Engineering 178, 104215</ref>. Experiments for a medium-sand-clay (kaolinite) seabed supporting a submerged breakwater showed that wave-induced liquefaction required a clay content greater than 40%<ref>Chavez, V., Mendoza, E., Silva, R., Silva, A. and Losada, M.A. 2017. An experimental method to verify the failure of coastal structures by wave induced liquefaction of clayey soils. Coastal Engineering 123: 1–10</ref>. These experiment also showed that the initial water content in the soil is relevant to the liquefaction potential as it affects the pressure build-up process. Higher pore water fractions allow the residual pore pressure to accumulate faster. | ||
Line 47: | Line 45: | ||
− | Fluid mud can be considered liquified soil. However, cohesive | + | ==Fluid mud== |
+ | Fluid mud – a dense colloidal suspension of fine, cohesive sediments (mainly clay minerals) – can be considered liquified soil. However, cohesive soils are not liquefied by a similar process of pore pressure build-up as non-cohesive or slightly cohesive soils. Cohesive soil (mud) behaves like a viscoelastic material whose viscosity decreases when a stress is applied that is strong enough (fluctuates) to break the electrochemical bounds that hold the constituent particles together<ref>McAnally, W.H., Friedrichs, C., Hamilton, D., Hayter, E., Shrestha, P., Rodriguez, H., Sheremet, A. and Teeter, A. 2007. Management of Fluid Mud in Estuaries, Bays, and Lakes. I: Present State of Understanding on Character and Behavior. J. Hydr. Eng. 133: 9-22</ref><ref>Robillard, D.J. Mehta, A.J. and Safak, I. 2023. Comments on wave-induced behavior of a coastal mud. Coastal Engineering 186, 104400</ref>. Cohesive soil has thixotropic properties: after stress removal, it gradually returns to its original state. A fluid mud layer is formed when floc-forming particles settle to the seabed in sufficiently high concentration. High concentrations of suspended fine sediment often originate from previous mud bed erosion by strong waves or currents or from convergence of residual fine sediment transport related to [[Tidal asymmetry and tidal basin morphodynamics|tidal asymmetry]] and [[estuarine circulation]] (see [[Estuarine turbidity maximum]]). Stirring of unconsolidated mud deposits by wave-induced pressure variations is another process that can contribute to the formation of fluid mud<ref name=M07>McAnally, W.H., Friedrichs, C., Hamilton, D., Hayter, E., Shrestha, P., Rodriguez, H., Sheremet, A. and Teeter, A. 2007. Management of Fluid Mud in Estuaries, Bays, and Lakes. I: Present State of Understanding on Character and Behavior. J. Hydr. Eng. 133: 9-22</ref>. The high stresses needed in case of advanced consolidation will more likely lead to sediment erosion and suspension rather than to fluidization. A fluid mud layer can be formed when the suspended sediment settles<ref>Winterwerp, J.C., de Boer, G.J., Greeuw, G. and van Maren, D.S. 2012. Mud-induced wave damping and wave-induced liquefaction. Coastal Engineering 64: 102–112</ref>. When a fluid mud layer flows along the seabed, the density (and viscosity) can increase by bed erosion or decrease by uptake (entrainment) of overlying water<ref>Winterwerp, J.C., Vroom, J., Wang, Z-B., Krebs, M., Hendriks, E.C.M., van Maren, D.S., Schrottke, K., Borgsmüller, C. and Schol, A. 2017. SPM response to tide and river flow in the hyper-turbid Ems River. Ocean Dynamics 67: 559–583</ref>. See [[Fluid mud]] and [[Dynamics of mud transport]]. | ||
==Preventing soil liquefaction== | ==Preventing soil liquefaction== | ||
− | The susceptibility of a soil to liquefaction can be checked through [https://en.wikipedia.org/wiki/Standard_penetration_test Standard Penetration Tests] or [https://en.wikipedia.org/wiki/Cone_penetration_test Cone Penetration Tests]. Liquefaction risks can be reduced by measures such as replacing soft soil (expensive) or by consolidation measures. Soil consolidation can be done for example by vibro compaction or by preloading the soft bed with coarse sediment<ref name=MS> Van der Meer, J. and Sigurdarson, S. 2017. Design and construction of berm breakwaters. Advanced Series on Ocean Engineering vol. 40, World Scientific Publ. Co., Singapore</ref>. Monitoring of the consolidation process should ensure that a stable situation is reached. | + | The susceptibility of a soil to liquefaction can be checked through [https://en.wikipedia.org/wiki/Standard_penetration_test Standard Penetration Tests] or [https://en.wikipedia.org/wiki/Cone_penetration_test Cone Penetration Tests]. Such tests also enable the determination of soil characteristics needed to model soil behavior (see Appendix). Liquefaction risks can be reduced by measures such as replacing soft soil (expensive) or by consolidation measures. Soil consolidation can be done for example by vibro compaction or by preloading the soft bed with coarse sediment<ref name=MS> Van der Meer, J. and Sigurdarson, S. 2017. Design and construction of berm breakwaters. Advanced Series on Ocean Engineering vol. 40, World Scientific Publ. Co., Singapore</ref>. Monitoring of the consolidation process should ensure that a stable situation is reached. |
==Related articles== | ==Related articles== | ||
:[[ Stability of rubble mound breakwaters and shore revetments]] | :[[ Stability of rubble mound breakwaters and shore revetments]] | ||
+ | |||
+ | |||
+ | ===Further reading=== | ||
+ | Sumer, B.M. 2009. Liquefaction Around Marine Structures. Advanced Series on Ocean Engineering : Volume 39. World Scientific Publ. Co. | ||
Line 79: | Line 82: | ||
As an example, we consider sea surface waves with period <math>T = 10 \, s</math> and wavelength <math>L = 70 \, m</math>. | As an example, we consider sea surface waves with period <math>T = 10 \, s</math> and wavelength <math>L = 70 \, m</math>. | ||
− | The fluctuating wave load applies over a large stretch of order <math>L/(2 \pi)</math> length, which precludes substantial lateral soil | + | The fluctuating wave load applies over a large stretch of order <math>L/(2 \pi)</math> length, which precludes substantial lateral soil dilatation. We therefore assume horizontal soil uniformity and focus in the following on the vertical compression <math>dV/dt</math> of a small soil volume element <math>V</math> resulting from the wave-induced varying pressure load at the seabed surface Eq. (1). The compression <math>dV/dt</math> is due to: |
# compression of the pore fluid, <math>\Large\frac{dV_1}{dt}\normalsize = V \Large\frac{n}{\beta}\frac{\partial p}{\partial t}\normalsize ,</math> where <math>p(z,t)</math> is the pore fluid pressure at depth <math>z</math> below the seafloor, <math>n</math> is the porosity and <math>\beta</math> is the bulk modulus of the pore water (including gaz). | # compression of the pore fluid, <math>\Large\frac{dV_1}{dt}\normalsize = V \Large\frac{n}{\beta}\frac{\partial p}{\partial t}\normalsize ,</math> where <math>p(z,t)</math> is the pore fluid pressure at depth <math>z</math> below the seafloor, <math>n</math> is the porosity and <math>\beta</math> is the bulk modulus of the pore water (including gaz). | ||
# pore water outflow, <math>\Large\frac{dV_2}{dt}\normalsize = V \Large\frac{\partial u}{\partial z}\normalsize = V \Large\frac{k}{\rho g}\frac{\partial^2 p}{\partial z^2}\normalsize , </math> where <math>u</math> is the pore water vertical flow velocity, and <math>k</math> is the soil permeability. The second equation follows from [https://en.wikipedia.org/wiki/Darcy%27s_law Darcy's law]. | # pore water outflow, <math>\Large\frac{dV_2}{dt}\normalsize = V \Large\frac{\partial u}{\partial z}\normalsize = V \Large\frac{k}{\rho g}\frac{\partial^2 p}{\partial z^2}\normalsize , </math> where <math>u</math> is the pore water vertical flow velocity, and <math>k</math> is the soil permeability. The second equation follows from [https://en.wikipedia.org/wiki/Darcy%27s_law Darcy's law]. | ||
− | The volume compression | + | The volume compression implies a change in the volumetric strain (also called 'bulk strain') <math>d\epsilon=\Large\frac{1}{V}\normalsize (dV_1+dV_2)</math>, |
− | <math>\Large\frac{d \epsilon}{dt}\normalsize = \Large\frac{n}{\beta}\frac{\partial p}{\partial t}\normalsize + \Large\frac{k}{\rho g}\frac{\partial^2 p}{\partial^2 | + | <math>\Large\frac{d \epsilon}{dt}\normalsize = \Large\frac{n}{\beta}\frac{\partial p}{\partial t}\normalsize + \Large\frac{k}{\rho g}\frac{\partial^2 p}{\partial z^2}\normalsize . \qquad (2)</math> |
− | The | + | The soil is considered a poroelastic medium for which [https://en.wikipedia.org/wiki/Hooke%27s_law Hooke's law] applies. The volumetric strain <math>\epsilon</math> (degree of soil compression) is then linearly related to the effective stress (inter-grain stress) <math>\sigma'</math> by the formula |
<math>\epsilon = -m_v \sigma' = -m_v (\sigma - p) , \qquad (3)</math> | <math>\epsilon = -m_v \sigma' = -m_v (\sigma - p) , \qquad (3)</math> | ||
Line 101: | Line 104: | ||
<math>p(x,z,t) = p_{seafloor} + \large\frac{m}{1+m}\normalsize p_0 \Bigg[- \cos \Big( 2 \pi (\large\frac{x}{L}-\frac{t}{T}\normalsize)\Big) + e^{-\Large\frac{z}{d}\normalsize} \cos \Big( 2 \pi \large (\frac{x}{L}-\frac{t}{T})+\frac{z}{d}\normalsize\Big) \Bigg] .\qquad (5)</math> | <math>p(x,z,t) = p_{seafloor} + \large\frac{m}{1+m}\normalsize p_0 \Bigg[- \cos \Big( 2 \pi (\large\frac{x}{L}-\frac{t}{T}\normalsize)\Big) + e^{-\Large\frac{z}{d}\normalsize} \cos \Big( 2 \pi \large (\frac{x}{L}-\frac{t}{T})+\frac{z}{d}\normalsize\Big) \Bigg] .\qquad (5)</math> | ||
− | This expression is only valid near the soil surface, as the amplitude <math>p_0</math> of the wave-induced pressure fluctuation is a gradually decreasing function of depth. The depth <math>d</math> is given by | + | This expression is only valid near the soil surface (<math>z << L</math>), as the amplitude <math>p_0</math> of the wave-induced pressure fluctuation is a gradually decreasing function of depth. The depth <math>d</math> is given by |
<math>d = \Big( \Large\frac{T \, c_v}{\pi}\normalsize \Big)^{1/2} = \Big( \Large\frac{T \, k \, \beta \, m}{\pi \, \rho \, n \, g \, (1+m)}\normalsize \Big)^{1/2} . \qquad (6)</math> | <math>d = \Big( \Large\frac{T \, c_v}{\pi}\normalsize \Big)^{1/2} = \Big( \Large\frac{T \, k \, \beta \, m}{\pi \, \rho \, n \, g \, (1+m)}\normalsize \Big)^{1/2} . \qquad (6)</math> | ||
Line 137: | Line 140: | ||
In the numerical example, the coefficient of consolidation (Eq. 4) takes the value <math>c_v \approx 0.007 \, m^2/s</math>. Taking <math>B = 10^{-3}</math> and <math>D = 10 \, m</math>, residual liquefaction occurs for wave heights exceeding 1.6 m. | In the numerical example, the coefficient of consolidation (Eq. 4) takes the value <math>c_v \approx 0.007 \, m^2/s</math>. Taking <math>B = 10^{-3}</math> and <math>D = 10 \, m</math>, residual liquefaction occurs for wave heights exceeding 1.6 m. | ||
− | |||
− | |||
− | |||
− | |||
Latest revision as of 17:38, 7 March 2024
Definition of Soil liquefaction:
When a water-saturated soil starts behaving as a fluid, losing stiffness and bearing capacity.
This is the common definition for Soil liquefaction, other definitions can be discussed in the article
|
Seabed soils which are not well consolidated are prone to liquefaction when subjected to pressure fluctuations. Soil liquefaction can be produced for example by earthquakes, but here we concentrate on submarine soils which are subjected to cyclic loading by waves. Structures built on freshly deposited submarine soils can collapse or sink when the seabed is liquified[1]. A simplified mathematical introduction to the theory of soil liquefaction is given in the Appendix.
Contents
Two types of soil liquefaction
Two types of wave-induced soil liquefaction can be distinguished: Transient liquefaction (also called instantaneous or momentary liquefaction) and residual liquefaction. Both types can occur in poorly drained loosely packed soils, for example soils containing a high percentage of fine silty material. In well-drained soils, for example a medium/coarse-sand seabed, pore pressure by cyclic loading cannot build up, but transient liquefaction is possible.
Transient liquefaction
The upper part of the seabed can be liquefied if the soil pore pressure near the bed surface is higher than the pressure exerted by the overlying water mass. This occurs when the pore pressure in the seabed follows (with some delay, depending on the soil permeability) the fluctuating water pressure exerted by surface waves. The soil top layer then experiences an upward force during the passage of the wave trough, which lifts sediment particles near the seabed surface out of the soil skeleton, see Fig. 1. The soil top layer (typically ten or a few tens of centimeters) becomes liquefied over a short period of time during which this condition prevails. One consequence is that objects deposited on the seabed can sink and become buried[3][1] or float up if the density of the object is less than that of the liquified soil.
Most observations of wave-induced transient liquefaction come from laboratory experiments. There are also some studies based on field observations, in the nearshore subtidal zone and in the intertidal zone. In the latter case it was found that even a small volume fraction of air in the soil can greatly dampen the transmission of wave-induced water pressure to the pore fluid[4].
Residual liquefaction
Soils prone to liquefaction are typically water-saturated, consisting of loosely packed fine sediments – fine sand, silt and some clay – with poor drainage capacity (low hydraulic conductivity). When such a soil is subjected to cyclic wave-induced loading and associated shear deformations, the soil grains tend to rearrange such that the soil skeleton is progressively compressed. (The reverse happens for a densely compacted soil.) The resulting reduction of the pore volume is associated with an increase of the pore water pressure. When the pore pressure exceeds the pressure exerted by the load of the overlying water and soil (pore pressure higher than the initial mean normal effective stress, taking soil cohesion into account), the contact friction points between soil grains are broken and the soil skeleton collapses. The soil grains become unbound and free to move, carried by the pore water (Fig. 2). The soil begins to act like a liquid, losing its stiffness and bearing capacity. Any structure built on this soil will sink or collapse[1].
Experiments in a wave flume have shown that the residual pore pressure generated by random waves can be much larger than that by regular waves with equivalent mean wave height and period. Low-frequency waves in particular can make a significant contribution to the residual pressure[5]. Other flume experiments have shown that soil liquefaction through excess pore pressure enhances sediment re-suspension[6]. Seabed liquefaction and seabed scour reinforce each other, as removal of top soil strengthens the upward pore pression gradient in the soil underneath[7]
Liquefaction-consolidation process
The onset of liquefaction occurs first at the surface of the bed, and rapidly spreads out across the soil depth, causing the entire soil to behave like a liquid. When the liquefaction reaches the impermeable base, the soil begins to compact. The large pores between particles in the liquefied state allow the expel (drainage) of pore water and wave-induced compaction. Soil grains fall out of the liquid state, settle through the pore water until they come into contact with each other. The compaction gradually progresses from the impermeable base in the upward direction, and the entire sequence of the liquefaction/compaction process comes to an end when the compaction reaches the soil surface[8][9]. After liquefaction, a loosely packed soil will not recover its initial state. The soil is compacted and much less susceptible to renewed liquefaction. The liquefaction-compaction process is depicted schematically in Fig. 3.
Influence of soil composition
Flume experiments show that the liquefaction potential strongly depends on the soil composition. Experiments by Gratchev et al. (2006[11]) revealed the significant influence of plasticity on soil liquefaction resistance. Soil vulnerable to liquefaction had an open microfabric in which clay aggregates mainly gathered at the silt particle contact points, forming low-strength 'clay bridges' that were easily destroyed during cyclic loading (Fig. 2). The contact friction between particles that maintained the soil skeleton structure was reduced by the clay aggregates, so grains could easily slip and rotate once the seabed was subjected to wave loading. Clay minerals thus enhance the plasticity and compressibility of the clay-silt seabed and increases the sensitivity of the seabed to liquefaction.
Experiments with kaolinite-fine sand and illite-fine sand showed that the susceptibility of a soil to liquefaction increases with increasing clay content if the clay content is low (0.5-5%), but decreases with a higher clay content[12]. At a clay content of more than 10-30%, depending on the type of clay, the silt-clay mixture was not susceptible to liquefaction. This can be explained by the fact that in case of high clay content the silt grains are completely encapsulated in the clay matrix and therefore cannot rearrange under cyclic shear stresses, causing resistance to liquefaction. Experiments by Kirca et al. (2014[13]) showed that the sensitivity of a sand-clay mixture to wave-induced liquefaction not only depends on the clay content but also on the sand grainsize. Liquefaction occurred at a much larger clay content for medium sand than for fine sand or silt. The presence of shell fragments also decreases the susceptibility of silt to liquefaction[10]. Experiments for a medium-sand-clay (kaolinite) seabed supporting a submerged breakwater showed that wave-induced liquefaction required a clay content greater than 40%[14]. These experiment also showed that the initial water content in the soil is relevant to the liquefaction potential as it affects the pressure build-up process. Higher pore water fractions allow the residual pore pressure to accumulate faster.
The soil in an actual field situation usually has a long history of wave action, and hence it normally is a consolidated stiff soil, with low sensitivity to wave-induced liquefaction. This does not hold for freshly deposited soils. This may be the case, for example, where a pipeline is laid in a trench, and the trench is then backfilled[1].
Fluid mud
Fluid mud – a dense colloidal suspension of fine, cohesive sediments (mainly clay minerals) – can be considered liquified soil. However, cohesive soils are not liquefied by a similar process of pore pressure build-up as non-cohesive or slightly cohesive soils. Cohesive soil (mud) behaves like a viscoelastic material whose viscosity decreases when a stress is applied that is strong enough (fluctuates) to break the electrochemical bounds that hold the constituent particles together[15][16]. Cohesive soil has thixotropic properties: after stress removal, it gradually returns to its original state. A fluid mud layer is formed when floc-forming particles settle to the seabed in sufficiently high concentration. High concentrations of suspended fine sediment often originate from previous mud bed erosion by strong waves or currents or from convergence of residual fine sediment transport related to tidal asymmetry and estuarine circulation (see Estuarine turbidity maximum). Stirring of unconsolidated mud deposits by wave-induced pressure variations is another process that can contribute to the formation of fluid mud[17]. The high stresses needed in case of advanced consolidation will more likely lead to sediment erosion and suspension rather than to fluidization. A fluid mud layer can be formed when the suspended sediment settles[18]. When a fluid mud layer flows along the seabed, the density (and viscosity) can increase by bed erosion or decrease by uptake (entrainment) of overlying water[19]. See Fluid mud and Dynamics of mud transport.
Preventing soil liquefaction
The susceptibility of a soil to liquefaction can be checked through Standard Penetration Tests or Cone Penetration Tests. Such tests also enable the determination of soil characteristics needed to model soil behavior (see Appendix). Liquefaction risks can be reduced by measures such as replacing soft soil (expensive) or by consolidation measures. Soil consolidation can be done for example by vibro compaction or by preloading the soft bed with coarse sediment[20]. Monitoring of the consolidation process should ensure that a stable situation is reached.
Related articles
Further reading
Sumer, B.M. 2009. Liquefaction Around Marine Structures. Advanced Series on Ocean Engineering : Volume 39. World Scientific Publ. Co.
Appendix
This appendix provides a mathematical introduction to the phenomena of transient and residual liquefaction. It is not intended as a model for estimating liquefaction conditions in practice.
Transient liquefaction
According to linear wave theory (see Shallow-water wave theory) the fluctuating pressure (vertical stress) exerted on the seabed by a propagating sinusoidal wave is given by
[math]p_{seafloor} (x,t) = \rho g h + p_0 \cos \Big( 2 \pi (\large\frac{x}{L}-\frac{t}{T}\normalsize)\Big) , \quad p_0 = \Large\frac{ \rho g a}{\cosh(2 \pi h/L)} \normalsize, \qquad (1) [/math]
where [math]x=[/math] coordinate wave propagation direction, [math]t=[/math] time, [math]g=[/math] gravitational acceleration, [math]\rho=[/math] seawater density, [math]a=[/math] wave amplitude, [math]L=[/math] wavelength, [math]T=[/math] wave period, [math]h=[/math] water depth.
We consider loose-packed homogeneous soil consisting of fine sand, with fractions of silt and clay. The soil characteristics are:
- [math]n[/math] porosity of the soil, i.e. the volume not occupied by solids, thus left for pore water and gaz (air)
- [math]k[/math] [m/s] soil permeability (= hydraulic conductivity), relating pore water flow and pressure gradient according to Darcy's law
- [math]G[/math] [Pa] shear modulus of soil (= modulus of rigidity = the ratio of shear stress and shear strain)
- [math]\nu[/math] Poisson ratio of soil (= amount of transversal elongation divided by the amount of vertical compression, related to the friction angle [math]\phi[/math])
- [math]\beta[/math] [Pa] bulk modulus of pore water and included air (= the ratio of pressure increase and relative volume decrease = inverse of compressibility). The water density-pressure relationship is [math]\rho = \rho_0 \exp((p-p_0)/\beta)[/math] and the dependence of [math]\beta[/math] on the air fraction [math]C[/math] is [math]1/\beta = 5 \, 10^{-9} \, (1-C) + C/p \;[/math].
Typical values for loosely packed soil of fine sand with a small fraction of silt an clay, saturated with water and including a small fraction (order 1‰) of included air, are[21]: [math]n=0.4, \; k=10^{-6} m/s, \; G=10^7 Pa, \; \beta=10^8 Pa, \; \nu=0.45[/math]. As an example, we consider sea surface waves with period [math]T = 10 \, s[/math] and wavelength [math]L = 70 \, m[/math].
The fluctuating wave load applies over a large stretch of order [math]L/(2 \pi)[/math] length, which precludes substantial lateral soil dilatation. We therefore assume horizontal soil uniformity and focus in the following on the vertical compression [math]dV/dt[/math] of a small soil volume element [math]V[/math] resulting from the wave-induced varying pressure load at the seabed surface Eq. (1). The compression [math]dV/dt[/math] is due to:
- compression of the pore fluid, [math]\Large\frac{dV_1}{dt}\normalsize = V \Large\frac{n}{\beta}\frac{\partial p}{\partial t}\normalsize ,[/math] where [math]p(z,t)[/math] is the pore fluid pressure at depth [math]z[/math] below the seafloor, [math]n[/math] is the porosity and [math]\beta[/math] is the bulk modulus of the pore water (including gaz).
- pore water outflow, [math]\Large\frac{dV_2}{dt}\normalsize = V \Large\frac{\partial u}{\partial z}\normalsize = V \Large\frac{k}{\rho g}\frac{\partial^2 p}{\partial z^2}\normalsize , [/math] where [math]u[/math] is the pore water vertical flow velocity, and [math]k[/math] is the soil permeability. The second equation follows from Darcy's law.
The volume compression implies a change in the volumetric strain (also called 'bulk strain') [math]d\epsilon=\Large\frac{1}{V}\normalsize (dV_1+dV_2)[/math],
[math]\Large\frac{d \epsilon}{dt}\normalsize = \Large\frac{n}{\beta}\frac{\partial p}{\partial t}\normalsize + \Large\frac{k}{\rho g}\frac{\partial^2 p}{\partial z^2}\normalsize . \qquad (2)[/math]
The soil is considered a poroelastic medium for which Hooke's law applies. The volumetric strain [math]\epsilon[/math] (degree of soil compression) is then linearly related to the effective stress (inter-grain stress) [math]\sigma'[/math] by the formula
[math]\epsilon = -m_v \sigma' = -m_v (\sigma - p) , \qquad (3)[/math]
where [math]p[/math] is the pore pressure and where [math]\sigma \approx p_{seafloor} + \alpha (\rho_{soil} - \rho) g z \;[/math] is the total stress applied to a soil volume [math]V[/math] at depth [math]z[/math] near the seabed surface ([math]z \lt \lt L [/math]). The factor [math]\alpha = (1+2K_0)/3[/math], where [math]K_0 = \nu / (1 - \nu) \approx 1- \sin \phi \;[/math] expresses the partial transfer of soil weight to horizontal stresses ([math]\phi =[/math] friction angle) [22]. Liquefaction occurs when the effective stress vanishes, [math]\sigma' = 0 [/math]. The compressibility modulus [math]m_v[/math] can be expressed as a function of the soil shear modulus [math]G[/math] and the Poisson ratio [math]\nu[/math], [math]\; m_v = \Large\frac{1-2 \nu}{2 G (1-\nu)}\normalsize \approx \Large\frac{\sin \phi}{2 G}\normalsize .[/math]
Using Hooke's law, Eq. (2) can be rewritten as a differential equation from which the pore pressure [math]p(z,t)[/math] can be solved:
[math]\Large\frac{\partial p}{\partial t}\normalsize = c_s \Large\frac{\partial \sigma}{\partial t}\normalsize + c_v \Large\frac{\partial^2 p}{\partial z^2}\normalsize , \quad c_s = \Large\frac{1}{1+m}\normalsize , \quad c_v = \Large\frac{k \beta}{\rho g n}\frac{m}{1+m}\normalsize , \qquad m = \Large\frac{n}{\beta m_v}\normalsize = 2 n \Large\frac{G}{\beta}\frac{1-\nu}{1-2 \nu}\normalsize . \qquad (4)[/math]
The time variation [math]\partial \sigma / \partial t [/math] of the total stress is due to the fluctuating pressure transmitted by the surface wave to the pore fluid. The solution of Eq. (4) for a homogeneous soil layer with pore pressure [math]p=p_{seafloor}[/math] at the seabed surface [math]z=0[/math] is
[math]p(x,z,t) = p_{seafloor} + \large\frac{m}{1+m}\normalsize p_0 \Bigg[- \cos \Big( 2 \pi (\large\frac{x}{L}-\frac{t}{T}\normalsize)\Big) + e^{-\Large\frac{z}{d}\normalsize} \cos \Big( 2 \pi \large (\frac{x}{L}-\frac{t}{T})+\frac{z}{d}\normalsize\Big) \Bigg] .\qquad (5)[/math]
This expression is only valid near the soil surface ([math]z \lt \lt L[/math]), as the amplitude [math]p_0[/math] of the wave-induced pressure fluctuation is a gradually decreasing function of depth. The depth [math]d[/math] is given by
[math]d = \Big( \Large\frac{T \, c_v}{\pi}\normalsize \Big)^{1/2} = \Big( \Large\frac{T \, k \, \beta \, m}{\pi \, \rho \, n \, g \, (1+m)}\normalsize \Big)^{1/2} . \qquad (6)[/math]
Substitution of the values of the numerical example gives [math]m \approx 0.4 , \; d \approx[/math] 16 cm. The pore pressure gradient just below the seafloor ([math]z \lt \lt d[/math]) can be approximated by
[math]\Large\frac{\partial p(x,z,t)}{\partial z}\normalsize \approx - p_0 \Large\frac{m}{m+1}\frac{\sqrt{2}}{d}\normalsize \cos \Big( 2 \pi \large ( \frac{x}{L}-\frac{t}{T} )-\frac{\pi}{4} \normalsize\Big) . \qquad (7)[/math]
For liquefaction, the wave-induced pore pressure gradient near the soil surface must be larger than the submerged weight of the soil grains. This condition reads
[math]\Large\frac{m \sqrt{2}}{1+m}\frac{p_0}{d}\normalsize \gt g (\rho_{soil} - \rho) . \qquad (8)[/math]
In the numerical example, considering a water depth of 5 m, liquefaction occurs if the wave height exceeds 1.3 m. The maximum depth of liquefaction can be found from Eq. (5) by equating [math]p(z,t) = p_{seafloor} + (\rho_{soil}-\rho) g z \;[/math]. Liquefaction of the top soil layer or formation of a sheet flow layer can occur already for lower waves due to lift forces generated by wave orbital flow along the seabed.
Equation (5) does not take into account the gradual build-up of pressure below the soil surface under sustained cyclic loading. The resulting residual liquefaction process is discussed below.
Residual liquefaction
For transient liquefaction, it has been assumed that the volumetric soil strain [math]\epsilon[/math] under wave loading is cyclic. According to the soil mechanics theory of Terzaghi (1925[23]), the strain [math]\epsilon[/math] can gradually increase under cyclic loading if the soil is saturated but poorly drained. We assume for simplicity that the total stress [math]\sigma[/math] increases throughout the soil in proportion to the number [math]N[/math] of wave cycles,
[math]\quad \Large\frac{d \sigma}{dt}\normalsize \propto \Large\frac{p_0}{T}\normalsize , \qquad (9)[/math].
where [math]p_0[/math] is the amplitude of the fluctuating vertical stress on the seafloor given in Eq. (1). We call [math]\overline{p}(z,t)[/math] the wave-averaged pore pressure component related to the fluctuating surface water pressure. Instead of Eq. (3) we now have the following equation for [math]\overline{p}(z,t)[/math] :
[math]\Large\frac{\partial \overline{p}}{\partial t}\normalsize= B \Large\frac{p_0}{T} + c_v \Large\frac{\partial^2 \overline{p}}{\partial z^2}\normalsize , \qquad (10)[/math]
where [math]B[/math] is a proportionality constant dependent on the soil characteristics, that must be determined experimentally. An approximate solution of Eq. (8) for a homogeneous soil layer of thickness [math]0.1 L \lt D \lt 0.3 L[/math] can be found by assuming a parabolic depth dependence of the pore pressure[21]. The wave-averaged boundary conditions are: for [math]z=0: \; \overline{p}(z,t)=0 \;[/math] (no pore pressure builds up at the seabed surface because pore water can flow out here) and for [math] \; z=D: \; \Large\frac{\partial \overline{p}}{\partial z}\normalsize =0 \;[/math]. The approximate solution is (exact when depth-averaged):
[math]\overline{p}(z,t) \approx z(2D-z)\Large\frac{B p_0}{2 c_v T}\normalsize \Big[ 1 - e^{-\Large\frac{3 c_v t}{D^2}}\normalsize \Big] . \qquad (11)[/math]
The dimensionless coefficient [math]B[/math] represents the rate of pore pressure increase due to soil compaction generated in one wave cycle by a unit applied maximum pressure load. For loosely packed fine sand this rate is of the order [math]B \approx 10^{-3}[/math]; for densely packed soil [math]B[/math] is much smaller[21]. The coefficient [math]B[/math] has been assumed constant in Eq. (10), but in reality it will gradually decrease during soil compaction. The formula (11) therefore overestimates values of the pore pressure derived from measurements (Fig. 3).
The condition for liquefaction near the seabed surface is
[math]\Large\frac{\partial p}{\partial z} |_{z=0} \normalsize = \Large\frac{BDp_0}{c_v T}\normalsize \gt g(\rho_{soil} -\rho) . \qquad (12)[/math]
In the numerical example, the coefficient of consolidation (Eq. 4) takes the value [math]c_v \approx 0.007 \, m^2/s[/math]. Taking [math]B = 10^{-3}[/math] and [math]D = 10 \, m[/math], residual liquefaction occurs for wave heights exceeding 1.6 m.
References
- ↑ 1.0 1.1 1.2 1.3 Sumer, B.M., Fredsøe, J., Christensen, S. and Lind, M.T. 1999. Sinking/floatation of pipelines and other objects in liquefied soil under waves. Coast. Eng. 38: 53–90 Cite error: Invalid
<ref>
tag; name "S99" defined multiple times with different content - ↑ de Groot, M. B., Bolton, M. D., Foray P., Meijers, P., Palmer, A. C., Sandven, R., Sawicki, A. and The, T. C. 2006 Physics of Liquefaction Phenomena around Marine Structures. Journal of Waterway, Port, Coastal, and Ocean Engineering 132: 227-243
- ↑ Klammler, H., Penko, A.M., Staples, T., Sheremet, A. and Calantoni, J. 2021. Observations and modeling of wave-induced burial and sediment entrainment: Likely importance of degree of liquefaction. Journal of Geophysical Research: Oceans 126, e2021JC017378
- ↑ Michallet, H., Mory, M. and Piedra-Cueva, I. 2009. Wave-induced pore pressure measurements near a coastal structure, J. Geophys. Res. 114, C06019
- ↑ Niu, J.W., Xu, J.S., Dong, P. and Li, G.X. 2019. Pore water pressure responses in silty sediment bed under random wave action. Sci. Rep. 9, 11685
- ↑ Jia, Y., Zhang, L., Zheng, J., Liu, X., Jeng, D-S. and Shan, H. 2014. Effects of wave-induced seabed liquefaction on sediment re-suspension in the Yellow River Delta. Ocean Engineering 89: 146–156
- ↑ Wang, L., Zhang, J., Jeng, D-S., Zhang, Q. and Chen, T. 2023. Experimental study on the dynamic response of a silty seabed under waves. Ocean Engineering 269, 113554
- ↑ Sumer, B.M., Hatipoglu, F., Fredsoe, J. and Sumer, S.K. 2006. The sequence of sediment behaviour during wave-induced liquefaction. Sedimentology 53: 611–629
- ↑ Sui, T., Sumer, B.M., Kirca, V.S.O., Carstensen, S., Zheng, J. and Fuhrman, D.R. 2023. Effect of history of wave exposure on seabed liquefaction. Coastal Engineering 183 (2023) 104307
- ↑ 10.0 10.1 Sui, T., Kirca, V.S.O, Sumer, B.M., Carstensen, S. and Fuhrman, D.R. 2022. Wave-induced liquefaction in a silt and seashell mixture. Coastal Engineering 178, 104215
- ↑ Gratchev, I.B., Sassa, K., Osipov, V.I. and Sokolov, V.N. 2006. The liquefaction of clayey soils under cyclic loading. Eng. Geol. 86: 70–84
- ↑ Zang, J., Jiang, Q., Jeng, D., Zhang, C., Chen, X. and Wang, L. 2020. Experimental Study on Mechanism of Wave-Induced Liquefaction of Sand-Clay Seabed. J. Mar. Sci. Eng. 8, 66
- ↑ Kirca, V., Sumer, B.M. and Fredsøe, J. 2014. Influence of clay content on wave-induced liquefaction, J. Waterw. Port Coast. Ocean Eng. ASCE 140, 04014024
- ↑ Chavez, V., Mendoza, E., Silva, R., Silva, A. and Losada, M.A. 2017. An experimental method to verify the failure of coastal structures by wave induced liquefaction of clayey soils. Coastal Engineering 123: 1–10
- ↑ McAnally, W.H., Friedrichs, C., Hamilton, D., Hayter, E., Shrestha, P., Rodriguez, H., Sheremet, A. and Teeter, A. 2007. Management of Fluid Mud in Estuaries, Bays, and Lakes. I: Present State of Understanding on Character and Behavior. J. Hydr. Eng. 133: 9-22
- ↑ Robillard, D.J. Mehta, A.J. and Safak, I. 2023. Comments on wave-induced behavior of a coastal mud. Coastal Engineering 186, 104400
- ↑ McAnally, W.H., Friedrichs, C., Hamilton, D., Hayter, E., Shrestha, P., Rodriguez, H., Sheremet, A. and Teeter, A. 2007. Management of Fluid Mud in Estuaries, Bays, and Lakes. I: Present State of Understanding on Character and Behavior. J. Hydr. Eng. 133: 9-22
- ↑ Winterwerp, J.C., de Boer, G.J., Greeuw, G. and van Maren, D.S. 2012. Mud-induced wave damping and wave-induced liquefaction. Coastal Engineering 64: 102–112
- ↑ Winterwerp, J.C., Vroom, J., Wang, Z-B., Krebs, M., Hendriks, E.C.M., van Maren, D.S., Schrottke, K., Borgsmüller, C. and Schol, A. 2017. SPM response to tide and river flow in the hyper-turbid Ems River. Ocean Dynamics 67: 559–583
- ↑ Van der Meer, J. and Sigurdarson, S. 2017. Design and construction of berm breakwaters. Advanced Series on Ocean Engineering vol. 40, World Scientific Publ. Co., Singapore
- ↑ 21.0 21.1 21.2 Verruijt, A. 2006. Offshore soil mechanics. Delft University Lecture Notes
- ↑ Yamamoto, T. 1981. Wave-induced pore pressures and effective stresses in inhomogeneous seabed foundations. Ocean Eng. 8: 1-16
- ↑ Terzaghi, K. 1925. Principles of Soil Mechanics: IV—Settlement and consolidation of clay. Engineering News-Record 95(22): 874-878
Please note that others may also have edited the contents of this article.
|