PHYSICAL MODELLING OF TSUNAMI INDUCED SEDIMENT TRANSPORT AND SCOUR

PHYSICAL MODELLING OF TSUNAMI INDUCED SEDIMENT TRANSPORT AND SCOUR Adedotun Moronkeji, University Of Missouri-Rolla O.H. Hinsdale Wave Research Laboratory (Tsunami Wave Basin), Oregon State University Faculty Mentor: Prof. Yin Lu (Julie) Young, Princeton University Abstract The experimental project focuses on the effect of tsunami runup and drawdown, which can mobilize substantial amount of sediments deposits. Consequently, this results in erosion and scour damage that can undermine building foundations and other coastal infrastructures. The project will help to improve the understanding of the effect of tsunamis on sediments deposits and it effects on structures. The ultimate goal of this project is to develop performance based tsunami engineering design methodology for coastal infrastructures. Maximum inundation limits, changes in coastline morphology, wave forces, and maximum scour around structures are all important factors that are needed for performance based tsunami engineering design methodology. However, these effects are not well understood. The objective of this project is to improve our understanding of tsunami induced sediment transport and scour, and wave-soilstructure interactions. Introduction Background and Objectives People are becoming increasingly aware of the effects of tsunami on our coastal line. As such, a performance based tsunami engineering design methodology for coastal infrastructures is necessary so that we do not witness such devastation and destruction of infrastructures and loss of life that occurred in the Indian Ocean tsunami in December 2004. High intensity wave runup and drawdown over a movable bed, and the associated scour that accompany are so unpredictable and it has been a major factor causing damage to infrastructures but they are not well understood. Sediment transport and scour are important because of the damage they can cause to structures. In addition to wave forces on the structure, tsunamis can mobilize significant portions of the soil substrate, leading to undermining of the structural foundation, slope instability failures, and even liquefaction failures. De Groot, M.B. et al (2006) identified the different kinds of liquefaction that can affect marine structure as complete and partial liquefaction. Liquefaction occurs when the normal effective stress becomes zero and this can occur in two ways, if the total stress remains the same but the pore pressure increases or when the total stress decreases but the pore pressure remains the same. Complete liquefaction has to do with the total loss of effective stress. Effective stress is that portion of the applied load that is

carried by the soil particles, it s the total stress minus the pore pressure, and partial liquefaction can be caused by two different types of loadings which are monotonic and cyclic loading. Some of the factors that influence liquefaction are the characteristics of the loading, the compressibility of the pore water, the sensitivity volume strain of the soil structure to the soil permeability and the distance of the drainage. The effect of liquefaction due to wave-seabed interaction was investigated by Chowdhury, B. et al (2006) and they found that the pore pressure within the seabed increases with increase in the wave period, permeability and degree of saturation of the sand bed. They also found that the most important factor affecting the depth to which liquefaction occurs is the degree of saturation, because it affects the compressibility of the pore fluid relative to the soil. Until now, there has not been much systematic study on the high intensity, transient wave runup and drawdown effects on a mobile bed. Physical simulations of solitary wave and cnoidal wave over a movable sand bed with an initial constant slope of 1:10 and 1:15 and with a reef are conducted at the tsunami wave basin of the O.H. Hinsdale wave research laboratory at the Oregon state University in Corvallis, Oregon. The wave basin is 48.8m long, 26.5m wide and 2.1m deep, the wave maker is a piston-type, electric motor with 29 wave boards, 2.0m high which are capable of producing regular, irregular, tsunami multidirectional and user defined wave types. The period ranges from 0.5 to 10 seconds with a maximum wave of 0.8m in 1m depth of water. The maximum stroke of the wave maker is 2.1m and the maximum velocity of the wave board is 2.0m/s. Two flumes each 32.75m long, 2.16m wide and 2m deep were constructed on the south end of the wave basin and filled with natural Oregon beach sand with mean grain diameter D 50 =0.2mm at a constant slope of 1:10 and 1:15 in the two flumes respectively. For this project, 1m and 1.1m depth of water were used. The results obtained are used to analyze tsunami-induced erosion and deposition. For the 1:10 slope case with the reef, a 45cm diameter cylinder will be embedded in the sand to investigate tsunami-induced scour around a cylindrical structure. Tonkin S. et al (2003) performed an experiment to study the effect of tsunami around a cylinder and discovered that most of the scour occurs at the end of the drawdown as a result of increase in pore pressure gradients rather than shear stresses. During the drawdown, the shear stress reduces because the flow velocity is also reducing. Their experiment took into consideration only solitary waves and scouring as a result of tsunami runup only. In their simple model for calculating tsunami flow speed, Jaffe, B.E., and Gelfenbuam, G.A. (2007) observed that the grain size of a tsunami deposit has a strong influence on the calculated flow speed whereas the deposit thickness is not a strong indicator of flow speed. However their model is only applicable under the conditions that deposit are formed from suspension and calculated flow speeds from deposits are formed under steady flow. The models implicit assumption is that all the source are from offshore and only the runup were considered, the effect of drawdown were not considered. The objective of this is to study the erosion and or deposition processes due to runup and drawdown of solitary and or cnoidal waves over a movable bed. The wave profile, flow speed, fluid pressure gradient, sediment flux rates, bottom morphology, and pore pressure gradients will all be measured. Soil remediation option will also be investigated to determine possible mitigation strategies for coastal area infrastructures.

Experimental approach Experimental Setup The set of experiments conducted were 4 different cases, Case I was 1:10 constant slope, Case II was the 1:10 slope with reef, case III was the 1:10 slope with a 45cm cylinder with reef, and Case IV was the 1:15 constant slope. Figure 1 and Figure 2 illustrate the experimental setup for the 1:10 constant slope flume. 10m 17m 18m 19m 20m 21m 22m 23m 24m 25m 26m 27m 28m 29m 30m 31m 32m 33m 34m 35m 36m 37m 41.5m A Wall 3 200mm Wave Maker N Wall 2 Wall 1 200mm 820mm ADV 1 ADV 2 ADV 3 ADV 4 ADV 5 ADV 6 ADV 7 520mm PPS 1-4 PPS 5-8 DS 1 DS 2 DS 3 DS 4 WG 2 WG 3 WG 4 WG 5 WG 6 WG 7 WG 8 WG 9 WG 10 WG 12 WG 1 WG 11 820mm 200mm Camera 9 Camera 8 Camera 5 II-II 15358mm Note: 1).Wall 1 & 2 are Flume wall and Wall 3 is Wave basin wall 2). Distance from wave maker to interior end of flume wall is 41500mm Downward looking ADVs Sideward looking ADVs 1118mm 1118mm 1118mm 1016mm 1016mm 32750mm Note 12007mm Resistance Wave Gauge (12) - WG 200mm 4920mm A 2160mm 2160mm 200mm ADV ( 4 above waterline and 3 below waterline- all mounted vertically to allow for easy vertical adjustment) Acoustic Wave Gauge (4) - DS Pore Pressure Sensor (8) - PPS Figure 1: Flume instrumentation plan view (1:10 constant slope) 10m 17m 18m 19m 20m 21m 22m 23m 24m 25m 26m 27m 28m 29m 30m 31m 32m 33m 34m 35m 36m 37m Water level 1:10 134mm 2000mm 1000mm Note: Flume wall height is 2000mm, Wave basin wall height is 2134mm Figure 2: Side elevation of flume (1:10 constant slope) The experiments were conducted in the two flumes each 32.75m long, 2.16m wide and 2m deep constructed on the south end of the wave basin and filled with natural Oregon beach sand with mean grain diameter D 50 =0.22mm at a constant slope of 1:10 and 1:15 in the two flumes respectively, a 1m depth of water was used for the 1:10 constant slope case and 1:15 constant slope (Case I and Case IV respectively) and a 1.1m depth of water was used for the 1:10 slope

case with reef and the cylinder experiments (Case II and III respectively). The offshore line extended to 27m away from the wave maker and the onshore line started at the 27m line, the 1:10 constant slope extend from 17m (offshore) to 37m (onshore) as shown in Figure 2, the 1:10 slope with reef and cylinder extend from 17m to 27m. Figure 3: Showing deployment of the sensors in the flume 12 ImTech Inc. constant-current linear wire wave gages with +/- 5V range was used in this experimental setup, the first was installed at 10m along the flume wall and the remaining 11 were installed starting at 17m with a meter interval between them along the flume wall, the wave gages report relative free surface elevation and accuracy is calibration-dependent. 4 D&A Instrument Co. model OBS-3Optical Backscatter Sensors (OBS) with a range of 0.02-2000FTU and 3 Nortek AS., model Vectrino+ Advanced Doppler Velocimeter (ADV) collocated are supported on the bridge, they can be adjusted vertically to allow for relocation to any desired locations. The location of the 4 OBS and 3 ADVs collocated on the bridge are shown in Figure 4. OBS sensors reports turbidity/suspended sediment concentration in water, the way OBS operates is by sending a beam of infrared light into the water and the suspended particles then scatters the beam and the quantity of light that is reflected back to the OBS sensor is measured by the OBS. OBS accuracy is calibration-dependent.

2560mm 2160mm 200mm 200mm Moving bridge Flume Wall 2 ADV 3 ADV 2 OBS 1 OBS 2 OBS 3 2000mm Flume Wall 1 ADV ADV 1 OBS 4 OBS 820mm 200mm 120mm 200mm 820mm Figure 4: Position of OBS and ADVs attached to the moving bridge. Note: 4 ADVs were used for Cases II-IV. 3 wall-mounted downward looking 3-D Acoustic Doppler Velocimeter (ADV) were vertically mounted at 24m, 25m and 26m mark below the waterline and 4 wall-mounted sideward looking 3-D Acoustic Doppler Velocimeter (ADV) were vertically mounted at 28m,29m,30m and 32m mark above the waterline so as to allow for adjustment of the vertical clearance from the top of the sand, the ADVs reports velocity, signal to noise ratio and along-beam correlation percentage, Range is set at +/- 1m nominal and accuracy is ±0.5% of measured value ±1mm/s, it is based on the Doppler shift effect and therefore it is a remote-sensing, three-dimensional velocity sensor. 4 Senix Corp, model TS-30S1-1V Distance sonic / ultrasonic wave gage (DS) were collocated with the 4 ADVs at 28m, 29m, 30m and 32m mark from the wave maker, the ultrasonic wave gages reports distance from transducer to boundary with a range between 10-427cm. There are 8 Druck/GE Inc., model PDCR81 Pore Pressure Sensor (PPS), 4 located horizontally at 25m and at 0.20m 0.65m vertically with 0.15m spacing in between them and another 4 located horizontally at 27m respectively and at 0.40m 0.85m vertically with 0.15m spacing in between them, the pore pressure sensors reports fluid pressure, the range is 0-5 psig and accuracy is ±0.2% below sea level (BSL). Accuracy is also dependent on signal conditioning module used. The instrumentation plan for Case II and III is similar to that of Case I except that the Pore Pressure Sensors (PPS) are placed at 27m and 29m respectively. Also 4 pressure transducers are mounted on the movable bridge wing. The primary function of a pressure transducer is to convert pressure into an analog electrical signal. 8 SeaTek Multi transducer array (MTA) reports distance from each transducer to boundary, the range is 5-420cm with a resolution of +/- 1mm were used to measure the bed profile after every 3 experimental test were ran, the purpose is to measure the change in the bed profile before and after every 3 runs. Accuracy is application/deploymentdependent. 3 underwater cameras were placed at the 3 windows respectively to capture the sediment transportation and scour through the windows, 2cm grid marks were placed on the windows to visualize the initial sediment level and the final sediment level after each experimental run and determine how much sediment transportation and scouring occurred during the runup and the drawdown process.

Figure 5: ADVs anchored to flume wall Figure 6: Underwater camera with the 2cm grid

Soil The soil used in this experiment is the natural Oregon beach sand. A sieve analysis that measures the grain size distribution of a soil by passing it through a series of carefully manufactured mesh of wires with a specified opening size was used to perform the particle size distribution for the Oregon beach sand. The grain size distribution curve is shown in Figure 7. The Oregon beach sand has a mean grain diameter D 50 =0.22mm. The coefficient of uniformity (C u ) for this soil was calculated as 1.77 using equation 1 and the coefficient of curvature (C c ) was calculated as 0.97 using equation 2. The soil has a low value of C u, which reflects that the soil is a poorly graded soil, which means soils with a narrow range of particle sizes. C u = D D 60 10 (Equation 1) ( ) 2 D30 C (Equation 2) c = D10 D60 Table 1 shows the physical properties of the Oregon beach sand. Direct shear test, vacuum triaxial test, permeability test and specific gravity of soil solids were conducted on the soil sample. 100 Oregon Beach Sand #200 #140 #60 #40 #20 #10 #4 90 80 Percent Passing 70 60 50 40 30 20 10 0 0.01 0.1 1 10 Sieve Size (mm) Oregon Beach Sand Figure 7: Grain size distribution curve

According to Sumer, B.S. et al (2007) field observations shows that very fine sand can be present in various relative densities on the ocean floor from D r =0-0.15 to 0.85-1 (very loose to very dense respectively). Paris, R. et al (2006) presented a description of the event in the Lhok Nga Bay and an interpretation of the tsunami sand deposits, mostly based upon grain-size analysis. They looked at successive deposition caused by consecutive runup and backwash, the topmost layer deposition was identified as caused by the backwash because it had increasing mean grainsize and reducing degree of sorting. They had only eye witness account and photograph after 3 weeks of the tsunami taking place, so speculations were used for the sediment transport and deposition that occurs during the tsunami. The physical properties of the soil sample are very important in determining its behavior during a tsunami. Table 1: Physical Properties of the Oregon beach Sand Mean grain size D 50 (mm) 0.22 Void ratio 0.72 Specific gravity 2.67 Porosity 0.42 Density (g/cm 3 ) 1.56 Phi angle( degrees) 34.3 Experimental Runs Solitary waves of 10cm, 30cm, 50cm and 60 cm and cnoidal waves of 30cm and 50cm with wave length of 12m and 8m respectively were generated by the piston-type, electric motor with 29 wave boards, and 2.0m high wave-maker for the 1:10 constant slope, Tables 2-5 shows all the experimental runs for the 3 different cases for the 1:10 slope and the one case for the 1:15 slope. 45 trial experimental runs were conducted for case I, 37 trial experimental runs were conducted for case II, 30 trial experimental runs were conducted for case III and 69 trial experimental runs were conducted for case IV. In all a total of 181 trial experimental runs were performed for all the slope cases. Table 2: Experimental runs for 1:10 constant slope (Case I) Case # Trial # Wave condition Water Depth (m) Wave Height (cm) Wave Length (m) M10 M10 Trial 01-09 M10_ST30 1 30 - M10 M10 Trial 10-12 M10_ST50 1 50 - M10 M10 Trial 13-21 M10_ST10 1 10 - M10 M10 Trial 22-30 M10_CN30 1 30 12 M10 M10 Trial 31-39 M10_ST50 1 50 - M10 M10 Trial 40-41 M10_CN50 1 50 12 M10 M10 Trial 42-43 M10_CN50 1 50 8 M10 M10 Trial 44-45 M10_ST60 1 60 -

Table 3: Experimental runs for 1:10 slope with reef (Case II) Case # Trial # Wave condition Water Depth (m) Wave Height (cm) Wave Length (m) RF10 RF10 Trial 01-09 RF10_ST10 1.1 10 - RF10 RF10 Trial 10-18 RF10_ST30 1.1 30 - RF10 RF10 Trial 19-27 RF10_CN30 1.1 30 12 RF10 RF10 Trial 27-30 RF10_ST50 1.1 50 - RF10 RF10 Trial 31-34 RF10_CN50 1.1 50 12 RF10 RF10 Trial 35-37 RF10_ST60 1.1 60 - Table 4: Experimental runs for 1:10 slope with cylinder (Case III) Case # Trial # Wave condition Water Depth (m) Wave Height (cm) Wave Length (m) CY10 CY10 Trial 01-09 CY10_ST10 1.1 10 - CY10 CY10 Trial 10-18 CY10_ST30 1.1 30 - CY10 CY10 Trial 19-27 CY10_CN30 1.1 30 12 CY10 CY10 Trial 28-30 CY10_ST50 1.1 50 - Table 5: Experimental runs for 1:15 constant slope (Case IV) Case # Trial # Wave condition Water Depth (m) Wave Height (cm) Wave Length (m) M15 M15 Trial 01-09 M15_ST10 1 10 - M15 M15 Trial 10-18 M15_ST30 1 30 - M15 M15 Trial 19-27 M15_CN30 1 30 12 M15 M15 Trial 28-42 M15_ST50 1 50 - M15 M15 Trial 43-51 M15_CN50 1 50 12 M15 M15 Trial 52-60 M15_ST20 1 20 - M15 M15 Trial 61-69 M15_ST60 1 60 - Results and Observations In their examination of solitary wave cross-shore sediment transport Kobayashi N. and Lawrence A.R. (2004) discovered that the wave motion and sediment transport were not really affected by the bed profile changes that occurred after running waves on the bed profile, they noticed that the effect of the strong downrush resulted in erosion on the foreshore and deposition seaward of wave run-down and concluded that more laboratory test in a larger wave tank will be needed to develop a suitable empirical formula to be able to predict accurately the cross-shore sediment transport and erosions. It was observed for the 1:15 constant slope that the waves breaking

occurred around 25 m, it was observed from video footages and MTA surveys carried out after several waves were ran that deposition occurs before the 25 m (offshore) where the waves breaking occurred as a result of strong downrush and that scour occurred after the 25 m (onshore) also a result of the downrush, this agrees somewhat with Kobayashi N. and Lawrence A.R. (2004) observations. Figure 8 shows the initial bed profile before any test trial was conducted, after trial # 6, after trial # 12 and then after trial # 21, it would be observed as expected that the deposition and scour becomes more obvious with each trial and at the end of trial # 21, the deposition and scour can be measured as approximately 0.5 m and 0.3 m respectively. It should be noted that the bed profile measurement were taken after every 3 trial runs. 2 1.5 Bed profile after trial # 0. For Exp. m15 2 1.5 Bed profile after trial # 6. For Exp. m15 MTA LBP Hand Survey Initial z [m] 1 z [m] 1 0.5 0.5 0 10 15 20 25 30 35 40 0 10 15 20 25 30 35 40 2 1.5 Bed profile after trial # 12. For Exp. m15 MTA LBP Hand Survey Initial 2 1.5 Bed profile after trial # 21. For Exp. m15 MTA LBP Hand Survey Initial z [m] 1 z [m] 1 0.5 0.5 0 10 15 20 25 30 35 40 0 10 15 20 25 30 35 40 Figure 8: Bed profile showing deposition and scour for 1:15 constant slope case after 4 different trials of 3 consecutive runs After the flume was drained, samples were taken from different locations in the flume to determine the void ratio. The result of the void ratio tests are given in Table 6. According to the results of the test, it can be concluded that the void ratio is within a range of 0.75 to 0.77. In the region where scour occurs, the void ratio increases to around 0.79 to 0.83,

which indicates that there exist excess pore pressure which uplift the soil skeleton and reduces the volume fraction of the solid sand particle in the mixture. Initially when the bed was configured it was observed that there was a lot of air bubbles on the surface of the soil and the soil is muddy which indicates that air occupies some of the pore spaces and the soil is not completely saturated by water, but after a few days it is observed that the water surface is cleaner which indicates that most of the air which were in the soil mixtures has dissipated and thus the soil mixture is more likely a two-phase porous media. A) B) Figure 9: Photographs showing partial saturation and complete saturation of the soil. A: Partial saturation of soil, B: Complete saturation of soil.

Table 6: Void ratio test weight of wet sand + container (gram) weight of wet sand S = 100% weight of dry sand + container (gram) weight of dry sand (gram) void ratio e= wgs S=100% Position Label 30 30R1 4.559 4.509 3.49 3.44 77.37% 30 30R2 4.514 4.464 3.48 3.43 74.93% 33 33R1 5.721 5.671 4.434 4.384 73.93% 33 33R2 4.363 4.313 3.376 3.326 73.55% 35 35R1 4.606 4.556 3.6111 3.5611 69.34% 35 35R2 4.328 4.278 3.37 3.32 71.40% 37 37R1 2.11 2.06 1.633 1.583 69.29% 37 37R2 2.785 2.735 2.173 2.123 68.54% 39 39R1 3.97 3.92 2.971 2.921 84.65% 39 39R2 3.522 3.472 2.771 2.721 67.04% 41 41R1 6.19 6.14 4.829 4.779 71.94% 41 41R2 4.661 4.611 3.677 3.627 67.31% 30 30L1 1.899 1.849 1.416 1.366 81.03% 30 30L2 2.68 2.63 2.055 2.005 74.15% 33 33L1 4.473 4.423 3.445 3.395 75.23% 33 33L2 4.03 3.98 3.062 3.012 79.45% 35 35L1 2.484 2.434 1.917 1.867 71.47% 35 35L2 2.389 2.339 1.828 1.778 74.08% 37 37L1 1.175 1.125 0.929 0.879 55.91% 37 37L2 2.73 2.68 2.24 2.19 52.05% 39 39L1 1.646 1.596 1.249 1.199 73.62% 39 39L2 2.907 2.857 2.223 2.173 75.58% 41 41L1 3.351 3.301 2.671 2.621 62.50% 41 41L2 5.073 5.023 4.035 3.985 64.89% CY shore 1 CY1 2.84 2.79 2.146 2.096 79.52% Grilli S.T. et al (1997) concluded that the most important parameter deciding the shape of breaking waves is the slope. It was observed through the videos footages from both the underwater cameras and handheld cameras that the location of wave breaking on both the 1:10 constant slope and the 1:15 constant slope were different, it was observed that the higher the wave heights the higher the breaking locations and maximum runup. Pore pressure sensor data are given for the 1:10 slope cylinder with reef case (Case IV) in figure 10 to show the changes in the pore pressure after different time intervals has elapsed during each wave run.

Figure 10: Pore pressure sensor graph from the 1:10 slope cylinder with reef test (Case IV).

Conclusions Analysis of the data are still on going but observations suggest that in the region where scour occurs, the void ratio increases which indicates that there exist excess pore pressure which uplift the soil skeleton and reduces the volume fraction of the solid sand particle in the mixture, also it was observed that the effect of the strong downrush resulted in erosion on the foreshore and deposition seaward of wave run-down as observed by Kobayashi N. and Lawrence A.R. (2004). The findings of this experiment will be more obvious after all the data has been processed and analyzed. Acknowledgment The funding to participate in this REU program was made possible by the Network for Earthquake Engineering Simulation (NEES) and NSF. Special thanks to Dr. Dan Cox, the director of the O.H. Hinsdale Wave Research Laboratory at Oregon State University, Alicia Lyman-Holt, the NEESReu coordinator at Oregon State University and all the faculty and Staff at O.H. Hinsdale Wave Research Laboratory at Oregon State University. I would also like to thank my faculty mentor, Dr. Yin Lu (Julie) Young, the PhD students who helped me in my research, Heng Xiao, Ting Tan and WaiChing Sun and my REU colleagues, Jennifer Krebs, Jason Miles and Brittany Snider. References Chowdury, B. et al. (2006). Laboratory Study of Liquefaction due to Wave-Seabed Interaction. Journal of Geotechnical and Geoenvironmental Engineering, Vol. 132, No. 7, pp. 842-851. de Groot, M. B. et al (2006). Physics of liquefaction phenomena around marine structures. Journal of Waterway, Port, Coastal, and Ocean Engineering/ ASCE/JULY/AUGUST 2006, pp. 227-243. Foster, D.L. et al (2006). Field evidence of pressure gradient induced incipient motion. Journal of Geophysical Research, Vol. III, C05004, doi: 10.1029/2004JC002863. Grilli S.T. et al (1997). Breaking criterion and characteristics for solitary wave on slopes. Journal of Waterway, Port, Coastal, and Ocean Engineering/ ASCE/MAY/JUNE 1997, pp. 102-112. Grilli S.T. et al (2004). Numerical modeling and Experiments for solitary wave shoaling and breaking over a Sloping Beach. The International Society of Offshore and Polar Engineers, pp. 306-312. Jaffe, B.E. and Gelfenbuam, G. (2007). A simple model for calculating tsunami flow speed from tsunami deposits. Sedimentary Geology, doi:10.1016/j.sedgeo.2007.01.013

Kobayashi, N. and Karjadi E.A. (1994). Solitary wave runup. Journal of Waterway, Port, Coastal, and Ocean Engineering, Vol. 120, No. 6, ASCE/November/December 1994, pp. 645-650. Kobayashi, N. and Johnson, B., D. (2001). Sand suspension, storage, advection, and settling in surf and swash zones. Journal of Geophysical Research, Vol. 106, No. C5, pp. 9363-9376. Kobayashi, N. and Lawrence, A.R. (2004). Cross-shore sediment transport under breaking solitary waves. Journal of Geophysical Research, Vol. 109, C03047, 1-13. Madsen, O.S. (1978). Wave-induced pore pressures and effective stresses in a porous bed. Geotechnique 28, No. 4, pp. 377-393. Madsen, O.S. and Durham, W.M. (1978). Pressure-induced subsurface sediment transport in the surf zone. Mei, C.C. and Foda, M.A. (1981). Wave-induced responses in a fluid-filled poro-elastic solid with a free surface-a boundary layer theory. Geotechnique 28, No. 4, pp. 377-393. Paris R. et al (2007). Coastal sedimentation associated with the December 26, 2004 tsunami in Lhok Nga, west Banda Aceh (Sumatra, Indonesia). International Journal of Marine Geology, Geochemistry and Geophysics 238, pp. 93-106. Sasaki, Y. and Tanaka, N. (2004). Situations of the Indian Ocean Tsunami at Dec. 2004 in Thailand, Sri Lanka and Mardives. Sumer B.M. et al (2007). Wave scout around a pile in sand, Medium dense, and Dense Silt. Journal of Waterway, Port, Coastal, and Ocean Engineering/ ASCE/JANUARY/FEBRUARY 2007, pp. 14-27. Tonkin, S. et al (2003). Tsunami scour around a cylinder. Journal of Fluid Mechanics, Vol. 496, pp. 165-192.