Ground Surface Displacement Research Articles

Parametric investigation of lateral spreading of gently sloping liquefied ground

This paper investigates the main parameters affecting the anticipated maximum surface displacements due to earthquake-induced lateral spreading of mildly sloping ground. The main tool used for this purpose is a numerical methodology employing a bounding surface plasticity model implemented in a finite difference code, which has been thoroughly validated against 16 published centrifuge lateral spreading experiments. This study shows that important problem parameters are the mean ground (surface) acceleration, the duration of strong shaking following the onset of liquefaction, the corrected SPT blowcount, the depth to the sliding plane, the inclination of the ground surface and the fines content of the liquefied soil layers. A new approximate multi-variable relation is proposed for the estimation of ground surface displacements due to lateral spreading in gently sloping ground, which includes the foregoing parameters. The form of the relation builds upon sliding block theory, but its final formulation is based on statistical analysis of the input data and the results from 120 parametric analyses performed with the validated numerical methodology. Comparison of the predictions of the proposed relation for ground surface displacement against pertinent field data (from 256 case histories) and centrifuge test measurements shows satisfactory accuracy. Furthermore, the variation of lateral displacements with depth is explored and distinct displacement patterns are proposed for uniform, 2-layer and 4-layer ground profiles.

Pixel Offset Technique for Measuring Local Large Ground Surface Displacement

Pixel Offset Technique for Measuring Local Large Ground Surface Displacement

Numerical models of caldera deformation: Effects of multiphase and multicomponent hydrothermal fluid flow

Ground surface displacement (GSD) in large calderas is often interpreted as resulting from magma intrusion at depth. Recent advances in geodetic measurements of GSD, notably interferometric synthetic aperture radar, reveal complex and multifaceted deformation patterns that often require complex source models to explain the observed GSD. Although hydrothermal fluids have been discussed as a possible deformation agent, very few quantitative studies addressing the effects of multiphase flow on crustal mechanics have been attempted. Recent increases in the power and availability of computing resources allow robust quantitative assessment of the complex time‐variant thermal interplay between aqueous fluid flow and crustal deformation. We carry out numerical simulations of multiphase (liquid‐gas), multicomponent (H2O–CO2) hydrothermal fluid flow and poroelastic deformation using a range of realistic physical parameters and processes. Hydrothermal fluid injection, circulation, and gas formation can generate complex, temporally and spatially varying patterns of GSD, with deformation rates, magnitudes, and geometries (including subsidence) similar to those observed in several large calderas. The potential for both rapid and gradual deformation resulting from magma‐derived fluids suggests that hydrothermal fluid circulation may help explain deformation episodes at calderas that have not culminated in magmatic eruption.

Hydrothermal fluid flow and deformation in large calderas: Inferences from numerical simulations

Inflation and deflation of large calderas is traditionally interpreted as being induced by volume change of a discrete source embedded in an elastic or viscoelastic half‐space, though it has also been suggested that hydrothermal fluids may play a role. To test the latter hypothesis, we carry out numerical simulations of hydrothermal fluid flow and poroelastic deformation in calderas by coupling two numerical codes: (1) TOUGH2 [Pruess et al., 1999], which simulates flow in porous or fractured media, and (2) BIOT2 [Hsieh, 1996], which simulates fluid flow and deformation in a linearly elastic porous medium. In the simulations, high‐temperature water (350°C) is injected at variable rates into a cylinder (radius 50 km, height 3–5 km). A sensitivity analysis indicates that small differences in the values of permeability and its anisotropy, the depth and rate of hydrothermal injection, and the values of the shear modulus may lead to significant variations in the magnitude, rate, and geometry of ground surface displacement, or uplift. Some of the simulated uplift rates are similar to observed uplift rates in large calderas, suggesting that the injection of aqueous fluids into the shallow crust may explain some of the deformation observed in calderas.

Investigating landslides with space-borne Synthetic Aperture Radar (SAR) interferometry

This paper is addressed to readers without advanced knowledge of remote sensing. It illustrates some current and potential uses of satellite Synthetic Aperture Radar interferometry (InSAR) for landslide assessment. Data acquired by SAR systems can provide 3D terrain models and be used to assist in regional scale investigations, e.g. aimed at evaluation of susceptibility of slopes to failure. Under favourable environmental conditions, the innovative Permanent Scatterers (PS) technique, which overcomes several limitations of conventional SAR differential interferometry (DInSAR) applications in landslide studies, is suitable for monitoring slope deformations with millimetric precision. The PS technique combines the wide-area coverage typical of satellite imagery with the capability of providing displacement data relative to individual image pixels. With the currently available radar satellites, however, only very slow ground surface displacements can be reliably detected and measured. The presented case study of a landslide from the Liechtenstein Alps indicates that the most attractive and reliable contribution provided by this remote sensing technique lies in the possibility of (i.) wide-area qualitative distinction between stable and unstable areas and (ii.) qualitative (relative) hazard zonation of large, slow landslides based on the identification of segments characterised by different movement rates. Since only the radar line of sight projection of the displacements can be detected, a quantitative exploitation of the PS data is possible only where sufficient ground truth is available. In site specific or single landslide investigations the PS data can represent a very useful complementary data source with respect to the information acquired through ground based observations and in situ surveying. However, the difficulties associated with the feasibility assessments of the applicability of SAR data to local scale problems, as well as with the interpretation of PS results, require a close collaboration between landslide experts and specialists in advanced processing of radar satellite data. The interpretation of the exact geotechnical significance of small, radar sensed ground surface deformations is challenging, especially where ground truth is lacking. Although any ground deformation is potentially of interest to an engineering geologist, detection of movements in both vertical and horizontal directions is needed in the case of landslides to evaluate slope failure mechanisms. With their high radar viewing angles, however, the current space-borne systems can detect only a fraction of the horizontal component of movement. It is expected that the upcoming SAR dedicated missions with new sensors and different acquisition geometries, combined with the rapid developments in the field of advanced radar data processing, will allow a full 3D reconstruction of deformation data and help to further reduce the current limitations of the PS and similar DInSAR approaches.

Discussion of “Single Piles in Lateral Spreads: Field Bending Moment Evaluation” by Ricardo Dobry, Tarek Abdoun, Thomas D. O’Rourke, and S.H. Goh

Discussion of “Single Piles in Lateral Spreads: Field Bending Moment Evaluation” by Ricardo Dobry, Tarek Abdoun, Thomas D. O’Rourke, and S.H. Goh

Pullout capacity of spatial anchors

This paper discusses the pullout capacity of spatial anchors in soil under applied vertical force. In field tests, the pullout forces were gradually increased and the ground surface displacements measured in two profiles perpendicular to each other. The laboratory and field tests were performed for several embedment depths and anchor diameter ratios in the same sand and under the same conditions. The anchor pulling was also laboratory‐tested so that the vertical anchor displacements were given and the corresponding force intensity measured. The finite element method was used for the pullout force computation in test cases. The relations between displacements and pullout forces obtained by the laboratory tests, field tests and numerical computations were statistically analysed. Owing to gradual convergence of pullout forces towards the limit value, the exponential function was adopted as an approximation curve. The two obtained constants of the function represent the significant mechanical characteristics. The first is limit pullout force and the second gives the total stiffness of the soil mechanical system.

On estimating the earthquake-induced changes in hydrogeological properties of the Choshuishi Alluvial Fan, Taiwan

The hydrological response of the Choshuishi alluvial fan to the 1999 Chi-Chi earthquake shows that the earthquake did impact the aquifer. The possible earthquake-induced changes in hydrogeological properties were investigated in this study. First, contour maps of the hydrologic anomaly, seismic factors, and vertical ground-surface displacement were compared qualitatively. Bull’s eye patterns were found on the contour maps of hydraulic conductivity, coseismic groundwater-level change and vertical ground-surface displacement but did not occur with other seismic factors. The more permeable zones of the aquifer were found to coincide with the locations of greater vertical ground-surface displacement and coseismic groundwater-level change in the 1999 Chi-Chi earthquake. This indicates that the change of the hydrogeologic properties of Choshuishi alluvial fan due to the 1999 Chi-Chi earthquake may have mainly occurred in the highly permeable zones. Fractal, cross semivariogram and cross correlogram analyses were performed to quantitatively measure the persistency, variability and similarity, respectively, of spatial hydrologic response, seismic factors and hydraulic conductivity. The groundwater-level change, earthquake intensity, and vertical ground-surface displacement were found to show antipersistent tendencies while other factors showed the opposite. Higher correlations were found between hydraulic conductivity and groundwater-level change in aquifers 2–1 and 2–2, and between hydraulic conductivity and vertical ground-surface displacement in aquifer 3. Changes in porosities and hydraulic conductivity were evaluated in the main aquifers of the Choshuishi alluvial fan based on the data of hydrologic anomaly and the vertical ground-surface displacement. While both approaches show that the 1999 Chi-Chi earthquake has impacted the Choshuishi alluvial fan by reducing its porosity and hydraulic conductivity, these changes were not significant relative to natural variation in hydraulic conductivity.

A One-body Model of the 1999 Chi-Chi, Taiwan, Earthquake

An Ms7.6 Chi-Chi earthquake, which ruptured the Chelungpu fault, struck central Taiwan on 20 September 1999 at 17:47 p.m. GMT. Observed data and inversion results show remarkable differences in source proper-ties between the northern and southern segments of the Chelungpu fault. In this study, the Chelungpu fault is divided into two individual segments, and each segment is approximated by a one-body spring-slider model in the presence of friction. Results show that the simple model can interpret the differences in displacements and predominant periods between the two segments. Meanwhile, the ground surface displacement of the fault is capable of reflecting behavior of the thoroughly ruptured area, consisting of numerous different asperities, while the predominant period is able to display the oscillations of a major asperity in the fault plane. However, based on ground surface data the simple model cannot explain the differences in velocities and accelerations between the two segments.

Liquefaction induced lateral spread analysis using the CIP method

Abstract This paper presents a CIP (cubic interpolated pseudoparticle) based numerical method for liquefaction induced lateral spread analysis in the framework of fluid dynamics. Previously, the authors presented similar methods using a SIMPLE-VOF based commercial code but were not able to extend such a code to simulate lateral spread of liquefied ground with an overlaying non-liquefied layer, which is the most common and more practical post liquefaction problem. In this study, the CIP method is used because it is able to treat solid, liquid and gas together, can correctly define the flow behavior at interfaces of multi-fluids and can also be used as a unified scheme for both compressible and incompressible materials. A Bingham model that takes the undrained strength of soils into account is used as the basic constitutive model. A CIP based numerical scheme, which has been successfully used in fluid dynamics problems, is modified by incorporating the Bingham viscosity and an implicit calculation procedure for pressure terms. The Poisson equation is used to compute the pressure over the whole domain. The current numerical method is validated and good agreement is produced in comparison with experimental results. A previously verified similitude for liquefied ground flow is also reproduced by this method. The method is, finally, used to simulate shaking table tests on a liquefied subsoil model with an overlaying non-liquefied layer. From the results of simulations, the numerical method is found to satisfactorily reproduce the time histories of ground surface velocity and displacement and depth distribution of displacement.

Displacement Flexibility Number for Multipropped Retaining Wall Design

The results from 30 nonlinear finite-element analyses of undrained deep excavation in stiff clay are used to support the use of a new displacement flexibility number in multipropped retaining wall design. The analyses address the effects of different initial stress regimes and various values of prop stiffness for the internal supports to the excavation. It is demonstrated that this flexibility number defines support systems that will displace to the same maximum lateral wall deflection and will result in the same profiles of vertical and horizontal ground surface displacement behind the wall. It is concluded that, as it is these movements that must be controlled to limit the damage to adjacent buildings, structures, and services, the new flexibility number gives the engineer more confidence in assessing possible support strategies to a given problem at a given site.

EPOLLS Model for Predicting Average Displacements on Lateral Spreads

A new empirical method, called the EPOLLS model, is developed for predicting ground surface displacements due to liquefaction-induced lateral spreading. Lateral spreading is defined as the mostly horizontal deformation of gently sloping ground as a result of soil liquefaction. In strong earthquakes, lateral spreads often cause considerable damage. The EPOLLS (Empirical Prediction of Liquefaction-Induced Lateral Spreading) model can be used to predict the average horizontal surface displacement that can occur on a potential lateral spread. The model is presented in three parts (designated as the Regional-EPOLLS, Site-EPOLLS, and Geotechnical-EPOLLS components) that allow for progressively better predictions with the addition of more site parameters. The model consists of simple algebraic equations with four to nine parameters that represent the seismic input, site topography, and subsurface conditions. The EPOLLS model was developed from a multiple linear regression analysis of data from 71 lateral spread case studies. The EPOLLS database, regression modeling, quality of the fit, and limitations on the use of the EPOLLS model are discussed.

パステルナーク・モデルによる2次元地表面変位解析

Classical Winkler Model has often been used for static soil-structure interaction analyses. The model has some shortcomings because of the discontinuity of adjacent spring displacements. Pasternak Model, as an improved soil model, was applied for the analysis of an elastic continuum on a rigid base in a plane strain condition. The characteristics of Pasternak Model were evaluated by comparing the displacements of Pasternak Model with those of finite element analysis. A simple explicit formulation of the ground surface displacements was proposed in case of uniformly distributed vertical loads for practical use.

Precision calculation of crustal deformation induced by radial steady laminellar flow of underground water from single well in multilayered structural aquifers

According to both the general formula of ground surface displacement by drainage from a well for radial permeable flow of underground water and the drawdown expressions for the flow in multilayered structural aquifers, we have derived the analytical expressions of surface displacement induced by steady flow withdrawal from a full penetrating well on phreatic water and confined water in multilayered structural aquifers and discussed the numerial integration scheme of these analytical expressions. And by means of Hermite’s quadrature formula with 20 nodes, we have made calculational programs and examples to show that the methods mentioned in this paper are effective. We think that these methods lay a foundation to study quantitatively crustal deformation due to groundwater drainage when we are engaged in high precision dynamic geodetic measurement on the area of steady flow of multilayered aquifers.

Uncertainties in earthquake magnitudes from surface fault displacement based on finite element modeling

Numerical simulation of normal-fault earthquakes was performed using dynamic finite element analysis, to evaluate uncertainties associated with empirical relationships for estimating earthquake magnitude. Simulations involved inducing slip on a steep normal fault and monitoring ground accelerations and displacements. Magnitudes of simulated earthquakes were then estimated using published empirical relationships and compared with magnitudes inferred from seismic slip energy. The results indicate that earthquake magnitudes inferred from ground-surface displacements, using empirical relationships between surface displacement and earthquake magnitude, are likely to underestimate magnitudes of relatively deep earthquakes. On the other hand, the same empirical relationships are likely to overestimate magnitudes of relatively shallow earthquakes.

兵庫県南部地震による野島断層周辺の地表面変動

The Southern Hyogo Prefectural Earthquake (M=7.2, epicenter: 34.36°N, 135.0.2°E, depth =17.9km) occurred on January 17, 1995, and made the Nojima fault appear on the ground surface on Awaji Island. The Geographical Survey Institute surveyed the fault just after the earthquake on January 18-22 and February 13-17. The fault was a right-lateral fault with a reverse component. The maximum displacement measured on the ground along the fault was 1.7 meters in the horizontal (rightlateral) and 1.3 meters in the vertical direction in Nojima-Hirabayashi. The detailed measurement of ground surface displacement around a fault has been only available in terms of the relative movement of one side of the fault against the other. The authors succeeded in measuring three dimensional displacement of 880 points around the Nojima fault by employing air photos taken before and after the earthquake. The characteristics of the displacement are: l) on the west side of the fault, the maximum horizontal displacement is 3.4 meters and directed toward the east, but the amount and direction of the vertical displacement change in a complex manner; 2) the horizontal displacement on the east side is more than one meter and directed toward the south or southeast. In addition, large displacement can be found even in areas more than one kilometer from the fault line. The displacement patterns around the fault are considered consistent with those caused by eastwest compression with eastward movement on the west side and southward movement on the east side: The ground survey identified the southern end of the fault at Toshima. However, the horizontal displacement around the fault can be considered to show invisible continuation of the fault down to Tonouchi

震害状況から推定される断層変位運動

The damage caused by the Southern Hyogo-ken Earthquake 1995 suffered not only wooden houses but also reinforced concrete structures such as highway, railway and modern building, that is unbelievable in the case of heavy vibration in large earthquake occuring at a far away position. The observation results of the damage make the auther to conclude the reason as follows. At least, from Kobe city to Nishinomiya city, there exsited an unfounded active fault in subsurface ground and it moved with right lateral sense. The fault activity caused the ground surface displacement. Namely, the earthquake damage is estimated to be owing to mono-lateral ground displacement with rebound motion due to the fault displacement.

The Effects of Seismic Pore Water Pressure on Ground Surface Motion

The effects of the gradual buildup of seismic pore water pressures and associated degradation of stiffness on the ground surface motion are examined for shallow saturated liquefiable deposits. Ground surface displacements and accelerations, the Fourier Transforms of the acceleration-time histories and layer gain factors are analyzed. The analysis is based on the ground surface and sub-surface accelerations and pore water pressures recorded at sites that either fully liquefied or built up considerable pore water pressures during strong earthquakes. The analysis shows that: (i) the seismic pore water pressures within the deposit and the ground surface motion are intimately related, (ii) relatively small seismic excess pore water pressures can cause a considerable lengthening of the predominant period of the ground surface motion, and (iii) these phenomena may lead to a limited maximum ground surface acceleration of approximately 0.2g, and a considerable increase of maximum ground displacements.

Three-dimensional modeling of the driving mechanism of the 1976 Tangshan earthquake in China

A three-dimensional creep model of the driving mechanisms of the 1976 Tangshan earthquake is presented, with consideration of both the body force and the regional tectonic force. The upper crust is assumed to be a visco-elastic body, the strain rate for the viscous component is assumed to be linearly proportional to the deviatoric stress, but the lower crust and the upper mantle respond to the stress as a non-linear Newtonian fluid obeying the power law of creep. This model is solved numerically using the principle of additively of elastic, plastic and viscous strains, as well as the initial strain method with a tangential modulus for the upper crust and with a secant modulus for both the lower crust and the upper mantle to represent the energy dissipation of creep flow. Computed results show that: 1) to match roughly the pre-seismic ground surface leveling and the co-seismic dip- and strike-slip, uplifting beneath the doming zone of the crust-mantle boundary plays a more important role than a horizontal driving mechanism; 2) the lower-crust fault first relaxes the strain induced by the uplift, resulting in stress accumulation in the upper crust, expecially at the front of the tip of the Tangshan fault lying above the crustal low-velocity zone; 3) significant co-seismic and post-seismic ground-surface displacements along the newly faulted belt in Tangshan result from a combination of elastic strain rebound and change in gravitational potential.

Dike-induced faulting in rift zones of Iceland and Afar

Geodetic data and field observations demonstrate that the emplacement of dikes in volcanic rift zones frequently generates normal faulting and graben subsidence at the Earth9s surface. Elastic modeling of the vertical ground-surface displacements above dikes and faults indicates that the extent of graben subsidence can be achieved only if fault slip extends virtually to or beyond the dike plane at depth. A mechanical model that includes dikes and frictional faults shows that dike opening tends to compress and lock faults located to either side of the dike. Therefore, slip extending into or beyond the dike cavity must occur either (1) on faults that intersect the dike near its top, above the zone of dike-induced compression, or (2) on faults that slip ahead of the dike as it propagates laterally. Data from Iceland indicate that slip occurred on deep faults that presumably slipped in advance of the laterally propagating dike.