1D Ground Response Analysis Research Articles

Site-Specific Study for Guwahati, Assam, NE India based on 1D Ground Response Analysis

Site-Specific Study for Guwahati, Assam, NE India based on 1D Ground Response Analysis

From ambient vibration data analysis to 1D ground-motion prediction of the Mj 5.9 and the Mj 6.5 Kumamoto earthquakes in the Kumamoto alluvial plain, Japan

We present horizontal ground motion predictions at a soft site in the Kumamoto alluvial plain for the Mj 5.9 and Mj 6.5 Kumamoto earthquakes of April 2016, in the framework of an international blind prediction exercise. Such predictions were obtained by leveraging all available information which included: (i) analysis of earthquake ground motions; (ii) processing of ambient vibration data (AMV); and (iii) 1D ground response analysis. Spectral analysis of earthquake ground-motion data were used to obtain empirical estimates of the prediction site amplification function, with evidence of an amplification peak at about 1.2 Hz. Horizontal-to-vertical spectral ratio analysis of AMV confirmed this resonance frequency and pointed out also a low-frequency resonance around 0.3 Hz at the prediction site. AMV were then processed by cross-correlation, modified spatial autocorrelation and high-resolution beamforming methods to retrieve the 1D shear-wave velocity (Vs) structure at the prediction site by joint inversion of surface-wave dispersion and ellipticity curves. The use of low frequency dispersion curve and ellipticity data allowed to retrieve a reference Vs profile down to few thousand meters depth which was then used to perform 1D equivalent-linear simulations of the M 5.9 event, and both equivalent-linear and nonlinear simulations of the M 6.5 event at the target site. Adopting quantitative goodness-of-fit metrics based on time–frequency representation of the signals, we obtained fair-to-good agreement between 1D predictions and observations for the Mj 6.5 earthquake and a poor agreement for the Mj 5.9 earthquake. In terms of acceleration response spectra, while ground-motion overpredictions were obtained for the Mj 5.9 event, simulated ground motions for the Mj 6.5 earthquake severely underestimate the observations, especially those obtained by the nonlinear approach.Graphical

Assessment of the prediction of ground motion parameters in 1D ground response analysis using data from seismic arrays and centrifuge experiments

Peak ground acceleration (PGA), peak ground velocity (PGV), and spectral acceleration are among the most widely used metrics to represent seismic hazard characteristics in practice. Several simplified seismic design procedures to evaluate liquefaction triggering, slope stability, or structural response have also proposed other ground motion parameters (GMPs) to represent a ground motion’s intensity, duration, and frequency content. To account for soil effects, these parameters can be obtained from the results of one-dimensional (1D) ground response analyses. Few studies have systematically evaluated the prediction of these additional parameters from the results of ground response analyses. This study presents an exhaustive review of the accuracy and precision of the prediction of 19 common GMPs from the results of 1D ground response analyses using vertical seismic arrays and centrifuge tests. Total stress nonlinear analyses are conducted using the software DEEPSOIL along with equivalent-linear analysis. A reference dataset composed of 10 sites and 306 ground motion recordings representing varying conditions of seismic excitation is employed. The findings of this study showed that while the models produced a reasonable approximation of spectral accelerations, a general tendency toward the over-prediction of most parameters was revealed. The results identified that the mean period ( Tm) yielded the lowest bias, and other parameters, such as the predominant spectral period ( To), PGA, and PGV, also offered some improvements over the other parameters included in this study.

How well can we predict earthquake site response so far? Machine learning vs physics-based modeling

In site-specific site-response assessments, observation-based site-specific approaches requiring a target–reference recording pair or a regional recording network cannot be implemented at many sites of interest. Thus, various estimation techniques have to be used. How effective are these techniques in predicting site-specific site responses (average over many earthquakes)? To address this question, we conduct a systematic comparison using a large data set which consists of detailed site metadata and Fourier outcrop linear site responses based on observations at 1725 K-NET and KiK-net sites. We first develop machine learning (i.e. random forest ( RF)) amplification models on a training data set (1580 sites). Then we test and compare their predictive powers at 145 independent testing sites with that of the one-dimensional (1D) ground response analysis (GRA). The standard deviation of residuals between observations and predictions, that is, between-site (site-to-site or inter-site) variability, is used as the benchmark. Results show that the machine learning model using a few predictor variables, surface roughness, peak frequency fP, HV, VS30, and depth Z2.5 achieves better performance than the physics-based modeling (GRA) using detailed 1D velocity profiles. This implies that machine learning can be more effective in using existing site information than 1D GRA which is inflicted by a high level of parametric and modeling uncertainties. This finding warrants the further exploration of machine learning in site effect characterization, especially on model transferability across different regions.

Site characterization at Treasure Island and Delaney Park downhole arrays by heterogeneous data assimilation

This article extends a recently proposed heterogeneous data assimilation technique for site characterization to estimate compression and shear wave velocity ( Vp and Vs, respectively) and damping at Treasure Island and Delaney Park downhole arrays. The adopted method is based on the joint inversion of earthquake acceleration time series and experimental surface wave dispersion data and includes physical constraints to improve the inverse problem’s well-posedness. We first use synthetic data at these two sites to refine the proposed approach and then apply the refined algorithm to real datasets available at the Treasure Island and Delaney Park downhole arrays. The joint inversion results show that the estimated Vs and Vp profiles are in very good agreement with measured profiles at these two sites. Our synthetic and real data experiment results suggest that Vp estimation from inversion at downhole arrays can be improved by integrating the water table depth information or the higher modes of the Rayleigh wave dispersion data. In the last part of this article, we compare the performance of the inverted profiles to other methods used to incorporate spatial variability and wave scattering effects in one-dimensional (1D) ground response analysis (GRA). The comparisons show that the joint inversion-based Vs and Vp profiles and damping ratios estimated in this article can better integrate the effects of spatial variability and wave scattering into 1D GRAs, especially at the Delaney Park downhole array, which is classified as a poorly modeled site using traditional 1D GRA.

New Insights into Long-Period (>1 s) Seismic Amplification Effects in Deep Sedimentary Basins: A Case of the Po Plain Basin of Northern Italy

ABSTRACTThis study investigates and quantifies the influence of the shallower deposits (down to few hundreds of meters) of the Po Plain sedimentary basin (northern Italy) on the long-period component (i.e., 1 s<T<3 s) of seismic ground motion, in which amplification effects due to the soft sediments above seismic bedrock were observed.A new seismostratigraphic model of the shallow deposits of the entire basin is provided with an unprecedented detail by taking advantage of recently acquired geophysical data. The seismostratigraphic model is used to simulate the ground motion amplification in the Po Plain by means of extensive 1D ground response analysis. Results are compared with seismic observations available at a number of sites equipped with borehole seismic stations, where earthquakes have been recorded both at the surface and at the seismic bedrock depth. Despite the general agreement with observations concerning the seismic resonance frequencies, our model may fail in capturing the amplitude of the actual seismic amplification of the basin in the long-period range. We observe that 3D basin effects related to surface waves generated at the edge of the basin may play a significant role in those zones where seismic hazard is controlled by distant sources. In these cases, 1D modeling leads to average underestimations of 30%, up to a maximum of 60%. The amplification functions need to be corrected for a basin-effects correction term, which in this case is provided by the ground-motion prediction equation of the study area. The corrected amplification functions agree with the empirical observations, overcoming the uneven distribution of the recording stations in strong-motion datasets.These results should be taken into account in future seismic microzonation studies in the Po Plain area, where the 1D approach is commonly adopted in ground response analyses, and in site-specific seismic hazard assessments aimed at the design of structures that are sensitive to the long-period component of seismic ground motion (e.g., long-span bridges and tall buildings).

Static and dynamic characterization and response analysis of soils from northern India

Experimental investigations on soil samples collected from the sites of Baraut, Noida and Mandi in North India have been performed to study the static and cyclic behavior of soil. Liquefaction potential of the sites near Himalayan range is also carried out. The SPT and downhole seismic tests have been performed for the estimation of static and dynamic properties of soil. Strain-controlled cyclic triaxial tests are also carried out on soil samples collected from the borehole at various depths up to 30 m for the evaluation of dynamic soil properties. The study examined the effects of parameters such as cyclic shear strain, loading frequency and overburden pressure on the cyclic behavior of the soil. One- and two-dimensional dynamic response analysis has been performed on three sites. The initiation of initial liquefaction ranged from 202 to 752 cycles for the Baraut site, 212 to 722 cycles for the Noida site, and 121 to 617 cycles for the Mandi sites. The final design response spectra of the sites are obtained from 1D ground response analysis. The results from the response analysis have been compared with the reported results. The soils from the present study areas are prone to liquefaction at 10 m to 15 m from the ground surface.

Integration of viscoplastic effects in a one-dimensional constitutive model for ground response analysis

During an earthquake, strain-rate effects affect both the stiffness and damping behaviour of soils, yet existing constitutive models for ground response analysis are typically formulated within a rate-independent framework. In this paper, a one-dimensional (1D) viscoplastic stress–strain model is presented to introduce strain rate effects in ground response analysis. Its constitutive equations are based on a model that uses a cubic spline fit of the modulus reduction curve and a coordinate transformation technique to match any input modulus reduction and damping curve. A viscous stress component is added to model the effect of strain rate on the mechanical behaviour of soils using a single input parameter. The model is able to reproduce the linear increase in shear strength with the logarithm of shear strain rate and allows the introduction of viscous effects in 1D ground response analysis with control over damping and modulus reduction behaviour. The model is implemented in a software for ground response analysis and used to predict the results of a centrifuge test modeling 1D wave propagation. The results show that the model accurately predicts the amplification and attenuation of shear waves, in a context where strain rates impact the response of the model significantly.

Estimated Empirical Correlations Between Shear Wave Velocity and SPT-N Value for Indore City Using NLR and ANN

In this paper, an attempt has been made to propose a generalized correlation between shear wave velocity (Vs) and SPT-N values for all soil types for generalized depth using the concepts of regression. Around 40 correlations have been compiled and applied for estimating the shear wave velocity (Vs) correlation for the region and also to address the uncertainty issue. The generated correlation is then used for the computation of Vs for one-dimensional (1D) ground response analysis at two different site locations in Indore region. 1D ground response analysis is performed using DEEPSOILv6.1. Soil profiles at site taken as input for analysis were collected from 245 boreholes at 26 different locations in the Madhya Pradesh region (one of the central states in India), and it can be concluded that Vs–N correlations estimated after reducing the errors applying the multilinear perceptron and generalized regression neural network models using MATLAB coding gave better results over the linear regression analysis applied for the site-specific ground response analysis.

A generalized VS–N correlation using various regression analysis and genetic algorithm

Shear wave velocity (VS) is an important parameter from geotechnical earthquake engineering perspective. In the absence of the direct measurement of VS, VS–N correlation serves an alternative for evaluating the shear wave velocity (VS) from standard penetration resistance value (SPT-N) at a particular region. In the present study, an attempt has been made to propose generalized VS–N correlation from the globally available correlations for all soil and specific soil type using various regression procedures and genetic algorithm (GA). Various regression procedures along with GA have been used to assess the sensitivity of generalized VS–N correlations in connection with regression methodology. 1D ground response analysis has been carried out at two bore-hole locations in Kolkata city using the VS obtained from generated correlations and site specific response spectra have been generated at those two locations. Finally, performance of some typical buildings have been studied by nonlinear static analysis procedure using those site specific response spectra developed for proposed VS–N relationships for all type of soil and specific type of soil. The performance of the considered typical buildings has been represented in terms of base shear and roof displacement. Chi square ( $$ \chi^{2} $$ ) test has also been conducted to check the confidence level of base shear and roof displacement obtained by using various developed VS–N relationships varies with the results obtained using available site specific relationships proposed by Nath (2016), and it reveals that VS–N relationships developed by GA perform better than the VS–N relationships developed by other regression method.

Seismic hazard assessment at micro level in Gandhinagar (the capital of Gujarat, India) considering soil effects

Gandhinagar City (the Capital of Gujarat, India) falls under Zone III on the seismic zoning map of India where an earthquake of magnitude 6 can be expected. It is a well established fact that the site amplification/ shaking and damage is large in soil covered areas. To estimate the effect of soil on ground motion and to estimate the strong ground motion parameters at surface, soil modeling and the ground response analysis have been conducted along uniformly distributed 14 boreholes drilled upto a depth of 50 m. The methodology is divided into three parts (i) Estimation of depth of Engineering Bed layer (EBL) (a layer with a shear wave velocity 400 m/s ≤ Vs ≤ 750 m/s, N value >80 and minimum soil variation below it) through soil modeling, (ii) Estimation of Ground Motion at EBL due to scenario earthquake at nearby active fault and (iii) Estimation of surface strong ground motion using 1D ground response analysis through SHAKE 2000 program. The EBL is found at a depth of 21–33 m (shallower in central part and deeper in northern and southern parts). The Near-Field scenario earthquake (Eq.) of magnitude Mw 6.0 has been considered along East Cambay Fault (normal fault, 60° dip) located at about ~ 20 km east and Far Field scenario Eq. of Mw 7.6 is considered along Kachchh Mainland Fault located ~270 km west. The Peak Ground acceleration (PGA) of 0.172–0.237 g have been estimated at surface due to near field earthquake scenario. The mean spectral acceleration maps for 0.1–0.4 s, 0.4–0.7 s, 0.7–1.0 s and 1.0–1.5 s have also been computed. The mean spectral acceleration for the period of 0.1–0.4 s has been varying from 0.330 g to 0.508 g, for period of 0.4–0.7 Sec, it has been varying from 0.151 g to 0.161 g and for period between 1.0 and 1.5 Sec, it has been found from 0.83 g to 0.09 g. The PGA is found increased by 5–38% in the first subsurface soil layer in Gandhinagar city. The PGA of the order of 0.059–0.072 g and peak Spectral acceleration of the order of 0.187–0.259 g have been computed (with predominant periods of ~0.1 s and 0.31 s) due to Far-Field Eq. scenario and are found less than Indian code. The PGA and Spectral acceleration (Sa) values are found higher than the Indian code in the period range of 0.1–0.4 s (one to four storey buildings) for Near Field Eq. Scenario.

GIS-based multi-criteria earthquake hazards evaluation using analytic hierarchy process for a nuclear power plant site, west Alexandria, Egypt

Nuclear power plants are designed to prevent the hazardous effects of the earthquakes and any external events to keep the safety of the plant. Ninety-one shallow seismic refraction profiles were performed to determine shear wave velocity of the engineering layers at the site of El Dabaa area that is situated to the northern coastline of Egypt for seismic hazard microzonation evaluation according to hazard index values. A microzonation is a procedure of delineating an area into individual zones having different ranks of numerous seismic hazards. This will aid in classifying areas of high seismic risk which is vigorous for industrial design of nuclear structures. The site response analysis requires the characterization of subsurface materials considering local subsurface profiles of the site. Site classification of the area under investigation was undertaken using P- and S-waves and available borehole data. The studied nuclear power plant site has been characterized as per NEHRP site classification using an average velocity of transverse wave (V s 30 ) of depth 30 m which acquired from seismic survey. This site was categorized into two site classes: the major one is “site class B,” and the minor one is “site class A.” The attenuation coefficient, the damping ratio and the liquefaction potential are geotechnical parameters which were derived from P- and S-waves, and have their major effects on the seismic hazard contribution. 1D ground response analysis was carried out in the places of seismic profiles inside the site for estimating the amount of ground quaking using peak ground acceleration (PGA), site amplification, predominant frequency and spectral accelerations on the surface of ground by the DEEPSOIL software package. Seven factors (criteria) deliberated to assess the earthquake hazard index map are: (1) the peak ground acceleration at the bedrock, (2) the amplification of the site, (3) the liquefaction potential, (4) the main frequency of the earthquake signal, (5) the average V s of the first 30 m from the ground surface, (6) the depth to the groundwater and (7) the depth to the bedrock. These features were exemplified in normalized maps after uniting them to 0–1 scores according to some criteria by the minimum and maximum values as linear scaling points. Multi-criteria evaluation is an application of multi-criteria decision analysis theory that used for developing a seismic hazard index map for a nuclear power plant site at El Dabaa area in ArcGIS 10.1 software. Two models of decision making were used in this work for seismic hazard microzonation. The analytic hierarchy process model was applied to conduct the relative weights of the criteria by pairwise comparison using Expert Choice Software. An earthquake hazard index map was combined using Weighted Linear Combination model of the raster weighted overlay tool of ArcGIS 10.1. The results indicated that most of the study site of the nuclear power plant is a region of low to moderate hazard; its values are ranging between 0.2 and 0.4.

Estimation of strong motion parameters in the coastal region of gujarat using geotechnical data

Seven boreholes are drilled in the coastal region of Gujarat; three in Kachchh (at Jangi, Mandvi and Mundra), one in Saurashtra (Jodiya) and three in mainland Gujarat(Dahej, Kamboi and Dholera) to depth of 27–80m for estimating the surface strong ground motion parameters. These parameters are required for seismic resistant design of structures in the areas. The adopted methodology comprises of three parts: (i) soil modeling and estimation of depth to Engineering Bed Layer (EBL) (a prominent subsurface soil layer with shear wave velocity of 450m/s to 760m/s with SPT N value of more than 80) (ii) Estimation of the ground motion at EBL (iii) estimation of ground motion at the surface by 1D ground response analysis using SHAKE program. The soil models are prepared from borehole data and shear wave velocity/ N-values. The ground motion at EBL is estimated using Stochastic Finite Fault source Modeling Technique by considering scenario earthquakes along two major faults of Kachchh (Kachchh Mainland Fault (Mw7.6) and Katrol Hill Fault (Mw7.5)), one of Saurashtra (North Kathiawar Fault (Mw 6.5)) and one of Mainland Gujarat (West Cambay Fault (Mw 6.0)). The surface ground motion is estimated by passing ground motion simulated at EBL through prepared soil models of each site using SHAKE program. The surface peak ground acceleration from 0.076g to 0.396g and peak spectral acceleration from 0.24g to 1.4g are estimated at seven coastal sites. Spectral accelerations are found higher than Bureau of Indian Standard (BIS) suggested values at Jangi (between 0.1 and 0.3s & 0.8–1.4s), Jodiya (0.1–0.3s), Mundra (0.3–0.6s) and Dholera (0.1–0.4s) sites, where the importance factor of 1.5 needs to be considered. The study suggests taking BIS values at Dahej, Kamboi and Mandvi coastal sites.

Implications of surface wave data measurement uncertainty on seismic ground response analysis

The extent of data uncertainty of surface-wave measurements may have significant consequences on 1D seismic response analysis and it may lead to inaccurate estimate of design ground motion. In this paper, an attempt has been made to quantify the data measurement uncertainty from a large repetition of the field data at two test sites and to determine bounds of data uncertainty. In the inversion with neighborhood algorithm, we generated profiles below the misfit value calculated from the uncertainty bound. Equivalent profiles have been selected covering the whole misfit value range and these profiles are subjected to equivalent linear 1D ground response analysis. The results of seismic response analysis are presented in the form of amplification spectra and response spectra which show remarkable variations. Significant variation is observed in peak frequency and peak amplification and it differs from one site to the other based on their measured data uncertainty bound. Peak Ground Acceleration (PGA) and peak spectral acceleration at both the sites reflect very high Coefficient of Variation (COV).

Effect of surface wave inversion non-uniqueness on 1D seismic ground response analysis

Surface wave methods are increasingly being used for geotechnical site characterization. The methodology is based on the dispersive characteristic of Rayleigh waves in vertically heterogeneous medium. Experimental dispersion curve is inverted to obtain one-dimensional shear-wave velocity profile by inverse problem solution. Uncertainty associated with this surface wave inversion has drawn much attention. Inverse problem solution can provide different equivalent shear-wave velocity profiles, which may lead to different seismic site response analysis. In this study, a neighborhood algorithm has been used for inversion of dispersion curve to get a set of equivalent shear-wave velocity profiles. These equivalent velocity profiles are then used for 1D ground response analysis for different input motion record of the same earthquake at different epicentral distances. Results show significant variation in amplification spectrum in terms of maximum amplification as well as peak frequency. The extent of this uncertainty largely depends on the characteristics of the ground motion records at different epicentral distances. A linear variation is observed between mean coefficients of variation of amplification spectrum and epicentral distance of ground motion records. A gradual increase in mean value of peak frequency and peak amplification with the epicentral distance is also observed.