No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
SHEAR-WAVE VELOCITY COMPILATION FOR NORTHRIDGE STRONG-MOTION RECORDING SITES
Abstract
Borehole and other geotechnical information collected at the strong-motion recording sites of the Northridge earthquake of January 17,1994 provide an important new basis for the characterization of local site conditions. These geotechnical data, when combined with analysis of strong-motion recordings, provide an empirical basis to evaluate site coefficients used in current versions of US building codes. Shear-wave-velocity estimates to a depth of 30 meters are derived for 176 strong-motion recording sites. The estimates are based on borehole shear-velocity logs, physical property logs, correlations with physical properties and digital geologic maps. Surface-wave velocity measurements and standard penetration data are compiled as additional constraints. These data as compiled from a variety of databases are presented via GIS maps and corresponding tables to facilitate use by other investigators.
... Evaluation of the site effect is very important in the prediction of strong motion based on GMPE. One of the methods usually used in site effect evaluation is based on the correlation between the site effects and Vs30 Borcherdt et al. [1]; Midorikawaet al. [12]. Midorikawa et al. [12] provided the following formulation for determining the acceleration amplification factor ...
... where ARA is the amplification for the bedrock of Vs = 745 m/s and AVs30 is the average shear wave velocity to a depth of 30 m. Borcherdt and Fumal [1] have provided acceleration-dependent amplification factors (AF) estimated through geotechnical and geophysical investigations at a number of strong-motion sites in the USA. He provided amplification factors for short and mid-period ranges for different shear wave velocities with respect to base rock of different shear wave velocities. ...
... So, following Midorikawaet al. [12] (Eq. 5) and Borcherdt [1] (Eq. 6), the amplification factors are calculated with respect to the base rock of Vs 745 m/s and 1378 m/s, respectively, using the shear wave velocities obtained at the 14 boreholes. ...
Abstract
Gandhinagar is the capital of Gujarat state in western India and a centre of intense economic and urban development activity, which requires a constant need for development and expansion. The city, having a total area of 57 km2, is situated on the bank of the river Sabarmati. Geologically, the city is covered by alluvium deposits, and the top layer of 80–100 m is composed of loose sandy and soft clayey soils. Most of the urban activities in the city are located on these subsoil strata. A geotechnical database is prepared at various locations of the city along with the geological and geomorphological details. In the framework of this study, 14 boreholes have been drilled, and detailed mapping of the city is done according to its lithology, soil properties, standard penetration test (N value) and shear wave velocity (Vs). With the use of the average shear wave velocity, the amplification factor and peak ground acceleration (PGA) at surface are estimated and further used for the assessment of seismic hazard. The seismic hazard analysis is done by using published ground motion prediction equations (GMPEs) developed for the Gujarat region. Four published GMPEs are used in a logic tree approach by taking magnitude 6.0 as the scenario earthquake for near-field case. The distance of the source from each borehole is calculated and used in GMPEs. Two scenario earthquakes representing Mw7.7, 2001 Bhuj earthquake and M6.5 from Narmada-Son region, are also considered as far-field earthquakes. From the detailed analysis, soil maps, SPT N value maps, Vs map and PGA map are prepared and placed on Geographic Information system (GIS) platform to interpret the complex data for reliable interpolation. The study shows that the area in the central and south-western parts of the city are more prone to high ground shaking and related damage during a major earthquake in Gujarat. The results can be considered to guide further urban development and possible infrastructural development of the city.
... Mean shear-wave velocity to a depth of 30 m (100 ft) as either measured or estimated for each site by Borcherdt and Fumal (2002) is used to characterize the conditions at each recording site. The measured values have been compiled from extensive borehole logging efforts since the earthquake, as compiled in digital databases by Bardet et al. (1998), Nigbor (1998), Gibbs et al. (1980;1996), Fumal (19811982a;1982b;1984), Boore, (pers. ...
... 1996) and digital geologic maps compiled from Tinsley and Fumal (1985), Yerkes (pers. commun, 1994), and Wills et al. (2001) provided the basis to develop estimates for sites with no borehole measurements (see Borcherdt and Fumal, 2002, for details). ...
... The GIS maps based on various digital geologic maps were compiled at large scales displaying borehole v s to 30 m as well as estimates of v s derived from SPT and surface-wave measurements. These maps together with recent unpublished measurements were used to develop estimates of shear-wave velocity at the strong-motion re- cording sites (Borcherdt and Fumal, 2002). A GIS map il- lustrating some of the various data sets is shown in Figure 1. ...
... Type 3 identifies those estimates derived from averages computed from measurements in other boreholes in materi- als with similar physical properties in southern California Fumal and Tinsley, 1985). Type 4 identifies those estimates derived from average velocity estimates de- rived for the corresponding geologic unit (Borcherdt and Fumal, 2002;Fumal, 1978). Type 1 or 2 estimates are available for nearly all of the sites with base accelerations exceeding 0.3g near or within the projected rupture surface. ...
Implications of theoretical, anelastic, seismic wave-field models for site-response estimates used in building code-provisions and in the design of critical facilities
... The GIS maps based on various digital geologic maps were compiled at large scales displaying borehole v s to 30 m as well as estimates of v s derived from SPT and surface-wave measurements. These maps together with recent unpublished measurements were used to develop estimates of shear-wave velocity at the strong-motion recording sites (Borcherdt and Fumal, 2002). A GIS map illustrating some of the various data sets is shown in Figure 1. ...
... Type 3 identifies those estimates derived from averages computed from measurements in other boreholes in materials with similar physical properties in southern California Fumal and Tinsley, 1985). Type 4 identifies those estimates derived from average velocity estimates derived for the corresponding geologic unit (Borcherdt and Fumal, 2002;Fumal, 1978). Type 1 or 2 estimates are available for nearly all of the sites with base accelerations exceeding 0.3g near or within the projected rupture surface. ...
... The GIS maps based on various digital geologic maps were compiled at large scales displaying borehole v s to 30 m as well as estimates of v s derived from SPT and surface-wave measurements. These maps together with recent unpublished measurements were used to develop estimates of shear-wave velocity at the strong-motion recording sites (Borcherdt and Fumal, 2002). A GIS map illustrating some of the various data sets is shown in Figure 1. ...
... Type 3 identifies those estimates derived from averages computed from measurements in other boreholes in materials with similar physical properties in southern California Fumal and Tinsley, 1985). Type 4 identifies those estimates derived from average velocity estimates derived for the corresponding geologic unit (Borcherdt and Fumal, 2002;Fumal, 1978). Type 1 or 2 estimates are available for nearly all of the sites with base accelerations exceeding 0.3g near or within the projected rupture surface. ...
Site-specific amplification factors, F-a and F-v used in current U.S. building codes decrease with increasing base acceleration level as implied by the Loma Prieta earthquake at 0.1g and extrapolated using numerical models and laboratory results. The Northridge earthquake recordings of 17 January 1994 and subsequent geotechnical data permit empirical estimates of amplification at base acceleration levels up to 0.5g. Distance measures and normalization procedures used to infer amplification ratios from soil-rock pairs in predetermined azimuth-distance bins significantly influence the dependence of amplification estimates on base acceleration. Factors inferred using a hypocentral distance norm do not show a statistically significant dependence on base acceleration. Factors inferred using norms implied by the attenuation functions of Abrahamson and Silva show a statistically significant decrease with increasing base acceleration. The decrease is statistically more significant for stiff clay and sandy soil (site class D) sites than for stiffer sites underlain by gravely soils and soft rock (site class C). The decrease in amplification with increasing base acceleration is more pronounced for the short-period amplification factor, F-a, than for the midperiod factor, F-v.
... The stations in the NGA database contain V S30 (the average shear wave velocity in the upper 30 m) and site type. The V S30 values of Table 2 are extracted from the preferred V S30 column of the NGA database, as based on the U.S. Geological Survey Northridge assignments (Borcherdt and Fumal, 2002) and on the California Geological Survey assignments (Wills and Silva, 1998;Wills et al., 2000). Other stations with unknown site type were assigned NEHRP class D with the average V S30 255 m=sec; this value is a reasonable average for soil sites in the area. ...
The stochastic finite-fault ground-motion modeling technique is modified to simulate the effects of a variable-stress parameter on the fault. The radiated source spectrum of each subsource that comprises the fault plane is multiplied by a correction spectrum that leaves the low-frequency portion of the spectrum intact and multiplies the high-frequency end of the spectrum by a constant proportional to the stress parameter of each subfault raised to the power of 2/3; this scaling behavior follows from the Brune source model. The modification causes the response spectra and time series of simulated traces to be sensitive to the stress parameter distribution on the fault surface. The approach is implemented using an inversion tool that effectively inverts observed response spectral data to derive the stress parameter distribution on the fault surface. It applies the Levenberg-Marquardt nonlinear inversion method to minimize differences of average (log) response spectra ordinates at high frequencies between observations and simulations at all stations. We perform a number of experiments to study the effects of fault-dip angle, iterations per station, initial guess of the stress distribution, and station distribution on the capabilities of the inversion tool. We also evaluate the ability of the inversion tool to resolve the relative stress parameters of multiple asperities. Application of the inversion tool to the data of the M 6 2004 Parkfield earthquake indicates that an asperity with a high stress parameter is located in the southeast end of the fault, at a depth greater than 4 km; another asperity is located in the center of the fault, but with a lower stress parameter. This distribution is in agreement with results by other researchers.
... Second, V S30 values were inferred from site geology for California stations that recorded the Northridge earthquake from the analysis by Borcherdt (2003, personal communication;Borcherdt and Fumal, 2002) and for other California stations from the surface geology assignments by Clahan (2004, 2006) (Table 2). ...
A key component of the NGA research project was the development of a strong-motion database with improved quality and content that could be used for ground-motion research as well as for engineering practice. Development of the NGA database was executed through the Lifelines program of the PEER Center with contributions from several research organizations and many individuals in the engineering and seismological communities. Currently, the data set consists of 3551 publicly available multi-component records from 173 shallow crustal earthquakes, ranging in magnitude from 4.2 to 7.9. Each acceleration time series has been corrected and filtered, and pseudo absolute spectral acceleration at multiple damping levels has been computed for each of the 3 components of the acceleration time series. The lowest limit of usable spectral frequency was determined based on the type of filter and the filter corner frequency. For NGA model development, the two horizontal acceleration components were further rotated to form the orientation-independent measure of horizontal ground motion (GMRotI50). In addition to the ground-motion parameters, a large and comprehensive list of metadata characterizing the recording conditions of each record was also developed. NGA data have been systematically checked and reviewed by experts and NGA developers.
... One especially significant effort in developing the NGA database was the estimation of V S30 at recording station sites based on correlation of measured V S30 values with mapped geology and other characteristics. Studies used in developing these estimates included estimation of V S30 values for 1994 Northridge, California, earthquake recording sites by the USGS (Borcherdt and Fumal 2002;Borcherdt 2003, personal communication) and V S30 values estimated for station sites throughout California by the California Geological Survey (CGS) (Wills and Clahan 2004, personal communication;Wills and Clahan 2006). Similar correlation studies were made to estimate V S30 at other recording sites, including sites that recorded the 1999 Chi-Chi, Taiwan, earthquake and aftershocks. ...
The "Next Generation of Ground-Motion Attenuation Models" (NGA) project is a multidisciplinary research program coordinated by the Lifelines Program of the Pacific Earthquake Engineering Research Center (PEER), in partnership with the U.S. Geological Survey and the Southern California Earthquake Center. The objective of the project is to develop new ground-motion prediction relations through a comprehensive and highly interactive research program. Five sets of ground-motion models were developed by teams working independently but interacting with one another throughout the development process. The development of ground-motion models was supported by other project components, which included (1) developing an updated and expanded PEER database of recorded ground motions, including supporting information on the strong-motion record processing, earthquake sources, travel path, and recording station site conditions; (2) conducting supporting research projects to provide guidance on the selected functional forms of the ground-motion models; and (3) conducting a program of interactions throughout the development process to provide input and reviews from both the scientific research and engineering user communities. An overview of the NGA project components, process, and products is presented in this paper.
... Horizontal-component response spectral data of three California earthquakes (M7.1 1989 Loma Prieta earthquake, M7.2 1992 Landers earthquake , and M6.7 1994 Northridge earthquake) were extracted from the NGA database (see " Section 6 " ). The V S30 values are extracted from the preferred V S30 column of the NGA database , as based on USGS Northridge assignments (Borcherdt and Fumal 2002 ) and CGS assignments (Wills and Silva 1998; Wills et al. 2000). The recorded signals' response spectra are first corrected for soil response to produce the corresponding spectra for V s30 = 620 m/s (NEHRP B/C boundary) site conditions, based on the site factor correction given by Boore and Atkinson (2007). ...
Stochastic finite fault modeling is used to derive the coseismic stress parameter distribution on the fault surface of three
well-recorded California earthquakes: M7.0, 1989, Loma Prieta; M7.3, 1992, Landers; and M6.7, 1994, Northridge. Classical waveform inversion techniques are inherently more powerful than stochastic modeling as a
means of deriving detailed source parameters. However, the application of stochastic methods to the source modeling problem
is useful to: (1) explore and calibrate the limitations and boundaries of stochastic modeling, (2) understand its relationship
to more deterministically based techniques, and (3) provide a view of the source radiation not available from deterministic
modeling. The stress parameter distribution for the M7.0 1989 Loma Prieta earthquake fault shows a concentration of stress in the lower part of the northwest side of the fault
and another concentration in the upper southeast side of the fault, with an average stress parameter of 80bars over the fault
surface. The stress parameter distribution for the M7.3 1992 Landers earthquake fault shows a gradual increase of stress starting from the southeast side of the fault, close
to the hypocenter, towards the center. The maximum stress occurs in the lower central part of the modeled fault surface. The
average stress parameter is 70bars for the Landers earthquake. The stress parameter distribution of the M6.7 1994 Northridge earthquake shows a concentration at the lower southeast end of the fault surface, extending toward the
center of the fault surface and stretching to the northwest end. The average stress parameter is 80bars for Northridge earthquake.
The stress parameter distributions derived in this study by stochastic finite-fault modeling of high-frequency motions show
considerable similarity to many of the slip distributions provided by different research groups for the same earthquakes,
suggesting that the derivation of stress parameter distribution on a fault surface by the method applied in this study is
reliable and closely tied to slip on the fault.
KeywordsStochastic finite-fault technique-Stressparameter distribution-Response spectra inversion
Comparative Analysis of Utilization of Salts for Stabilization
of Black Cotton Soil .
Vs30, defined as the average seismic shear-wave velocity from the surface to a depth of 30 meters, has found wide-spread use as a parameter to characterize site response for simplified earthquake resistant design, as implemented in building codes worldwide. Vs30, as initially introduced by the author in 1992 for the US 1994 NEHRP Building Code provisions, provides unambiguous definitions of site classes and site coefficients for site-dependent response spectra based on correlations derived from extensive borehole logging and comparative ground-motion measurement programs in California. Subsequent use of Vs 30 for development of strong ground motion prediction equations (GMPEs) and measurement of extensive sets of Vs borehole data have confirmed the previous empirical correlations and established correlations of Vs30 with Vsz at other depths. These correlations provide closed form expressions to predict Vs30 at a large number of additional sites and further justify Vs30 as a parameter to characterize site response for simplified building codes, GMPEs, ShakeMap, and seismic hazard mapping.
An empirical model for estimating the horizontal pseudo absolute spectral accelerations (PSA) generated by shallow crustal earthquakes is presented in this paper. The model was selected to be simple and the model parameters were estimated using the recordings gathered as part of the New Generation Attenuation (NGA) project. These parameters are presented for sites with an average shear wave velocity in the upper 30 m, V S30 900 m / s, and for sites with 450 m / s V S30 900 m / s. Site-specific dynamic response calculations are recommended for estimating spectral ordinates for sites with V S30 180 m / s. Parameters for sites with 180 m / s V S30 450 m / s are not included in this paper. The median values of peak horizontal ground acceleration (PGA) and PSA for short periods are on the order of 15% to 20% lower for strike slip events and 30% to 40% lower for reverse events than those calculated using pre-NGA relationships. The differences decrease significantly at longer periods. The minimum values of the standard error terms (for moment magnitude, M 7.5) are about 15% to 30% larger and the maximum values of the standard error terms (for M 5) are about 2% to 12% larger than the pre-NGA values.
Recent borehole‐geotechnical data and strong‐motion measurements constitute a new empirical basis to account for local geological conditions in earthquake‐resistant design and site‐dependent, building‐code provisions. They provide new unambiguous definitions of site classes and rigorous empirical estimates of site‐dependent amplification factors in terms of mean shear‐wave velocity. A simple four‐step methodology for estimating site‐dependent response spectra is specified herein. Alternative techniques and commentary are presented for each step to facilitate application of the methodology for different purposes. Justification for the methodology is provided in terms of definitions for the new site classes and derivations of simple empirical equations for amplification as a function of mean shear‐wave velocity and input ground‐motion level. These new results provide a rigorous framework for improving estimates of site‐dependent response spectra for design, site‐dependent building‐code provisions, ...
Site-response coefficients, Fa and Fv, used in U.S. building code provisions are based on empirical data for motions up to 0.1 g. For larger motions they are based on theoretical and laboratory results. The Northridge earthquake of 17 January 1994 provided a significant new set of empirical data up to 0.5 g. These data together with recent site characterizations based on shear-wave velocity measurements provide empirical estimates of the site coefficients at base accelerations up to 0.5 g for Site Classes C and D. These empirical estimates of Fa and Fnu; as well as their decrease with increasing base acceleration level are consistent at the 95 percent confidence level with those in present building code provisions, with the exception of estimates for Fa at levels of 0.1 and 0.2 g, which are less than the lower confidence bound by amounts up to 13 percent. The site-coefficient estimates are consistent at the 95 percent confidence level with those of several other investigators for base accelerations greater than 0.3 g. These consistencies and present code procedures indicate that changes in the site coefficients are not warranted. Empirical results for base accelerations greater than 0.2 g confirm the need for both a short- and a mid- or long-period site coefficient to characterize site response for purposes of estimating site-specific design spectra.
Site-specific amplification factors, F-a and F-v used in current U.S. building codes decrease with increasing base acceleration level as implied by the Loma Prieta earthquake at 0.1g and extrapolated using numerical models and laboratory results. The Northridge earthquake recordings of 17 January 1994 and subsequent geotechnical data permit empirical estimates of amplification at base acceleration levels up to 0.5g. Distance measures and normalization procedures used to infer amplification ratios from soil-rock pairs in predetermined azimuth-distance bins significantly influence the dependence of amplification estimates on base acceleration. Factors inferred using a hypocentral distance norm do not show a statistically significant dependence on base acceleration. Factors inferred using norms implied by the attenuation functions of Abrahamson and Silva show a statistically significant decrease with increasing base acceleration. The decrease is statistically more significant for stiff clay and sandy soil (site class D) sites than for stiffer sites underlain by gravely soils and soft rock (site class C). The decrease in amplification with increasing base acceleration is more pronounced for the short-period amplification factor, F-a, than for the midperiod factor, F-v.
This paper presents a summary of uncorrected peak ground accelerations recorded during the Northridge, California, earthquake of 17 January 1994 and a preliminary analysis of these data. The presented contours of recorded accelerations agree well with observed patterns of damage. The paper also addresses the issue of how ‘unusual’ and ‘unexpected’ the recorded accelerations are relative to earlier predictions.