This report documents the development of a probabilistic model to represent the occurrence rate a... more This report documents the development of a probabilistic model to represent the occurrence rate and characteristics of future hurricanes capable of producing significant surge inundation along the Mississippi coast, using available hurricane data and statistical tools that have been developed for the offshore oil industry. The report also documents the generation of a suite of synthetic storms, and associated recurrence rates, which provide an efficient representation of the population of possible future hurricanes and their characteristics, for use as inputs to numerical wind, wave, and surge models. These synthetic storms are generated by means of a JPM-OS (Joint-Probability Method—Optimal Sampling) scheme, which is also described in the report.
Gabriel R. Toro
Carola Di Alessandro
Linda Al Atik
The three operating nuclear plants (Diab... more Gabriel R. Toro
Carola Di Alessandro
Linda Al Atik
The three operating nuclear plants (Diablo Canyon, Palo Verde, and Columbia Generating Station) in the western United States recently performed SSHAC Level 3 seismic hazard studies in response to a Request for Information by the Nuclear Regulatory Commission, following the accident at the Fukushima Dai-ichi nuclear facility. The treatment of zero-distance kappa, referred to as kappa_0 and commonly attributed to material damping and scattering in the shallow crust, was given extensive consideration in these studies. Available ground motion prediction equations (GMPEs) do not typically include kappa_0 as a prediction parameter and are developed for an average kappa_0 of the host region. Kappa scaling is routinely applied to adjust for the differences in average kappa between the GMPEs host regions and the target regions. The impact of kappa scaling on the results of probabilistic seismic hazard analyses is significant for nuclear and other facilities that are sensitive to high frequency ground motions (frequencies greater than about 10 Hz). There are several available approaches for deriving kappa scaling factors to GMPEs, which all require estimating kappa_0 at the target site. It is difficult to constrain the target kappa_0 empirically due to the scarcity of ground-motion data from hard-rock sites in ground-motion databases.
The hazard studies for the three nuclear power plants had different data, faced different challenges in the estimation of kappa_0, and used different methods for the estimation of the effect of kappa_0 on the site-specific ground motions. This presentation summarizes the approaches used for the evaluation of kappa_0 and for their incorporation in the probabilistic seismic hazard analysis. Emphasis is given to the quantification of the kappa_0 uncertainty, and on the evaluation of its impact to the resulting seismic hazard at the different sites.
Multiple alternative explanations of large earthquakes in the eastern US imply uncertainty in the... more Multiple alternative explanations of large earthquakes in the eastern US imply uncertainty in the calculated seismic hazard. A study conducted by the Electric Power Research Institute has developed and applied a technique to quantify and document alternative scientific interpretations of the causes and characteristics of large earthquakes, using a Bayesian framework. The method documents uncertainties in the sources of earthquakes, in the parameters of probability distributions describing their characteristics, and in the ground shaking that will accompany such events. The evaluation of hazard uncertainty follows a logic tree approach, with end branches representing specific alternative interpretations. Efficient computational procedures allow many end branches (~50,000) to be handled easily. An important part of the analysis is the interpretation of uncertainty in hazard represented by the 50,000 end branches; conditional distributions allow us to isolate effects of parameter uncertainties and to identify the major contributors to the total uncertainty in hazard. One of these is the range in expert interpretations of the causes of earthquakes. The method allows explicit documentation of expert uncertainties in scientific interpretations, and communicates the resulting uncertainties in earthquake hazard in a convenient form for a risk-based decision process.
This paper considers the uncertainty in the shear wave velocity (Vs) of soil and rock profiles fo... more This paper considers the uncertainty in the shear wave velocity (Vs) of soil and rock profiles for use in earthquake site response calculations. This uncertainty is an important contributor to uncertainty in site response, which in turn is an important contributor to uncertainty in earthquake ground motions and in seismic hazard. The paper begins with a discussion of the different types of uncertainty and how they are characterized in probabilistic seismic hazard analysis, and how this differentiation is particularly ambiguous in the case of soil properties. This is followed by a description of the probabilistic models of Vs that are most commonly used in engineering practice, for both generic and site-specific applications. In site-specific applications, the uncertainty in Vs (which is measured by the logarithmic standard deviation or by the coefficient of variation of Vs) is lower than in generic applications, but other elements of the profile model are also different. Next, the p...
The probabilistic analysis of seismogenic faulting follows the same general methodology as the mo... more The probabilistic analysis of seismogenic faulting follows the same general methodology as the more common seismic hazard analysis for shaking. The main difference lies in the influence model that describes the effects at the site, given the occurrence of an earthquake of a certain magnitude at a certain location. This paper describes the various elements of a probabilistic model of seismogenic faulting and its effect on the waste canisters, the required inputs, and sources of data. The exposition will highlight the similarities between faulting and shaking studies (where a wide body of experience is available), as well as the differences (where the assumptions may have to be refined or new models may have to be developed).
... Variability in Site-Specific Seismic Ground-Motion Design Predictions. by CJ Roblee , M.ASCE ... more ... Variability in Site-Specific Seismic Ground-Motion Design Predictions. by CJ Roblee , M.ASCE , WJ Silva , GR Toro , AMASCE , and N. Abrahamson pp. ... ASCE Subject Headings: Site evaluation. Ground motion. Seismic effects. Earthquakes. Seismic design. Predictions. ...
Ground motions for rock are calculated using the attenuation equations recently developed by EPRI... more Ground motions for rock are calculated using the attenuation equations recently developed by EPRI (EPRI, 1993), as slightly revised and extended by Toro et al. (1997). These attenuation equations are of the form with coefficients given by Table 2 of Toro et al. (1997). In the above equation, M is moment magnitude, R is horizontal distance, e is epistemic uncertainty, and a is aleatory uncertainty. The EPRI attenuation equations include the effect of crustal structure and contain a thorough treatment of epistemic and aleatory uncertainty in source characteristics, path effects, and near-site anelastic attenuation (kappa). Aleatory uncertainty is treated as magnitude- and distance-dependent (see Toro et al., 1997 for details). Epistemic uncertainty (i.e., e in Equation 4-1) is treated as magnitude-dependent and is modeled in the seismic-hazard calculations by using
This report documents the development of a probabilistic model to represent the occurrence rate a... more This report documents the development of a probabilistic model to represent the occurrence rate and characteristics of future hurricanes capable of producing significant surge inundation along the Mississippi coast, using available hurricane data and statistical tools that have been developed for the offshore oil industry. The report also documents the generation of a suite of synthetic storms, and associated recurrence rates, which provide an efficient representation of the population of possible future hurricanes and their characteristics, for use as inputs to numerical wind, wave, and surge models. These synthetic storms are generated by means of a JPM-OS (Joint-Probability Method—Optimal Sampling) scheme, which is also described in the report.
Gabriel R. Toro
Carola Di Alessandro
Linda Al Atik
The three operating nuclear plants (Diab... more Gabriel R. Toro
Carola Di Alessandro
Linda Al Atik
The three operating nuclear plants (Diablo Canyon, Palo Verde, and Columbia Generating Station) in the western United States recently performed SSHAC Level 3 seismic hazard studies in response to a Request for Information by the Nuclear Regulatory Commission, following the accident at the Fukushima Dai-ichi nuclear facility. The treatment of zero-distance kappa, referred to as kappa_0 and commonly attributed to material damping and scattering in the shallow crust, was given extensive consideration in these studies. Available ground motion prediction equations (GMPEs) do not typically include kappa_0 as a prediction parameter and are developed for an average kappa_0 of the host region. Kappa scaling is routinely applied to adjust for the differences in average kappa between the GMPEs host regions and the target regions. The impact of kappa scaling on the results of probabilistic seismic hazard analyses is significant for nuclear and other facilities that are sensitive to high frequency ground motions (frequencies greater than about 10 Hz). There are several available approaches for deriving kappa scaling factors to GMPEs, which all require estimating kappa_0 at the target site. It is difficult to constrain the target kappa_0 empirically due to the scarcity of ground-motion data from hard-rock sites in ground-motion databases.
The hazard studies for the three nuclear power plants had different data, faced different challenges in the estimation of kappa_0, and used different methods for the estimation of the effect of kappa_0 on the site-specific ground motions. This presentation summarizes the approaches used for the evaluation of kappa_0 and for their incorporation in the probabilistic seismic hazard analysis. Emphasis is given to the quantification of the kappa_0 uncertainty, and on the evaluation of its impact to the resulting seismic hazard at the different sites.
Multiple alternative explanations of large earthquakes in the eastern US imply uncertainty in the... more Multiple alternative explanations of large earthquakes in the eastern US imply uncertainty in the calculated seismic hazard. A study conducted by the Electric Power Research Institute has developed and applied a technique to quantify and document alternative scientific interpretations of the causes and characteristics of large earthquakes, using a Bayesian framework. The method documents uncertainties in the sources of earthquakes, in the parameters of probability distributions describing their characteristics, and in the ground shaking that will accompany such events. The evaluation of hazard uncertainty follows a logic tree approach, with end branches representing specific alternative interpretations. Efficient computational procedures allow many end branches (~50,000) to be handled easily. An important part of the analysis is the interpretation of uncertainty in hazard represented by the 50,000 end branches; conditional distributions allow us to isolate effects of parameter uncertainties and to identify the major contributors to the total uncertainty in hazard. One of these is the range in expert interpretations of the causes of earthquakes. The method allows explicit documentation of expert uncertainties in scientific interpretations, and communicates the resulting uncertainties in earthquake hazard in a convenient form for a risk-based decision process.
This paper considers the uncertainty in the shear wave velocity (Vs) of soil and rock profiles fo... more This paper considers the uncertainty in the shear wave velocity (Vs) of soil and rock profiles for use in earthquake site response calculations. This uncertainty is an important contributor to uncertainty in site response, which in turn is an important contributor to uncertainty in earthquake ground motions and in seismic hazard. The paper begins with a discussion of the different types of uncertainty and how they are characterized in probabilistic seismic hazard analysis, and how this differentiation is particularly ambiguous in the case of soil properties. This is followed by a description of the probabilistic models of Vs that are most commonly used in engineering practice, for both generic and site-specific applications. In site-specific applications, the uncertainty in Vs (which is measured by the logarithmic standard deviation or by the coefficient of variation of Vs) is lower than in generic applications, but other elements of the profile model are also different. Next, the p...
The probabilistic analysis of seismogenic faulting follows the same general methodology as the mo... more The probabilistic analysis of seismogenic faulting follows the same general methodology as the more common seismic hazard analysis for shaking. The main difference lies in the influence model that describes the effects at the site, given the occurrence of an earthquake of a certain magnitude at a certain location. This paper describes the various elements of a probabilistic model of seismogenic faulting and its effect on the waste canisters, the required inputs, and sources of data. The exposition will highlight the similarities between faulting and shaking studies (where a wide body of experience is available), as well as the differences (where the assumptions may have to be refined or new models may have to be developed).
... Variability in Site-Specific Seismic Ground-Motion Design Predictions. by CJ Roblee , M.ASCE ... more ... Variability in Site-Specific Seismic Ground-Motion Design Predictions. by CJ Roblee , M.ASCE , WJ Silva , GR Toro , AMASCE , and N. Abrahamson pp. ... ASCE Subject Headings: Site evaluation. Ground motion. Seismic effects. Earthquakes. Seismic design. Predictions. ...
Ground motions for rock are calculated using the attenuation equations recently developed by EPRI... more Ground motions for rock are calculated using the attenuation equations recently developed by EPRI (EPRI, 1993), as slightly revised and extended by Toro et al. (1997). These attenuation equations are of the form with coefficients given by Table 2 of Toro et al. (1997). In the above equation, M is moment magnitude, R is horizontal distance, e is epistemic uncertainty, and a is aleatory uncertainty. The EPRI attenuation equations include the effect of crustal structure and contain a thorough treatment of epistemic and aleatory uncertainty in source characteristics, path effects, and near-site anelastic attenuation (kappa). Aleatory uncertainty is treated as magnitude- and distance-dependent (see Toro et al., 1997 for details). Epistemic uncertainty (i.e., e in Equation 4-1) is treated as magnitude-dependent and is modeled in the seismic-hazard calculations by using
Uploads
Papers by Gabriel Toro
Carola Di Alessandro
Linda Al Atik
The three operating nuclear plants (Diablo Canyon, Palo Verde, and Columbia Generating Station) in the western United States recently performed SSHAC Level 3 seismic hazard studies in response to a Request for Information by the Nuclear Regulatory Commission, following the accident at the Fukushima Dai-ichi nuclear facility. The treatment of zero-distance kappa, referred to as kappa_0 and commonly attributed to material damping and scattering in the shallow crust, was given extensive consideration in these studies. Available ground motion prediction equations (GMPEs) do not typically include kappa_0 as a prediction parameter and are developed for an average kappa_0 of the host region. Kappa scaling is routinely applied to adjust for the differences in average kappa between the GMPEs host regions and the target regions. The impact of kappa scaling on the results of probabilistic seismic hazard analyses is significant for nuclear and other facilities that are sensitive to high frequency ground motions (frequencies greater than about 10 Hz). There are several available approaches for deriving kappa scaling factors to GMPEs, which all require estimating kappa_0 at the target site. It is difficult to constrain the target kappa_0 empirically due to the scarcity of ground-motion data from hard-rock sites in ground-motion databases.
The hazard studies for the three nuclear power plants had different data, faced different challenges in the estimation of kappa_0, and used different methods for the estimation of the effect of kappa_0 on the site-specific ground motions. This presentation summarizes the approaches used for the evaluation of kappa_0 and for their incorporation in the probabilistic seismic hazard analysis. Emphasis is given to the quantification of the kappa_0 uncertainty, and on the evaluation of its impact to the resulting seismic hazard at the different sites.
Carola Di Alessandro
Linda Al Atik
The three operating nuclear plants (Diablo Canyon, Palo Verde, and Columbia Generating Station) in the western United States recently performed SSHAC Level 3 seismic hazard studies in response to a Request for Information by the Nuclear Regulatory Commission, following the accident at the Fukushima Dai-ichi nuclear facility. The treatment of zero-distance kappa, referred to as kappa_0 and commonly attributed to material damping and scattering in the shallow crust, was given extensive consideration in these studies. Available ground motion prediction equations (GMPEs) do not typically include kappa_0 as a prediction parameter and are developed for an average kappa_0 of the host region. Kappa scaling is routinely applied to adjust for the differences in average kappa between the GMPEs host regions and the target regions. The impact of kappa scaling on the results of probabilistic seismic hazard analyses is significant for nuclear and other facilities that are sensitive to high frequency ground motions (frequencies greater than about 10 Hz). There are several available approaches for deriving kappa scaling factors to GMPEs, which all require estimating kappa_0 at the target site. It is difficult to constrain the target kappa_0 empirically due to the scarcity of ground-motion data from hard-rock sites in ground-motion databases.
The hazard studies for the three nuclear power plants had different data, faced different challenges in the estimation of kappa_0, and used different methods for the estimation of the effect of kappa_0 on the site-specific ground motions. This presentation summarizes the approaches used for the evaluation of kappa_0 and for their incorporation in the probabilistic seismic hazard analysis. Emphasis is given to the quantification of the kappa_0 uncertainty, and on the evaluation of its impact to the resulting seismic hazard at the different sites.