geologic data in defining the regional rate of earthquake

magnitude earthquakes on the San Andreas fault than on

activity.

the Hayward fault and the relatively greater influence of

magnitude on long-period motions than short-period

(c) Figure E-15 illustrates the generic logic

motions).

tree for seismic source characterization used for the PSHA.

As shown, the study incorporated uncertainty and

(c) Magnitude contributions to the mean

alternative hypotheses and parameter values for

hazard curves are illustrated in Figure E-19. The

segmentation, maximum rupture length (influencing

contributions of higher magnitudes increase both with

maximum earthquake magnitude), maximum magnitude

increasing period of vibration and with increasing return

estimate correlations, recurrence approach (alternatives of

period (RP).

using seismicity data and tectonic convergence rate data for

source zones), recurrence rates and b-values, and

(d) Analyses of two of the components of the

magnitude distribution model for recurrence assessments

seismic hazard model that contribute to the uncertainty in

(characteristic for faults and characteristic and exponential

the hazard curves are illustrated in Figures E-20 and E-21.

for area sources).

From Figure E-20 it can be seen that much of the

uncertainty in the hazard curves is associated with

uncertainties as to the appropriate attenuation relationship.

(2) Ground motion attenuation characterization.

The uncertainty in the hazard associated with different

Three different sets of rock ground motion attenuation

models of earthquake recurrence for the San Andreas fault

relationships for response spectral acceleration at different

(different segmentation models) (Figure E-21) is small,

periods of vibration (5 % damping) as well as for peak

particularly at lower frequencies of exceedance.

acceleration were utilized. Median values for these

relationships (for magnitudes 5, 6, and 7) are illustrated in

(e) Equal hazard response spectra (expressed

Figure E-16 for peak acceleration and spectral acceleration

in the form of tripartite plots) constructed from the mean

at two periods of vibration. Each set of these relationships

hazard results are shown in Figure E-22 for return periods

also has its associated model of uncertainty (dispersion)

varying from 100 to 2000 years.

around the median curves. The dispersion relationships for

the preferred model (designated Caltrans, 1991, in Figure

E-16 are summarized in Table E-1. (The attenuation

model designated Caltrans, 1991, is the relationship of

(1) Seismic Source Characterization. The site

Sadigh et al., 1993). Note that this model predicts

location is shown in Figure E-23 and is in southern Illinois

increasing dispersion for decreasing magnitude and

on the Ohio River. The dominant source zone for this site

increasing period of vibration, based on analysis of ground

is the Iapetan Continental Rifts source zone (ICR), which

motion data. The three sets of attenuation relationships

represents an interconnected system of partially developed

comprise three additional branches that are added to the

and failed continental rifts that lie within the mid-continent

logic tree in Figure E-15.

region of the United States and includes the New Madrid

source zone (NSZ) where the large 1811 and 1812

(3) PSHA Results

earthquakes occurred. The extent of ICR is shown by the

heavy line in Figure E-23 along with source zones outside

(a) Typical results of the PSHA are illustrated

ICR and the historical seismicity. Modeling of earthquake

in Figure E-17 in terms of the hazard curves obtained for

recurrence within the dominant ICR can be summarized as

peak acceleration and response spectral acceleration at two

follows:

periods of vibration. The distribution about the mean

hazard curve represents the uncertainty in seismic source

(a) The recurrence rate for large (1811-1812

characterization and ground motion attenuation

type) earthquakes in NSZ is modeled based on paleoseismic

characterization modeled in the logic tree.

evidence. As shown in Figure E-24, the paleoseismic-

determined rate of these earthquakes exceed the rate of

(b) Figure E-18 shows the contributions of

large earthquakes predicted from the historical seismicity.

different seismic sources to the hazard (sources are shown

in Figures E-10 and E-11). As shown, the Hayward fault,

(b) The recurrence rate for smaller

which is closest to the site, dominates the hazard for PGA

earthquakes in ICR is determined by the historical

and spectral values at low periods of vibration, but the San

seismicity. Two basic models are used within a logic tree

Andreas fault contribution increases with increasing

framework for defining subzones for characterizing

vibrational period (reflecting the potential for larger

E-19

Integrated Publishing, Inc. |