ML20133G407

Final Rept, Seismic Hazard at Songs. W/One Oversize Drawing

ML20133G407
Person / Time
Site:	San Onofre
Issue date:	08/25/1995
From:	RISK ENGINEERING, INC.
To:
Shared Package
ML20133G394	List: ML20133G390 ML20133G407
References
NUDOCS 9701160035
Download: ML20133G407 (340)

Text

..---- -. _ __--- -- -- - - - -..-._ _. - - _ - - _ - - -

1 i

i SEISMIC HAZARD AT i

SAN ONOFRE

, NUCLEAR GENERATING STATION t

Prepared for I

Southern California Edison Co.

2244 Walnut Grove Avenue

, Rosemead, California 91770 1

i l

! by

! Risk Engineering,Inc.

j 4155 Darley Avenue, Suite A l Boulder, Colorado 80303 i

i

FINAL REPORT

!; AUGUST 25,1995 I

i i

REPORTS /SONOS/932544D.RPT 4

9701160035 970114 1 PDR ADOCK 05000361 p PDR J

i

)

TABLE OF CONTENT l

Section 1 INTR ODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 - 1

1.1 REFERENCES

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 -3 Section 2 SEISMIC HAZARD METHODOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1

2.1 INTRODUCTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 2.2 BASIC SEISMIC HAZARD MODEL . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1 2.3 TREATMENT OF UNCERTAINTY . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-6 2.4 SUMM ARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2- 8

2.5 REFERENCES

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2- 8 Section 3 SEIS MIC S OURCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3- 1 Section 4 SEIS MICITY PARAMETERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1

4.1 REFERENCES

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4 Section 5 GROUND MOTION A'ITENUATION FUNCTIONS . . . . . . . . . . . . . . . . . . . . 5-1 Section 6 SEIS MIC HAZARD RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1 Section 7 CONCLUS ION S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7- 1 APPENDIX A: Seismic Source Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . A-1 APPENDIX B: Maximum Magnitude Distributions . . . . . . . . . . . . . . . . . . . . . . . B-1 APPENDIX C: Earthquake Recurrence Relationships . . . . . . . . . . . . . . . . . . . . . . C-1 APPENDIX D: Attenuation Relationships . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D- 1 APPENDIX E: Time Histories for Fragility Analysis . . . . . . . . . . . . . . . . . . . . . . E-1 REPORTS / SONGS /9325-94D.RPT 1 I

a

LIST OF TABLES Table 4-1. Mean values of seismicity parameters . . . . . . . . . . . . . . . . . . . . . . . . . 4-5 Table 4-2. Distribution of values of seismicity parameters . . . . . . . . . . . . . . . . . . 4-6 Table 5-1. Weights of attenuation equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-3 Table 6-1. Horizontal ground motions at various probabilities of exceedence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-6 Table 6-2. Mean vertical ground motions at SSE level probabilities of exceedence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-7 l

4 I

l l

l l

i REPORTS / SONGS /9325-94D.RIrr il l

d i

1 L LIST OF FIGURES i

! Figure 2-1. Seismic hazard computational model . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10

Figure 2-2. Logic tree representation of uncertain parameters . . . . . . . . . . . . . . . . . . 2-11 j Figure 3-1. Active faults in southem California . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2 Figure 3-2. Area sources in southern California included . . . . . . . . . . . . . . . . . . . . . . 3-3 Figure 3-3. Fault corridors used for analysis of historical seismicity . . . . . . . . . . . . . . . 3-4 Figure 4-1. Epicenters of all earthquakes in southern Califomia catalog . . . . . . . . . . . 4-7

+

Figure 4-2. Epicenters of dependent earthquakes in southem Califomia . . . . . . . . . . . 4-8 Figure 4-3. Epicenters of mainshocks in southern California . . . . . . . . . . . . . . . . . . . . 4-9' Figure 4-4. Coronado fault seismicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-10 Figure 4-5. Elsinore fault seismicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4- 11 i 3 Figure 4-6. Newport-Inglewood fault seismicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-12

Figure 4-7. Newport-Inglewood-SCOZD fault seismicity . . . . . . . . . . . . . . . . . . . . . 4-13 j Figure 4-8. Palos Verdes fault seismicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-14

i Figure 4-9. Rose Canyon fault seismicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-15 l Figure 4-10. Rose Canyon-SCOZD fault historical seismicity . . . . . . . . . . . . . . . . . . . 4-16  ;

l Figure 4-11. San Andreas fault seismicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-17

! Figure 4-12. San Diego Fault seismicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-18 l l Figure 4-13. San Jacinto fault seismicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-19

> Figure 4-14. Santa Catalina fault seismicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 20 I Figure 4-15. LA Basin Source A blind thrust seismicity . . . . . . . . . . . . . . . . . . . . . . . 4-21 j Figure 4-16. LA Basin Source B blind thrust seismicity . . . . . . . . . . . . . . . . . . . . . . . . 4-22 '

Figure 4-17. Historical seismicity in Central LA Basin source . . . . . . . . . . . . . . . . . . . 4-23

Figure 4-18. Historical seismicity in Central LA Basin and Peninsular Ranges . . . . . . . 4-24 i Figure 4-19. Historical seismicity in Peninsular Ranges source . . . . . . . . . . . . . . . . . . 4-25 i

Figure 4-20. Historical seismicity in Offshore Basin source . . . . . . . . . . . . . . . . . . . . . 4-26 Figure 5-1. Predicted values of SA (25 Hz) vs. distance . . . . . . . . . . . . . . . . . . . . . . . 5-4 Figure 5-2. Predicted values of SA (10 Hz) vs. distance . . . . . . . . . . . . . . . . . . . . . . . 5-5 I

Figure 5-3. Predicted values of SA (5 Hz) vs. distance . . . . . . . . . . . . . . . . . . . . . . . . 5-6

Figure 5-4. Predicted values of S A (2.5 Hz) vs. distance . . . . . . . . . . . . . . . . . . . . . . . 5-7 Figure 5-5. Predicted values of SA (1Hz) vs. distance . . . . . . . . . . . . . . . . . . . . . . . . . 5-8 Figure 5-6. Predicted values of SA (0.5 Hz) vs. distance . . . . . . . . . . . . . . . . . . . . . . . 5-9 i Figure 6-1. SA (10 Hz) hazard by fault . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8 Figure 6-2. S A (10 Hz) hazard by fault . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-9 Figure 6-3. SA (10 Hz) hazard by fault . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-10 Figure 6-4. SA (10 Hz) hazard by blind thrust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-11 t Figure 6-5. S A (10 Hz) hazard by area source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12 Figure 6-6. S A (1 Hz) hazard by fault . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-13 l Figure 6-7. SA (1 Hz) hazard by fault . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 14

Figure 6-8. S A (1 Hz) hazard by fault . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-15 Figure 6-9. SA (1 Hz) hazard by blind thrust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-16

~

Figure 6-10. SA (I Hz) hazard by area source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-17 1

REPORTS / SONGS /9325 94D.RIT iii r

4 1

1 I

l, Figure 6-11. SA (10 Hz) hazard, sensitivity to attenuation equation . . . . . . . . . . . . . . 6-18 Figure 6-12. SA (I Hz) hazard, sensitivity to attenuation equation . . . . . . . . . . . . . . . 6-19 Figure 6-13. SA (10 Hz) hazard, sensitivity to slip rate . . . . . . . . . . . . . . . . . . . . . . . . 6-20 Figure 6-14. SA (1 Hz) hazard, sensitivity to slip rate . . . . . . . . . . . . . . . . . . . . . . . . . 6-21 Figure 6-15. SA (10 Hz) hazard, sensitivity to b-values . . . . . . . . . . . . . . . . . . . . . . . 6-22 Figure 616. SA (1 Hz) hazard, sensitivity to b-values . . . . . . . . . . . . . . . . . . . . . . . . 6-23 Figure 6-17. SA (10 Hz) hazard, sensitivity to depth . . . . . . . . . . . . . . . . . . . . . . . . . . 6 24 Figure 6-18. SA (I Hz) hazard, sensitivity to depth . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-25 Figure 6-19. SA (10 Hz) hazard, sensitivity to maximum magnitude . . . . . . . . . . . . . . 6-26 Figure 6-20. SA (1 Hz) hazard, sensitivity to maximum magnitude . . . . . . . . . . . . . . . 6-27 Figure 6-21. Fractiles and mean of total SA (25 Hz) hazard . . . . . . . . . . . . . . . . . . . . 6-28 Figure 6-22. Fractiles and mean of total SA (10 Hz) hazard . . . . . . . . . . . . . . . . . . . . 6-29 Figure 6-23. Fractiles and mean of total SA (5 Hz) hazard . . . . . . . . . . . . . . . . . . . . . 6-30 Figure 6-24. Fractiles and mean of total SA (2.5 Hz) hazard . . . . . . . . . . . . . . . . . . . . 6-31 Figure 6-25. Fractiles and mean of total SA (1 Hz) hazard . . . . . . . . . . . . . . . . . . . . . 6-32 Figure 6-26. Fractiles and mean of total SA (0.5 Hz) hazard . . . . . . . . . . . . . . . . . . . . 6-33 Figure 6-27. Fractiles and mean of uniform hazard spectra,1.4x10" APE . . . . . . . . . . 6-34 Figure 6-28. Fractiles and mean of uniform hazard spectra,1.7x104 APE . . . . . . . . . . 6-35 Figure 6-29. Mean spectra for 1.7x104 and 1.4x10d annual probabilities . . . . . . . . . . . 6-36 Figure 6-30. Venical uniform hazard spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-37  ;

Figure 6-31. Comparison of SSE and 2xSSE level vertical spectral shapes . . . . . . . . . 6-38 Figure 6-32. Comparison of vertical standard errors for Campbell and Sadigh . . . . . . . 6-39 l Figure 6-33. Companson of vertical spectral shapes using full and simplified analysis . 6-40 l Figure 6-34. Comparison of vertical spectral shapes at the SSE and 2xSSE levels . . . . 6-41 REPORTS / SONGS /9325-94D.RFT iv Risk Engineering, Inc.

1 e

1 I Section 1 INTRODUCTION I

i 4

nis study investigates the probabilistic hazani of earthquake-induced ground shaking at the San Onofre Nuclear Generating Station (SONGS), California. nese results will be used to guide

{ decisions regarding seismic safety and levels of seismic evaluation at the facility. An express purpose of this study is to follow the methodology developed by several recent studies of seismic j hazard at nuclear facilities in the U.S., so that the same insights reganiing seismic hazards and risk j mitigation can be used at SONGS as have been gained from other studies. These other studies make l explicit representation of the uncertainty in seismic hazard caused by multiple, alternative j hypotheses of the causes and characteristics of earthquakes.

a i

!, De design of this study has been to use one team (Geomatrix Consultants, Inc.) to derive seismic

sources (see Appendices A through C), a separate team (Woodward-Clyde Consultants, Inc.) to i

derive ground motion attenuation equations (see Appendices D and E), and a third team (Risk Engineering, Inc.) to perform historical seismicity analysis, integrate the study, and compute seismic

] hazard (as reported in the main body of this report). In addition, a peer review team consisting of:

Dr. Norman Abrahamson -- consultant, Pmf. Keiiti Aki -- University of Southern California, and Prof. Clarence Allen - Califomia Institute of Technology has been involved from the beginning of the project to review ongoing studies and comment on j preliminary results and reports. This arrangement has provided a strong scientific and engineering j basis for the interpretations made here.

)

i Recent studies of seismic hazard in the central and eastem United States (CEUS) have been j completed by the Electric Power Research Institute, funded by the Seismicity Owners Group

\ (EPRI/SOG) (1), and by the lawrence Livermore National Laboratory (LLNL), funded by the U.S.

! Nuclear Regulatory Commission (2). These studies represent major efforts to characterize the j seismic hazard for nuclear power plants in the CEUS, and use the most recent, up-to-date j understandings of seismicity and ground motion relations to the region. Als.o, the Pacific Gas and i

REPORTS /SONGSS325-94D.wr 11 Risk Engineering, Inc.

4

Electric Co. (3) performed a major study of seismic hazard at the Diablo Canyon Power Plant in California, specifically to incorporate alternative hypotheses on tectonics, seismicity, and ground motion values into the decision process reganiing seismic hazards.

A geneal Acdydon of faults, area sources, and parameters is not available for southern California that expresses quantitative uncertainties in interpretations (although such a description is being developed by the Southern California Earthquake Center). Therefore a description of this type is developed by Geomatrix Consultants, Inc. (see Appendices A through C), treating earthquake occurrences both on faults and in area sources (where specific faults have not been identified).

Following the methodology of the other studies indicated above, multiple interpre-tations are medcad for faults and area sources, in order to characterize uncertainty in the seismic hazard that results from uncertainty in earthquake characteristics.

SONGS is located at latitude 33.369* north and longitude 117.554' west. Stmetures at the site are founded on stiff soils overlying bedrock. Mathematical functions describing earthquake ground motion dependence on magnitude and distance in southern California are derived by Woodward-Clyde Consultants, Inc. (see Appendix D). 'Ihese functions are used here to describe ground motion, its randomness, and its uncertainty. Consistent with other state-of-the-art seismic hazard studies, we use these functions to derive the distribution of seismic hazard for spectral accelerations (SA) at frequencies from 25 Hz to 0.5 Hz. We also show constant hazard spectra to demonstrate typical spectral amplitudes and shapes that are appropriate for earthquake ground motions of interest.

Section 2 of this report summarizes the calculational methodology for seismic hazard analysis used here, which is a standard methodology used in virtually all studies of this type. Section 3 summarizes the seismic sources (including faults) that were examined in this study, and Section 4 documents the analysis of historical seismicity that was conducted to estimate seismicity parameters for these sources. Section 5 summarizes the attenuation equations used to estimate SA for the study. ,

Section 6 reports the results of the study, including the dominant sources of uncertainty in seismic REPORTS / SONGS /9325-94D.RfT 1-2 Risk Engineering, Inc.

l

_I

- . - - __ -_ _ _ _ - . - _ _ - . _ . = _ . - . - . - - _. - . _ _ . . . -.

j hazard. Finally, Section 7 presents conclusions of the study and some important qualifications to j these results.

4

.i

1.1 REFERENCES

l 1. SeisnicHazardMethodologyfor the Central and Eartern United States. Technical Report

NP-4726-A, Electric Power Research Institute, July 1986. Revised,1988. Vol.1 Part 1

Methodology, Vol.1, Part 2

Theory, Vol. 2: EQHAZARD Programmer's Manual, Vol.

3: EQHAZARD User's Manual, Vol. 4: Applications, Vols. 5 through 10: Tectonic Interpretations, Vol.11: Nuclear Regulatory Commission Safety Review.

2. D.L. Bernreuter, J.B. Savy, R.W. Mensing and J.C. Chen. Seismic Hazard Characterization J

of69 Plant Sites East of the Rocky Mountains. Technical Report NUREG/CR5250, UCID- .

21517, U.S. Nuclear Regulatory Commission,1988.

l

3. Pacific Gas and Electric Co. Final Report of the Diablo Canyon Long Term Seismic

Program, Docket Nos. 50-275 and 50-323, July,1988.

4 i

i

?

1 l

l l

i i

4 2

REPORTS /SONGSS325-94D.RFr 1-3 Risk Engineering,Inc.

Section 2 SEISMIC HAZARD METHODOLOGY

2.1 INTRODUCTION

• Ihis Section describes the methodology used to calculate seismic hazard in this study, in a general way. Specific inputs to the methodology are described in subsequent secdons.

State-of-the-art seismic hazard studies c=hd* ground-motion acd=+ probabilities using earth-science hypcd-:e:5 about the causes and characteristics of earthquakes in the region being studied.

Scientific uncertainty about the causes of earthquakes and about the physical characteristics of potentially active tectonic features lead to uncertainties in the inputs to the seismic hazard r=hdadons. 'Ihese uncertainties are quantUied using the tectonic interpretations developed by earth scientists familiar with the region. These experts evaluate the likelihood associated with attemative tectonic features and with alternative characteristics of these potential sources.

These and other uncertainties, for example on the ground motion attenuation equations, are carried through the entire analysis. The result of the analysis is a suite of hazard curves and their associated weights; these curves quantify the seismic hazard at the site and its uncertainty.

2.2 BASIC SEISMIC HAZARD MODEL 2.2.1 Overview The methodology to calculate seismic hazard at a site is well established in the literature O M 4.5).

Calculation of the hazard requires specification of three inputs:

1. Source geometry: the geographic description of the seismic source. A seismic source is a portion of the earth's crust, associated with a fault, with a concentration of historic seismicity, or with a specific tectonic feature (other than a fault), that may be capable of producing earthquakes. Source geometry determines the probability distribution of distance from the earthquake to the site:f,(r).

REPORTS /SONOSS325-94D.RPr 2-1 Risk Engineering,Inc. I l

1 I

)

2. Seismicity: the rate of occurrence v, and magnitude distribution 4g (m) of earthquakes occurring in eacit source I. Magnitude is usually characterized by the moment magnitude scale M in California and the Rocky Mountain region, and by the body-wave magnitude m, in the central and eastern U.S. (CEUS).

3. Attenuation functions: a relationship that allows the estimation of ground motion at the site as a function of earthquake magnitude and *===. incorporating known effects of surficial soils on seismic motions.

These inputs are illustrated in Figure 2-1, parts a through c. Figure 2-la shows the geometry of a seismic source. From the source's geometry, f,/r), can be derived. The density function on magnitudefw/m)is often specified as the doubly truncated exponential distribution for area sources and the chehissc magnitude distribution (6) for faults, as illustrated in Figure 2-lb. Seismicity for a source with the exponential magnitude distribution is completely specified by the minimum magnitude m, and per-ras a and b. F--rai a is a measure of seismic activity, b is a measure of relative frequency of large versus small events, and log [ v,f,/m)] is proportional to a + b m for m, < m s m, For the characteristic magnitude distribution, it is necessary in addition to specify the " characteristic" part of the distribution, i.e. the magnitude range of earsquakes that act in a characteristic way, and the annual rate of occurrence of magnitudes in that range.

De gmund motion is modeled by an attenuation function, as illustrated in Figure 2-Ic. Attenuation functions are usually of the form In[A] =f(M,R) + e, where A is ground-motion amplitude, M is i magnitude, R is distance, and e is a random variable that represents w aner. De attenuation function is used to calculate G,Ja*) = F/A > a*/m,rJ: the probability that the ground-motion amplitude j is larger than a*, for a given M and R. The seismic hazani over all sources is calculated as a summation:

i l

P[A > a n Smetl- v, [ [ P[A > a *l'.n,r] fgn(m) f,(n(r) dm dr (2-1)

REPORTS / SONGS /9325-94D.RFr 2-2 Risk Engineering,Inc.

in which the summation is performed over all seismic sources i and in which the probability is calculated per unit time.

2.2.2 Tectanic and Selemicity Internretariane The specificatiors of potential sources of future earthquakes is the first step in the evaluation of earthquake hazards. Seismic sources indicate zhtm canhquakes may occur; analysis of historical neismicity within those defined sources indicates the probabilities of occurrence and characteristics of future canhquakes (i.e., a magnitude distribution is derived from historical data within the source, once the source is defined).

A seismic source is taken by definition to be a fault or area with a single probability of being active, a single magnitude distribution, and a single distribution for maximum magnitude. Within a seismic source the seismicity is usually taken to be spatially homogenous, i.e., canhquakes ue assumed to be equally likely to occur at any location within the source. Some studies (e.g. the EPRl/SOG study) use spatially-varying seismicity, but this generalization is not adopted here.

In general, seismic sources are defined based on faults, tectonic features, or other evidence 1

(including, in some cases, merely a spatial cluster of historical seismicity). Because of this  !

derivation there is, conceptually, some causal association of canhquakes within a source: they are l

releasing crustal stresses of the same orientation and amplitude, and/or they are caused by slip on faults with the same general depth, orientation, and sense of slip. Because of these similarities the delineation is consistent with the seismic source definition with regard to maximum magnitude and probability of activity.

2.2.3 Seiemicitv Parameters Seismicity parameters for canhquake sources are estimated using the rate of tectonic slip for active faults, and using historical seismicity for area sources. The rate of slip on faults is important because multiple methods of estimation can be applied, including measured offsets of datable horizons, crustal strain measurements or inferences, mechanistic tectonic block models of crustal plates, and palcoseismicity studies. For area sources, canhquake catalogs are analyzed to collect REPORTS / SONGS /9325-94D.RFr 2-3 Risk Engineering, Inc.

__ . ___ _ . _ ~ . _ .

all seismic events that have occurred within each source. For each magnitude level, periods of completeness are picked and the rate of occurrence for that magnitude level is calculated as the number of events divided by the time of complete observation. 'Ihese data are then fit using the maximum-likelihood procedum (2) to obtain estimates of a and b.

When the characteristic magnitude distribution is used, the rate of occurrence of events with the characteristic size must generally be estimated using data other than historical seismicity. This is the case because there are few places in the U.S. where a sufficient number of cycles of seismicity have been observed historically to calculate a rate of characteristic events from observations. For some faults (e.g. the San Andreas), palcoseismic evidence gives some indication of the rate of occurrence of the characteristic canhquakes.

Maximum magnitude distributions are estimated using a combination of techniques (e.g., fl 2).

Among these are fault length magnitude relations, comparison with other regions of similar charm.batics, macidaration of geophysical characteristics that relate to m and consideration of the amount ofinformation known about the region under consideration. Ultimately the choice of m.,,, distribution should be made by analysts familiar with the region.

The choice of minimum magnitude m,is based on the characteristics of small earthquakes (i.e., on how damaging are the ground motions associated with these earthquakes), analysis of structural response for the facilities being studied, and field observations of structural performance during low-intensity ground motions. Convention in current studies is to use a moment magnitude of 5.0 for m,, which is supported by studies of the damageability of ground motions from small magnitude earthquakes (10, .11). These studies were made for generic nuclear power plant structures and equipment, and there is no reason to believe that they would not be applicable to SONGS.

REPORTS / SONGS /9325-94D.RPT 2-4 Risk Engineering, Inc.

2.2.4 Ground Motion Attenuntion Enuntions Equations eeimating seismic ground motion are required for the seismic hazard calculations. These are selected using ground motion studies conducted in the region, available strong motion and Analagicd data, and inferences from characteristics of earthquakes. Equations are selected for all measures of interest for the study, which are spectral accelerations (SA) corresponding to 5%

damping for frequencies in the range of 25 Hz to 0.5 Hz. Ground motion estimates exhibit randomness, and the standard assumption in seismic hazard analyses is to characterize the randomness using a log-normal distribution with a specified standard deviation of In[ ground motion]. Typically the value of owwvaries as a function of structural frequency, and it may also vary with magnitude of the earthquake.

2.2.5 Calculations Equation 2-1 is formulated using the assumption that earthquakes (most particularly, successive earthquakes) are independent in size and location. In all seismic hazard applications, primary interest is focused on computing probabilities for high (rare) ground motions. As a result, the probability of two exWes in time t is negligible. De same argument holds when considering hazard at a site from multiple sources. Hus, the summation on the right side of Equation 2-1 --

~

which is the rate of earthquakes with A > a* -- is a good approximation to the probability of exceeding amplitude a* in time t. His is why Equation 2-1 is an approximation (but an accurate one), not a strict equality.

The calculation of hazard from all sources is performed for multiple values of a* in order to generate the hazard curve, which gives the annual probability of exceedence as a function of a*.

This calculation is performed in the current study for six different measures of ground motion: SA at frequencies (25,10,5,2.5,1, and 0.5 Hz), all at 5% damping.

2.3 TREATMENT OF UNCERTAINTY I State-of-the-art seismic hazard studies distinguish between two types of variability: randomness and l

uncertainty. " Randomness" is probabilistic variability that results from natural physical processes.  !

he size, location and time of the next earthquake on a fault and the details of the ground motion 1

REPORTS / SONGS /9325-94D.RFr 2-5 Risk Engineering,Inc.

i 4

i j are examples of random events. In concept, these elements cannot be predicted even with collection 3

j of additional data, so the randomness component of variability is irreducible. The second category of variability is " uncertainty," which is the statistical or modeling variability that result from lack

,4 of knowledge about the true state of nature. In principle, this variability can be reduced with the collection of additional data.

a

)

t

• Ihese two types of variability are treated differently in advanced seismic hazard studies, as follows.

Integration is carried out over probabilistic variabilities to get a single hazard curve (as indicated by Equation 2-1). Modeling uncertainties are expressed by multiple assumptions, hypotheses, or

parameter values.

t

't i

There are uncertainties associated with each of the three inputs to the seismic-hazard evaluation, as I follows:

i 1

j

• Uncertainty about seismic sources and faults (i.e. which tectonic features in a region are

actually earthquake sources) arises because there are multiple hypotheses about the causes of earthquakes and hecaw there is incomplete knowledge about the physical characteristics l of tectonic features. Uncertainty may also arise about the geometry of a seismic source.

a 1

e Uncertainty in seismicity is generally divided into uncertainty in maximum magnitude and i

uncertainty in seismicity parameters y and b. Uncertainty about m_, the maximum i magnitude that a given source can generate, arises for the same reasons described above. '

, Ferimatas of m are obtained from physical characteristics of the source and from historical 5

seismicity. Uncertainty in seismicity parameters vand b arises from statistical uncertainty l and from uncertainty about the accuracy of various catalogs of historical seismicity available j with which to estimate parameters. For the characteristic magnitude distribution, additional e uncertainties are the magnitude range of the characteristic event and its annual rate of I occurrence.

1 1

l REPORTS /SONGSS325-94D.RPr 2-6 Risk Engineering,Inc. ,

. 1 i ,

1

s a i

J e

Uncertainty in the attenuation functions arises from altemative h3 potheses about the ground i i i motion characteristics associated with canhquakes. This uncenainty often is large, I l panicularly in areas where few direct recordings of strong motion are available.

j These multiple interpretations are used to calculate attemative seismic hazard values according to f

Equation 2-1, resulting in a suite of hazani curves. The weight assigned to each seismic hazard j curve is calculated from the probabilities given to each of the uncenain inputs used to calculate it; l the final weight is calculated as the product of the probabilities of the input variables. From the l 1

suite of hazard curves, each with an associated weight, fractile curves or a mean seismic hazard i curve are derived.

1 i

l In order to organize and display the multiple hypotheses, assumptions, parameter values and their I j possible combinations, a logic tree approach is used in this study. Iagic nees are a convenient means to express alternative interpietations and their probabilities. Each node of the logic tree j represents one source of uncenainty. The branches emanating from one node represent possible l ahemative values of a parameter. The probability astigned to a branch represents the likelihood of l the parameter value associated with that branch, and these parameter values (and probabilities) may l depend on values of the preceding parameters.

l 1

{ 'Ihe logic tree in Figure 2-2 illustrates the treatment of parameter uncenainty. There is one hazard f enrve associated with each terminal node; this hazard curve corresponds to cenain sources being a active, each active source having a certain m.,,, and cenain seismicity parameters, and a cenain i anenuation function being the " correct" attenuation model. The probability associated with that end j branch is the product of the probabilities of all branches traversed to reach that end branch.

i

)<

Imgic trees are a convenient way of organizing the uncertainties incorporated into a seismic hazard 1

i analysis, and of documenting them as well. Examples are contained in the separate repon on seismic sources.

REPORTS /SONGSM25-94D.RPr 2-7 Risk Engineering, Inc.

2.4

SUMMARY

Probabilistic seismic hazard analysis requires as input a delineation of seismic sources, a  !

WW of seismicity characteristics for those sources consisting of magnitude distributions and associated parameters, and a selection of ground motion attenuation equations for the region of l

interest. In concept all possible carthquakes in the region are modeled, as are the associated ground motions. Uncertainties in active faulting, areal sources, characteristics of seismicity, and ground motion are incorporated explicitly as multiple alternative hypotheses. The effects of these uncensinties are represented as uncensinty in the hazard curves, and sensitivity studies show the j influence of each input uncertainty on the resulting calculated hazards. 'Ihus the hazard analysis is an overall methodology that can represent both randomness and uncenainty in canhquake occurrences, characteristics, and ground motions, for the purpose of decision-making regarding seismic risk mitigation.

2.5 REFERENCES

1. C.A. Cornell. " Engineering Seismic Risk Analysis," Bulletin of the SeismologicalSociety ofAmerica,58(5):1583-1606, October 1%8.

2. C.A. Cornell. Dynamic Waves in Civil Engineering, Chapter 27: "Probabilistic Analysis of Damage to Structures Under Seismic Loads," Wiley Interscience,1971.

3. A. Der Kiureghian and A.H.S. Ang. A Line Source ModelforSeismic HazardRisk Analysis, Technical Repon UILU ENG-75-2023, University of Illinois, October 1975, pp.134.

4. R.K. McGuire. FORTRAN Caputer Programfor Seismic Risk Analysis, Open-File Repon 76-67, U.S. Geological Survey,1976, pp.1-90.

5. R.K. McGuire. FRISK: Computer Programfor Seismic Risk Analysis Using Faults as Earthquake Sources, U.S. Geological Survey, Open-File Report 78-1007,70 p.,1978.

6. D. Schwanz and K.J. Coppersmith. " Fault Behavior and Characteristic Eanhquakes:

Examples from the Wasatch and San Andreas Fault Zones," Journal of Geophysical Research,89:5681-5698,1984

7. D.H. Weichen. " Estimation of the Eanhquake Recurrence Parameters for Unequal Observations Periods for Different Magnitudes," Bulletin of the Seismological Society of America,70:1337-1346,1980.

REPORTS / SONGS /9325-94D.RFr 2-8 Risk Engineering, Inc.

8.

K.J. Coppersmith, A.C. Johnston, and W.J. Arabasz. " Assessment of Maximum Earthquake Magnitudes in the Eastern United States," Earthquake Notes,57-1:12,1986.

9. G.A. Bollinger, M.S. sin >l, and M.C. Chapman. " Maximum Magnitude Estimation for an Intraplate Setting -- Example: The Giles County, Virginia, Seismic Zone," Seismological Researchletters,63:2, April. June,1992.

10. M.W. McCann and J.W. Reed. Selection of a Lower Bound Magnitudefor Seismic Hazard Assessment, Technical Report, Electric Power Research Institute, Palo Alto, California, 1988.

11. M.W. McCann and J.W. Reed, editors. Engineering Characterization ofSmall Magnitude Earthquain, Electric Power Research Institute, Palo Alto, California,1988, 12.

R.K. McGuire and W.J. Arabasz. "An Introduction to Probabilistic Seismic Hazard Analysis," in S.H. Ward, ed., Geotechnical and Environmental Geophysics, Society oi Exp.

Geophys., Tulsa,1990.

REPORTS / SONGS /9325 94D.RFT 2-9 Risk Engineering,Inc.

I A. Solemic Source 1 (Earthquake locations in space lead lo a distribution of epicentral ggi, p,ygg l e

distances fg(rlm) fRI'!*}

l i

Distance r

8. Magnitude distribution and rate of occurrence for Source I: IM ('"}

fy (m), Pg mo m as Magnitude m m as 7 ,

C. Ground motion estimation: $'o$d a' -  !

Ogg ,,,(s') (log oc le) r Distance (log scale)

D. Probability analysis:

P[ A > s' in time t)11 = r((0g,,,(a9f(m) g 4 g R fb s.

\

\ \ I P(A > a* In t]/t N  %

(log 89 ale)

\ il around Motion Level a* 1 (log scese)

Figure 2-1. Seismic hazard computational model (modified from.12). j i

REPORTS / SONGS /9325-94D.RPT 2-10 Risk Engineering,Inc.

j

a SflOMICITY NAEAna

'PARAMttgns ' GROUND MAXIMUM NOTION AN ALY SIS CONSIN ATION C As ts:

OF ACTIVE M A GNIT UD E S FUNCTIONS SOURCES C1 st M3,e1

/

o' e's i t* 3 M g4 8I WI et C 1. 3 3, M t,g g C' #

4,

, O .

C COME. Cf f ** _,

C1 st Ms,es e A o,

s .

' e, ...

s d

1 I

Figure 2-2. Logic tree representation of uncenain parameters in the EPRI/SOG methodology.

t 1

i REPORTS / SONGS /9325 ND.RPT 2-11 Risk Engineering, Inc.

i i

f Section 3 SEISMIC SOURCES

, '!his Section summarizes the seismic sources used in this study for calculation of seismic hazard at SONGS. 'Ihe seismic sources were delineated by cwwmtrix Consultants, Inc., and are documented I in detailin Appendix A.

4 Figure 3-1 shows the active faults identified as possible sources of earthquakes in southern California, and Figure 3-2 indicates the area sources in the vicinity of SONGS that were used to l

l represent seismicity that occurs away from known faults. All of these sources were investigated to 1 determine their possible contribution to seismic hazard at SONGS, as described in Section 6.

!s

, To analyze historical seismicity as presented in Section 5, Geomatrix defined corridors around each a

fault, as shown in Figure 3-3. This allowed historical seismicity to be collected around each fault, j for comparison of observed rates of activity to rates predicted from fault slip rate and the

! characteristic carthquake model. These corridors also allowed seismicity to be assigned to the j identified faults, so that the remaining historical seismicity could be modeled using the area sources I

(Figure 3-2). These analyses are presented in Section 4.

i i

4 i

4 i

1 1

EEPORTSISONGSS325-94D.RPT 3-1 Risk Engineering,Inc.

g_m_,my h%.--A.E_ha.4 W4 W.- --L-A- * - *"^- --' -'

l SOUTHERN CALIFORNIA FAULTS z

3 I N 3 36*

b E

$ h

? s+

a 35'  %

s-A y  !? %e """

~ a *%, ~

a 34* gh m-~.

k -

e.  % <' %,N *.

=

• ~ E 33' < '4,

.. ;s s g% \, L

~

j- '

s 32,

,R -122' -121* -120' -119' -118' -117' -116* -I15* -114*  ;

l SOUTHERN CALIFORNIA AREA SOURCES 9

o ,

5 E i t

O y 36'

~

h W

.s>

g 35' H

a w 5-O 8 5

0 34' qb " "*8"

!L 2

9 m 5.' h c;-

[

H. \%, E

{o. 33' 5 g- *tT 5- -

R b "

\

!. g s e e p

32*

" -122'

-121' -120* -119' -118" -117* -116* -115* -114*

i 1

a g, SOUTHERN CALIFORNIA FAULT CORRIDORS C

Y u -

g 36' 5 s it 8

g 35' l

2 -

w H.

3- 34- qV 9, .

5 \^*  %

0-  %

• e \ >,

- l 33-D \h T

E !:

1-g\r ~ ~~~ s '

=

5' EI 32* ,

\S

~

-122* -121* -120* -119* -118* -117* -116* -115* -114' 3- E e v ET

.a

Section 4 SEISMICITY PARAMETERS To derive seismicity p.r. ores for the faults and area sources shown in Figures 3-1 and 3-2, three earthquake catalogs were used. 'Ihe first is the catalog of Ellsworth (USGS), who studied all events with M > 6 and huJued epicentral locations and magnitude estimates. The second cattlog was obtained from the National Oceanic and Atmospheric Administration, and includes earthquakes in southern California prior to 1932. The third catalog was obtained from the California Institute of

Technology; it includes events in southern California in 1932 and later years. These three catalogs

! were combined, with the Ellsworth locations and magnitudes given preference over those from the other two catalogs because he has studied earthquakes with M>6 in California and synthesized l location and magnitude estimates from other sources. A plot of the epicenters in the catalog is shown in Figure 4-1. Within each catalog, instrumental magnitudes and intensities were used to characterize each event. Each earthquake is converted to a consistent magnitude measurement M, or moment magnitude. The Gutenberg and Richter (1956) relationship was used to convert MMI to M: M = 1 + 2/3 MMI. The instrumental magnitudes were converted to M using equations i

derived imm a graph by Boore and Joyner (1982). In total,13,844 events with M23 are present in 3 the catalog.

i l Seismic hazard is calculated for mainshocks only (this is a common element of the EPRI/SOG, LLNL, and PG&E studies, for example). The use of mainshocks only is standard in seismic hazard

analyses and allows these results to be compared on a consistent basis to seismic hazard results i produced for other nuclear plant sites. Except at low ground motion levels, the additional hazard ll from aftershocks is small for structures that are not damaged in the mainshock, but including

.6 Accis as additional mainshocks will significantly overestimate the hazard by up to a factor of 2 (Merz and Cornell,1973). To identify aftershocks and other dependent events, we adopted the algorithm of Reasenberg (1985) and applied it to the southern California catalog. This resulted in 7051 events being identified as aftershocks or other dependent events; these are plotted in Figure 4-2. The remaining 6793 mainshocks are shown in Figure 4-3.

I REPORTS /SONGSM25-94D.RPT 4-1 Risk Engineering, Inc.

l For both the fault corridors and the area sources described in the previous section, an analysis was I conducted to determine rates of activity and b-values for each seismogenic zone. This analysis proceeded with the following steps:  ;

l i

1. For each seismogenic zone, determine canhquakes that fall within the boundaries of that l i

zone.

2. For specific magnitude ranges, adopt the times of complete reponing described by Engdahl I and Rinehan (1), and determine the number of earthquakes observed in that magnitude range over the time of complete reponing. ,

)

3. Use the maximum-likelihood procedure of Weichen (2) to calculate an activity rate and b value for seismicity in the zone. )

For these calculations, preliminary estimates of the upper-bound magnitude were used; this is i sufficient because the calculated activity rates and b values are insensitive to the choice of M. I value.

The calcula*d historical rates of activity were used in two ways. For the faults, seismicity within the fault corridors was compared to the rate of activity predicted using fault slip rate, as determined by Geomatrix (see Appendix A). Figures 4-4 through 4-17 indicate the historical seismicity and maximum likelihood fits, as follows:

Coronado fault Figure 4-4 Elsinore fault Figure 4-5 Newpon-Inglewood fault Figure 4-6 Newpon-Inglewood-SCOZD fault Figure 4-7 Palos Verdes fault Figure 4-8 Rose Canyon fault Figure 4-9 Rose Canyon-SCOZD fault Figure 4-10 San Andreas fault Figure 4-11 San Diego fault Figure 4-12 San Jacinto fault Figure 4-13 REPORTS / SONGS /9325-94D.RPr 4-2 Risk Engineering, Inc.

Santa Catalina fault Figure 4-14 LA Basin Source A Figure 4-15 LA Basin Source B Figure 4-16 It is important to note the methodology used to obtain the seismicity in the blind thrust regions. LA Basin Sources A and B seismicity counts were obtained after the extraction of events by other regions. No detailed study was done to correlated earthquakes to fault or blind thrust.

Also shown on these figures are the predicted seismicity rates using the fault slip rates, as documented in the Appendix A by Geomatrix. This predicted seismicity assumes a characteristic magnitude model. 'Ihe comparison indicates that historical seismicity is generally within the range of the predicted rates of activity derived from the slip rate model; some faults indicate higher rates of activity historically, and some lower.

For area sources, the historical seismicity was used to estimate rates of activity and b-values for seismic hazard calculations. Summaries are shown on Figures 4-18 through 4-22 as follows:

Central Los Angeles Basin source Figure 4-17 Centrallos Angeles Basin and Peninsular Range source Figure 4-18 Peninsular Range source Figure 4-19 Offshore Basin source Figure 4-20 For these area sources the seismicit/ associated with the fault corridors was first removed, so that it would not be double counted in estimating the rates of activity.

Table 4-1 summarizes the mean seismicity parameters for faults and area sources, as derived from slip rate for the faults and from historical seismicity for the area sources. Uncertainty in rates, b-values, and % were incorporated into the hazard analysis; details of the uncertainty distributions for faults are documented in the Appendix A, and for the area sources are documented in Table 4-2.

REPORTS /SONGSM25-94D.RFr 4-3 Risk Engineering,Inc.

4.1 REFERENCES

1. E.R. Engdahl and W.A. Rinehan. " Seismicity Map of North America Project," in D.B.

Slemmons, E.R. Engdahl, M.D. Zoback and D.D. Blackwell, eds., Neotectonics of North America, Boulder, Colorado, Geological Society of America, Decade Map Volume 1,1991.

2. D.H. Weichert. " Estimation of the Eanhquake Recurrence Parameters for Unequal Observations Periods for Different Magnitudes," Bulletin of the Seismological Society of America,70:1337-1346,1980.

3. D.M. Boore and W.B. Joyncr. "The Empirical Prediction of Ground Motion," Bulletin of j the Seismological Society ofAmerica, 72:545,1982. \

4. B. Gutenberg and C.F. Richter. " Earthquake Magnitude, Intensity, Energy and i Acceleration," Bulletin of the Seismological Society ofAmerica, 32, p.163-191,1956.

5. C.A. Cornell and H. A. Merz. "Aftershocks in Engineering Seismic Risk Analyris,"

Proceedings Fifth World Conference Earthquake Engineering, Rome, Italy.

l l

REPORTS / SONGS /9325-94D.RFr 4-4 Risk Engineering, Inc.

l TABLE 4-1 Mean Values of Seismicity Parameters for Faults and Area Sources SOURCE vu b-value M_

l Coronado fauh , 0.0439 0.80 7.01 Elsinore fault 0.0854 0.80 6.95 Newport-Inglewood fault 0.0065 0.80 6.74 Newport-Inglewood- 0.0236 0.80 6.86 SCOZD fault 1

Palos Verdes fault 0.0385 0.80 6.81 Rose Canyon fault 0.0132 0.80 6.66 i

Rose Canyon-SCOZD fault 0.0207 0.80 6.79 San Andreas fault 0.2463 0.80 7.63 San Diego fault 0.0103 0.80 7.14 San Jacinto fault 0.1615 0.80 7.06 Santa Catalina fault 0.0103 0.80 6.84 LA Basin Source A 0.0090 0.80 6.16 ,

LA Basin Source B 0.0065 0.80 6.60 Central Los Angeles Basin 0.0053 1.02 6.1 source Central Los Angeles Basin 0.0117 1.03 6.0 and Peninsular Range source Peninsular Ranges source 0.0064 1.03 6.0 Offshore Basin source 0.0038 0.83 6.0 Note: v ,is s the annual rate of earthquakes with M a 5.0.

REPORTSISONGS)9325-94D.Rvr 4-5 Risk Engineering, Inc.

t l

TABLE 4-2 Distribution of Values of Seismicity Parameters for Area Sources l

SOURCE v,.. b-value vs , b-value M, weight weight Centrallos Angeles 0.0221 .6935 .0278 5.5 .2 Basin source 0.0171 .6935 .1111 6.0 .4 0.0121 .6935 .0278 6.5 .4 0.0069 1.0229 .1111 0.0053 1.0229 .4444 0.0037 1.0229 .1111 0.0021 1.3523 .0278 e 0.0016 1.3523 .1111 0.0012 1.3523 .0278 CentralLos Angeles 0.0307 .8054 .0278 5.5 .2 Basin and Peninsular 0.0257 .8054 .1111 6.0 .6 Range source 0.0208 .8054 .0278 6.5 .2 0.0139 1.0267 .1111 0.0117 1.0267 .4444 0.0095 1.0267 .1111 0.0063 1.2480 .0278 0.0053 1.2480 .1111 0.0043 1.2480 .0278 Peninsular Ranges 0.0236 .7309 .0278 5.5 .2 source 0.0186 .7309 .1111 6.0 .6 0.0135 .7309 .0278 6.5 .2 0.0081 1.0297 .1111 0.0064 1.0297 .4444 0.0047 1.0297 .1111 0.0028 1.3285 .0278 0.0022 1.3285 .1111 0.0016 1.3285 .0278 Offshore Basin 0.0282 .3553 .0278 5.5 .2 source 0.0182 .3553 .1111 6.0 .6 0.0083 .3553 .0278 6.5 .2 0.0059 .8266 .1111 0.0038 .8266 .4444 0.0017 .8266 .1111 0.0012 1.2979 .0278 0.0008 1.2979 .1111 0.0004 1.2979 .0278 REPORTS / SONGS /9325-94D.RFT 4-6 Risk Engineering,Inc.

Mainshocks and Dependent Earthquakes n

s 8 3e.

s po w .. w g . ; .

s-

! w .. . ,: >r.

y _e g aet, -

e -pr.

s . -

.,g ....g

-;bh. s%Iti'

34 F k.

e. =- -

.y

~. .

w .,p-e ,= 0 v

3 g, de ip e

. .:.e'd b* M

$ O 5.0-6.0 j.

~

o 4o-50 .: " s* ..e I * ** M m!d a_ .

. s.y]-] x,

- . j

- <40 ,-

. 4 l 4s 3 32 _1 0. u, ._== =#. .

1

@. -122* -121* -120' -119' -118* -117' -116* -115* -114' l

E

" Figure 4-1. Epicenters of all earthquakes in southern California catalog. .

Dependent Earthquakes a

8 5 36' c g . --

g .., a # @+ -

.nn -

g. ..

m s. .

e ,3

. ben _ -

$ 34' Q- g M. 8*4

. y 9 IJ . ,' e, 9~-..%.

Msy >= 8.0 pp y . .. .y 33'

@ 7.0-8.0 .S E , */ ,,

@ 6.0-7.0 0 S. , , 9

$ ' 0 5.0-6.0 *

• g

~ o 4.0-5.0

• x k . < 4.0 * .q ,

,% f _, p 3 32

-122* -121* -120' -119' -118* -117* -116* -115' -114' 4.'

E' Figure 4-2. Epicenters of dependent earthquakes in southern California.

.D

1 Mainshocks l 1 ,

5 5

36'

.$.f'

• h

. o. '

.A:: .

k.*k.3-

..~ .;. %.m ~

z y7 1 4

e 34 g -- - - --= -

E

.i g i a

. 0.  ; 3*, , .,. ,. .

h >= 8.0 . o * . t

@ 7.0-8.0 .

h, h k' *

~

c;-

33'

@ 6.0-7.0 .

k 'Sh.

a45

[ 25Iw. IM da ace b ( ,

• O 5.0-6.0 .

,,. y.* yA*.

N O ,*g *-

j g o 4.0-5.0 . . .

. ,g , , m

s. *M %. 3 k . < 4.0 0

• ct*** 0*.

3. 32' , -

j -122* -121* -120' -119' -118* -117* -116* -115* -114' E Figure 4-3. Epicenters of mainshocks in southem Califomia.  !

h i"i w

a 9

o CORONADO FAULT i . .

u i i I

0 -

5 $ _- b = 1.20 -

A B

c  !

c

> 6 -

N ~

. 1 u w -

.' o -

o , 'N. ~ t

% Q -

g w

~

g [

, 'g o

g 4 "s : ', \ .

= \

t:

n .

m b

~

~

= ~ , , ,

J.

\ ,

D. 3 4 5 6 7 8 9

= l 3,

=

Magnitude (M)

?*

a Figure 4-4. Coronado fault historical seismicity along with the .85, median, ,

P and .15 fractiles of the seismicity estimated from slip rate. l

=

h 9 ELSINORE FAULT S

3 $_ . . . . .

u . .

g . _

O -

N 2 $  ; b = 1.16 -

A 3

a o -

> b -

N -

E .

u  % .

N. N L O - %

. - g _

~ n - -

e 6 -

M - -

N .

g

.\.  :

a E - N

< h  : -

!Le, .

\

E. -

1 M

~

. l g - , , , .. ,

p 3 4 5 6 7 8 9 m

3.

a Magnitude (M)

E' Figure 4-5. Elsinore fault historical seismicity along with the .85, median, P and .15 fractiles of the seismicity estimated from slip rate.

i 4

5 8 NEWPORT-INGLEWOOD FAULT a

3 @ . . . . .

u ~

I o -

t 2

S: b = .78 -

A B

c o -

> 6 -

m -

E o --

e w --

~

o __ ,

G . o 3 n a

g s

~

=

o -

N w . N.

m N.s c -.

"6 _ \ _

= \

w .

n -

= b

- . . . '. . i.\. .

$. 3 4 5 6 7 8 9 3.

=

Magnitude (M) 5

= Figure 4-6. Newport-Inglewood fault historical seismicity along with the .85, median, D and .15 Fractiles of the scismicity estimated from slip rate.

i a

en O

NEWPORT-INGLEWOOD-SCOZD FAULT i i a i i

• o .

xE 8 .79  ;

A 5

e o - --

> 6 -

o m -

5

, ~~

O -

w .

. o 3 n -N. o es g 6 -

s .

$ \

c ,

h : '

- i 5.,.

r N -

t ne , l

> h

= - . . .

\

j..

3 4 5 6 7 8 9

3. Magnitude (M) l

= ,

D ,

Figure 4-7. Newport-Inglewood-SCO2D fault historical seismicity along with I a

- the .85, median, and .15 fractiles of the seismicity estimated from slip rate.  ;

a m

S z

PALOS VERDES FAULT 3 e .

e k

o -

2 $ .-

b = .79 -

A B

C o -

w <*  : _

s o __ -,

s '

• ~ '

m 6 - ~.

g - _

, \ -

s c:

\ .

kne c .

\

W  : '. k -

m m 6

=

1 D.

= 3 4 5 6 7 8 9 m

a.

= Magnitude (M) 5 a Figure 4-8. Palos Verdes fault historical seismicity, along with the D .85, median, and .15 fractiles of the seismicity estimated from slip rate.

... ,,. m s

2 E -

= sy-:I y

a og gw b^

53 b

5' -

.- - :g.,8

<C .

@U 14 / 'E .&

....., o Z / 8 'E o .

.' o g .e l ,' -

w E .5 *

, 2 2$

/ ,

& st d 0 g .

a na / ,'

2 8s rn ^ id deS m

O M

"4 5

E 4.

w Q}

l 0 l Z$

m 101 001 1-01 t-0I c-01 WI

}q < S1U3A3 JO 91EM {EDUUV REPORTS / SONGS /9325-94D.Rrrr 4-15 Risk Engineering, Inc.

l

4 5

O M

m ROSE CANYON-SCOZD FAULT 2 -

y S ~

k U

2 $ -

b = .83 -

A  :

D . -

c o -

6'_-

p -

u  % .

O N  : _-

2 n '

as 6 -

N, -

M -

~.

g __

, \

c ~. '

"6 -

, \ _

g - -

c;- -

w

. .\

h 1

,T ,

k' 3 4 5 6 7 8 9 i n

3- Magnitude (M)

!=

E Figure 4-10. Rose Canyon-SCOZD fault historical seismicity along with P the .85, median, and .15 fractiles of the seismicity estimated from slip rate.

, , . . ... . .... ... m s

2 E. .- -

oo

- - E .e.

m u j 7e a ,- ,....... ug g

H *

%y J / '

du b -

& "G.]

/ ,

.! c m /: =.  :

f 'S .f 3-4 -

e o '.g n .g x

/ .

e s .e t

/ ' ,S l

.~k e

5g a.

/ ,

2 2R em 15 m ,

m 29 mo 5

5.$ l e

II vi a9 o

s m

10I 00I i-01 t-01 c-0I WI W < stuGAg JO 9)BM {EDUUV j REPORTS / SONGS /9325-94D.RPT 4-17 Risk Engineering,Inc.

a 5

9 SAN DIEGO FAULT 3 5 M  : -

i  :

~

p _ -

5 5 g-_ b = 1.17 -

A -

13 a

o -

> 6 - -

N

• g o - -

o , o o

$A

~

l N.

s

?

as _

__' 'N. '

6 , \ -

' ~

hu

g. _

., \

pe -

N 1

= - , , , ,.

. \ ,

]-. 3 4 5 6 7 8 9 3.

=

Magnitude (M)

P a Figure 4-12. San Diego fault historical seismicity along with

. the .85, median, and .15 fractiles of the seismicity estimated from slip rate.

o

?E o*3 REoT n;c3 - esu kPry3vg8v,4

%3'*og gc a~

ow y>ocB A 2 1 6- 6n 6- $ _5 3 .

t h

e 8

,5 F mgu i

e r 4 ,

de i

a 4 _

,n 1- _

a3

n. _

d S a

1 5 n S fa r

a c J

5 ,

. A _

ci t

i t n .~ N _

e o M ,-

l J

s f ou a g

a - A _

fl t

hh t

n i ~

C I

ei s t .

s t eo u d

6 ,

~ .

N .-

ir s i c e T .

ma

.~ o .

O _

il c ( -

is yt e si M

)

.. F em s _ \ A U

t i -

c i

mt ay i 7 ,:

. . * . - ' . '\

L t

ea \ T .

dl o . .

fn .

\

r og I . \

b mw s t i = _

l i h .

p 8 ,

8 .

r 4 a

t _

e 9 ,  ; .

IE M

o j SANTA CATALINA FAULT B -

o p . . . . .

2 a

2 $ b = 1.31 -

A g l-c o - -

> 6 -

m -

A w '

O -

Q

?a n 6 --

W -

- s j '

c .

g J$ -

\

g. -

,\

av -

& L '\

g - , , , . ,

= 3 4 5 6 7 8 9 m

3- Magnitude (M) .

to E

Figure 4-14. Santa Catalina fault historical seismicity along with the .85, median, and .15 fractiles of the seismicity estimated from slip rate.

t o PE o3,{oh

=gg . bu oIu3vio54is ca M%, w P. >ocD A 2 1

b 6 ~E ~6- E E 3 -  :: _~: E - - :r  :

t F

i hg

_ eur

_ 8e

,5 -4 m5 1 L A

_ e 4 ,

_ d iI a

,n A _ _ B aB na

_ a _

A ds i n

S 1

I 5 S fu r

o 5 , .

N a r S

~o

_ c c t e O

_ i l

eA s

b M

a -

U

- ol i g - R f n n o C

t h d i -

et s rh t

u 6 N E eu d \

e i

s s mh i

t i ( .\

A cs \

it y or t

M

)

B e i L

s c t a I N

i l ms a e 7 ,

t s e

i D _

dm .

f r i c

i T _

ot mya b H _

s l

l o

= R _

U p n i

g 8 ,

3 9

r aw S t

ei

. t h

T _

9 - - _  : .

l . ' -:

l l 5

a.

v 8

m LA BASIN SOURCE B BLIND THRUST b 5 g - . . . . .

2 -

O j '

2 $2  ;

b = .76 -

A _

D c --

c - >

> 6 m -  ;

U O -

o , o ca 6 -

M ~

N. ,

g ' - i

[ s.

n --

N c

w k6 ~

\ -

4-pe

\ .

7 L

\, ,

5' 3 4 5 6 7 8 9 i R

E- Magnitude (M) to E

Figure 4-16. LA Basin Source B blind thrust historical seismicity along with

- the .85, median, and .15 fractiles of the seismicity estimated from slip rate.

Dy 5}R5%h Dg"=: whb 0gg3w5O54 4cc5E' Mbo o* W>cc3 A 2

.b.

~6m -b ~6-E 5 3 .

- -  : 5  :: - - ~

.=  :._

i F

g ur 4 ,

e 4-1 C 7

E H N i

s t

o T

r i

c 5 ,

R a

l A

s e M L i

s a m

g n

L i

c i

t i

t A

y u 6 ,

B i

n d C

e e A S

n (

I t

r a

l

)

M N L S A 7 ,

O B

a s U R

i n

s o

u b C E

r =

c _

e 1 .

8 , . .

0 .

2 .

9 ~ ~ _ . . .

, !  ;'!  ;! .il l, ,.

'1 l!.

.i i  ! i+ !

El 8

g CENTRAL LA BASIN AND PENINSULAR RANGES SOURCE 5 -

g $_ i i . . i g -

U 5 h _= b = 1.03 -

A  :

c ~

6 - -

m -

(  %  : o

$g g -

g -

c

, k1 -

c;- .

n 1

~

D . , , ,

5- 3 4 5 6 7 8 9

} Magnitude (M) to y Figure 4-18. Historical seismicity in Central LA Basin and Pt:ninsular Ranges source.

D

I4 5

O g PENINSULAR RANGES SOURCE 5 -

y $_: < . . . .

g _

p . .

5 2 $_- b = 1.03  :

A _

5 a __

b .- -

m __

bE x -

c h  :

a.

g b

' ~

5- 3 4 5 6 7 8 9 R

} Magnitude (M) to y Figure 4-19. Historical seismicity in Peninsular Ranges source.

.D

i u

8 z OFFSHORE BASIN SOURCE a ~

b o y

g .

0 -

2 $ r b = .83 -

A  :

B . -

c -

o -

> 6~^

l N

• p w  : _

O -

M -

3 , o gg as 6 c

=

kh  ::

Q* _

y -

M L -

4 -

3 4 5

6 7

8 5- 9 R

g. Magnitude (M) 5 .

y Figure 4-20. Historical seismicity in Offshore Basin source.

.D

i a

d

. Section 5 1

i, GROUND MOTION ATTENUATION FUNCTIONS i ,

i This Section summarizes the ground motion attenuation equations used in this study to estimate f ground shaking at SONGS as a function of earthquake magnitude and distance. Details of the l functions are given in a separate report prepared by Woodward-Clyde Consultants (Appendix D). l a

l Five horizontal attenuation equations were evaluated based on comparisons of the five equations l

{ with strong motion data. All five equations estimate ground motion at the surface of a stifT soil column, and therefore are appropriate to use directly to estimate seismic hazard for SONGS. The

! equations are identified in Table 5-1.

Weights were assigned to the equations based on how well they fit a strong motion data set representative of southern California. The assigned weights are indicated in Table 5-1, and details of the wup risons are given in Appendix D. Table 5-1 also indicates how the standard error varies (with frequency T and, in some cases, magnitude M). Specific values of the standani error are described in Appendix D.

These equations were used in the seismic hazard analysis as mutually-exclusive alternatives. That is, if one equation applied to a particular fault or area source, it also applied to all others. The weights used in the hazani analysis are those shown in Table 5-1.

Plots of the predicted ground motions for the six frequencies of interest (SA at 25,10,5, 2.5,1, and 0.5 Hz) are shown in Figures 5-1 through 5-7. For these plots a depth of 5 km was assumed for the causative fault, so that the curves could be compared on a reasonable basis. ,

I Appendix D shows residuals (observed minus predicted values) for the five attenuation equations and the data set used for comparison. There is some indication that the median residual tends to be negative (i.e. indicates overpiediction) for spectral response. To the extent that this is the case, the resulting seismic hazard curves will be conservative. ,

REPORTS / SONGS /9325 94D.RPr 5-1 Risk Engineering, Inc.

Peak accelerations fnxn recent earthquakes have tended to exceed the standard empirical attenuation relations for distances greater than 40 km. This trend has suggested that the existing empirical attenuation relations may be biased tow ard under-prediction of ground motion at large distances (>

40 km).

Since the hazard at SONGS is dominated by nearby events (as will be shown in section 6), any potential bias in the attenuation relations will only have a significant effect on the hazard for low ground motion levels (e.g. < 0.lg). The hazard at such low ground motion levels does not affect the risk at SONGS. The anenuation relations were evaluated against empirical data in the distance range of 0-25 km (see section 3 of Appendix D) because sources in this close distance range dominate the hazard. The attenuation relations are shown in Appendix D to be adequate for these short distances.

In addition to the horizontal equations, one venical anenuation equation (Campbell,1990) was used to obtain a spectral shape for vertical ground motions. Details of this function are given in Appendix D, and the calculated spectral shapes are presented in the next section.

REPORTS / SONGS /9325-94D. RET 52 Risk Engineering,Inc.

l TABLE 51 Weights of Attenuation Equations A'ITENUATION RELATIONSHIP WEIGHT (U RANDOM ERR 1 Idriss Stiff Soil Site 0.15 f(M.T)

Abrahamson Soil Site 0.20 f(M,T)

Sadigh Soil Site 0.20 f(M,T)

Boore, Joyner and Fumal Average Site 0.25 f(T)

Class B & C Campbell Soil Site 0.20 f(T) for SA f(M.T) for PG A cu See Appendix D for the derivation of these weights.

REPORTS /SONGSS325 94D.RYT 53 Risk Engineering, Inc.

. _ _ _ _ ._ . ._._ ~_ - --. . . _ _ - - - - - - - . - - - . - _ . _ - _ _

l l

i i

l l

l ATTENUATION COMPARISONS b: '

Idriss  :

. . . .. Abrahamson

. - Sadigh .

- - -Boore, Joyner & Fumal  ;

- - Campbell -

^

O & -

l

)

e ____  :

o

-=-

• u  :-

':,D y , q,5  :

,5 ...... .\-N g -

g- .

u -== -

o -

% g  !

4 5-

'Nhssy 'M = 5.0 N .

gw  :  :

\ .\ s , s.  :

" ,S s, s

',N us N

s N

( -

% \', 's '

m N 1-  :.

y,'.

T ,  !.

N ', :

\']

N Depth = 5 km \-

E- , . . . , , , , .

Horizontal Distance (km) i Figure 5-1. Predicted values of S A (25 Hz) vs distance.

l REPORTS /SONGSS325-94D.RPr 5-4 Risk Engineering,Inc.

-l 5

i 4

I

! ATTENUATION COMPARISONS

a,, . . .. . . , , , , _

S Idriss -

i  : - - - - Abrahamann j . - Sadigh .

- - -Boore, Joyner & Fumal j - - Campbell -

i g

v l

g

.j

!:'-'-..s

-Y M = 7.5  :

, k  ;

s s, h N. -

2 8

-_ ' x. 'D- .

N M = 5.0 N j

< E'"

' s'\\ ,

\ _

- i "\,\ ,

)

$  : -

• g% si -

i =

( -

N ks, ' ,

x t.

DO x

g s, .

5 -

e x \ ,s s

- 1_ . s i

~

\'i a 1

\:

N Depth = 5 km m

6 ' '

100 101 'b2 1

Horizontal Distance (km)

Figure 5-2. Predicted values of S A (10 Hz) vs distance.

REPORTSl50NGS)9325-94D.RYT 55 Risk Engineering, Inc.

a i

i j

ATTENUATION COMPARISONS

.' Idriss  :

- - - - Abrahamaan

. - Sadigh .

- - -Boore, Joyner & Fumal

- - Campbell -

1 --

e 3,

j $ -  : -

% %s\. M = 7.5 i

o , .- .

. g --

7..m-- . '

1 g  : ._ s .%. N .

! * . 's o s . .

u \

? u j < 6

~  : \s'h -

4  : \ .M = 5.0 Ns'\N  !

- Ns \, l w -

Ns3 .

I 3;

m E . s

,s N

) .

. N i N - - N j  % \ ', s l 4

m 6

\

Depth = 5 km -

6 . . ....., . . . . ...,

100 101 102 Horizontal Distance (km)

Figure 5-3. Predicted values of SA (5 Hz) vs distance.

REPORTS / SONGS)9325 94D.RPr 56 Risk Engineering,Inc.

3 1

J h

b l ATTENUATION COMPARISONS 1

1 i

Idriss

- - Abrahamson  :

Sadish .

1

-Boore, Joyner & Fumal

- Campbell -

4 ~

s 3

v &

9 M Wh-b w M = 7.5 -

ce

.e

'hs x.  !

j

~

s3 g .

a

c - _%  % - .

g --

.., d ,

< A j - -

~

6

~ : .NN NN.g'-

i

. M = 5.0

'h

. \Y 8  : 'h s N m

A -

%sNs -

~

s

~

\, ' s s

N 6 D

.  %' i

- Depth = 5 km m

6 Horizontal Distance (km)

Figure 5-4. Predicted values of SA (2.5 Hz) vs distance.

REPORTS / SONGS /9325-94D.RPT 5-7 Risk Engineering,Inc.

I l

l l

1 ATTENUATION COMPARISONS "o,

Idriss -

- - - - Abrahamaan

- - Sadigh  :

- - -Boore, Joyner & Fumal

- - Campbell -

a S -

vno

! - . . --*, o. i s M = 7.5 -

8 -

.g -

o 'D % s g -

sg .

8 N'-

< s "_-= 5 \',:

g -

N N N'y

{ -

'. 's M = 5.0  %^

l

, s -

1

%g '

- s h -

,\ 's s

..N  :

N .

,\ s .

- '(N.' N Depth = 5 km \ D . .  !

100 101 102 Horizontal Distance (km)

Figure 5-5. Predicted values of S A (1 Hz) vs distance.

REPORTstsonas>9325 940.nvr 58 Risk Engineering,Inc.

ATTENUATION COMPARISONS Iddss -

- - - - Abrahamson

. - Sadigh .

- - -Boore, Joyner & Fumal

- - Campbell -

o "8 9  ?

e  :

.3 m = ,

y -- - - - - ~M ; -

. M = 7.5 g --- s 4; s. ,

To . 'N.

'N. -

o '% %

~

6

'\ _

2  : N,\ . ~

- N ,

8  : NN 's : -

2 - . _ _ _

m - -- % s, s

'N s N

m ._.

-.,'s N~ s. s s

s M = 5.0 9  % s 's .

o -  : -

s s -

.N sN s

s N. N .s s

N Depth = 5 km \- y ,

6

- . . . . . . ..N .

,N x Horizontal Distance (km)

Figure 5-6. Predicted values of S A (0.5 Hz) vs distance.

REPORTS / SONGS /9325-94D.RPT 5-9 Risk Engineering, Inc.

1 Section 6  ;

SEISMIC HAZAR.D RESULTS l

Horizontal Ground Motions )

l l

This Section reports the seismic hazard results calculated with the inputs described in Sections 3 through 5. These results were obtained with the computer program FRISK 88M, which incorporates uncertainties in inputs to seismic hazard analyses and produces explicit hazard curves for each combination of uncertain parameters. The calculations are, in all ways, equivalent to the calculations performed under other modern seismic hazard studies (e.g., EPRI/SOG, LLNL, and PG&E),

including the effect of fault mpture length and three-dimensional geometry.

Figures 6-1 through 6-4 illustrate the contribution to SA(10 Hz) hazard by fault. As would be expected, the hypothesis of a nearby active fault (either connected to the Newport-Inglewood-SCOZD or the Rose Canyon-SCOZD faults) dominates the hazard for the larger ground motions (SA>0.15g). At lower ground motions the San Andreas, Elsinore, and San Jacinto faults contribute most to the hazard (Figure 6-1). The hazards from other faults and area sources are plotted on 1

Figures 6-2 through 6-4 to improve the readability of the plots. The area sources (Figure 6-5) do not I contribute to much of the hazard compared to the faults.

1 Figures 6-6 through 6-10 show a similar comparison for SA (1 Hz) hazard. The same faults dominate the hazard, although the San Andreas contributes more at moderate ground motions (spectral accelerations up to 0.lg).

Sensitivity to attenuation equation is shown in Figures 6-11 and 6-12 for 10 Hz SA and 1 Hz SA, respectively, for the Rose Canyon-SCOZD fault. This is a small contributor to the total uncertainty at low ground motions, but is a moderate contributor at the higher accelerations.

Sensitivities to seismicity parameters are illustrated in Figures 6-13 through 6-20 as follows:

Sensitivity to slip rates: Figures 6-13 and 6-14 REPORTS / SONGS /9325-94D.RPT 6-1 Risk Engineering, Inc.

i

(

l, i

i Sensitivity to b-values: . . . . . . . . . . Figures 6-15 and 6-16 4

i Sensitivity to depth: . . . . . . . . . . . . Figures 6-17 and 6-18 l Sensitivity to M.: . . . . . . . . . . . . Figures 6-19 and 6-20 l i l A11 of these sensitivities are shown for the Rose Canyon-SCOZD fault, for both 10 Hz SA and 1 Hz l:

j SA. The sensitivity to slip rate (Figures 6-11 and 6-12) indicates an important contribution of slip l rate uncertainty to the hazard uncertainty. As expected, a change of a factor of 3 in slip rate (total l i range) results in a factor of 3 change in seismic hazard. Changes in b-values do not result in much change in hazard, and this is illustrated in Figures 6-15 and 6-16. He total depth of the seismogenic .

l zone has a moderate influence on seismic hazard (Figures 6-17 and 6-18), with a 15 km depth  ;

resulting in 50% more earthquakes (and 50% more hazard) than a 10 km depth. Finally, sensitivity to M. W*< a strong importance (Figures 6-19 and 6-20). A common seismic hazard result is seen in these plots, which is that a higher value of M. results in lower hazard. The reason is that fault activity is characterized by slip rate. For a fixed value of slip rate, a lower value of M. means that more earthquakes must occur (v must be higher) to cause that slip rate, and this higher value  !

of v causes higher seismic hazard. l l

Figures 6-21 and 6-26 show the total hazard over all faults and area sources, with uncertainties I

c==d by uncertainties in attenuation equations and seismicity parameters. These plots are for SA 1 at 25,10,5,2.5,1, and 0.' 5Hz, respectively. The uncertainty in annual probability of exceedence i for 10 Hz SA (15% to 85% fractile) is a factor or three to four, which is lower than typical uncertainties in the central and eastem U.S. This reflects the greater knowledge about faults and  !

activity in southern California.

He hazattiresults are presented in a different format in Figures 6-27 and 6-28. These are fractiles j 4

of spectra for frequencies of 25 to 0.5 Hz for annual probabilities of 1.4x10" and 1.7x10 . Rese probabilities were chosen because they approximately correspond to the annual pmbabilities of I

exceedence for the SSE spectrum (anchored to PGA = 0.67g) and the 2xSSE spectrum (anchored to 1.34g). Figure 6-29 shows mean spectra for annual probabilities of exceedence of 1.4x10" and I 1.7x104. The numerical values for these spectra are shown in Table 6-1.

REPORTSISONGS)9325-94D.RYT 6-2 Risk Engineering,Inc.

The mean magnitude M and distance R that cause exceedences of ground motions at specified amplitudes was also investigated to gain an understanding of the characteristics of canhquakes that dominate the seismic hazard. These parameters were calculated for both 10 Hz and 1 Hz motions, and for several amplitudes. The following table shows the calculated values and how they vary with structural frequency and amplitude. The choice of amplitudes was made to obtain d

results for hazards at approximately 1.4x10 and 1.7x10' annual probability of exceedence.

As indicated by the table, the mean magnitude M increases for higher levels of shaking, and the mean distance R decreases, for both 10 Hz and 1 Hz.

Level Frea, PSA (e) B R (km)

SSE 10 Hz 1.2 6.7 9.3 SSE 1 Hz 1 7.0 17.0 2xSSE 10 Hz 3 6.9 8.7 2xSSE 1 Hz 2 7.2 20.2 Vertical Ground Motions l

The attenuation of venical component ground motion has not been studied as extensively as the j horizontal component. As a result, there is only one applicable venical attenuation relation for i

spectral values that has been published

Campbell (1990). Unlike the horizontal component, for which 5 different attenuation models were used in the hazard analysis, the venical attenuation is represented only by the Campbell (1990) model.

4 Figure 6-30 shows the uniform hazard spectra at the SSE level acceleration at 5% damping for the venical ground motions. The peak spectral acceleration occurs at 10 Hz. The uncertainties represented by the 85th and 15th fractiles are relatively small because only one attenuation j equation is used to predict the ground motion. The numerical values for the mean are contained l

in Table 6-2.

REPORTS /SONGSS325-94D.RFr 6-3 Risk Engineering, Inc.

The Campbell (1990) model for vertical spectral values has very large standard errors for the high frequency (e.g.10 Hz) response spectra. This large standard error at 10 Hz has an impact on the high frequency spectral shape for the 2xSSE level. The vertical spectral shapes I i

(normalized over 1-10 Hz) from the hazard study are shown in Figure 6-31. There is a  !

significant difference in the spectral shapes between the SSE and 2xSSE levels.

This difference in the spectral shapes is primarily due to the standard errors in the Campbell (1990) model. To demonstrate this, we have conducted a simplified analysis computing the hazard for the SCOZD only. (The SCOZD dominates the high frequency hazard at the SSE and 2xSSE levels at the site.)

An attenuation relation for the venical component for rock sites was developed by Sadigh et al.

(1993). This model presented standard errors that were developed for both soil and rock sites as is commonly done in developing attemation relations. (For example, the Sadigh (1994) model used for the horizontal component uses standard errors that were developed from a combined set of soil and rock data, but with different median attenuation relations for soil and rock sites.)

i Therefore, the vertical component standard enors developed by Sadigh et al (1993) are applicable to soil sites as well as rock sites. Since this model is based on a much larger data set than Campbell (1990), the standard errors should be more accurate (and more stable). These standard '

errors are compared to the Campbell (1990) standard errors in Figure 6-32. Preliminary results of new attenuation relations for the vertical component by Abrahamson and Silva (1995) have found standard errors for the vertical component that are similar to the Sadigh et al values.

To test the sensitivity of the spectral shapes at the SSE and 2xSSE levels to the standard errors, the hazard from the OZD was computed using the Campbell (1990) median attenuation relation vi:th two different standard error models. In the first case, the standard errors published by Campbell (1990) are used; in the second case, the Sadigh et al (1993) standard errors for the vertical component are used.

REPORTS / SONGS /9325-94D.RPT 6-4 Risk Engineering, Inc.

The resulting spectral shapes at the SSE level are shown in Figure 6-32 for the full hazard analysis, and for the two simplified analyses. The simplified analyses give similar spectral shapes as the full hazard analysis (Figure 6-33) indicating that it is reasonable to use the simplified analysis for this sensitivity study.

The comparison of the spectral shapes at the 2xSSE level is shown in Figure 6-34. Using the Sadigh et al standard errors, the spectral shapes for the 2xSSE level is similar to the spectral shape for the SSE level.

Based on this comparison, the vertical spectral shape for the SSE level computed in the hazard analysis is used for both the SSE and the 2xSSE levels.

i REPORTS / SONGS /9325-94D.RPT 6-5 Risk Engineering,Inc.

TABLE 6-1

} Horizontal Ground Motions at Various Probabilities of Exceedence 8

y PROBABILITY SA(25 Hz) SA(10 Hz) SA(5 Hz) SA(2.5 Hz) SA(1 Hz) SA(0.5 Hz) g IE-8 mean 2.712 5.315 6.962 6.816 4.938 2.109

'h median 2.002 3.050 4.132 4.049 3.727 1.913 9

IE-5 mean 1.198 2.157 3.044 2.714 1.550 .858 median 1.072 1.913 2.662 2.491 1.526 .832 2E-5 mean 1.071 1.919 2.696 2.443 1.376 .758 median .973 1.726 2.415 2.266 1.368 .736 IE-4 mean .795 1.402 1.964 1.799 1.007 .543 median .730 1.311 1.804 1.693 1.018 .528 h 2E-4 mean .674 1.195 1.652 1.542 .857 .458 median .630 1.129 1.544 1.467 .866 .446 IE-3 mean .423 .729 1.029 .985 .544 .283 median .400 .702 .976 .952 .548 .278 2E-3 mean .334 .552 .783 .755 .423 .221 h

e median '.320 .531 .750 .733 .431 .217 4

M 1.715x10 mean 1.516 2.813 3.972 3.461 2.060 1.128 a

}. (2xSSE) median 1.293 2.384 3.280 ' 3.034 1.933 1.081 h, 1.386xlO mean .735 1.301 1.810 1.673 .934 .504 h (SSE) median .681 1.227 1.676 1.583 .945 .488 a

Accelerations are in g

I 4

TABLE 6-2 Mean Vertical Ground Motions at SSE Level Probabilit!es of Exceedence

] FREQ SA(af 0.5 Hz .247

}

i 1 Hz .439 i

2.5Hz .783 5 Hz 1.312 10 Hz 1.544 25 Hz .845

• Vertical SA levels were chosen at an annual probability level of 1.386x10d, corresponding to the horizontal SSE.

REPORTS / SONGS /9325-94D.RFr 67 Risk Engineering, Inc.

i i

i i

i FAULT SENSITIVITIES,10 Hz i

~

! San Andreas i

- - - -San Jacinto  :

- New.-Ing, Rose Can. -

! - - -Elsinore -

- -Palos Verdes g  !$! _ . , -Coronado 1 E  ? - [' ,

y .

~

a N

o E

~  :% . ' N> b j '

~ N, s

x -

- \ 'is'(s . s  :

~ s .

$ ' \\

, \ ,

t h '. \} -s 5 -

2" '

ss .

\

\\ \

J 1 .

t \-:

_: i

\

j\s\ \:..

~

s.

.g g

\\ ,\

10 2 10-1 100 10 Hz Spectral Acceleration (g)

Figure 6-1. SA (10 Hz) hazard by fault.

REPORTS / SONGS /9325-94D.RPT 6-8 Risk Engineering, Inc.

FAULT SENSITIVITIES,10 Hz

~ .

E Aguana-Agua Tibia i

- - - -Santa Catalina  : )

- . - San Clemente .

l

- - -Cucamonga -

- -Hollywood-Raymond c s  : -Malibu-Santa Monica-y _

- -Peralta Hills o

}u

~ ~~

~

k  %

's

~

N -

O s s .

":::. . . N  :

3 -5( N ,s

.c

,N s, \ \

-\s-E -

c- \'x .x a b m .

\, .

.\

\g' \ 4 1

\\  :

i .\ \ y

\\

\.

.\

m .

M , , \\, y A '. ....

10 2 104 100 10 Hz Spectral Acceleration (g)

Figure 6-2. SA (10 Hz) hazard by fault.

REPORTS /SONGSS325-94D.RPT 6-9 Risk Engineering, Inc.

FAULT SENSITIVITIES,10 Hz

San Miguel  :

- - - -La Nacion

- San Diego

- - -Sierra Madre -

- - -Te.1-:+:4 o M  ?

~

-Vallecitos  :

o

- -Whittier -

~

s g s m g s

\ 1 I'~ N s  :

l x

'. 'N  :  !

\

y

• w N.

\

\ - i

' s\ I

• E N a a ~ \

O ~

ss \

A

. t  !

\ \\\

a g - \ . \ -

N.

- 5

\

\

h i \\  :

5

\\\

\

~

-., \. g\ -

\

gg). .

6 '

\,\ . t ...,

10-2 10-1 100 10 Hz Spectral Acceleration (g)

Figure 6-3. SA (10 Hz) hazard by fault.

1 1

REPORTS / SONGS /9325-94D.R_wr 6-10 Risk Engineering, Inc.

- _ - . . - = - _ . - - _ . _ . -. . - - - - _ _ _ . - . - _ . - _ - .

BLIND THRUST SENSITIVITIES,10 Hz

~

E LA Basin Source A i

- - - -LA Basin Source B :

u C:

o 9 .

>(

N $- _

% i  !

C  :  :

y -

.g -

3 b Q - -

2 -

.- i A -

~

s -

b -

6 10-2 101 100 10 Hz Spectral Acceleration (g)

Figure 6-4. S A (10 Hz) hazard by blind thnist.

REPORTS / SONGS /9325-94D.RFr 6 11 Risk Engineering, Inc.

i

! l l

4 AREA SENSITIVITIES,10 Hz

, e ig

! ~ s i g

! Central LA Basin  !

3  : - - - -Peninsular Ranges  :

) - -

- Offshore Basin -

1 4 l ent

< c 6 o _

. c _

! c _

u -

! x

W 6  : -

i w -

C  :. . .  :

r u

x . -.

l 3

.c

[ N.'g [

.E o g- -

'K A

l .. i 1  : g  :

5a i

g .~

a .

b -

\  :

n

. \. .

4 6

\ . . . .

10-2 10-1 100

10 Hz Spectral Acceleration (g) 1 l Figure 6-5. SA (10 Hz) hazard by area source.

l REPORTS /SONGSM25-94D.Ryr 6-12 Risk Engineering, Inc.

FAULT SENSITIVITIES,1 Hz

h. .

San Andreas

- - - -San Jacinto  :

- New.-Ing, Rose Can. -

- - -Idminore

- - -Palos Verdes c

u k  :

-Coronado -

8 '

.~ s 1 s

$ g. --

m g .

w ~

o %N g*\ .

\'b. -

'J -

%\N -

h  ? \ .

\ \

- '. \\

3 -

'. N.s \ -

'. \

~

~

"g h

~

?

\ .

g, \ '

e . \

6 . \g \

10-2 10-1 100 1 Hz Spectral Acceleration (g)

Figure 6-6. SA (I Hz) hazard by fault.

REPORTS / SONGS /9325-94D.RPT 6-13 Risk Engineering, Inc.

1 1

l 1

I FAULT SENSITIVITIES,1 Hz I

~ :

_ Aguana-Agua Tibia :

. - - - -Santa Catalina  :

- . - . San Clemente -

- - -Cucamonga -

- - -Hollywood-Raymond M  : -Malibu-Santa Monica-0 - -Peralta Hills 5

~

Eo w s

DC  %

N $ N w - EN -

s I

. o N. s -

x . N. s

.j

^ (N \,\ -

\\

gN ,

i e

m

_ N' g .

C g.

sg .\ .

A L

-  :- \  :

~

\g  :

\.s .  !  !

6

\\s.\.

\ ', i i

10 2 10-1 100 1 Hz Spectral Acceleration (g)

Figure 6-7. SA (1 Hz) hazard by f ault.

REPORTS /SONGSS325-94D.RFr 6-14 Risk Engineering, Inc.

1 l

. l l

FAULT SENSITIVITIES,1 Hz  !

1

San Miguel i 1

- - - -La Nacion  :

- . San Diego -

- - -Sierra Madre

- - -Temescal b l -Vallecitos i

. - -Whittier  :

~

u s N

M 's E~

's -

o "N s N

~

s ,

A N s y

. -N, N

N

\g .

. =t

.c .

- g '

s s b~

\. \\ -

[

N ,\\

\\\

N \ -

c -

. Ng .

k 1

~

N s N \ss -

N  :

N  %-  :  !

\ \

\\

\ '. -

k '

. \ \, . i\ . . . ,

1&2 1&1 100 1 Hz Spectral Acceleration (g) l I

l Figure 6-8. SA (I Hz) hazard by fault.

REPORTS / SONGS /9325-94D.RIT 6-15 Risk Engineering, Inc.

1 1

BLIND THRUST SENSITIVITIES,1 Hz h; , ,

l LA Basin Source A .

- - - -LA Basin Source B :

c 6 u .

e o

~

u -

M N "6 -

m -

o -

i se ~

1

• 3

~

l 3 .

.E 9 e o

~

r w  :

A  :. .

• g -

3 .

c '

c ,

4 y o  :

9 '

o s 10 2 101 100 1 Hz Spectral Acceleration (g)

Figure 6-9. SA (1 Hz) hazard by blind thrust.

REPORTS / SONGS /9325-94D.RPT 6 16 Risk Engineering, Inc.

i i

AREA SENSITIVITIES,1 Hz

) .

J  : Central LA Basin

- - - -Peninsular Ranges  :

i - -

- Offshore Basin -

i c 6 s o  : .

f h .

3o 4

M M

w Q

4 o -

>, s. -

~

7 g -

D a

N. g .

, 4  % - . .

P ~

!  : 'N  :

! o -

d

'N -

'h ..

i

. 1 -

'h..

- \, -

1 1

\

4 6 ';h

10-2 10-1 100 1 Hz Spectral Acceleration (g)

.i Figure 6-10. SA (I Hz) hazard by area source.

REPORTS / SONGS /9325-94D.RPT 6-17 Risk Engineering, Inc.

ATTENUATION SENSITIVITIES,10 Hz g

~ .

~

Rose Canyon-SCOZD Fau,lt ,

. .15 Idriss [

.20 Campbell  :

- - - .20 Abrahamaan

- .20 Sadigh -

.25 Boore, Joyner o $3 : & Fumal -

~

N o .

k 8 * * ~~. , &N s

M M ~

w Q

s. .

N  :

O %N D k b

.c 'h .

se i [ \  :

E %g.

s, g -

%\

~

~

M 1  :

5Y

~ v\- i

! h. \ i E

10-2 10-1 100 i

10 Hz Spectral Acceleration (g)

Figure 6-11. SA (10 Hz) hazard, sensitivity to attenuation equation.

REPORTS / SONGS /9325-94D.RPT 6 18 Risk Engineering, Inc. I

l 1

i ATTENUATION SENSITIVITIES,1 Hz g~

Rose Canyon-SCOZD Fa, ult ,

.15 Idriss  !

.20 campbell  :

- - - .20 Abrahamson -

- .20 Sadigh -

.25 Boore, Joyner

$l [ & Fumal  :

8 .

S  : .

~

o s b

~

'N, -

Ns., s, .

g N \. .

2 N \

~

,8 m N .\

o 6 s .

N ',\

- \ -

$ \ -

C \ . I L

- :_ \

\ _

g  :

~

E \ \g 1&2 10-1 100 1 Hz Spectral Acceleration (g) l Figure 6-12. S A (I Hz) hazard, sensitivity to attenuation equation. l i

REPORTS / SONGS /9325-94D.RPT 6-19 Risk Engineering, Inc. l i

l

l f

SLIP RATE SENSITIVITIES,10 Hz g

~

Rose Canyon-SCOZD Fau,lt ,

.1 SR = 3.0 mm/p :

- .1 SR = 2.1 mm/p :

- - - .6 SR = 1.5 mm/p

.2 SR = 1.0 mm/p G

E o

5 .- ~ ,

~ . .

.._ - ~

u -

.'s s y g '

'N's I

-:N?s s i i o -

s.s D .

NN  !

3

N o -

- N .s .

l a

4 C E -

' \\ _

- E '\  !

A

'h\ l e \

' \g

s

• c . .

c

.,\

4 *\

L -

. .\ -

.,\:

\\ .

.y o

10-2 10-1 100 10 Hz Spectral Acceleration (g)

Figure 6-13. SA (10 Hz) hazard, sensitivity to slip rate.

REPORTS / SONGS /9325-94D.Rrrr 6-20 Risk Engineering, Inc.

l 1

l 1

l i

I

,1 SLIP RATE SENSITIVITIES,1 Hz

g. Rose Canyon-SCOZD Fault

~

.1 SR o 3.0 mm/yr i

- .1 SR = 2.1 mm/yr :

- - - .6 SR = 1.5 mm/yr -

i .2 SR = 1.0 mm/yr -

! E - -

j g - -

i c -

o -

j k ~'l~~~ ,

l

g ~. . _ ~ ~ ,',

l j.

E.1

.N.'s w -

N s -

O j  : 'JN's '

i

! 3 -

,' s s -

y -

. .N -

i ',

J -

'\\ .

' \

l '$ E

~

') .\ -

a g  : ,

s  :

-  : . .\ -

g -

' \\

5 -

\ -

Q '\\

b  :

'. \.

'\g a

'.t.t 10-2 10-1 100 1 Hz Spectral Acceleration (g)

Figure 6-14. SA (1 Hz) hazard, sensitivity to slip rate.

REPORTS / SONGS /9325-94D.RFT 6-21 Risk Engineering, Inc.

l 4

4 i b-VALUE SENSITIVITIES,10 Hz i

Rose Canyon-SCOZD Fau,1t I g!

~

.2 b = .65 4

- - - .6 b = .80  : l l - -

-.2 b = .95 -

i 6 -

8 -

c -

o i o -

5 "6

~

o D .

• c Ia

• A "b o -

T:s .

c -

c 4

1 ._ _.

m 6

10-2 10-1 100 10 Hz Spectral Acceleration (g)

Figure 6-15. SA (10 Hz) hazard, sensitivity to b-values.

REPORTS / SONGS /9325-94D.RPr 6-22 Risk Engineering, Inc.

i J

1 i

i b-VALUE SENSITIVITIES,1 Hz

&. Rose Canyon-SCOZD Fault ,

.2 b = .65

,  : - - - .6 b = .80 .

- l l

- .2 b = .95 -

i l c 6 1 o  : -

c

o .

1 o -

i M I M h -

w -  :  :

o -

l D -

1 4 .

) *E Jo i m 1

- 6 -

W .

b -

I e, .

c .

a l

l 4 b

~

6

. . . . . . .., l 10-2 101 100 1 Hz Spectral Acceleration (g)

Figure 6-16. SA (I Hz) hazard, sensitivity to b-value.

REPORTS / SONGS /9325-94D.RFr 6-23 Risk Engineering, Inc.

1 i

4 i 1 i

i DEPTH SENSITIVITIES,10 Hz

~ .

Rose C,anyon-SCOZD Fau,lt , ,,

. --- .4 15 km  :

.6 10 km  :

j -

c 6 u -

! c:  :

y

..~. .. .- l i o g ~. .

f N $ '. _

o '

' ~

~

2 *c 3as -

f% b~ -

4 -

s g  : ',

s - -

g - ' -

< g .

I -

l '. l l  : .:

i 10-2 101 100 1

! 10 Hz Spectral Acceleration (g)

Figure 6-17. SA (10 Hz) hazard, sensitivity to depth.

i REPORTS @NGSS325-94D.RPT 6-24 Risk Engineering,Inc.

DEPTH SENSITIVITIES,1 Hz g_

Ro. se Canyon-SCOZD Fault

---.4 15 km  :

.6 10 km  !

c b u -

E .

~

}o g

~. . 1 W Q -

~.

w -

E  : l o _ '.,  :

y '

l n _ ,

y .

e -

.c  % ' l w

2 - l '. .

i.

% '. j a -

b  : '. -

l in

• 6 i 10-2 101 100 1 Hz Spectral Acceleration (g)

Figure 6-18. SA (I Hz) hazard, sensitivity to depth.

REPORTS / SONGS /9325-94D.RPT 6-25 Risk Engineering, Inc.

1 i

i MAXIMUM MAGNITUDE SENSITIVITIES,10 Hz j g Rose Canyon-SCOZD Pault

~

l

.12 M , = 6.5  !

: - - - .20 M x = 6.6  :

- .20 M x = 6.7 -

~ - -

.21 M x = 6.8 -

i 5~ -

.10 M x = 6.9

.05 M x = 7.1 i g

.07 M x = 7.2 .

.05 M x = 7.5 i

ko YE 5l.lL

'-q%,

k c.

@ 6

~ .

N @'.,  :

~ N k. -

o

.'% s N s Q g, ,

j

' q.s sN -V, x .

3

, sN , N*N ,

. i

'N.

'xx',

N-

.e a- \\..

0

.N .x .

' ~

.Y  %'  :

a .%.% -

E -

'V.  %{,

.\\

L _

' .T - l 6

.g\

10-2 10-1 100 10 Hz Spectral Acceleration (g)

Figure 6-19. S A (10 Hz) hazard, sensitivity to maximum magnitude.

I REPORTS / SONGS /9325-94D.Rn 6-26 Risk Engineering, Inc.

l l

E MAXIMUM MAGNITUDE SENSITIVITIES,1 Hz g.

Ro. se C.anyo, n-SCOZD Fau.lt. .

.12 M a = 6.5  :

- - - .20 M = 6.6  :

- .20 M = 6.7 -

.21 M x = 6.8

.10 M x = 6.9 e b i -

- 05 M x = 7.1 1

8

• - - 07 M a = 7.2

.05 M a = 7.5 ko t'

~~

N M ,

N

'.*' :~:

6 N"%.- '

o

~

$4- NN.

• Q (-

~

x NN ss'N'- N

' q.

5  %.

• E

. s\ . N' -

m . N. N,

$w 6

~

. 'N

.N s. k'%

~

~

'I -

g . .

c -

^. -

I

\] ]-

6 10-2 10-1 100 1 Hz Spectral Acceleration (g)

Figure 6-20. SA (I Hz) hazard, sensitivity to maximum magnitude.

REPORTSISONGS)9325-94D.RYr 6-27 Risk Engineering, Inc.

- -.=. .- . .. . . - . - _ - _ - - - - - . . .. - - . . . . _ .

1 4

4 i

J 4

l

.i i; TOTAL HAZARD, 25 Hz r . . . , ,i, . . . .

1

.85 Fractile  :

: Mean . l

- s

- s . _ . Mdan -

. N.

s s - - - .15 Fractile N

6

~

l 8 '

N  !

E - N

. N

. .N i

w

. s -

i W E -

\

,N

% g

. o -

\

j . N 4

i

.x

g -

s I

p - .

,.\ -

1 m \

b .\

i Q -

. \ -

2 -  :

A  : . \. .\

\  :

~

g .

', \ \ .

c .

.\ .

4 1 -

'\s

~ 1

. . \s

.\g .

..s b .\y s

- . . . . ., . . . ...,t 10-2 101 100 25 Hz Spectral Acceleration (g)

Figure 6-21. Fractiles and mean of total SA (25 Hz) hazard.

REPORTstsoNoss325-940.Rvr 6-28 Risk Engineering, Inc.

TOTAL HAZARD,10 Hz

.85 Fractile

, Mean  :

. - . Median -

., 's s - - - .15 Fractile -

6

. s -

3  : , s  :

a '. N -

l c _ -

N

]O -

'. N N

u -

'. N x l W E _

'. N N _

g . .

N x

'. N c _ .

N -

gg

\

\

$w E~

\

s

i m<  : . -

i

. .\  :

4 m -

'. \ \ .

i\ .

'\\

b~  :

g- )

- .\.\ :-

<. .' i

.\..

i m

b '\

10-2 101 100 10 Hz Spectral Acceleration (g)

Figure 6-22. Fractiles and mean of total SA (10 Hz) hazard.

REPORTS / SONGS /9325-94D Rn 6-29 Risk Engineering, Inc.

i 1

1 1

, 1 TOTAL HAZARD, 5 Hz

~

.85 Fractile  :

L, Mean  :

's s . _

. Mdan -

~~

's s - - - .15 Fractile -

' s -

ds - -

N -  !

o - -

o .  %

g '.

s  ;

'.s N -

u -

. N -

M . \

N $ -

', N _

l w -

E - N o .

x p '

s \ -

4 N

• 5

, . \

m -

• g

.c Q ~ .

2 -

,\ \  ;

A -

.\ -

e ~

.\\-

2

.\s

• \-

b

'. \.

'i

~

9 o

10-2 10-1 100 5 Hz Spectral Acceleration (g)

Figure 6-23. Fractiles and mean of total SA (5 Hz) hazard.

REPORTS / SONGS /9325-94D.RFr 6-30 Risk Engineering, Inc.

TOTAL HAZARD, 2.5 Hz

.85 Fractile  :

Mean

's s . - . Median -

.. ' 's '

. . . .15 Fractile -

~. .

6 7 ., s -

8 ,

s  :

a o

- s s

1 -

N g .

', N .

M

\

W E _

N N

E ,

o -

, N

* \

g y

\ -

\

.c a , \

$, E r

'. \

m . . \

- - ', \

a

\\-

, 5 -

', \-

4 1 -

.\3

.\

~

r 6

10-2 10-1 100 2.5 Hz Spectral Acceleration (g)

Figure 6-24. Fractiles and mean of total S A (2.5 Hz) hazard.

REPORTS / SONGS /9325-94D.RFr 6-31 Risk Engineering, Inc.

l I

4 I

l l TOTAL HAZARD,1 Hz

!, E_

)  : - -

.85 Fractile  !

Mean  :

i s

- Median -

s s - - - .15 Fractile -

s 4

o 6

- 's s -

, l 4

a o

. Ns T '

- N .

3 -

N u

M N -

w b -

'. N -

C s .

\ -

D '

.g

'. \ -

. . \

,o .

a .

\

b  % - . \ .

O i '

\ i A .

. \

a n -

'. \ -

. i

\ .

\

L-

'. \ -

i \ .

g .

m i

\~

6 i

t 10-2 10-1 100 1 Hz Spectral Acceleration (g)

Figure 6-25. Fractiles and mean of total SA (I Hz) hazird.

REPORTS / SONGS /9325-94D.RFr 6-32 Risk Engineering, Inc.

l l

1 i

l l

l TOTAL HAZARD, 0.5 Hz ,

S.

~

~

i

.85 Fractile

Mean  :

. _ . Median -

( - - - .15 Fractile i

& _'s N

_ i g -  :-

s  : t c . s l c .

.s - s

'. \

s g -

x '

W E - '. \

g _

w -  : '

\  :

C  :

3 ',\ g '

l 3

'.\

[

s

$w Ea

\

s A ',

~

\ l e

• n .

\

. \ -

4

'. \ l h j . \ i

. '. \  :

i

\ .

i g

~

g . .

', \

L 10 2 10-1 100 0.5 Hz Spectral Acceleration (g) l l

Figure 6-26. Fractiles and mean of total S A (0.5 Hz) hazard.

i i

REPORTS / SONGS /9325-94D.RFT 6-33 Risk Engineering, Inc.

l l

UNIFORM HAZARD SPECTRA l 3, ,

. . . . 5. % Damoine - . . .. .

l Prob. = 1.4x104 (SSE) i 3 n g -

s -

o / s

'c n

~

/ ~N s 2

e

/ < -

Ns s

- / , -

8 O h -

/

/ , '.\\ '

.N 4 -

,\ -

g / , .

~

g -

6 m .

.85 Fractile a Mean

_ . _ . Median

.15 Fractile 3

EWumt ie a i >t i e if iia 10-1 100 101 102 Frequency (Hz)

Figure 6-27. Fractiles and mean of uniform hazard spectra, 1.4x10"(SSE) anr.ual probability of exceedence.

REPORTS / SONGS /9325-94D.RPr 6-34 Risk Engineering, Inc.

1 l

l l

l UNIFORM HAZARD SPECTRA a_ . ..

5% Damnine -

i Prob. = 1.7x104 (2xSSE) f

- 's .

N f' . \

'/.,N ,

'. \

3 n l / ,'

- - 'N,

\

c .

/ '\ \

\ -

o j ',

j U f ,

, ' ,\

o * ,

1 Tl &

5 .

<l -

2 C/) -

.85 Fractile a Mean

_ . _._ . Median

.15 Fractile E

101 100 101 102 Frequency (Hz) i 1

Figure 6-28. Fractiles and mean of uniform hazard spectra, 4

1.7x10 (2xSSE) annual probability of exceedence. I 1

REPORTS /SONGSS325-94D.RFr 6-35 Risk Engineering, Inc. l

v MEAN UNIFORM HAZARD SPECTRA

, 3, ,

. . . . . ,5% Damni,n. _e a .

] _

Prob. = 1.7x104 (2xSSE) 1 9

v a Prob. = 1.4x1 a

} SSE) -

o

'd 9 j

g "

i o i

j &

~

u -

4 -

q ~

~

3 i

g 4

n.

r.o -

l 4

i N -

S r

S . . . . , .1 ,

10-1 100 101 102 Frequency (Hz)

Figure 6-29. Mean spectra for 1.7x10' and 1.4x10" annual probabilities of exceedence.

REPORTSISONGSS325-94D.Rn 6-36 Risk Engineering, Inc.

l i

UNIFORM HAZARD SPECTRA ,

a-5% Dampine. Vertical Ground Motions Prob. = 1.4x1(H (SSE) -

3 " -

c -

I O ,

'N l 3 s' \ l

/,,. \

'E g /< ',N 1

~

8 - / ,' ',\  !

/,* . .

/, ' -

$ /

/a "8 -

O.

• m -

/ , .

/ ,'

/o -

.85 Fractile a

Mean

. _ . Median ~

- - - .15 Fractile 3

E . .,... . ... . .

Frequency (Hz) l Figure 6-30. Fractiles and mean of uniform hazard spectra for vertical ground d

motions,1.4x10 (SSE) annual probability of exceedence.

RFPORTS/ SONGS /9325-94D.RPr 6-37 Risk Engineering, Inc.

3.5

SSE (Risk Eng) 2xSSE (Risk Eng) 3 2.5

/ i.

\.

h2 i

i g

1.5

\

1 0.5

. Y 0

0.1 1 10 100 Frequency (Hz)

Figure 6-31. Comparison of vertical spectral shapes (normalized 1-10 Hz) for the SSE level and the 2xSSE level.

REPORTS / SONGS /9325 94D.RFr 6-38 Risk Engineering,Inc.

l l

I l

1 1

1.2 f,,!

l l l 3

A. Il l ll,l l l ill

\\ l!  !  !  !

.] l' .

i I t I

l l l.

4 .

i! !

0.8 e .

iV

- ii

+

l'!!  :  ;

i, m i t,

z . I ,*

a  :

i ii l l: l/

$ l l i c 0.6-W

_ . . . .. . . . . . . . j . . . . .. . . .%..............................;...j.........., .j . . . . . . . . . . . . . . ./

m  : I ;i I l!

E -

3 j 5 ' \ i' m l li i i  !' 4 i' j',

0.4

'l,il

i I! il I

i - Campbell (1990) M4.7-6.1 l

l j Campbell (1990) M6.2-7.8 i ii 0.2 l;

-- Sadigh et al. (1993) M< 6.0

i 4

'; ....... Sadigh et al (1993) M>6.5 i ji l I >l Ill l i  !;!  ! l .

0 0.1 1 10 100 Frequency (Hz)

Figure 6-32. Comparisons of standard errors for Campbell (1990) and Sadigh (1993) vertical attenuation equations.

REPORTS / SONGS /9325 94D.RFr 6-39 Risk Engineering,Inc.

3.5 SSE (Risk Eng) 3- SSE (C90 - OZD only)

SEE (c00 w/ modified sigma, OZD only) 2.5 ' -

3 i

.5 2 .

is s.

k .

1.5 - i' ig i

i 1

l

[

0.5 /' .i fi O

0.1 1 10 100 Frequency (Hz)

Figure 6-33. Comparison of venical spectral shapes at the SSE level using the full hazard analysis and simplified analysis including only the SCOZD.

REPORTS / SONGS /9325-94D.RPT 6-40 Risk Engineering,Inc.

l l

l 3.5 SSE (Risk Eng) 2xSSE (Risk Eng) i 3- --

2xSEE (c00 w/ modified sigma, OZD only) l 1

- 1 l 2.5 ,' '

, ,\

.~ i :

.B - ,

2 -

i

\

I t .

31.5-. '

l 'h v) _

.s 1 .

i

/

0.5

. Y,'

0 0.1 1 10 100 l Frequency (Hz)

Figure 6-34. Comparison of venical spectral shapes at the SSE and 2xSSE levels.

REPORTS /SONGSS325-94D.Pyr 6-41 Risk Engineering,Inc.

4

Section 7 3 '

i CONCLUSIONS l

j This study presents seismic hazard results that represent the annual frequency of exceedence of l

} various ground motion levels at SONGS, and the uncertainty in the annual frequency of exceedence.

i These results are represented as a family of fractile seismic hazard curves, and as uniform-hazard spectra corresponding approximately to the SSE and 2xSSE levels. The uncertainties in hazard f

i derive from uncertainties on input assumptions regarding seismic sources, seismicity parameters, i

and ground motion attenuation equations. 'Dius the analysis presented here is state-of-the-art, h~=_= it ir-yorates and presents uncertainties in the major factors affecting seismic hazard in j the region around the site.

(

j The tectonic interpretations and seismicity p. rim-s were derived by Geomatrix Consultants, Inc., ,

and are 4x - a.ted in Appendix A. They consist of faults and area sources in southern California j that might contribute to hazard, and the parameters defining those sources (slip rate, b-values, t

i geometry, and maximum magnitude). For area sources, an analysis of historical seismicity was I

conducted to establish rates of activity and b-values.

Attenuation equations were derived by Woodward-Clyde Consultants and are documented in Appendix D. Five equations were selected to estimate spectral velocity at 25,10,5,2.5,1, and 0.5 Hz. These were weighted by whyaring the predictions to observations of strong motion in southern Califomia and determining how well the equations fit the data. All five equations predict ground motion on stiff soil, which is appropriate for application at SONGS.

The methodology used in this study follows closely that used in other state-of-the-art studies of seismic hazards at nuclear plant sites in the U.S. The derivation of seismic sources is specified by the earth science experts; an analysis of historical seismicity is performed to aid in estimation of l

seismicity parameters; and all relevant theories and data on earthquake causes and characteristics in southern California were examined and incorporated into the interpretations.

REPORTS /SONGSS325-94D.RPr 7-1 Risk Engineering, Inc.

l 1

As discussed in the Intmduction, this cudy used three teams to develop the seismic hazard results, one for attenuation equation, one for seismic source descriptions, and one for integration and hazard calculations. In addition, an expert review panel was assembled and has provided feedback during the entire project, including reviews ofintermediate results and reports and the final report. Thus

{ we feel that the results presented here have a strong basis and are appropriate for use in the IPEEE PRA for SONGS. Regarding the analysis of earthquake data, we have not conducted an extensive i

evaluation of the earthquake catalogs used in this study and described in Section 4, to address issues

! such as the accuracy of specific event locations and magnitudes, the conversion of intensity to ,

magnitude, and the completion of earthquake coverage represented by the catalogs. 'Ihe available 4

I catalogs have been scrutinized closely, e.g. by Ellsworth of the USGS, and it is eyy opiate to use these data bases as presented. The catalogs have simply been accepted and used with theirlisted j values of magnitude and location. Similarly, the specific soil conditions at SONGS have not been modeled in detail with the ground motion attenuation equations adopted here. The attenuation equations use generic factors to estimate the dynamic response of stiff soils. Site-specific studies '

of soil response under earthquake loads might yit.ld results different from those used here, with a l l corresponding effect on the hazard results. Also, as pointed out regarding the comparison of p+fsw and data in Appendix D, there is an apparent tendency for the attenuation equations to slightly over-predict spectral response, which if correct would result in the hazard values reported here being slightly conservative. Apart from this point, we have no reason to believe that more detailed studies in any of these areas would preferentia'.ly either increase or decrease the hazard results calculated here.

j i

e REPORTS / SONGS /9325-94D.RIrr 7-2 Risk Engineering,Inc.

i i

i i

APPENDIX A Seismic Source Characterization Prepared by Geomatrix Consultants, Inc.

San Francisco, California

e l

l

--- --- ~_..__ _____, Ww,--- n _ _ _ '- fem _ _ , , _

'#M M ar- 1 + , _, "' " * " " emww,,

TABLE OF CONTENTS fagt

1.0 INTRODUCTION

A-1 1.1 PURPOSE AND OBJECTIVES A-1 1.2 METHODOLOGY A-2 1.2.1 Development of Seismic Hazard Model A-3 1.2.1.1 Seismic Source Characterization A-3 2.0 REGIONAL TECmMIC SETTING A-10 3.0 SEISMIC SOURCE CHARACTERIZATION A-11 3.1 OFFSHORE FAULTS A-11 3.1.1 Newport-Inglewood-South Coast Offshore Zone of Deformation-Rose Canyon Fault Zone A-13 3.1.1.1 Segmentation and Rupture 1.cngths A-14 3.1.1.2 Slip Rate A-17 3.1.1.3 Alternate Source Characterization Models A-18 3.1.2 Palos Verties Fault Zone A-20 3.1.2.1 Segmentation and Rupture Lengths A-22 3.1.2.2 Slip Rate A-22 3.1.3 Coronado Bank Fault Zone A-23 3.1.4 San Diego Trough Fault A-25 3.1.5 Santa Catalina Escarpment Fault A-25 3.1.6 San Clemente-San Isidro Fault Zone A-26 3.1.7 San Mateo 'Ihrust Fault A-28 3.2 ONSHORE FAULTS A-30 3.2.1 Elsinore Fault Zone A-30 3.2.2 Whittier Fault A-32 3.2.3 Aguanga-Agua Tibia-Earthquake Valley Faults A-33 3.2.4 San Jacinto Fault Zone A-34 3.2.5 San Andreas Fault Zone A-37 3.2.6 Malibu Coast-Santa Monica Fault Zone A-40 3.2.7 Hollywood-Raymond Fault Zone A-41 3.2.8 Sierra Madre Fault Zone A-42 3.2.9 Cucamonga Fault Zone A-43 3.2.10 Peralta Hills-Norwalk Fault Zone A-44 axeuwreroc (i) w y is i m

TABLE OF CONTENTS (continued) 3.2.11 Temescal Fault A-45 3.2.12 La Nacion Fault Zone A-46 3.2.13 Cristianitos Fault A-47 3.3 BURIED OR BLIND FAULT SOURCE ZONES A-47 3.3.1 les Angeles Basin Source Zone A A-49 3.3.2 Los Angeles Basin Soume Zone B A-52 3.3.3 Assessment of the Potern::d For Unknown Blind Thrust Faults Near San Onofre Nuclear Generating Station A-54 3.4 REGIONAL AREAL SOURCE ZONES A-55 3.4.1 Peninsular Ranges Source Zone A-56 3.4.2 Centrallos Angeles Basin Source Zone A-57 3.4.3 Offshore Basin Source Zone A-57

4.0 REFERENCES

. A-59 LIST OF TABLES Table A-1 Estimated Fault Rupture Parameters and Weights LIST OF FIGURES Figure A-la Fault sources included in seismic source model Figure A-lb Areal source zones included in seismic source monel Figure A-2 Logic tree showing alternative models for the Newpon-Inglewood-SCOZD-Rose Canyon fault zones Figure A-3 The Newpon-Inglewood-Rose Canyon fault zone, showing major fault segments (modified from Fischer and others,19'82)

Figure A-4 Cross sections of the Newpon Inglewood-Rose Canyon fault zone from digitally processed seismic reflection data. Locations are shown on figure

4. (from Fischer and Mills,1991)

Figure A-5 Previous interpretations of the Newpon-Inglewood fault zone from 1972 -

1988 aums4carr.arr. Toc (ii) w y is.i m

. _ -. _. _. - . . - .....- _-.. -_-_. . ~ . - . . - . - - _ _ . . . -. . . _ - . . -

l i

i I

l TABLE OF CONTENTS (continued)  !

a i l 1

1 LIST OF FIGURES (continued)

Figure A-6 Geologic setting of the inner margin from the Palos Verdes Peninsula to l the Silver Strand (from Fischer and Mills,1991) l I Figure A-7 Cross section along JEBCO digitally processed seisinic reflection line number 49-102 off San Mateo Point. Location of cross section shown on Figure 4 as a dashed line (from Fischer and Mills,1991)

Figure A-8 Idealized section of upper lithosphere, illustrating features of regional and local strain partitioning. Regionally partitioned structures originate at crustal depths where large-moment-release canhquakes nucleate and should be characterized independently as separate seismic sources. locally partitioned structures are not necessarily separate seismic sources and i

should be characterized collectively to assess underlying seismic sources (from lettis and Hanson,1991)

LIST OF PLATES Plate 1 Seismicity and Quaternary fault map southern California and northern Baja California mun6ucarr arr. roc (iii) wy is, i9w

J +45.J 4 d. 4-a4-4 4W -M Mw..AS A4 h W-4W a e W.JmM4WAeA-- "

"^4 pse A"

"h8' ' * - " ~ - -

'M AmM C.* k. e e # % #-. 4 Fh ' s4 4m&.4%MA as m A =

l I

l 4 I f I I

i r

l

, 1 l

I t'

e i t

4 e

i e

J e

f 1

4 4

5 I

i..

J 1

4 8

s .

5

't I

t d

d

.1 1

k, 4

t W

d 5

l 1

i J

I l

l 1

-..- ,. - -, , .,--, .c . .-_.- . . - , .--.- , , , , . , , , .. , . . ,.- --,, , . . , - - - -- - ..,.- . , ,, , . . , ,. ,. ,-

d i

1 l

l t

1.0 INTRODUCTION

i l 1.1 PURPOSE AND OBJECTIVES i

i j This document is the final Seismic Source Characterization report for the probabilistic seismic l hazard analysis (PSHA) at the San Onofre Nuclear Generating Station (SONGS) site. The l probabilistic hazard results will provide input to a probabilistic risk assessment in support of J

the Individual Plant Examination for External Events (IPEEE) program.

r q

4 This report focuses on first-order seismic source data, characteristics of seismic sources, and l 1

assessments of the relative credibility of alternative interpretations of the data. Two principal 4

seismic source types that can affect ground motion hazards at the site are included: fault sources and regional areal source zones. Individual Quaternary faults and fault zones that may

{

have earthquake potential are identified on Figures la and b, and the available data related to  !

the seismic source characterization are discussed in detail for each active or potentially active j source. 'Ihe fault sources represent local sources of earthquakes, and they will localize

seismicity in the seismic hazard analysis. In the case of the postulated blind thrust fault I sources in the Los Angeles basin, where the geometry of the structure is uncertain or where l

! alternative geometries have been proposed, these sources are represented on the seismic source

map as local areal fault source zones (Figs. A-la and A-lb). The regional areal somre zones I serve as background sources in the hazard analysis, within which earthquakes are assumed to occur randomly. Besides representing observed seismicity that cannot be readily associated with faults, the regional source zones are also a means of representing for the hazard analysis the potential for earthquake occurrence on faults and geologic structures that are buried, blind, or otherwise unknown.

The characterization of seismic sources provided in this study relies to a large extent on geologic information. This is consistent with the methodologies that have emerged in probabilistic seismic hazard analysis in the past several years (e.g., Coppersmith,1991) for assessments in the western United States. Seismicity and geodetic data are used to assess the coummarr-arrnr A.1 my is, im

I locations and probability of activity of seismic sources and to provide constraints on the likely l l

slip rate values for faults, particularly those faults in the offshore region for which there is little or no slip rate information.

A key part of the assessment of seismic source characteristics is the inclusion of uncertainty j in the assessment of seismic source parameters using logic trees (e.g., National Research l Council,1988). In using this approach, individual hypotheses and parameter values each are 1 assigned a relative weight that expresses the degree of credibility of that value in light of the available data (Fig. 2). This approach has now been used in a wide variety of projects,

! including seismic hazard analyses for numerous nuclear power plant sites, major bridge sites, and other critical facilities. The key seismic source characterization elements for each crustal source are presented in Table 1. Assessments are made for significant parameter values i

including probability of activity, total length, rupture length, rupture width, slip rate, and  !

maximum magnitude. The range of possible values of these parameters and the relative credibility of particular parameter values are given, with supporting documentation in the text.

These judgments were made by a team of Geomatrix geologists and seismologists (Kevin l

Coppersmith, Kathryn Hanson, Donald Wells, and Laurel DiSilvestro) based on a review of these data. The objective of the assessments is ro represent the range of models and parameter I values, and their relative weight, that might be developed by the larger scientific community.

Preliminary values were presented at project review meetings, and the final source characterization model reflects suggestions of participants at these meetings and written comments by the review panel members. The goal of this evaluation is to make explicit the I available data for individual seismic sources and to explain how the assessment of seismic source characteristics is derived from these data sets. Uncertainties in these parameters are identified and quantified in Section 3.0 and summarized in Table A 1.

1.2 METHODOLOGY In the following sections, we briefly discuss the methodology used to identify and characterize seismic sources, aunararr.arrar A-2  % 15. m l

4

4 i

1.2.1 Development of Seismic Hazard Model 1

The seismic hazard at a site' is a function of the location and geometry of potential sources of 4

1 futme earthquakes, the frequency of occurrence of various size earthquakes on these sources, i

and the characteristics of seismic wave propagation in the region. In the methodology i described here, these elements are analyzed within a probabilistic framework that addresses  !

4 both the randomness of the earthquake process and the uncertainty in modeling the process.

The seismic hazard model consists of two basic components, a model of the sources of l

,' potential future earthquakes and a model of the effects of future earthquakes at the site. A a

j discussion of the first of these two components is given below.

{ 1.2.1.1 Seismic Source Characterization 4

j A seismic source model provides a description of potential future earthquakes in terms of their i spatial distribution, the rate of seismic activity, and the relative frequency of various size events. The steps involved in source characterization are (1) definition of regions within the

crust that are potential sources of future earthquakes, (2) assessment of the source geometry,

! (3) assessment of the maximum size of future earthquakes possible on each source, and (4)

); assessment of recurrence rates for earthquakes of various sizes.

i

Source Definition - A seismic source represents a region of the earth's crust where the i

! characteristics of earthquake activity are recognized to be different from those of the adjacent j crust. Seismic sources are identified on the basis of geologic, seismologic, and geophysical data. An understanding of the regional tectonics, local Quatemary history, and seismicity of an area leads to the identification of geologic structures that may be seismic sources. To this end, developing tectonic models for crustal deformation and assessing the tectonic rule of individual geologic structures are essential for both identifying potential sources and assessing their characteristics. Geologic studies can be used to assess the location, timing, and style of crustal deformation. Associating geologic stmetures with historical or instrumental seismicity may clarify their role within the present tectonic stress regime. Characteristics of seismic energy release, such as focal depths and source mechanisms, also can aid in identifying potential sources.

<oawzerr.nrrm A-3 m u. nm

Because earthquakes occur as a result of differential slip on faults, modeling of seismic sources as individual faults is the most physically realistic method for seismic hazard analysis. Under favorable conditions, individual faults can be identified and treated as distinct seismic sources.

Active faults are usually identified on the basis of geomorphic expression and stratigraphic displacements, but can also be identified by lineations of seismicity or by geophysical measurements. For example, the results of marine seismic reflection surveys have been successfully used to identify active faulting. A fault model for individual sources allows the use of geologic data on fault behavior to characterize earthquake activity because the use of seismicity data alone may not be sufficient for source modeling.

An understanding of the local tectonics can provide a basis for identifying seismic sources in areas where individual faults cannot be identified at the surface. Recent earthquakes in California (Coalinga,1983; Whittier Narrows,1987) occurred on thrust faults located beneath active folds. Geologic structures that show evidence of active deformation can be identified using techniques similar to those used to identify active faults. Such structures can be identified as seismic sources where the location of the actual fault plane is uncertain. For such structures, quantitative structural geology techniques (e.g., Suppe,1983) may provide a basis for estimating the location of buried faults.

In regions where no distinguishing geologic features can be identified, seismicity is usually modeled as occurring randomly within large areal background sources. Here again, geologic and tectonic data can be used to identify blocks of the earth's crust that are expected to have fairly homogeneous characteristics. The extent of such regions can serve as the basis for defining the boundaries of regional areal sources of distributed seismicity.

Probability of Activitv 'Ihe assessment of crustal fault activity reflects our judgment of the likelihood that the structure is seismogenic, or active, within the present tectonic regime and will, therefore, localize seismicity above the levels occurring randomly within the regional source zones. Our assessment of activity is based on such factors as association with historical seismicity, evidence for late Quaternary fault displacements, geomorphic evidence for aummarr.arrnr A-4 w is. im I

?

1 l

I  ;

E I

geologically recent deformation, association with neighboring structures showing evidence for 1

Quaternary activity, pm-Quaternary history of deformation, and orientation relative to the i

present stress field. Evidence for one or more late Quaternary fault displacement and l l geologically recent deformation is considered the strongest indicator of activity. However, as noted below, the geologic record is inadequate in many areas for either evaluating or l

l precluding fault activity. A spatial association with seismicity such that there is a clear j

lineation along the structure also is an indicator of activity. However, the absence of a spatial  !

seismicity pattern provides only a weak argument against activity.

i 3

j For this study, a fault is classified as " active" and is considered to be a seismic source if there I

is evidence for Holocene (<10 ka) or latest Pleistocene (<20 ka) surface displacement, recunent late Quaternary (<780 ka) surface displacement, and/or if it is associated with a large-magnitude historical earthquake. This definition is consistent with existing (10 CFR Pan 100, Appendix A) and proposed (Appendix B) Nuclear Regulatory Commission criteria for fault capability. If there is questionable evidence for late Quaternary surface displacement and the age of the most recent event is not known, or if the fault is questionably associated with a large-magnitude historical canhquake, the fault is considered "potentially active" and a probability of activity is assigned to the structure based on available data. Quaternary faults for which there is evidence suggesting no displacement having occurred during the late Quaternary (approximately less than 780 ka) are not considered to be potential seismogenic soumes; for these faults, the probability of activity (i.e., the likelihood that they will localize seismicity above the rate due to the background areal source) is considered to be zero.

Source Geometry - A description of the geometry of a seismic source is necessary to evaluate the distances from the site at which future canhquakes could occur. In addition, source geometry can place physical constraints on the maximum size canhquake that can occur on l the source.

Seismic sources defined as faults are modeled in the analysis as segmented planar features.

Eanhquake ruptures on fault sources are modeled as rupture areas (based on estimated rupture munwoner.nrrnt A-5 u r n. w

s i

length and downdip width), with the size of rupture defined on the basis of empirical

{ relationships between earthquake magnitude and mpture size (e.g., Wyss,1979).

I For faults, the maximum canhquake magnitude is related to fault geometry and fault behavior through an assessment of the maximum dimensions (i.e., length and downdip width) of a single

) rupture. Fault segmentation provides a means of identifying the ponions of the fault zone l likely to rupture during individual earthquakes (Schwartz and Coppersmith,1986; Schwartz, j 1988). Studies of coseismic fault ruptures worldwide indicate that faults typically do not i

rupture their entire length during individual earthquakes. Rather, they rupture individual i

1 segments or limited numbers of adjacent segments, and through time, these segments may i rupture repeatedly through several seismic cycles.

I The identification of future rupture segments is difficult, and methodologies for using

j. segmentation modeling to estimate future ruptures are in the early stages of development. The j best types of data that provide information on segmentation are those that quantify differences j in behavior along the length of a fault during its most recent and previous seismic cycles. In l addition to observations of historical canhquakes, paleoseismic data regarding the timing of j

i past events, slip per event and its distribution along the length of a fault, and slip rate are

! critical for defining segments and for modeling canhquake recurrence (Schwartz,1988). Major changes in the cumulative amount and sense of slip, the strike of the fault, the trace i

complexity, the occurrence of significant lithologic changes, and the presence of transverse geolo'gic structures may supply additional information that can be used to recognize fault i

j segments (e.g., Knuepfer,1989). Our assessment of potential rupture lengths for seismic sources is based on the available data regarding these characteristics.

t i

j The downdip width of a fault can only be assessed indirectly. Estimates of the downdip width typically rely on observations of canhquake focal depths and physical considerations such as the thickness of the seismogenic crust and the dip of the fault plane. The distribution of observed focal depths ofinstrumentally recorded seismicity provides the best indication of the depth of the seismogenic crust in panicular regions (Sibson, 1982; 1984).

a u w anrr.nvr n t A6  % n. m w r -- y, we -r

- - - . - . . - . - - - - - _ - - . - - _ - - . . . - - . . . - . ~ - . -

1 i

] For seismic sources defined as geologic structures suspected to contain faults, the distribution i

of canhquakes can be modeled as rupture surfaces occurring on multiple fault planes distributed throughout the source volume if the general trend of such planes is known or can j be inferred. Alternatively, earthquake locations can be modeled as random point sources within the source volume if the orientation of potential fault planes is unknown. The spatial i distribution of seismicity within large areal sources can be modeled in a similar fashion.

1

! Maximum Earthauake Marnitude - A key assumption in seismic hazard analysis is that each j seismic source is associated with a maximum canhquake. Maximum earthquakes are usually 1

l assessed in two principal ways: (1) by estimating the maximum dimensions of future ruptures and relating those dimensions to magnitude; and/or (2) by considering the size of the largest historical earthquakes associated with the source and with tectonically analogous sources.

Assessment of a maximum magnitude is ultimately a judgment that incorporates an understanding of specific fault characteristics, the regional tectonic environment, the similarity to other faults in the region, and data on regional seismicity. For many faults these data are poorly constrained, and even for the faults for which there are a great deal of data, uncertainties can be high.

A common approach to estimating maximum earthquakes involves estimating the maximum rupture dimensions for a fault of interest ano, using the empirical correlations developed through analysis of coseismic fault ruptures worldwide (e.g., Bonilla and others,1984; Bonilla and Buchanan,1970; Mark and Bonilla,1977; Slemmons, 1977,1982; Slemmons and others, 1989; Wesnousky,1986; Khromovskikh,1989; Wells and Coppersmith,1994), calculating the magnitudes associated with those dimensions. Empirical regressions have been developed of magnitude on the length of rupture, the area (length times width) of rupture, and the amount of fault displacement. Because these relationships are subject to some uncenainty, using a number of magnitude estimation techniques can result in more reliable estimates of maximum magnitude than applying a single relation.

ovouu. owr-wrm A-7  % is, a

In this analysis, the distribution of maximum magnitude values for fault sources veill reflect assessments of postulated rupture dimensions (i.e., rupture length, downdip fault width, slip per event). The thickness of seismogenic crust is well constrained for onshore regions of southern California due to the density and quality of seismic networks. Rese data allow for reasonable estimates of the downdip width of most structures. He assessment of rupture length is less well constrained. To capture the uncertainty in the estimated rupture length, we incorporate a range of values based on single- and multiple-segment rupture scenarios. When displacement per event data are available or can be inferred from scarp heights, we have also l i

used these data to estimate maximum magnitude values. For this study, we use empirical regressions that have been developed by Wells and Coppersmith (1994), based on a recently 4

updated comprehensive data base of historical surface ruptures. Empirical relationships 4 relating magnitude to subsurface length and magnitude to rupture area are used to assess each j seismic source. In the absence of displacement per event data, these two relationships are given equal weight. Where paleoseismic evidence for slip per event is available, these data

are included in the analysis, with relatively equal weight given to the three approaches. The i

final result of the analysis is a probabilistic distribution of maximum magnitude for each  !

source (Appendix B) that reflects the uncertainties in rupture parameters and judgments about these parameters.

I j Slio Rate - Fault slip mte provides a fundamental constraint on the average rate of seismic

, moment release and earthquake recurrence. Slip rate has the advantage of spanning a longer j time period than the historical record, but there can be uncertainties both in measuring displacement and in determining the ages of geologie units displaced.

i Because our analysis is intended to assess the potendal for future earthquake activity, we typically are most interested in the slip rates during a geologically recent period (e.g., late Quaternary-Holocene). Our assessment of slip rate incorporates uncertainties in the cumulative amount of deformation and in the time period over which the total deformation occurred. We incorporate available information on the amount and age of displaced Quaternary units. Late Quaternary-Holocene slip rates, however, are not well constrained for many of the potential axenarr.arr.rxT A-8  % u. w

seismic sources identified in this study. Regional tectonic and kinematic models that incorporate recent geodetic data (i.e., Feigt and others,1993; Larson,1993) provide additional constraints on the deformation budgets across the study area that can be used to assess slip rates on structures for which there are little or no specific slip rate data.

Earthouake Recurrence - Earthquake recurrence is represented in terms of the rate of seismic activity and the relative frequency of various-magnitude earthquakes. Recurrence rates are j estimated from historical seismicity, from geologic data on rates of fault movement, and from

{ palcoseismic data on the timing of large prehistoric events.

I 4

For large areal sources, historical seismicity is usually d to estimate canhquake recurrence i

rates. In analyzing the earthquake catalog, it is important to translate the data into a common magnitude scale consistent with the magnitude scale used in the ground motion models and to account for completeness in earthquake reporting as a function of time and location. Once these are established, straightforward statistical techniques can be used to estimate earthquake j recurrence parameters (e.g., Weichert,1980). For areal sources, the truncated exponential j recurrence model (Cornell and Van Marke,1969), based on Gutenberg and Richter's (1954) 1

recurrence law,is used. The resulting relationships are then extrapolated out to the maximum

} magnitude for the seismic source to provide recurrence estimates for the full range of f magnitudes considered in the analysis. Uncertainty in the recurrence parameters is estimated i

! statistically from the seismicity data.

I' 4

j For sources defmed as individual faults, the available historical seismicity is usually

{ insufficient to characterize the canhquake recurrence. Geologic data can be used to evaluate

! the rate of fault slip, and this, in turn, can be used to estimate the rate of seismic energy f release, leading to the rate of earthquake recurrence. In addition, paleoseismic studies can lead i

l to dating of large prehistoric events. Predictions of recurrence rates for larger events from fault-specific geologic data have been shown to match well with observed historical rates on l a regional basis (Youngs and Coppersmith,1985; Youngs and others,1992).

1 l

l axmasumer arr.rxr A-9  % is, w l

)

1

The above techniques provide the basis for specifying the recurrence rate of the largest earthouakes c., a source. The recurrence rate for small and intermediate size events is estimated by estrapolating from the largest events using a recurrence model. Initially the expnentis.1 model was used (e.g., Anderson,1979). However, recent advances in underttanding the earthquake generation process indicate that earthquake recurrence on individual faults may not conform to the exponential model developed from regional observations. Instead, individual faults or fault segments may tend to rupture in what have been termed " characteristic" size events at or near the maximum size earthquake (Schwartz and Coppersmith,1984). 'Ihis has led to the development of fault-specific recurrence models such as the characteristic size recurrence model of Youngs and Coppersmith (1985a,1985b).

Recurrence relationships have been developed for fault sources significant to the SONGS site, based on a characteristic earthquake model (Youngs and Coppersmith,1985a,1985b) using combinations of estimated slip rates, maximum magnitude distributions, and a b-value of 0.8 0.15. These recurrence plots are provided in Appendix C and are compared with observed instrumenta! seismicity occuring within narrow corridors surrounding each fault.

2.0 REGIONAL TECTONIC SE'ITING The San Onofre Nuclear Generating Station (SONGS) lies within a broad deforming region between the interiors of the Pacific and North America plates. According to the NUVEL-1 plate motion model (DeMets and others,1990), which incorporates spreading rates in the Gulf of California (DeMets and others,1987) and along the East Pacific Rise and Pacific-Antarctica Rise (DeMets and others,1990), the rate of relative Pacific-North America motion in southern California is approximately 46 1 mm/yr and oriented about N41'W. Relative motion between the plates is characterized by transpressive dextral shear and is accommodated largely by dextral strike slip centered along the San Andreas fault system and faults in the borderlands of southern and Baja California (Fig. A-2), and to a lesser degree, by a component of Basin and Range extension parallel to the plate boundary, extension in the Gulf of California, and coussumer arr.rxr A 10 wy is, im

i contractional structures in the Transverse Ranges and Los Angeles basin region (Zoback and others,1981; Weldon and Humphreys,1986; Argus and Gordon,1988; Stein and Yeats,1989).

3.0 SEISMIC SOURCE CHARACTERIZATION Seismic sources include all structures that have some potential for causing strong ground shaking at the San Onofre Nuclear Generating Station (SONGS). These structums include i faults located in southern California; northem Baja Califomia, Mexico; and in the borderland offshom fmm these agions (Plate 1). We consider sources within a radius of about 100 km of the site (Fig. A-2). Potential rupture scenarios are developed for each fault to identify the range of magnitudes expected to result fmm potential earthquakes on these faults. These rupture scenarios are based on fault segmentation, fault slip rate, historical earthquakes, and depth of seismicity, as discussed in the following sections. The faults are discussed in two sections, offshom and onshore structures. In addition, we consider alternative models for a random background source zone for SONGS that includes the agion between major faults located to the east and to the west of the facility.

3.1 OFFSHORE FAULTS 1

Four major subparallel, northwest-trending, right-lateral strike-slip fault zones occur in the offshore region of southern California and northern Baja California, Mexico. Westward from the coast, these zones are the Newport-Inglewood/ South Coast Offshore Zone of Deformation (SCOZD)/ Rose Canyon, Palos Verdes-Coronado Bank-Agua Blanca, San Diego Trough, and San Clemente-San Isidro fault zones (Plate 1). The San Clemente-San Isidro fault is not exposed onshore; the easternmost three zones are exposed onshore in southern California and

- northern Baja California, Mexico. The location and geometry of faults in the offshom region are known primarily from interpretation of seismic reflection data (e.g., I2gg and Kennedy, 1991; Fischer and Mills,1991, Fischer and Lee,1992, Fischer and others,1992). The seismic reflection data and submarine topography show that Quaternary sediments are displaced along each of these faults (Vedder and others,1986; Clarke and others,1987)~.

aum34carr.arrnr A-11 w r u.

• i Although no large magnitude (greater than magnitude 6) earthquakes are known to have occurred on offshore faults along southern California in historical times, seismicity studies show diffuse linear trends that correspond to the location of faults mapped in the offshore region (Legg,1980). The diffuse nature of epicentral locations in the offsho e regions results, at least in part, from the uncertainty in location for many of these earthquakes. The large location errors result from the poor azimuthal distribution of seismograph stations used to locate offshore earthquakes legg (1980) notes, however, that some earthquakes do not occur along the major northwest-trending faults, and that other faults also may be active. Focal mechanisms for offshore earthquakes have strike-slip, reverse, and normal solutions (legg and Kennedy,1991). These mechanisms, as well as the orientation of and type of displacement on individual fault traces in the offshore region, show that deformation across the continental borderland occurs by dextral strike-slip faulting (Legg and Kennedy,1991). On the basis of association of seismicity and the displacement of Quaternary sediments along these faults, the San Clemente-San Isidro, San Diego Trough, Palos Verdes Coronado Bank, and Newport-Inglewood-Rose Canyon fault zones are considered to be active faults. The average depth of seismicity along offshore faults is poorly known, but is likely similar to the average depth of seismicity associated with other strike-slip faults in California (approximately 12 km; Hill and others,1990). We assume that these faults are high angle, and that the maximum depth of seismicity (95% cutoff of focal depths) defines the downdip rupture width. Accordingly, we assign downdip widths of 12 and 15 km, with equal weight for ruptures along offshore faults. l The rate of activity for most of the offshore faults is not well constrained. Studies of Very Long Baseline Interferometry (VLBI) sites in southern California indicate that significant slip occurs on offshore faults along the southern California borderland. Sauber (1989) evaluated displacement of VLBI sites in southern California and estimates that 5.9 2.9 mm/yr of slip occurs in a N23'W 4' orientation on offshore faults. Ward (1990) also evaluated VLBI sites and estimates that a minimum of 4.5 mm/yr of slip occurs on offshore faults. More recent analysis of Global Positioning System (GPS) and VLBI observations also indicate that the deformation budget must include 5 mm/yr between the offshore islands and the mainland (Feigl and others,1993). On the basis of five years of GPS observations, Larson (1993) axmas4sarr.arr.rrr A-12  % is. m

i i

i t

calculates that San Clemente Island is moving relative to San Diego (Soledad) at the rate of f

5.9 1.8 mm/yr at a direction of N38' 20'W. Catalina Island has a slower rate relative to

San Diego (3.5 1.8 mm/yr), but at a similar dimetion, N28' 20'W. In addition to fault-1 l specific slip rate data compiled from offshore observations and palcoseismic studies of onshore

! extensions of faults, we use these geodetic data to provide regional constraints on the range of likely slip rate values for faults along the southem California borderland.

4 1

4 3.1.1 Newport Inglewood-South Coast Offshore Zone of Deformation-Rose Canyon Fault Zone l

A zone of right-slip faults along the inner coastal margin can be traced from the Santa Monica

!~ Mountains on the northern border of the Los Angeles basin to Baja Califomia, a distance of approximately 200 km. From north to south, this system of faults includes the onshore i Newport-Inglewood fault zone, the South Coast Offshore Zone of Deformation (SCOZD), and i the Rose Canyon fault (Plate 1 Fig. A-3). This zone of faults is thought to accommodate a significant portion of the total slip between the North America and Pacific plates in southern California (Anderson and others,1989). Because more than half of the total length of the Newport-Inglewood-SCOZD-Rose Canyon fault zone lies offshore, and most of the onshore sections of the fault lie within urbanized regions of San Diego, Orange, and los Angeles l counties, assessment of the Quaternary history, activity, and continuity of structures within this zone of faults has been difficult. However, recent paleoseismic investigations along the onshore faults (e.g., Rockwell and others,1992; Freeman and others,1992) and mapping of the geology of the inner margin using new (1990) high-resolution and deep-penetration seismic reflection data (e.g., Southern California Edison,1988; Fischer and Mills,1991; Fischer and others,1992) provide better information for evaluating the location, segmentation, recency, and slip rates of faults within this zone. These studies documented Holocene activity on both onshore and offshore segments within this zone of faults. The occurrence of many small canhquakes along the fault and the 1933 M36.3 earthquake, which occurred the along Newport-Inglewood fault zone south of Long Beach, also demonstrate that faults within this zone are seismically active and capable of producing large-magnitude earthquakes. Southern gunmoner arr.m A-13 wy u. m

.- . - .. = _ -- . .. . -

California Edison (1988) characterized this zone as capable of generating significant shaking 4 at the SONGS site.

I 3.1.1.1 Segmentation and Rupture Lengths  !

l The Newpon-Inglewood fault in the 14s Angeles basin consists of a series of shon, i discontinuous, nonhwest-trending, en echelon right-lateral faults; relatively shallow drag fold anticlines; and subsidiary normal and reverse faults that extend approximately 70 km from the Santa Monica Mountains to offshore Newpon Beach. This narrow belt of deformation is the result of movement along a major through-going, right-slip fault in basement rocks (Harding, 1973; Clarke and others,1987). The geometry, continuity, and recency of activity of the onshore s:ction of the Newpon-Inglewood fault has been reviewed by Bryant (1985),

Hauksson md Gross (1991), and Freeman and others (1992). On the basis of these studies and l mapping by Jennings (1992), we identify two segments in the onshore region. These segments ,

are separa'ed by a 2+ km-wide restraining bend nonh of Long Beach. The nonhern segment extends from the restraining bend nonhward to the end of the mapped traces (approximately 30 km). The southern segment extends southward from the restraining bend to Newpon Beach (approximately 40 km). The March 11, 1933 canhquake (Ms 6.3) ruptured most of this segment, based on the locations of aftershocks shown by Hauksson and Gross (1991). This canhquake produced right-lateral slip of about 50 cm along 25 km (Hauksson and Gross, 1991). A wave-form analysis of teleseismic body wave records from this canhquake (Woodward-Clyde,1979) indicates a focal depth of 10 km for this canhquake. Hauksson (1987) notes that the focal mechanisms show different stress fields operating along the nonh and south segments of the Newport-Inglewood fault. The nonh segment has the minimum principal stress venical (consistent with) reverse faulting, whereas the south segment has the intermediate principal stress venical, indicating strike-slip faulting.

Although the continuity and recency of activity of fault segments in the offshore region is less well known, the locations of individual fault traces and the overall fault zone geomeuy has been identified from seismic reflection studies. As indicated in a series of interpreted cross sections from offshore seismic data, the structure of the fault zone is typically a positive flower coupowrarr.rrr A-14  % is. i9w

i J

stmeture (Fischer and Mills,1991) (Fig. A-4). The linear trend of wrench related fold and

flower structums recognized in the offshore are similar in size and character to the faulted

anticlinal structures along the Newpon Inglewood trend in the Los Angeles basin (Crouch and Bachman,1989). As discussed by Fischer and Mills (1991), several varying interpretations of the offshore zone of faulting were published during the period between 1972 and 1991 (Fig.

l A-5). Southern Califomia Edison (1988) recognized three structural reaches (segments) within  :

] the offshore zone of faulting that exhibited differing degrees of activity and thus differing canhquake potential. They identified these segments as the Newpon-Inglewood zone of l 4

j deformation, the South Coast offshore zone of deformation, and the Rose Canyon fault zone.

i Fischer and Mills (1991) also identify three major segments and several subsegments along the i

t offshore section of the Newpon Inglewood-Rose Canyon fault zone (Fig. A-3). The three

{ offshore segments, the Dana Point, San Onofre, and Oceanside segments of the Newpon-3 Inglewood fault, as shown on Figure 3, approximately correspond to the " South Coast Offshore j Zone of Deformation" as defined by Southern California Edison (1988). We follow Southern 1

! California Edison (1988) in referring to the offshore segments of this fault zone as the South Coast Offshore Zone of Deformation (SCOZD). However, as noted in the following discussions, we incorporated the fault segmentation model presented by Fischer and Mills l (1991) in our assessment of likely rupture scenarios. In the following paragraph we briefly l describe the offshore segments as defined by Fischer and Mills (1991). The offshore zone of l deformation as shown on Plate 1 is based on the mapping of Fischer and Mills (1991).

{

l

The SCOZD-nonh segment, which coincides with the Dana Point segment as originally defined 1

j by Fischer and Mills (1991), extends from Newpon Beach to about 10 km southeast of San Onofre (43 km) (Fig. A-3). As originally defined, the Dana Point segment included the San j Onofre subsegment. More recently, Fischer and others (1992) redefined this subsegment as l

a separate segment (Fig. A-3). The nonhern end of the SCOZD-north segment is located at i a left step and bifurcation of the fault south of Newport Beach. There is an abrupt decrease in seismicity south of this left step. This location is approximately 5 km north of the r.onhern

.{ terminus of the South Coast Offshore Zone of Deformation identified by Southern California

{ Edison (1988). The southern end of this segment is located at a 2-km-wide left step at Las l (otan654carr RFr.TXT A-15  % is,i994 i

- - - - - . - . . . - - - - - - - - - - . - . . . . . . . - - - . ~ . _ _

Pulgas Canyon; this left step is also associated with a cluster of microcanhquakes. Within this segment the fault zone is typically a narrow (>500-m wide), positive flower structure that l

consists of two or three separate splays (Fischer and Mills,1991; Fischer and others,1992) l (Fig. A-2a - A-2c). Greater structural complexity is noted along the southern pan of this 1

segment, where thrust complexes are present west of the fault and a more south-trending reactivated (?) fault (the San Onofre-Oceanside fault) diverges from the main trace of the fault in the vicinity of San Mateo Point. Recency of fault activity appears to decmase southward from Newport Beach, where Holocene faulting and related seafloor bowing are present, to Dana Point, where the most recent activity was about 5,500 yr ago, to San Mateo Point, where Holocene reflectors are not displaced (Fischer and others,1992). South of San Mateo Point, however, the San Onofre segnient, as defined by Fischer and others (1992), exhibits evidence of late Holocene displacement.

The SCOZD-south segment, which coincides with the Oceanside segment, extends south for 32 km from the 2-km stepover at Pulgas Canyon. Between Carlsbad and Encinitas, the southern 13 km of this segment overlaps with the Del Mar segment of the Rose Canyon fault, forming a 2.5- to 4.5-km-wide restraining left step. Along this overstep, the trend of the fault zone changes 25' from N40'W along the SCOZD to N15'W along the Rose Canyon fault (Fischer and Mills,1991). The southern end of the SCOZD-south segment is characterized by a series of horsetail-like, diverging splays. However, within the overstep, the fault tone is a very broad flower structure (Fig. A-2d), suggesting that the SCOZD (considered pan of the Newport Inglewood fault zone by Fischer and Mills) and the Rose Canyon fault are one fault zone (Fischer and Mills,1991; Fischer and others,1992). Activity along the Oceanside (SCOZD-south) segment is early to mid-(?) Holocene, except near the nonhern left step, where late Holocene activity is present (Fischer and Mills,1991).

The Del Mar segment of the Rose Canyon fault lies inboard of the SCOZD-south segment and extends 32 km between Carlsbad and La Jolla. The southem end of this segment is located at a 40 restraining bend near La Jolla (Fischer and Mills,1991). Typically, the fault zone is made up of two to three horsetail splays. Activity along the Rose Canyon fault in the nonhem owasewr.er.m A-16  % is. im

i i i l

I '

i Del Mar segment is late Pleistocene-early Holocene (7) in the Carlsbad area (Fischer and Mills, t

1991). South of this area activity appears to be early (?) Holocene or younger (Fischer and

. others,1983; Webb,1984).

The Mission Bay segment of the Rose Canyon fault extends southward from La Jolla for a distance of 24 to 30 km (Anderson and others,1989; Jennings,1992). Fischer and Mills l (1991) define a slightly shorter (17-km) segment south of La Jolla that they refer to as the San Diego segment.

l In the north part of San Diego Bay, the Rose Canyon fault branches into a number of more l

north-trending fault splays (Kennedy,1975; Kennedy and others,1978: Clarke and others, 1987). This segment of the fault is referred to as the Silver Strand segment of the Rose ,

Canyon fault by Fischer and Mills (1991). According to legg (1985) and Legg and others (1991), the southern Rose Canyon zone may connect to the Pescadero fault near the international border and become part of the Agua Blanca system.

3.1.1.2 Slip Rate Studies of recency of activity and slip rate have been completed for the onshore segment of the Rose Canyon fault near San Diego and for the onshore segments of the Newport and  ;

Inglewood faults north of Newport Beach. Geomorphic analysis and trenching of one strand  :

of the Rose Canyon fault in San Diego indicate that the fault has undergone dominantly right-slip displacement during the Holocene (Lindvall and others,1990; Rockwell and Lindvall,  !

1990; Rockwell and others,1992). Rockwell and others (1992) calculate a minimum Holocene slip rate of 1.07 0.03 mm/yr based on trenching studies of this strand of the Rose Canyon fault. They suggest a maximum slip rate of 2.0 mm/yr based on the geomorphic expression of this fault trace. Further, because other Holocene active strands of the. Rose Canyon fault may exist to the east or west of the fault trace that was trenched, they note that the slip rate across the entire fault zone may be higher. Rockwell and others (1991) suggest that it is reasonable to assume a minimum Holocene slip rate of 1.5 mm/yr for the Rose Canyon fault in the vicinity of San Diego.

j awas4mer nrr.Txr A-17  % is.i m

The bte Pleistocene-Holocene slip rate of the Newpon-Inglewood fault near Los Angeles is not well constrained. A venical slip rate of 0.12 to 0.6 mm/yr is derived fmm estimates of venical displacements along the northern segment, and a vertical slip rate of 0.1 to 1.2 mm/yr 1 is derived from displacements of rock formations along the southern segment (Ziony and I Yerkes,1985). Clark and others (1984) cite a late Quaternary venical slip rate of 0.6 mm/yr. I The horizontal slip rate on the Newpon-Inglewood fault zone is not well defm' ed, with estimates ranging from 0.1 to 6 mm/yr, and preferred estimates typically of approximately 1 mm/yr (Wesnousky,1986). Fischer and Mills (1991) estimate a long-term slip rate of 1.3 to 2.1 mm/yr, based on offset of middle-late Miocene age structural features along the Newpon-Inglewood fault zone in the southern Los Angeles basin. They report that slip rates estimated by Guptill and Heath (1981), Bird and Rosenstock (1984), and themselves fmm offsets determined by Yeats (1973) and Wright (1991) rang fmm 0.4 to 0.8 mm/yr. Freeman

)

and others (1992) note that data from more recent sediments suggest a return period of about 2,000 to 3,000 yr for large Holocene events, which is consistent with the long-term (post-late Miocene and early Pliocene) horizontal slip rate of 0.5 mm/yr esdmated by Guptill and Heath (1981). 'Ihe studies conducted by Woodward-Clyde (1979) also indicate that the venical to horizontal ratio of tectonic displacement is highly variable and is estimated to be about 1:20 where least affected by local folding (Freeman and others,1992). In the offshore region, the Quaternary horizontal slip rate along the SCOZD is between 0.8 and 1.3 mm/yr, based on submarine canyons displaced along the fault zone near La Jolla and Carlsbad (Fischer and Mills,1991).

3.1.1.3 Alternate Source Characterization Models Changes in style, recency, and rate of activity along the Newpon-Inglewood-SCOZD-Rose Canyon fault, as described above, suggest that this fault zone should be subdivided into differ-ent fault segments with different source parameters. The available fault behavioral data, how-ever, do not clearly demonstrate that the major faults (i.e., the onshore Newpon-Inglewood fault zone, the SCOZD, and the Rose Canyon fault zone) behave as independent fault sources.

Therefore,in this analysis we consider two segmentation models that are given equal weight (Figure A-2).

awasarrer.rxT A-18 wy u. p

l l

Model A assumes that the onshore Newpon-Inglewood fault zone and the South Coast Offshore Zone of Deformation are pan of the same fault zone, and that ruptures may propagate l across the small stepover between the two faults. In this model, the more major 2.5- to 4.5 km wide Carlsbad-Encinitas stepover between the southern segment of the SCOZD and the Rose Canyon faults is assumed to act as a barrier to surface ruptures and, therefore, these two faults are considered to be independent fault sources. Recent dynamic modeling studies by Ilarris (1992) suggest that 5-km-wide stepovers, having either releasing or restraining geometries, are effective barriers to spontaneously propagating ruptures. Single- and multiple-segment rupture scenarios are considered in this model, with the highest weights given to single-segment ruptures of 32 km (0.3) and 43 km (0.4). Although the available data for recency along the different segments suggest that rupture of multiple segments has not occurred during the mid- to late Holocene, we give some weight (0.2) to a 75-km rupture scenario that would involve either the entire SCOZD or the southern onshore segment of the Newpon-Inglewood and the nonhern pan of the SCOZD nonh segment (excluding the San Onofre subsegment). We give a small probability (0.1) to a 116-km-rupttue that would include the southern Newpon-Inglewood fault and the entire SCOZD.

In Model A, the Rose Canyon fault zone is treated as an independent fault source.

Paleoseismic data suggest that the slip rate for the Rose Canyon fault may be slightly higher than that of the Newpon-Inglewood-SCOZD. Single- and multiple-segment mptures are considered, with the mast weight (0.5) given to a rupture of the Del Mar segment (34 km) and less weight given to a shoner rupture involving only the San Diego segment (18 km [0.2]) or to a longer rupture involving both segments (52 km [0.3]).

Model B assumes that the onshore segment of the Newpon-Inglewood fault zone is an independent fault source that is characterized by a lower rate of activity. In this model, the SCOZD-Rose Canyon fault zone is considered to be a single fault zone as interpreted by Fischer and Mills (1991) and Fischer and Lee (1992). The rupture scenarios for the SCOZD-Rose Canyon are similar to those described for the Newport Inglewood-SCOZD zone above, except that an intermediate rupture length of 52 km (representing rupture of either the SCOZD-aupamerer.nir A-19 wy is. im

n south and Del Mar segments or the Del Mar and San Diego segments) is given some weight '

(0.1).

For both models, a range of slip rate values and weights (Table A 1 and Figure A-2) is I provided for each fault source based on the estimated slip rates discussed above. The maximum magnitude distributions for the fault sources considered in these models (Appendix B) reflect the range and weights assigned to rupture dimensional data given in Table A-1.

3.1.2 Palos V(rdes Fault Zone j The Palos Verdes fault is a system of right-lateral and right-oblique slip faults extending approximately 115 km from Santa Monica Bay across the Palos Verdes Peninsula, l southeastward onto the San Pedro shelf (Clarke and others,1985) (Plate 1). The fault is onshom for about 15 km, where it forms the eastern boundary of the Palos Verdes uplift. The Palos Verdes fault exhibits evidence for both vertical (up-on-the-south) and horizontal components of marement. Faulting began in late Miocene (7 Ma) during a penod of regional l extension, and as much as 1,800 m of vertical separation is observed on the surface of the Lower Cretaceous Catalina Schist (basement) across the fault zone (Yerkes and others,1965; Nardin and Henyey,1978; Fischer and otim,1987; Wright,1991). Beginning in the middle to late Pliocene, however, the Palos Verdes Hills were uplifted by reverse motion along the Palos Verdes fault (Fischer and others,1987).

Alternative interpretations of the style of faulting on the contemporary tectonic setting currently are being postulated for the Palos Verdes fault. Earlier workers concluded that significant dextral motion had occurred along the fault in addition to the venical separation (Yerkes and others,1965; Nardin and Henyey,1978). Fischer and others (1987) characterize the fault as a northwest-trending, right-reverse oblique fault. They interpret the Palos Verdes uplift as a positive flower structure along the southern fiank of the fault zone. In contrast, Davis and others (1989) model the Palos Verdes fault as a west-dipping back thrust sitting above a large blind thru:t. Sha (1993) interprets the Palos Verdes fault as a west-dipping back thrust off the Compton thrust that tmncates the Palos Verdes fault at a depth of about 5 cxemm.mm A-20  % is, w

4 1

l.

i km. Ward and Valensise (1994) interpret the geometry of the central part of the fault as an

oblique reverse-slip fault associated with a restraining bend geometry along a right-slip fault i

zone. 'Ihey model a 19-km-long faalt dipping 67' at a 6- to 12 km depth beneath the j peninsula. Recent geomorphic and structural studies on the peninsula and on the San Pedro j shelf (Fischer and Lee,1992; Stephenson and others, in review) are more consistent with the l fault being a throughgoing right-slip fault at or near the surface. We adopt this interpretation

! in our analysis. The altemative interpretation of the fault as a reactivated blind thrust fault is 3 addressed in the los Angeles basin Torrance-Wilmington fold trend source zone, as discussed i in a later section.

l l The location, geometry, and activity of the fault traces in both onshore and offshore regions i have been identified from studies of seismic reflection profiles (Nardin and Henyey,1978;

1. egg and Kennedy,1991; Vedder and others,1986; Clarke and others,1987; Fischer and

others,1987; Stephenson and others, in review). The fault can be divided into at least thr

e

{ major segments: a northern segment under Santa Monica Bay, a central segment associated  !

! with the Palos Verdes Hills, and a southern segment under San Pedro Bay. The northern segment is a kcad zone of faulting in Santa Monica Bay that is bounded on the south by the  !

j intersection with the Redonda fault (Clarke and others,1987). Fischer and others (1987) note I that there is a lack of recent seismicity and Holocene reflector offsets in the Santa Monica Bay I

j segment. Convincing evidence for late Holocene to modern faulting is documented on the southern segment of the fault from the Los Angeles Harbor area southeastward to the outer l ,

shelf margin (Fischer and others,1987). Over the San Pedro Shelf, the zone is 1 to 2 km wide

{

and consists of a complex pattern of en echelon and anastomozing fault splays. Surface expression of recent deformation along the onshore central segment of the fault is masked by urban and industrial development and by geologic processes such as landsliding. Interpretation of seismic-reflection and gecmorphic data indicate that the onshore Palos Verdes fault zone consists of at least three major, high-angle fault strands that have accommodated predominantly strike-slip displacement in the late Pleistocene (Stephenson and others, in review).

an.warr.arr.rrr A-21 wy u. m

I l

l 3.1.2.1 Segmentation and Rupture Lengths We consider several rupture scenarios, based on changes in fault geometry and available data on recency and activity. Two recendy active branches of the fault are mapped on either side of Lasuen Knoll at the southern end of the Palos Verdes fault zone (Fischer and others,1987; Fischer and Lee,1992). In order to consider both branches as active sources, we incorporate two equally weighted models into the source model. Model A permits rupttres :o extend ,

1 along the fault splay east of Lasuen Knoll, and Model B accounts for ruptures that terminate along the western splay. In both models the shortest rupture length considered (27 km)is the relatively continuous zone oilate Holocene to modem faulting between the slight restraimng bend in the fault trend near Los Angeles Harbor and the intersection of the fault splays at the I north end of Lasuen Knoll. We give this rupture length scenario a weight of 0.3. A rupture length of 46 km (the length of relatively continuous recent faulting in the offshore region south of the Los Angeles Harbor to Lasuen Knoll) is given a weight of 0.4 in both models. This rupture length also corresponds to a rupture extending from the Redondo Canyon fault intersection to the intersection of the fault splays on the north side of Lasuen Knoll. We also consider rupture of the entire fault zone south of the Redondo Canyon intersection: this scenario, which is given a weight of 0.2, results in a ruptme length of 85 km (Model A) or 70 km (Model B). A small probability (0,1) is given to rupture of the entire zone in each model: 115 km (Model A) and 95 km (Model B). The maximum magnitude distributions for Model A and Model B (Appendix A) reflect the range and weights assigned to postulated rupture dimensions (length and downdip width) given in Table A-1.

3.1.2.2 Slip Rate Recent studies of the Palos Verdes fault zone in both onshore and offshore areas provide new data on the style and rate of recent deformation of this fault. Several studies estimate rates of vertical displacement on the fault, but provide little information on the horizontal component of slip. Estimated vertical rates range from 0.2 to 0.7 mm/p (Clark and others, 1984). The estimated Pleistocene rate of uplift of the Palos Verdes Hills, based on marine terrace studies,is ~0.35 mm/yr (Ponti and Lajoie,1992; Muhs and others,1992a). This rate of uplift is in general agreement with Holocene vertical rates of 0.1 to 0.4 mm/yr estimated for ax n au w ir m m A-22 wy is. im

i i

i l

offshore portions of the Palos Verdes fault in San Pedro Bay (Fischer and others,1987). The horizontal component of slip has recently been directly measured where the fault crosses the Palos Verdes Peninsula. Stephenson and others (in review) show that the ancestral channel '

of the Los Angeles River is incised into an 80 to 120 ka surface and is offset about 300 m. I l

This gives a right-lateral slip rate of 2.5 to 3.8 mm/yr. The approximately 10:1 horizontal to i

vertical ratio of slip inferred from tnese slip rate estimates indicates that the Palos Verdes fault l

1: predominantly a right-slip fault. The estimated rate of horizontal slip is consistent with recent geodetic observations that suggest that 3.5 1.8 mm/yr of slip in a direction of N28' 20'W is occurring in the region between Catalina Island and San Diego (Larson,1993). On the basis of these data, we use estimated slip rate values and weights of 2 mm/yr (0.2),3 mm/yr (0.6), and 4 mm/yr (0.2).

3.1.3 Coronado Bank Fault Zone The Coronado Bank fault zone lies along the southward projection of the Palos Verdes fault zone in the offshore region to the west of the SCOZD-Rose Canyon fault zone (Plate 1). He Coronado and Palos Verdes faults appear to be part of the same regional system of faults, but because of the lack of evidence for continuous late Quaternary activity between the two structures, we treat the two fault zones as independent structures. The Coronado Bank fault zone extends south from near Lasuen Knoll for approximately 215 km to the intersection with the offshore extension of the Agua Blanca fault zone west of Ensenada, Mexico (Legg and Kennedy,1991; Vedder and others,1986; Clarke and others,1987; Anderson and others, 1989).

On the basis of variations in fault geometry and complexity, we identify a number of segments along the Coronado Bank fault zone. Between Lasuen Knoll and the region west of La Jolla, the Coronado fault is mapped along a narrow, linear zone extending approximately 75 km.

As indicated on Jennings (1992), this segment of the fault appears to have been more recently active than segments to the south. The somhern boundary of the segment is located at an extensional fault bend. The Coronado Bank fault zone to the south is characterized by greatly increased fault complexity. From the latitude of La Jolla to south of the international border, mumarrerm A-23  % u. m

l over a distance of about 50 km, the Coronado Bank fault zone consists of a series of subparallel, down-to-the-east fault scarps. South of the international border, the fault zone is ,

I divided into two segments. The northern segment comprises one or two traces mapped over 1 a length of approximately 62 km. A possible southern boundary for this segment occurs wl.ere  !

several faults intersect or merge with the Coronado Bank fault zone. The segment to the south is approximately 30 km long, and comprises multiple subparallel fault traces. The southem boundary of this segment occurs at the intersection with the offshore extension of the Agua Blanca fault zone.

On the basis of regional considerations, the Coronado Bank fault zone is judged to be an active structure within the offshore region, which is transferring slip from the Aqua Blanca fault zone north to the Palos Verdes fault zone. In the absence of detailed information regarding the rate and recency of activity along the fault, we rely on the geometric observations described above to estimate likely rupture lengths. We give the highest weight (0.6) to a rupture scenario involving a 50-km-long segment of the fault (i.e, the more recently active southern part of the northern segment or the more complex zone of faulting west of La Jolla). We also consider multiple segment ruptures of 90 km (0.3) and 125 km (0.1). The maximum magnitude distribution for this fault (Appendix B) reflects the range and weights assigned to estimated rupture length and downdip width as given in Table A-1. We use a range of slip rates comparable to those applied to the Palo Verdes fault zone: 2.0 mm/yr (0.4),3.0 mm/yr (0.4),

and 4 0 mm/yr (0.2). These values are consistent with regional deformation budgets that are constrained by recent geodetic observations and estimated slip rates for the onshore traces of the Agua Blanca fault zone. On the basis of regional considerations,5 to 6 mm/yr of slip on the Agua Blanca fault appears to feed directly into the Coronado Bank fault (about 4 mm/yr) and Descanso-Rose Canyon faults (Rockwell and others,1987).

3.1.4 San Diego Trough Fault The San Diego Trough fault is subparallel to and lies west of the Coronado Liank fault (Plate 1). It extends from the latitude of Encinitas southward for approximately 175 km to the intersection with the Maximinos fault west of Punta Banda, Mexico. The San Diego Trough axmuumwr wrrrr A-24  % is. m

! fault consists of at least two northwest trending segments; the northern segment has a linear i

trace over a distance of approximately 115 km, and the southern segment is discontinuous over

! a distance of appmximately 60 km. The 1986 Ms 5.8 Oceanside earthquake is integreted by j Hauksson and Jones (1988) to have occurred on a thrust fault near a left bend in the San i

Diego Trough fault. legg and Kennedy (1991) state that it is not certain if this earthquake

}

l was associated with a left bend in the predominantly right-slip San Diego Trough fault or with northeast-southwest-directed contraction and reactivation of the Thirtymile Bank escarpment l as a thrust fault. The southern segment of the San Diego Trough fault terminates at the intersection with the offshore extension of the Maximinos fault. The Maximinos fault strikes l

l west-northwest and appears to merge with the Agua Blanca fault onshore to the southeast of j Punta Banda. Because many fault segments in the offshore region appear to be approximately

) 50 to 60 km long, we assign a high probability (0.6) to rupture of a 60-km-long segment. The j entire length of the San Diego Trough fault zone is fairly linear, however. Thus, we assign j small probabilities to rupture of a longer segment (115 km,0.3) and rupture of the entire fault (175 km, 0.1). 'Ihe maximum magnitude distribution (Appendix A) reflects the range of 1

] rupture dimensions derived from combinations of these rupture lengths and widths given in i Table A-1.

]

J i The Maximinos fault has a maximum late Quaternary slip rate of 1.65 0.15 mm/yr (Rockwell and others,1987). Anderson and others (1989) assign a slip rate of 1 mm/yr to the San Diego Trough fault because it appears to receive slip only from the Maximinos fault.

legg (1980) notes that, while the San Diego Trough appears to displace Quaternary deposits, it is relatively aseismic. We assign the highest probability to a slip rate of 1.0 mm/yr (0.6) for the San Diego Trough fault on the basis of the slip rate calculated for the Maximinos fault.

We also assign small probabilities to a lower rate of 0.5 mm/yr (0.2) and a higher rate of 1.5 mm/yr (0.2).

3.1.5 Santa Catalina Escarpment Fault Quatemary activity along the fault bordering the Santa Catalina Escarpment south of Santa Catalina Island is shown by Jennings (1992). Given the northwest orientation of this fault and coums .oarr.arr.rxT A-25 wy is p

the confirmed activity of other nonhwest-trending strike-slip faults to the east and west of this fault, we consider this fault to be active with a probability of 0.5. Rupture lengths of 40 km and 60 km reflect possible segmentation of the fault zone, as indicated by the recency assessments shown on Jennings (1992). The maximum magnitude distribution (Appendix B) reflects the range of rupture dimensions derived from combinations of these rupture lengths and widths given in Table A-1. Assuming that slip along the San Diego Trough may be translated nonhward to the Santa Catalina Escarpment fault, we use the same values for the estimated slip rate: 0.5 mm/yr (0.2),1.0 mm/yr (0.6), and 1.5 mm/yr (0.2).

3.1.6 San Clemente-San Isidro Fault Zone The San Clemente-San Isidro fault zone includes a series of nonhwest-trending, dominantly right-slip faults mapped in the offshore region from the southern Channel Islands (Santa i Barbara-Santa Catalina Islands) southward along the coast of Califomia and Baja California, '

Mexico (Plate 1). The nonhem termination of the San Clemente fault is located near Santa Barbara Island at a 15-km-wide extensional step to the Santa Cruz-Santa Catalina Ridge fault zone (12gg and Kennedy,1991). The San Isidro fault extends along the coast of Baja Califomia, Mexico, south of the intersection with the Santo Tomas fault, indicating a minimum total length of at least 340 km for the San Clemen;e-San Isidro fault zone.

Along the southem Califomia coast, the nonhwest-trending San Clemente fault zone lies along a prominent escarpment in the vicinity of and south of San Clemente Island (Vedder and others,1986). Possible segment boundaries are defined at 2- to 3-km-wide extensional steps near San Clemente Island and at the latitude of the intemat'onal bonier. From nonh to south, I these three segments are approximately 54,76, and 54 km long. To the south, the fault follows a west-nonhwest trend through an eastward compressional bend for approximately 36 km. Slip on this transitional segment between the San Clemente and San Isidro fault zones occurs both on strike-slip and reverse faults (Legg and Kennedy,1991). The San Isidro fault accommodates strike-slip displacement along a nonh-northwest trend. Possible segment boundaries for the San Isidro fault zone are located at 1- to 2-km wide extensional steps west of Ensenada and Punta San Jose, and define segment lengths of approximately 20 to 60 km.

mwau.oarr.arrxxr A-26  % is. im

4 j We assign the highest probabilities to rupture of full segments located offshore of southern i

California (54 km,0.4 and 76 km,0.3); lower probabilities are assigned to possible multiple-segment ruptures (130 km, 0.2 and 184 km, 0.1). 'Ihe maximum magnitude distdbution (Appendix B) reflects the range of these rupture dimensions derived from combinations of rupture lengths and widths given in Table A-1.

The Quaternary history and slip rate of the San Clemente-San Isidro fault zone is poorly constrained because the fault lies entirely offshore. The late Pleistocene-Quaternary slip rate i

is estimated to be 2 4 mm/yr, based on offset submarine fans and canyons (Legg and Kennedy, 1991). Anderson and others (1989) suggest a range of possible slip rates from 0.5 to 5 mm/yr, with a preferred estimate of 5 mm/yr. We assign the highest probability to slip rates of 1.5 mm/yr (0.4) and 3 mm/yr (0.3) because the slip across the other offshore faults appears to accommodate as much as 3.5 to 5.5 mm/yr (Rockwell and others,1987; Rockwell and others, 1991; Rockwell, personal communication,1993) of the estimated 6 mm/yr total slip rate for all faults west of the Elsinore fault (Sauber,1989). These values are also consistent with recent geodetic data that suggest that right-slip faults in the region between the San Clemente and Catalina Islands (i.e., the San Clemente End San Diego Trough faults) may accommodate approximately 2.4 +3.6, -1.1 mm/yr of movement (Larson,1993). We assign smaller probabilities to possible minimum and maximum slip rates of 0.5 (0.2) and 4.0 (0.1) mm/yr.

Legg (1980) notes a rough correlation between epicentral locations of small earthquakes and the San Clemente-San Isidro fault zone. Further, Legg notes that the location of epicenters along the coast of northern Baja California, Mexico, indicates that the San Isidro fault does not terminate at, but continues southward of, the offshore extension of the Agua Blanca fault zone. Because earthquake locations are more poorly constrained with increasing distance offshore, the association of epicenters with the San Clemente-San Isidro fault zone is not well known compared to other southern California borderland faults.

ax.Am merwrnr A-27 %is,im i

i j

i j 3.1.7 San Mateo Thrust Fault i

4 Between San Mateo Point and Oceanside, a 30-km-long fold and thrust belt underlies the l continental slope seaward of the Newpon-Inglewood-SCOZD fault zone (Crouch and 4

, Bachman,1989; Fischer and Mills,1991) (Fig. A-6). Within this defonned belt, fold axes and thrust faults are between 2.5 and 8 km west of, and parallel to, the main trace of the Newpon.

l Inglewood-SCOZD. Crouch and Bachman (1989) and Crouch and Suppe (1993) interpret thrust faults within this belt as rooted into an older regional detachment that has become I i reactivated locally. They consider development of this fold-thrust belt to be the result of 1

nonheast-southwest shortening that is normal to, and decoupled from, the nonhwest-trending i

j strike slip deformation along the Newpon-Inglewood-SCOZD (Crouch and Bachman,1989; l l Crouch and Suppe,1993). Mills and Fischer (1991) note that this " blind" thrust ramp may extend as far south as Encinitas. They state that the thrust ramp may represent a major

! basement discontinuity offshore of San Onofre near the left-stepping break that separates the Dana Point and Oceanside segments of the Newpon-Inglewood fault zone. They describe a

{

{ series of fault-propagation folds and thrusts extending upward from the thmst ramp into the

! overlying Neogene sequence. Venical slip rates on these thrusts are estimated to be 0.08 to 0.5 mm/yr for the Pliocene and 0.01 mm/yr for Quaternary time (Mills and Fischer,1991).

4 1

i Fischer and Mills (1991) separate structures in this zone into an inner thrust-fault-fold l

l complex, which is probably a pan of the flower structure of the Newpon-Inglewood-SCOZD, and an outer thrust-fold complex (Fig. A-7). To the nonhwest, both fault-fold complexes do j not appear to extend north of the San Joaquin high, and to the southeast, the thrusts are apparently terminated or offset by the San Onofre-Oceanside fault (Fig. A-6). Fischer and j Mills (1991) note that the outer thrust complex appears to be cut by the main thrust fault of j the inner complex. They infer that the thrust faults along the inner margin are active, as is  ;

evidenced by their surficial topographic expression and the displacement of Quatemary j i reflectors.

4 l A number of factors need to be considered in assessing the seiscogenic potential of these thrust faults. Strain panitioning, a geologic process by which oblique strain or crustal axm aumarr-arr m A-28  % 35. m

i i

i

{ shonening in the lower crust or lithospheric mantle partitions into nearly pure tangential and i

normal components of strain in the mid-to-upper crust, commonly produces subparallel, coeval strike-slip and dip-slip faults (Tapponnier and others,1989; lettis and Hanson,1991). The occurrence and scale of paninoning is important for correctly identifying and characterizing i

potential seismic sources.12ttis and Hanson (1991) divide the strain panitioning scenarios into two types based on scale (Fig. A-8)

local strain partitioning within <3 to 6 km, and 1 regional strain panitioning within >3 to 6 km. The important distinction is that local strain panitioning occurs within a relatively narrow zone, and thereby, the fault zone would coalesce i

j downdip into a single fault at relatively shallow depths; partitioning at regional scales would

{ imply that the faults would coalesce at greater depths within the crust, or perhaps within the ductile pan of the lithosphere. Observations of canhquake focal depths in most of the westem United States now that large canhquakes nucleate at depths greater than about 5 to 7 km, which is shown on Figure A-8 as the region of large-moment release. As Lettis and Hanson point out, this observation, coupled with the defm' ition of local and regional strain panitioning, would suggest that the two scenarios have different implications to seismic source characterization. In the case of local panitioning, the individual faults in the shallow pan of the crust should not be treated as separate seismic sources, and the sense of slip on the fault at depth should be assessed by adding together the components of slip across the entire fault zone at the surface. In the case of regional panitioning, the individual faults each are separate seismic sources and should be characterized individually.

As shown on Figure A-7, thrust faults within the inner thrust-fold complex probably coalesce with the main trace of the Newpon-Inglewood-SCOZD within the upper 4 to 5 km of the crust, and are therefore considered to be due to local strain panitioning. The outer thrust faults, which appear to represent reactivation of an older detachment fault, likely are truncated by the main inner thrust fault within the upper 3 km of the crust and, therefore, would not extend to seismogenic depth. On the basis of these observations, we do not consider any of the thrust faults within the San Mateo thrust zone to be independent seismic sources that will contribute to the seismic hazard of SONGS.

aummmarrnr A-29 w u. m

3.2 ONSHORE FAULTS Onshore in southern Califomia, most of the dextral displacement between the Pacific and North American plates is accommodated by the San Andreas fault, the San Jacinto fault, and the Elsinore-Laguna Salada faults. All of these faults lie within 100 km of SONGS and are considered as seismic sources in this analysis. The nature and potential of these faults have been addressed in numerous studies and are summarized in the following sections. Several 1

other faults in southern Califomia and northern Baja Califomia, Mexico, also appear to have l

been active during the late Quaternary. Thus, these faults also may have some potential for causing stmng ground shaking in *.he SONGS region and are considered as potential seismic 1 sources.

3.2.1 Elsinore Fault Zone Tibe northwest-trending Elsinore fault extends over 255 km from the Los Angeles basin in southern California southeasterly across the intemational border into Mexico as the Laguna Salada fault (Lamar and Rockwell,1986) (Plate 1). Displacement is dominantly right-lateral strike-slip, although vertical slip has occurred along sections of the Laguna Salada fault zone (Mueller and Rockwell,1984; Millman and Rockwell,1986; Rockwell and Pinault,1986). The

)

entire fault is classified as active, on the basis of results of numerous studies (Jennings,1992).

1 Rockwell (1989) discusses the behavior, geomeny, and extent of individual segments of the Elsinore fault. The segments used in this study are modified slightly from Rockwell (1989) to reflect the results of more recent studies (Rockwell, personal communication,1994). The segments are, fmm northwest to southeast: Glen Ivy (35 km), Wildomar (Temescal) (55 km),

Julian (80 km), Coyote Mountains (60 km), and Laguna Salada (38 km). The following descriptions note only the primary geometric characteristics of the segment boundaries; additional factors, such as changes in geomorphic expression of the fault, recency of faulting, continuity of fault traces, and total bedrock displacement, are discussed by Rockwell (1989).

The northern boundary of the Glen Ivy segment is located at a 3- to 4-km-wide releasing bend slightly south of the junction of the Whittier and Chino faults. The boundary between the punmarr.arr.TxT A-30 wy is, im

-- - . . - - - - , - - - - - _ . - ~ - - . _ ..--- --- -

1 i

}

i I

Glen Ivy and Wildomar segments is located at a 3- to 4-km wide releasing step at Lake Elsinore. The boundary between the Wildomar and Julian segments is located at a 2- to 3-km-j wide releasing step south of Agua Tibia Mountain. The boundary between the Julian and  !

4 1

i Coyote Mountain segments is located at a 4- to 5-km-wide restraining bend between the Laguna and Vallecito mountains. The boundary between the Coyote Mountain and Laguna Salada segments is located at major restraining bend /stepover south of Coyote Mountain. l i

These fault segments range from 35 to 80 km long. )

l Palcoseismic investigations along the Glen Ivy segment document minimum displacements of j 0.35 and 0.5 meters for events that ruptured to the surface (Brake and Rockwell,1987). l Trenching investigations at Glen Ivy marsh (Rockwell and others,1986) show that five, and

! probably six, surface-rupture events have occurred since about 1060, yielding an average j recurrence interval of 150 to 200 yr. Along the Coyote Mountain segment north of the j international border, palcoseismic investigations suggest repeated late Holocene surface-faulting events with maximum displacements of 2+ meters per event (Rockwell,1990). The results of 3 these studies suggest that at least two characteristic earthquakes occur along the Elsinore fault;

these events occur during rupture along 35 km and 55- to 60-km long segments. The rupture l scenarios considered for this study are based on rupture of the individual segments closest to l the site

the Julian segment (80 km), the Wildomar (Temescal) (50 km), and the Glen Ivy (35

] km). Because these segments (35,50, and 80 km) are equally well defined, we assign equal-4 probability (0.3) to these three rupture lengths. Finally we assign a small probability (0.1) to a potential rupture of all segments (130 km).

i i

The average depth of the seismogenic zone varies from approximately 10 km along the northern and southern reaches of the fault (Glen Ivy, Wildomar, and Coyote Mountain segments) to 16 km along the central reach of the fault (Wildomar and Julian segments) (Hill and others,1990). On the basis of the average width of the seismogenic zone along the entire Elsinore fault, we assume the following downdip rupture widths and probabilities: 12 km (0.4),

14 km (0.4), and 16 km (0.2). The maximum magnitude distribution (Appendix B) reflects coupa narr.an.rrr A-31  % n.

• _ _ _ .._ . . . .__...___-__._._.__.____._.___._____._m l

1 L 1

the range of these rupture dimensions derived from combinations of rupture lengths and widths i

j given in Table A-1.

t

, Recent studies at several sites suggest a late Quaternary slip rate of about 5 to 6 mm/yr i

(Pinault and Rockwell,1984; Millman and Rockwell,1986; Vaughan,1987) for the Glen Ivy, '

Julian, and Coyote Mountain segments of the Elsinore fault zone. For the slip tate on the Glen Ivy segment, Millman and Rockwell (1986) calculate a range of 2.6 to 9.3 mm/yr, with a best estimate of 5.3 to 5.9 mm/yr for displaced alluvial fan deposits. Vaughn and Rockwell (1986) l estimate a slip rate of 3 to 6 mm/yr over the past 600 ka for the southem end of the Wildomar  !

j segmeta, based on displaced palcosoils developed on alluvial fan deposits. Rockwell (personal i j  !'

communication,1993) estimates a slip rate of 4.2 4.9, -0.5 mm/yr for the past 2,400 yr based I on displacement of a buried channel on the Wildomar segment. On the basis of this range of

calculated slip rates, we assign the following slip rates and probabilities for the Elsinore fault

3.0 mm/yr (0.2), 5.0 mm/yr (0.6), and 7.0 mm/yr (0.2).

)

. 3.2.2 Whittier Fault j The Whittier fault extends 32 km from the Whittier Narrows on the north to the Santa Ana v

j River on the south. The southem end of the Whittier fault is defined by a releasing bend to j the Glen Ivy segment of the Elsinore fault (Rockwell,1989). Near Yorba Linda, Rockwell and 3 others (1992) report a minimum dextral slip rate of 2.5 to 3.0 mm/yr, based on laterally i j displaced channels incised into a dated alluvial fan. Three-dimensional trenching by Gath and i

others (1992) in the Olinda oil field also established a minimum slip rate for the fault zone. j i

j They measure a rate of about I to 1.5 mm/yr on one of four strands of the fault. Gath and l others (1992) infer a minimum rate for the fault zone of about 2 mm/yr.

I j The dates of the past two events and the amount of slip in the most recent event have been i

1 established at Olinda oil field by three-dimensional trenching (Patterson and Rockwell,1993).

j The timing of the most recent and penultimate events is estimated at between 1.4 and 2.2 ka

} and about 3 to 3.1 ka, respectively. At least 1.9 m of dextral slip was produced in the most a

j recent event, based on direct measurement of a laterally displaced channel. This represents i

axmumarr.nrr.m A-32  % is, im I

l

i I j a minimum value for slip per event because additional slip may have occurred on another

trace.

i l We model the Whittier fault as an active fault and consider rupture scenarios involving the j entire 32-km length of the fault (0.5) and 15 km (0.5). We use the palcoseismicity data to estimate a range of slip rate values as follows: 1.5 mm/yr (0.2), 2.5 mm/yr (0.6), and 3.5

{ mm/yr (0.2). The maximum magnitude distribution for the Whittier fault (Appendix B) l includes consideration of the paleoseismic evidence for slip per event outlined in Table A-1.

k

3.2.3 Aguanga-Agus Tibia Earthquake Valley Faults The Aguanga, Agua Tibia, and Eanhquake Valley faults lie approximately 8 to 10 km east of j the Julian segment of the Elsinore fault (Plate 1). On the basis of the geometry of the fsults l and the observed bedrock displacements, it appears that right-lateral slip on the Elsinore fault l zone was accommodated by these faults during the Teniary, but has been accommodated j primarily by the main Elsinora hult (Julian segment) during the Quaternary (Rogers,1965; j Rockwell, personal communication, 1993). The total length of the Eanhquake i Valley-Aguanga-Agua Tibia faults is approximately 84 km.

i 1

I j The Aguanga fault trends nonhwest and is approximately 10 km long. Quaternary deposits are displaced by the fault, but the age of these deposits is uncenain (Jennings,1992). The

! Agua Tibia fault includes several northwest-trending faults mapped by Rogers (1965) and

{ Jennings (1992). These faults are not known to displace Quaternary deposits. The Agua Tibia f faults are mapped over a minimum length of 14 km; the Aguanga-Agua Tibia faults are

[ mapped over a total length of 27 km.

The Eanhquake Valley fault is active and is dominantly right-slip, on the basis of the presence of deflected drainages, shutter ridges, hillside benches, vegetation lineaments, and scarps in alluvium (Magistrale and Rockwell,1990). The surface trace of the fault is approximately 42 km long. Seismicity associated with the fault extends nonhwestward for another 18 km (Magistrale and Rockwell,1990; Rockwell, personal communication,1993). Evidence for

~

m u m marr.arr nr A-33 w is,i m

Holocene activity is present along the northernmost 23 km of the surface trace of the Eanhquake Valley fault, but has not been observed along the fault to the southeast (Jennings, 1992). On the basis of the geomorphic expression of Quaternary faulting, we assign higher probabilities for rupture lengths of 10 km (0.3) and 23 km (0.5), and a lower probability (0.2) for rupture of the entire Eanhquake Valley fault or for rupture of the nonhem subsurface segment and the nonhern surface segment (42 km). 'Ihe maximum depth of seismicity along the adjacent segment of the Elsinore fault is about 12 to 17 km (Hill and others,1990). We assume similar depths for potential rupture along the Aguanga, Agua Tibia, and Earthquake Valley faults, with a higher probability (0.6) for a maximum width of 12 km and a lower probability (0.4) for a maximum width of 17 km. The maximum magnitude distribution (Appendix B) reflects the range of these rupture dimensions derived from combinations of rupture lengths and widths given in Table A-1.

No estimate of the slip rate has been published for the Eanhquake Valley fault. Rockwell (personal communication,1993) suggests that the fault accommodates a ponion of the total slip along the Elsinore fault zone (on the order of 1 mm/yr). On the basis of the total slip for the Elsinore fault zone (4.2 mm/yr along the Wildomar segment), we estimate potential slip rates and the respective probabilities for the Eanhquake Valley fault to be 0.5 mm/yr (0.3),1.0 mm/yr (0.4), and 2.0 mm/yr (0.3).

3.2.4 San Jacinto Fault Zone The San Jacinto fault zone comprises a series of northwest-trending, right-lateral strike-slip faults mapped from Cajon Pass southward through the Salton Trough to El Centro (Plate 1).

The San Jacinto fault zone includes, from nonh to south, the Claremont, Casa Loma, Clark, Coyote Creek, and Superstition Mountain faults. The entire fault zone, except for the Superstition Mountain fault, shows evidence for Holocene activity (Jennings,1992). The northward termination of the San Jacinto fault zone occurs where the Claremont fault intersects the San Andreas fault near Cajon Pass. The Claremont fault is 78 km long. The southern end of the Claremont fault overlaps the Casa Loma fault and the nonhern pan of the Clark fault to form a 2- to 4-km-wide, approximately 30-km-long extensional stepover. The Clark fault ax.Awaarr arr.TxT A-34  % u. im

- - - . - . - - . - . -- . .- ~ . - - - . _ . - ..- - - - - -

6

- extends for approximately 90 km beyond the stepover region. The southern end of the Clark i fault also overlaps with the nonhem section of the Coyote Creek fault. The fauhs are i

! subparallel for approximately 44 km. The overlap region between these faults forms an a

extensional step that is approximately 2 to 9 km wide. He overlap region is structurally complex, and it is not clear how much slip transfers between the Coyote Creek and Claremont l

faults. It is possible that some slip may transfer between the faults along a series of l northeast-trending cross faults.

4 j The Coyote Creek fault is approximately 76 km long. The southern half of the Coyote Creek j fault ruptured during the M 6.6 Borrego Mountain earthquake on April 9,1968. A 31-km rupture occurred along the surface trace of the fault, the aftershocks delineate a rupture zone that was 40 km long, and the maximum surface displacement was abou; 0.4 m (Wells and Coppersmith,1994). The northern end of the Superstition Mountain fault is located less than 2 km east of the southern end of the Coyote Creek fault, across a restraining step. The

~ '

Superstition Mountain fault is approximately 27 km long, and the most recent displacement on the fault appears to be late Quaternary. The Superstition Hills fault is located approximately 4 to 7 km east of and subparallel to the Superstition Mountain fault. In the 1987 M 6.6 Superstition Hills earthquake,27 km of surface rupture occurred along this fault, averaging about 0.5 m, with a maximum surface slip of 0.9 m. The southern end of the San Jacinto fault zone is located at the stepover between the Superstition Mountain-Superstition Hills and Imperial faults.

l On the basis of geologic and seismologic data, Sanders (1989) identified 20 principal fault segments, ranging in length from 7 to 35 km, comprising the 242-km long San Jacinto faul.

zone. Anderson and others (1989) identified only nine segments, ranging in length from 17 to 55 km. He Working Group on California Earthquake Probabilities (WGCEP,1988) divided the fault into five segments using information on fault geometry, historical seismicity, and slip rate. The segments were named, from north to south, San Bernardino Valley, San Jacinto Valley, Anza, Borrego Mountain, and Superstition Hill.

aumuonn annr A-35  % is. im 1

Our segmentation model is based on the WGCEP (1988) model with some modifications. We select three possible rupture scenarios for earthquakes on the central part of the San Jacinto fault zone closest to SONGS. The shortest length (40 km) is consistent with the approximate length of the San Jacinto Valley segment in the stepover region between the Claremont and Casa I.oma faults. This length is also consistent with the rupture length of the 1968 earthquake on the Coyote Creek fault. A rupture length of 90 km is based on the potential rupture of the Anza segment that includes the entire length of the Clark fau?t. We assign probabilities of 0.4 and 0.5, respectively, to these rupture lengths. We allow a small probability (0.1) for a rupture scenario involving both segments (130 km).

The average width of the seismogenic zone varies considerably along the San Jacinto fault zone. The average width is approximately 15 to 17 km along the Claremont fault strand, approximately 15 km along the Clark fault, and approximately 10 km along the Coyote Creek fault (Hill and others,1990). The average downdip widths and probabilities selected for the San Jacinto fault (13 km (0.5),15 km (0.4), and 17 km (0.1)) are based largely on the widths of the seismogenic zones along the segments closest to SONGS (i.e., the Claremont and Clark fault strands). The maximum magnitude distribution (Appendix B) reflects the range of these rupture dimensions derived from combinations of rupture lengths and widths given in Table A-1.

Sharp (1981) calculated a minimum mid-Quatemary slip rate of 8 to 12 mm/yr for displaced gravels along the central part of the Clark fault near Anza. Rockwell and others (1990) calculated the following slip rates (mm/yr) for the same section of the Clark fault: 9.2 2  ;

since 9.5 ka; 11 +9, -5 since 14 ka; 12 +9, -5 since 17 ka; 13 +10, -6 since 50 ka; and 12 minimum since 700 ka. These slip mte estimates were based on ponded sediments and displaced fan deposits (Merifield and others,1987; Rockwell and others,1990). Because the  ;

Clark fault is geomorphically well defined and all slip occurs along a single fault trace at I Anza, the slip estimates at this location are considered to be the most reliable for the San ,

J Jacinto fault zone (Rockwell, personal communication,1993). l ll axmunmmm A-36 ur ns. sm

4 l

1 l

1 Sharp (1981) also evaluated slip rates for the Coyote Creek fault from displaced shorelines of I Lake Cahuilla and displaced stream channels. He calculated a slip rate of 2.8 to 5.0 mm/yr for the past 510 yr, and a slip rate of 1 to 2 mm/yr for the period from 400 to 6000 before present. Wesnousky and others (1989) calculated a minimum slip rate of 1.7 to 3.2 mm/yr for the past 1,931 yr for one fault strand on the Claremont segment. In the studies on the Coyote Creek and Claremont faults, the calculated slip rates are considered minimum rates because additional slip may occur on additional fault strands. Thus, because the slip rates for the Coyote Creek and Clammont faults am minimum rates and because the Clark fault lies the closest to SONGS, the slip rate for the Clark fault is considered appropriate for our analysis.

On the basis of the slip rate estimates for the Clark fault from Rockwell and others (1990), we  !

select the following rates and probabilities: 7 mm/yr (0.1),9 mm/yr (0.5), and 12 mm/yr (0.4).

1 Palcoseismic investigations along the Superstition Hills fault indicate that during the past 300 yr, the average interval between large surface-faulting events has been between about 150 and 300 yr.

i 3.2.5 San Andreas Fault Zone The San Andreas fault zone is the dominant tectonic stru:ture in California. Displacement on the San Andreas fault zone is nearly pure right-lateral strike-slip, and the fault is active, as shown by historical surface ruptures, microseismicity and earthquake focal mechanisms, and surface geomorphology (Jennings,1992). For this analysis we consider only the southern part j of the fault zone, consisting of the following three segments, as defined by the 1988 Working Group on California Earthquake Probabilities (WGCEP,1988): the Coachella Valley segment, l

the San Bernardino Mountains segment, and the Mcjave segment.

The Coachella Valley segment includes the southernmost section of the San Andreas fault l

zone; this segment is approximately 115 km long. The southern boundary of the Coachella segment is located at the junction with the Brawley seismic zone near the Salton Sea. The northern boundary of the segment is located north of Palm Springs near San Gorgonio Pass at an extensional bend (where the fault rotates to a west northwest trend). The San Bernardino axesearrmm A-37 ur is, im l

- . . . _ ~ - . _. - . _ _ _ _ - ... - . - . . . _ .-. - . . .

i i

3 i

Mountains segment is an approximately 85-km-long, structurally complex zone that lies between the Coachella Valley and Mojave segments. The Mojave segment extends 135 km northwestward from the southern end of the 1857 rupture a few km north of Cajon Creek to

] Three Points. The northern margin of this segment is a transitional boundary defined by an j merease m slip from 3 to 7 m in the 1857 event.

i We consider single- and multi-segment rupture scenarios for the southern San Andreas fault j zone that are consistent with historical and paleoseismic observations. The largest historical earthquakes on the San Andreas fault, the 1857 Fort Tejon and the 1906 San Francisco earthquakes, ruptured over a distance of 297 and 432 km, respectively (Wells and Coppersmith,1994). Thus, we conclude that a single, ~335-km-long rupture involving all three

of the southern San Andreas fault segments is possible. Paleoseismic data at several sites j along the southern part of the fault zone suggest, however, that ruptures involving the San

) Bernardino Mountains and Coachella Valley segments, individually and together, have occurred l more frequently. On the basis of these data, we assign the highest probability (0.6) to a potential 200-km-long rupture involving both the San Bernardino Mountains and Coachella i Valley segments, and lower probabUities to potential rupture of the Coachella Valley segment j (115 km) (0.1) and the combined southem three segments, (335 km) (0.3).

Rupture widths for the southern San Andreas fault zone are based on thicknesses of the seismogenic crust inferred from seismicity data. The Coachella segment of the San Andreas

! fault zone is generally aseismic. Sparse seismicity has occurred at depths of 3 to 10 km (Hill and others,1990). The depth of the seismogenic zone is poorly defined along this segment of the fault. The average widths of the seismogenic zone along the San Bernardino and 1

Mojave segments appear to be about 12 km; thus, we assume that the average width of the ,

seismogenic zone along the Coachella Valley segment also may be as wide as 12 km. On this I

basis, we assign a higher probability to a rupture width of 12 km (0.6) and a lower probability to a rupture width of 10 km (0.4).

<owasu>nrr.arrnr A-38  % is. m

The maximum magnitude distribution developed for the southern San Andreas fault zone  !

l l (Appendix B) reflects the estimated values and weights assigned to the potential rupture l dimensions (length and width) as summarized in Table A-1. Recent geologic and geophysical measurements suggest that the historically dormant southern segment of the San Andreas fault is cunently locked and slips primarily during great earthquakes (Raleigh and others,1982; Lindh,1983; Sykes and Nishenko,1984; and Sieh and Williams,1990). If the rate of strain accumulation along this segment has been srcady during the past three centuries, an average of 6 to 8 m of surficial fault slip could be expected during a future large earthquake (Sieh and Williams,1990). For a potential average surface displacement of 6 to 8 m, the expected magnitude is M 7.7 to 7.8, based on relationships between average surface displacement and magnitude (Wells and Coppersmith,1994). Williams (1989) identifies single-event displacements of about 2 to 3 m at Salt Creek on the Coachella Valley segment. Based on the previously mentioned relationships, these displacements att consistent with earthquakes of M 7.0 to 7.2. The estimated magnitudes (M 7.0 to 7.8) for the observed palcoseismic slip (2 to 3 m) and the potential accumulated slip (6 to 8 m) provide an additional assessment of the magnitude range for potential earthquakes on the Coachella Valley segment.

l i

The southern San Andreas has a relatively high probability for a major earthquake in the near future, based on statistical analyses of the fault's paleoseismic record (Sykes and Nishenko, i

1984; Wesnousky,1986). Paleoseismic trenching investigations at sites along the San Andreas fault in the Carrizo Plain to the Salton Trough (Sieh and Jahns,1984; Weldon and Sieh,1985; and Sieh,1986) demonstrate tha: large earthquakes recur every 150 to 300 yr, depending on the proximity of the site to segment boundaries. Although the southemmost 200 km of the San Andreas fault has been dormant during the historical period, studies of the prehistoric

earthquake history of the fault along this segment of the fault at the Indio site led Sieh (1986) to conclude that this segment of the fault generates a large earthquake at least once every 200 to 300 yr. The last earthquake at the k.dio site occurred about 300 yr ago (Sieh,1986).

Weldon and Sieh (1985) estimate a recurrence time of about 250 yr for large earthquakes along the San Andreas fault at Cajon Pass, with the last earthquake possibly occurring during the early 18th century (250 yr ago). More extensive and precise earthqtrake event chronologies axe 54mner arrxxT A-39  % is. m

i 1

are being developed at several sites along the southem San Andreas fault zone. The more specific recurrence history available at these sites is being used to develop recurrence and segmentation models (WGSCEP,in progress). The earthquake recurrence rates obtained from fault slip rates (Table A-1) are consistent with these estimated repeat times.

Palcoseismic studies show that during the Holocene, the San Andreas fault has accommodated about 36 mm/yr of slip in the Carrizo Plain (Sieh and Jahns,1984), about 24 mm/yr at Cajon Pass (Weldon and Sieh,1985), and about 30 mm/yr in the Salton Trough (Sieh,1986). The latter estimate by Sieh (1986) is for the Coachella Valley segment; the other rates were determined to the north along the San Bernardino, Mojave, and Carrizo segments. Wani (1990) evaluated VLBI measurements and calculated that the slip rate of the San Andreas fault in southern California is 251.1 mm/yr at N52 W. On the basis of this range of geologic and geodetic slip rates, we assign the following rates and probabilities for the southern segments  !

of the fault zone: 25 mm/yr (0.6) and 30 mm/yr (0.4).

3.2.6 Malibu Coast Santa Monica Fault Zone The Malibu Coast and Santa Monica faults are major elements of an east-trending zone of deformation that marks the southern front of the Transverse Ranges along the south flank of I the Santa Monica and San Gabriel Mountains. Other elements of the zone include the Raymond Hill, Hollywood, and Cucamonga faults. The zone extends east onshore for about 100 km, and all faults are nonh-dipping reverse left-oblique faults (Yerkes and 1.ce,1987).

Composite focal mechanisms show that displacement is reverse left-lateral oblique along these faults (Real,1987). Recent palcoseismic investigations along the Santa Monica fault west of the Newport-Inglewood fault and along the Hollywood fault to the east (Dolan and others, 1993) suggest that there is a significant component of lateral displacement along thets faults.

Ongoing studies of subsurface data show that the Quaternary dip slip rate is very low (R. Yeats, Oregon State University, personal communication, May 1994). We, therefore, model these faults as active north-dipping reverse left-lateral oblique-slip faults. There appears to be a fundamental structural change near the intersection of this system of faults and the Newport-Inglewood fault zone (J. Dolan, Caltech, personal communication, May 1994). On coumumrr.arrur A-40 my is, im

i j the basis of these data and the left stepover geometry between the Santa Monica and 1

]

Hollywood faults, which is accommodated by the northern extension of the Newport-l Inglewood fault zone (Hummon and others,1994), we treat structures east and west of the Newport-Inglewood fault zone as independent structures. l l West of the intersection of this fault system with the shoreline at Santa Monica, two or more subparallel elements of the zone are incognized. He northern fault, the Malibu Coast fault,

is designated as a late Quaternary fault that can be traced onshore for a distance of L approximately 25 km. The southern fault, the Santa Monica fault, is aligned. with a i

1 north-dipping reverse fault at the mouth of Potrero Canyon that cuts upper Pleistocene terrace

/

j deposits (Yerkes and Lee,1987; McGill,1989). Other faults in this area locally are associated j with groundwater barriers and topographic scarps in Pleistocene deposits (Hill and others, 1977). Offshore, the Santa Monica fault is shown as a late Quaternary fault that extends an j additional 20 km to the west (Jennings,1992).

I 4

1 The total length of the Malibu Coast-Santa Monica fault zone is approximately 54 km. Based

{ on the apparent variation in recency along this fault zone, as shown by Jennings (1992), we s

consider rupture scenarios of 25 and 30 km with equal weight. Rese rupture lengths represent

the approximate lengths of the individual mapped faults in this zone. The maximum t

j magnitude distribution (Appendix B) reflects the range of rupture dimensions derived from l combinations of these rupture lengths and widths given in Table A-1. Ziony and Yerkes 1

(1985) give a range of 0.04 to 0.4 mm/yr for the slip rate for faults along the northern margin

of the los Angeles basin. The amount of left slip on these faults is not known, but is likely j to be significantly girater than the rates cited by Ziony and Yerkes (1985) (J. Dolan, Caltech, 4

j personal communication, May 1994). Given the uncertainties in slip rate data for this fault l zone, we use a range of values: 0.04 mm/yr (0.2),0.4 mm/yr (0.4), and 2.0 mm/yr (0.4).

1 i

i 3.2.7 Hollywood Raymond Fault Zone 1

j Recent investigations along the Hollywood fault indicate that the fault (1) has experienced j significant post-17,000-year-old displacement, (2) may exhibit a strong component of left-4 l axesearrer.TxT A-41  % is.i m I

4 i

1 i

lateral strike-slip displacement, and (3) appears not to have experienced a ground-rupturing

) earthquake in at least several thousand years, indicating long recurrence intervals (Dolan and others,1993). These studies indicate that recent deformation along the Hollywood fault is  ;

I concentrated in a narrow zone just south of, and parallel to, the mountain front. Two to five l

steeply dipping (60'N) fault strands displace the alluvium-granodiorite contact at least 60 m i i

down to the south.

i l The Raymond fault is predominantly a left-lateral strike slip fault (Jones and othen,1990), l l although geomorphic expression reflects a strong component of convergence during the j Quaternary (Crook and others,1987; Real,1987). The Raymond fault shows well-defined  ;

evidence of a late Quaternary history of repeated fault movements. Palcoseismic trenching l j

l investigations provided evidence of at least five, and possibly three more, seismic events along this fault in the past 36 ka (Crook and others,1987). On the basis of these studies, Crook and

]_

others (1987) infer an average recurrence interval of about 3,000 yr, with an average vertical displacement of 0.4 m per event.

I I

We characterize the Hollywood and Raymond faults as active (probability of 1.0),

north-dipping, reverse left-lateral oblique faults. The Hollywood-Raymond faults extend for a distance of approximately 40 km (Plate 1). On the basis of the continuity of traces, as shown by Jennings (1992), we give the highest weight (0.6) to a 20-km-long rupture scenario (0.6) and less weight to a rupture involving both the Hollywood and Raymond faults (0.4).

The maximum magnitude distribution (Appendix B) reflects the range of nipture dimensions derived from combinations of these rupture lengths and widths given in Table A-1. The range in vrbes of slip rate assigned to the Malibu Coast and Santa Monica faults is also used for the Hollywood-Raymond Hills fault zone.

3.2.8 Sierra Madre Fault Zone The 85-km-long Sierra Madre fault zone consists of a series of arcuate 13- to 20-km-long northeast-dipping, reverse left-oblique fault segments (Proctor and others,1972; Crook and others,1987). The most active segment of this fault zone is the westernmost section, adjacent ax.w3earr.arrnr A-42  % is. im

1 I

l l

to the faults that broke during the 1971 San Fernando earthquake (Crook and others,1987).

The distribution of aftershocks from the 1971 San Femando and the 1991 Sierra Madre canhquakes shows that the downdip width of these segments is about 23 to 28 km (Hauksson, 1992). The age of activity, as indicated by the occurrence of Holocene faulting, decreases toward the east. Bonilla (1973) estimates the repeat time for earthquakes along the San Fernando segment to be 200 yr. Evidence of late Pleistocene, but no Holocene, faulting is observed along the central and eastern segments of the fault zone. Crook and others (1987) infer from these observations that the recurrence interval between major earthquakes on these segments is longer than about 5,000 yr. They also estimate that the local magnitude (ML) of the largest credible earthquake that could occur on the Sierra Madre fault zone is 7, based on the apparent subdivision of the fault zone into separate arcuate segments about 15 km long.

l We characterize the Sierra Madre fault zone as active with a probability of 1.0. We consider single- and multi-segment rupture scenarios to estimate likely rupture lengths. Given the apparent segmentation of the fault, we give the most weight to a single 17-km (0.3) or a 30-km two-segment (0.4) rupture. We give less weight to a multi-segment rupture involving three segments (45 km, 0.2 weight) or the entire fault zone (85 km, 0.1 weight). The maximum magnitude distribution (Appendix B) reflects the range of estimated rupture lengths and widths given in Table A-1. We use a range of estimated slip rate values as follows: 0.5 mm/yr (0.3), 3.0 mm/yr (0.4), and 4.0 mm/yr (0.3).

3.2.9 Cucamonga Fault Zone The Cucamonga fault is a major, east-striking nonh-dipping thrust fault that extends for 25 km along the south front of the eastern San Gabriel Mountains. The Cucamonga fault appears to be accommodating some percentage of the convergence between the Transverse Ranges and Peninsular Ranges that occurs along the San Jacinto fault. Faulting within the Cucamonga fault zone has recurred episodically during the Quaternary (Matti and others,1992). Young faulting is expressed as a series of fault scarps in late Pleistocene-Holocene alluvial fans at the mountain front. The most recent faulting may have occuned mainly in the eastern 15 km of the fault zone and not throughout its entire 25-km length (Matti and others,1992). On the axwarrm.rrr A-43  % is. im

basis of trenching and fault scarp profiling, Morton and Matti (1987) measured about 36 m of net surface displacement in the past 11 ka to 13 ka and calculate a slip rate for the fault of 4.5 to 5.5 mm/yr.  ;

We characterize the Cucamonga fault as active with a probability of 1.0. Given the evidence for more recent activity on the eastern 15 km of the fault zone, we consider two rupture scenarios: 15 km (0.6) and 25 km (the entire length of the fault) (0.4). The maximum magnitude distribution (Appendix B) reflects the range of rupture dimensions derived from combinations of these rupture lengths and widths given in Table A-1. On the basis of the recent paleoseismic data, we use a range of slip rate values and weights as follows: 4.0 mm/yr (0.2), 5.0 mm/yr (0.6), and 6.0 mm/yr (0.2).

3.2.10 Peralta Hills-Norwalk Fault The north-dipping Peralta Hills fault, and a possible northward extension of the fault shown ,

as the Norwalk fault on Plate 1, appears to be an active thrust fault (s) (J. Dolan, Caltech, written communication, May 1994; E. Gath,12ighton and Associates, personal communication, June 1994). Detailed investigations to assess the Pleistocene and Holocene behavior of the Peralta Hills fault have not been conducted. Preliminary investigations of a strike slip fault that may be a Par fault on the hanging wall of the Peralta Hills fault show evidence of recurrent Holocene movement (E. Gath, personal communication, June 1994). The Norwalk fault is inferred to be the surface expression of a recently active fault on the basis of an abmpt 6-m high scarp that projects northward from the Peralta Hills fault.

Given the evidence for possible Holocer'.: activity, we characterize the Peralta Hills fault as active with a high probability (0.8). Tne total length of the Peralta Hills and Norwalk faults is 25 km. Based on the discontinuity and lengths of the mapped surface fault (s), we consider rupture lengths of 10 km (0.6),15 km (0.2), and 25 km (0.2). The fault is modeled as a 60 NE-dipping fault. The maximum magnitude distribution (Appendix B) reflects the postulated rupture dimensions based on combinations of rupture lengths and widths given in Table A-1. The slip rate distribution is relatively broad because of the uncertainties in the slip munmarrer.TxT A-44  % u. w

l rate for this fault. The following range of slip rate values is used: 0.1 mm/yr (0.1),0.2 mm/yr i (0.3),0.5 mm/yr (0.5), and 1.5 mm/yr (0.1).

3.2.11 Temescal Fault The Temescal fault includes a series of nonh- to nonheast-trending faults and lineaments located to the west of the Elsinore fault near Romona. The fault is mapped on a nonheast trend in bedrock exposures for 9.6 km between the nonh end of Pamo Valley and the Elsinore fault (Rogers,1965). The Temescal fault may extend southward beneath Pamo Valley to connect with nonh-south-trending structures mapped directly south of Romona (Jennings, 1992). If these features are continuous, the total length of the fault may be 29 km. Dames and Moore (1984) reviewed published data on the Temescal fault and Pamo Valley lineament and conducted reconnaissance field studies along the trend of the proposed fault and lineament.

They conclude that there was no evidence for Quaternary faulting between Pamo Valley and '

the Elsinore fault, and that the lineament may have formed due to erosion along existing fractures, joints, or a shear zone in Mesozoic bedrock. Furthermore, Dames and Moore (1984) did not identify exposures of bedrock faulting along the trend of the Temescal fault or the Pamo Valley lineament, nor did they find any descriptions of fault exposures reponed by previous workers. Thus, while the possibility of Quaternary faulting has not been precluded, l there is no clear evidence that the Temescal fault exists. On the other hand, Rockwell (personal communicaric ,1993) repons that a moderate magnitude (approximately 4.0) canhquake occurred near the trace of the Temescal fault during 1992. On this basis, we assign a low probability (0.2) of activity to a potential fault in Pamo Valley. We assume potential rupture lengths of 10 and 29 km, based on the extent of the mapped bedrock fault and the proposed extent of the fault from the Elsinom fault to south of Romona. We assign a high probability (0.7) for potential ruptures alor4 the nonhem segment (10 km) and a low probability (0.3) for potential ruptures of the entire fault (29 km). No data on the expected slip type of the proposed fault have been reponed; however, the topography of Pamo Valley is similar to that of Basin and Range valleys dominated by extensional processes (West, personal communication,1992).

p u e s a rr.nrr n r A-45 w y is,i m

The maximum depth of seismicity along the Elsinore fault near the intersection with the Temescal fault is about 13 to 17 km (Hill and others,1990). We assume similar depths for potential rupture along the Temescal fault, with a higher probability (0.7) for a maximum width of 13 km and a lower probability (0.3) for a maximum width of 17 km. The maximum magnitude distribution (Appendix B) reflects the range of rupture dimensions based on combinations of these rupture lengths and widths.

No slip rate estimates are published for the Temescal fault; however, if the fault is active, it must have very low rates of activity because there is no geomorphic expression of faulting along the trend of the proposed fault. We assume a large range of possible slip rates for the Temescal fault, with the largest probabilities assigned to the lower values: 0.01 mm/yr (0.3),

0.05 mm/yr (0.3),0.1 mm/yr (0.3), and 0.5 mm/yr (0.1).

3.2.12 La Nacion Fault Zone The La Nacion fault zone includes a series of discontinuous, north-south trending, down-to-the-west normal faults located southeast of San Diego (Fig. 2). The total length of the fault zone is approximately 27 km; individual faults are as long as 13 km. Additional strands of the la Nacion fault zene may occur south of the intemational border; however, studies have not been undertaken to identify potential faults in northernmost Baja California.

The recency of activity and potential slip rate for the La Nacion fault zone have been the subject of numerous studies, which are summarized in Saul (1979) and Kahle (1988). The results of these studies suggest that displacement has not occurred along the primary strands of the La Nacion fault zone during the past 13,000 yr. Funhermore, the geomorphic expression of the mapped faults generally is very weak, the faults are discontinuous, and most of the faults do not appear to displace late Pleistocene marine and fluvial terrace deposits (Kahle,1988). We note, however, that while late Pleistocene and Holocene deposits are not known to be displaced along the fault zone, there is evidence to preclude Holocene displacement at only a few locations. No slip rate measurements for the La Nacion fault have been published. Kahle (1988) concludes, however, that if the La Nacion fault zone is active,

. axm34carrerm A-46  % is, wu

i

] it must have a very low slip rate. Because the fault zon is not considered to be active (Kahle, j 1988; Jennings,1992), we assign a low probability of activity (0.1). We assign a high 4

{ probability (0.7) for rupture of the longest individual fault (13 km), and a low probability for

, ruptum of the entire mapped fault zone (27 km,0.3). The width of potential ruptures is assumed to be 12 to 15 km. The maximum magnitude distribution (Appendix A) mflects the 1

l range of rupture dimensions derived from combinations of the rupture lengths and widths. We i

assume a wide range of potential slip rates, with the highest probabilities assigned to the lower values (0.01 mm/yr,0.4; 0.05 mm/yr,0.4; and 0.1 mm/yr,0.2).

t i

l 3.2.13 Cristianitos Fault j The Cristianitos fault is a nonh-nonhwest trending, nonhwest-dipping fault that marks the j boundary between thick Neogene sediments to the nonh and thin Upper Cretaceous-Paleogene I marginal deposits of the ancient foreart basin (Monon and Miller,1981; Moyle,1973; Vedder i and others,1957). The fault, which lies within 0.8 km of SONGS, can be traced inland for l

a distance of about 35 km. Crouch and Suppe (1993) interpret the Cristianitos fault as a normal fault that formed above a detachment during the extensional development of the Los Angeles basin. Substantial evidence has been presented to indicate that the fault has not been active during the past 125 ka (Southern California Edison,1988; Shlemon,1992). In addition, i j there is no evidence to indicate that the fault has been active during the past 500 ka (Southern l California Edison,1988). This fault, therefore, is not considered to be an active or potentially active fault.

i

3.3 BURIED OR BLIND FAULT SOURCE ZONES l The occurrence of several moderate- to large magnitude canhquakes (e.g., the 1983 Coalinga, j 1987 Whittier Narrows,1989 Imma Prieta, and 1994 Nonhridge canhquakes) has demonstrated 3

i that large, damaging canhquakes can occur without surface rupture. In this analysis, we j consider two potential source zones in which blind thrust faults (i.e, faults that do not interrupt j the surface) have been inferred from interpretation of structural, stratigraphic, and seismologic

data.

l j ax. Ape 4marr.arr.TxT A-47  % u. im i

i

l l

Despite the absence of surface faulting, the Coalinga, Whittier Narrows, and Northridge l canhquakes all were associated with coseismic tectonic deformation, as observed geodetically, and all of the events occurred in association with young folds that geologic evidence would indicate had undergone late Quaternary uplift (e.g., King and Stein,1983; Bullard and l_ettis, 1993; Anderson,1990). These observations suggest that the " blind" faults that generated the earthquakes are not truly " blind" in the sense that they have no surface expression. Rather than brittle surface faulting, the signature for this type of seismic source is young fold deformation, which reflects the cumulative effect of multiple earthquakes. In a study of reverse-faulting earthquakes around the world, Wells and I.ettis (1990) found that in nearly all cases, blind earthquakes are associated with surface deformation and occur within areas of recognizable young folding. The most commonly observed deformation is coseismic uplift and long-term uplift associated with anticlinc.] fold development.

l l

Recent years have seen the application of quantitative appmaches that relate fold shape, fault geometry, and fault slip (e.g., Suppe,1983; Suppe and others,1992; Shaw and Suppe,1992).

Interest in this approach was particularly spawned by the occurrence of the Whittier Narrows canhquake and spatial association of that event with an interpreted blind thrust ramp (Davis and others,1989). Since then, other buried thrust ramps have been interpreted beneath the Los Angeles Basin (Davis and others,1989; Hauksson,1990; Hauksson,1992; Shaw,1993).

Blind potential seismic sources can be identified by a combination of subsurface interpretations (e.g., balanced cross sections, seismic rcilection) coupled with evidence for geologically young deformation (e.g., folding of late Quatemary deposits and surfaces). Identification and characterization of the seismic potential of active blind thmst faults of the Los Angeles basin is an ongoing and evolving topic of research. The geometry and slip rates for these structures are not well constrained and alternative models have been presented by various researchers (e.g., Davis and others,1989; Shaw,1993). Given the distance of these sources to the SONGS site and the uncertainties in the geometries of these potential fault sources, we have generalized potential blind thmst sources into two source zones, described below.

counm. owr.ernr A-48 %aw

4 i

1 3.3.1 Los Angeles Basin Source Zone A The 1987 Whittier Nanows canhquake (Mt5.9) appears to have occurred on a nord-dipping blind thrust fault in the northeastern Los Angeles basin directly north of the Montebello Hills j (Hauksson and Jones,1989). Geodetic measurements mdicate that uplift of the Elysian Park l anticlinorium occurred during and/or immediately following the canhquake (Lin and Stein,

! 1989). Davis and others (1989) conclude that the causative structure responsible for this s

!. canhquake is a blind thrust ramp beneath the Los Angeles basin that is referred to as the

! Elysian Park thrust. Davis and others (1989) inte pret the uplift of the Santa Monica Mountains anticlinorium to be due to a crustal scale fault propagation fold above the Eylsian l

4 Park thrust, which is estimated to have a slip rate of 3.9 to 5.9 mm/yr since 2 to 3 Ma (Davis

! and Namson,1994).

l Hauksson (1990) also includes the Whittier Narrows blind thrust fault in a regional fold and l thrust belt along the eastern and nonhern margins of the Los Angeles basin that he refers to I

l' as the Elysian Park fold and thrust belt. The boundaries of this broad zone are based on the

{ locations of mapped folds and thrust-faulting canhquakes (Davis and others,1989; Hauksson,

) 1990,1992). 'Ihe location of the fold and thrust belt identified by Hauksson (1992), however, i differs from the interpretation of Davis and Namson (1994) in that Davis and Namson locate l - the Elysian Park thrust system nonh of the Santa Monica-Malibu fault zone in the western pan of the los Angeles basin. Hauksson notes that thrust or reverse focal mechanisms are mostly l located to the south of the nonh-dipping reverse faults, such as the Santa Monica and 4

j Hollywood faults, and form broad clusters along the flanks of the basin. He attributes this

{ spatial coincidence of the folding and thrust faulting to fault-bend or fault propagation folding l models (e.g., Suppe,1985). Southeast of Whittier Narrows, thrust-faulting mechanisms j delineate a zone of thrust faulting dipping 25-35 north. Hauksson states that it is not possible 1

l to determine, on the basis of hypocentral distribution and focal mechanisms alone, if this zone 1

i consists of one continuous fault or many small abutting or overlapping faults. Hauksson

{t (1990,1992) continues the Elysian Park fold and thrust belt to the west of Whittier Narrows,

{ beneath downtown les Angeles, and into Santa Monica Bay. He postulates that this fold and I

4 .

i

axmumrr.arr.rxr A-49  % u. m.

4 J

- . - .~. ,_ ,. . - , .- -, , , _ .-

i l

l thrust belt may be offset or segmented by the northern end of the Newport-Inglewood fault (Hauksson,1990), but that segmentation of this zone is poorly understood.

Bullani and lettis (1993) conducted detailed mapping of Quaternary deposits and geomorphic surfaces and quantitative morphometric analyses of landscape elements within the Montebello Hills and Monterey Park Hills to characterize the style and patterns of Quaternary deformation associated with the blind fault responsible for the Whittier Narrows canhquake. The results of these studies suppon recognition of a south-vergent, southward propagating thrust fault that consists of two 6- to 8-km-long segments. Movement on this thrust fault has produced uplift and folding of Quaternary deposits and surfaces in the region. Maximum rates of Pleistocene uplift range from 0.1 to 0.25 mm/yr, with slightly higher upper range values (0.5 to 0.6 mm/yr) for the latest Pleistocene and Holocene (Bullard and Lettis,1993). Bullard and Lettis note that deformation in the Whittier Narrows area occurs in a transitional zone between right-lateral strike-slip defonnation on the Whittier-Elsinore fault system and left-oblique contractional deformation on the Santa Monica-Raymond fault system. They postulate that crustal shortening in this area represents either (1) an area of local compression within a series of progressively smaller, left en echelon restraining stepovers in a regional right-lateral strike-slip system or (2) progressive onset of convergence as lateral slip along the Whittier-Elsinore fault system dies out northward and is accommodated by more westerly trending reverse and left-oblique slip faults along the southern margin of the Transverse Ranges.

Additional structures have been identified to the west of the Montebello Hills and Monterey Park Hills. Hummon and others (1994) postulate that a potentially seismogenic blind thrust fault, which they refer to as the Wilshire fault, underlies and causes the Wilshire arch, a Quaternary fold in the Hollywood area,just west of downtown Los Angeles, California. This 9-km-long fault is limited to the west by the Newport-Inglewood fault (Hauksson,1990) and to the east by the MacArthur Park fault (Dolan and Sieh,1992). Estimated slip rates for this structure vary depending upon the model used to describe the geometry. A fault bend fold model indicates a reverse-slip rate of 1.5 to 1.9 mm/yr on a 10 to 15 north-dipping fault, whereas a three-dimensional clastic-dislocation model indicates a right< reverse slip rate of 2.6 coums4mrrerm A-50  % is, m4

to 3.2 mm/yr on a 30 to 35 -dipping fault. Using the dislocation parameters determined from the dislocation modeling (length 9.0 km, width 1.8 km, depth to tip 2.8 km) they calculate a M 5.7 for this structure. This estimate, which is based on a postulated rupture of only the Wilshire fault, is a minimum (Hummon and others,1994). They postulate that larger-magnitude events would result from earthquake rupture of more than one segment, but provide no details of what such rupture scenarios would entail.

The Los Angeles basin source zone A outlined in this study encompasses the zone of west-northwest-trending fold axis in the northeastem part of the Los Angeles basin. This source zone extends west to the Newport-Inglewood fault trend and includes the thrust fault (s)in the vicinity of Whittier Narrows and additional structures such as the postulated Wilshire fault (Hummon and others,1994). The Elysian Park thmst, as modeled by Davis and others (1989) and Davis and Namson (1994), is not considered a significant seismic source for the SONGS site for the following reasons. Although, the eastern end of the Elysian Park thrust lies within source zone A, the majority of the Elysian Park thrust system, as shown by Davis and Namson (1994), lies beyond the 100-km-radius area of concern for the SONGS site. In modeling the Elysian Park thmst, Davis and others (1989) require that the Whittier fault be inactive. As noted above, the Whittier fault is a major active strike-slip fault.

The model used in this study incorporates rates and patterns of Quaternary deformation. As noted by Hauksson (1992), the high earthquake potential of the concealed faults in the two fold belts outlined by Hauksson (1990) is the result of their combined high slip rate and long length. The high slip rates, which are based on the Davis and others (1989) model, are long-term slip rates averaged over 2 to 3 Ma. Detailed Quatemary studies (i.e., Bullard and lettis, 1993) support neither the model of a long continuous structure nor the high slip rates in the vicinity of the Elysian Park thrust near Whittier Narmws. The regional background areal source zone for the Los Angeles basin that allows for random earthquakes ranging from M 5.5 to M 6.5 will capture Whittier Narrows-type events on blind thrust faults outside the Elysian Park fold trend zone outlined here, for which there presently are insufficient data for fault specific source characterization. -

aunamer wrnr A-5I w n. nm 1

l

.-.. . . - . . - . - . - - - -- - - --- - ...- - - . - - . _ ~ .

l Based on the occurrence of the 1987 Whittier Narrows canhquake and the associated evidence of Quaternary defonnation, we consider this source zone to be active with a probability of 1.0.

Given the uncertainties in the geometry and location of the possible fault (s) within this zone, we treat the source as a local areal source zone. On the basis of the segmentation arguments presented by Bullard and Lettis (1993) and the modeling results of Hummon and others (1994),

we allow for rupture lengths of 8 km (0.3),16 km (0.5), and 28 km (0.2). The longest rupture

, length allows for rupture of the entire zone from near Whittier Narrows to the Newpon-Inglewood fault. The maximum magnitude distribution (Appendix B) reflects the range of rupture dimensions derived from combinations of these rupture lengths and widths given in Table A-1. Using the geometry of the fault as modeled by Davis and others (1989) (M nonh-dipping ramp) and the uplift rates of 0.1 to 0.25 implies a Quaternary slip rate ranging from 0.2 to 0.6 mm/yr. The highest uplift values of 0.5 and 0.6 mm/yr allowed by the data for the youngest surfaces suggest slip rates of as much as 1.4 mm/yr on an underlying t' rust fault.

i These rates, which are based on Quaternary rates of deformation, are considerably less than i

the long-term slip rates estimated by Davis and others (1989). These rates, which are judged to be more representative of the behavior of the underlying fault in the current tectonic setting, provide the basis for the estimated range of slip rate values used in this analysis. We use the following range of slip rate values and weights: 0.1 mm/yr (0.1),0.2 mm/yr (0.5),0.5 mm/yr (0.3), and 1.5 mm/yr (0.1).

3.3.2 Los Angeles Basin Source Zone B

, Seismotectonic studies by Hauksson and others (1988), Hauksson and Saldivar (1989),

i Hauksson (1990) and balanced cross sections by Davis and others (1989) and Shaw (1993) interpret blind thrust faults in the western pan of the Los Angeles Basin near the Palos Verdes Hills and Torrence Wilmington fold trend. Davis and others (1989) model a 25' east-verging L thrust ramp beneath the Palos Verdes Hills to account for the uplift of the Palos Verdes  ;

l anticlinorium. Shaw (1993) shows the Palos Verdes fault as a west-dipping back thrust i extending to a depth of 5 km, where it intersects the Compton thrust. The 60-km long Torrance-Wilmington fold and thrust belt, as defined by Hauksson (1990) based on fold axes I

j and thrust focal mechanisms, extends from offshore Newpon Beach, across the Palos Verdes l aunes4mrr arr.m A-52 w is, im

I i

k I Peninsula,into Santa Monica Bay. He notes that the structure of this system is more complex l and less understood than the Elysian Park system. In Santa Monica Bay, a zone of thrust I

! faults dips north 30 to 40, whereas to the southwest (onshore), the hypocenters show a f steeply dipping zone beneath the Palos Verdes fault trace. A zone of thrust faults in San Pedro i

! Bay dips southwest at 20 to 30 (Hauksson,1992). Hauksson (1992) postulates that the l south-dipping thrust fault in San Pedro Bay may be a back thrust, suggesting that a deeper east- or nonheast-dipping primary thrust fault may underlie it.

I j Given the uncertainties in the locations and geometries of possible buried thrust faults in this region, we treat this region as an areal source zone that may include either an east-nonheast-l dipping fault coincident with the Torrance-Wilmington anticline or a west-dipping thrust fault i

under the Palos Verdes uplift. The low probability (0.2) of activity assigned to this souret

i i zone reflects our judgment of the likelihood that significant deformation in this region results I from activity on buried thrust faults. We give higher weight (0.8) to a preferred model that K

{ characterizes the Palos Verdes and Newport-Inglewood fault zones as strike-slip or oblique-slip

faults and the rest of zone B as a regional areal source zone with a magnitude distribution 4 similar to that described for the central Los Angeles basin areal source zone. As described in

)

the discussion of the Palos Verdes fault zone, the uplift of the Palos Verdes Hills is more t

j likely due to local compression along a restraining bend in a strike-slip fault zone. The steeply l dipping fault zone defined by recent seismicity (Hauksson,1992) also is consistent with a i

i strike-slip fault. Models that require an east-dipping ramp under the westem pan of the Los l Angeles basin imply uplift in the west basin that is not expressed geologically (D. Ponti, U.S.

Geological Survey, personal communication, March 1994). On the basis of the geomorphic l

! position of marine deposits in the Torrance and Long Beach plains, venical tectonism during 4 the past 600 ka in this area has been negligible (Ponti and Lajoie,1992). Significant l Pleistocene uplift, as evidenced by elevated marine terraces, occurs chiefly along the linear 4

alignment of anticlines associated with the Newport-Inglewood fault (Lajoie and others,1992).

1 i

l i

l axxmu.onrr.arrm A-53  % u. m

I 4

Segmentation scenarios for a blind thrust model are highly speculative. We assign an equal i

weight (0.4) to rupture lengths of 20 km (based on the length of the most significant uplift 1

along the Palos Verdes Hills) and 40 km. The 40-km rupture scenario for the west-dipping blind fault includes the possible extension of the zone of uplift onto the San Pedro shelf region. The 40-km length also reflects the approximate length of the Torrance-Wilmington anticline east of the Palos Verdes fault. The least weight (0.2) is given to a 60-km rupture.

j The maximum magnitude distribution (Appendix B) reflects the range of rupture dimensions derived from combinations of these rupture lengths and widths given in Table A-1. Maximum slip rates for blind thrusts in this zone assume that uplift of the Palos Verdes Hills

(-0.35 m/kyr)is due to slip on a 25'-dipping fault, as modeled by Davis and others (1989). The resulting slip rate of 0.8 mm/yr is considerably less than the long-term average rate estimated by Davis and others (1989). Much lower slip rates are indicated for the region east of the i Palos Verdes Hills, where maximum uplift (at rates of -0.5 m/kyr) ia occuning along domes I

l within the Newport-Inglewood fault zone. As discussed in the previous section, we use a j range of slip rates based on relatively well-constrained rates of Quaternary uplift and defonnation. For this source zone, the following slip rates and weights are applied: 0.01 )

l mm/yr (0.2),0.3 mm/yr (0.4), and 0.8 mm/yr (0.4).

4 1

3.3.3 Assessment of the Potential for Unknown Blind Thrust Faults Near San Onofre l Nuclear Generating Station j Our assessment of the potential for unknown blind thrust faults near the SONGS site is based l on consideration of the tectonic setting and style and rate of Quaternary deformation in the i

{ region. SONG site lies near the southwestern boundary of the Peninsular Ranges geomorphic 4

province. The site lies within a relatively stable structural block bound by major nonhwest-j trending strike-slip faults. The tectonic setting of the site is significantly different from the I

1 complex tectonic regime of the Los Angeles basin that is marked by north-south convergence

associated with geometry of the " big bend" in the San Andreas fault. This difference is

{ reflected in the markedly different rate of canhquake occurrences between the two regions and 4 a more diffuse spatial pattern of seismicity than the linear patterns associated with strike-slip i

faults to the south (Plate 1).

j j towm4sarr.nrrnr A-54  % is. im

l The presence or absence of blind thrust faults in a region is indicated by the presence or absence of significant uplift and folding of late Quaternary deposits and geomorphic surfaces i

(e.g, Stein and Yeats,1989). Information regarding the nature and rate of Quaternary j deformation along the coastal region in the vicinity of SONGS is provided by marine terrace investigations. Mapping of marine terraces along the western flank of the San Joaquin Hills

{ to the north of the site indicates a uniform uplift rate of 0.25 m/kyr for the past 80 to ~120 ka

! (Barrie and others,1992). Lajoie and others (1992) estimate a similar long-term average uplift rate of 0.19 m/kyr for the coastal region between San Onofre Bluff and Torrey Pines north of Soledad Mountain in San Diego. They note that the strandlines between these two points are approximately horizontal, indicating that there has been no significant crustal tilt perpendicular to the coastline during much of Quaternary time. Also, there is no indication from the marine i terrace studies of significant tilt parallel to the coastline during much of Quaternary time l (Lajoie and others,1992b). The Pleistocene slip rates in this region (0.19 to 0.25 m/kyr) are

{ comparable to uplift rates for other fault-bounded structural blocks within regions dominated by right lateral crustal displacements in coastal California, which are 0.1 to 0.3 m/kyr (Muhs and others,1992b).

i j The marine terrace data and other mapping indicates that geologically yoang folds, such as j those associated with known blind thrusts have not been mapped or identified in the nearby vicinity of the SONGS site. On the basis of the apparent lack of late Quaternary anticlinal l fold development, we conclude that there are no seismogenic blind thrust faults in the site

! region that are capable of generating significant earthquakes. In the hazard analysis, we allow the possibility of unknown sources, including smaller-scale blind thrust faults, within the

}

required areal source zones.

3.4 REGIONAL AREAL SOURCE ZONES i

! Several areal source zones are identified to account for the possibility that potential sources j of seismicity significant to SONGS may occur on small, unidentified faults and/or faults that i are not expressed at the surface. The boundaries of these zones are purposely drawn to l

cote 5earr arr.TxT A-55  % is. im 4

I

i exclude the neighboring faults, blind fault sources, and their associated seismicuy. Other reeerit seismic hazard analyses in California have shown that the inclusion of these types of areal zones is imponant to the proper representation of the full range of uncertainty regarding potential canhquake sources (e.g., Geomatrix,1992; Youngs and others,1992).

As noted above, random canhquakes that occur on faults that do not rupture to the surface are referred to as " blind faults" (Yeats,1986; Stein and Yeats,1989). Studies of moderate- to large-magnitude (e.g., greater than M 6 to 6 ) historical canhquakes that did not ruptuit to the surface indicate that these earthquakes are associated with Quaternary surface deformation,  !

such as active folding or regional uplift (Wells and lettis,1990). Therefore, these " hidden" canhquakes are expected to occur in regions of active deformation, such as uplift and folding (Stein and Yeats,1989). We have modeled these sources as blind fault sources in the analysis.

Background seismicity occurring in regions without such deformation may be constrained in magnitude to below that which would typically cause surface rupture. The magnitude '

threshold for surface-faulting events is usually considered to be about M 6 to 6 (Tocher, 1958; Bonilla and others,1984; Bonilla,1988). Recent moderate-magnitude canhquakes in the San Francisco area (Greenville, Hall's Valley, and Coyote Lake) were accompanied by very minor surface slip, suggesting a slightly lower threshold of surface faulting of about M l 5 to 6. A recent compilation of data for 275 crustal canhquakes shows that the magnitude l

at which there is a 0.5 probability of surface rupture is about M, 6.0; at M, 5.5 the probability is 0.2; at M,6.5 the probability is 0.8. On the basis of these observations, we consider a

! range of values of M 5.5 to 6.5 for the estimated maximum magnitude for the areal source zones identified in this study (Table A-1). Higher weight is assigned to M 6.5 in the Los Angeles basin to reflect the greater potential for unknown blind thrust faults.

3.4.1 Peninsular Ranges Source Zone l The Peninsular Ranges source zone lies between two major strike-slip faults, the Newport-l Inglewood-South Coast Offshore Zone of Deformation-Rose Canyon fault zones to the west and the Elsinore fault zone to the east. The northern border of this zone coincides with the inferred boundary between the northeastern Los Angeles basin-inner California borderland rift aumenrr.arrm A-56 wy is, m

and the lithotectonic belts of the Peninsular Ranges to the south, as defined by Crouch and Suppe (1993). This boundary marks the nonhwestern subsurface limit of Peninsular Ranges basement rock and overlying Cretaceous-Paleogene fore-arc strata that extend beneath Miocene and younger cover along the Anaheim nose (Yerkes and others,1965; Yeats,1973; Wright, 1991). In order to reflect uniform network seismicity coverage, the international bonier is selected as the southern boundary of this zone.

3.4.2 Central Los Angeles Basin Source Zone The Central Los Angeles basin source zone is bordemd by blind thrust source zone A to the nonh, the Newport-Inglewood fault zone to the west, and the Whittier-Elsinore fault zones to l the east. The southern boundary of this zone coincides with the inferred boundary between the nonheastern Los Angeles basin inner California borderland rift and the lithotectonic belts of the Peninsular Ranges to the south, as described above. In the central Los Angeles basin, upper Miocene and younger strata reach thicknesses in excess of 6 km (Yerkes and others, i

1965).

I This areal source zone lies within a region of active crustal shonening related to the geometry of the big bend in the San Andreas fault system. The rate of historical seismicity in this source area is significantly higher than that of the Peninsular Ranges source zone to the south.

On the basis of differences in the basement geology, tectonic setting, and observed seismicity, we give more weight (0.8) to a model in which this source zone is differentiated T _m the Peninsular Ranges source zone. We also give some weight (0.2) to a model that combines both the Los Angeles basin and Peninsular Ranges zones into a single source zone, comparable to the current Southern California Eanhquake Center (SCEC) model.

3.4.3 Offshore Basin Source Zone The Offshore Basin source zone includes the offshore region west of the Newpon-Inglewood-SCOZ.D-Rose Canyon to the Palos Verdes-Coronado fault zones. The nonhern boundary of this zone is selected at the boundary with the source zone B and the southern boundary is ax e m m wr.wrxxr A-57 wy is. i9w

drawn at the international border to avoid complications of variable seismicity catalogs in '

regions north and south of the border.

1 i

I mL^nmonn.mm A-58 m ss.n m

l 4

4.0 REFERENCES

l' Anderson, J.G.,1979, Estimating the' seismicity from geological structure for seismic risk i studies: Bulletin of the Seismological Society of America, v. 69, p.135-158. r j Anderson, J.G., Rockwell, T., and Agnew, D.,1989, Past and possible future earthquakes of significance to the San Diego region: Eanhquake Spectra, v. 5, p. 299-335.

j Andenon, R.S.,1990, Evolution of the nonhern Santa Cruz Mountains by advection of crust

past a San Andreas fault bend: Science, v. 249, p. 397-401.

Argus, D.F., and Gordon, R.G.,1988, Sierra Nevada-Nonh America motion from VLBI and

{ paleomagnetic data--implications for the kinematics of the Basin and Range, Colorado i Plateau, and Califomia Coast Ranges: EOS Transactions, American Geophysical Union, j v.69,p.1418.

i l Barrie, D., Totnall, T., and Gath, E.,1992, Neotectonic uplift and ages of Pleistocene marine i terraces, San Joaquin Hills, Orange County, California, in Heath P.G., and Lewis, W.L.

l (eds.), The Regressive Pleistocene Shoreline Coastal Southern California: South Coast Geological Society, Inc.,1992 Annual Field Trip Guide Book No. 20, p.115-122.

3 Bird, P., and Rosenstock, R.W.,1984, Kinematics of present crust and mantle flow in southem .

J California: Geological Society of America Bulletin, v. 95, p. 946-957.

4 j Bonilla, M.G.,1973, Trench exposures across surface fault ruptures associated with San Fernando canhquake in San Fernando, Califomia, canhquake of February 9,1971: U.S.

, Department of Commerce, Washington D.C., v. 3, p.173-182.

r

. Bonilla, M.G.,1988, Minimum canhquake magnitude associated with coseismic surface

! faulting: Bulletin of the Association of Engineering Geologists, v. 25, p.17-29.

j Bonilla, M.G., and Buchanan, J.M.,1970, Interim repon on worldwide historic surface  ;

faulting

U.S. Geological Survey Open-File Repon 70-34,32 p.

I j Bonilla, M.G., Mark, R.K., and Lienkaemper, J.J.,1984, Statistical relations among earthquake magnitude, surface rupture length, and surface fault displacement: Bulletin of the

! Seismological Society of America, v. 74, no. 6, p. 2379-2411.

I i

l Brake, J.F., and Rockwell, T.K.,1987, Magnitude of slip from historical and prehistorical l

! canhquakes on the Elsinore fault, Glen Ivy Marsh, southern California (abs.): Geological j

Society of America Abstracts with Programs, v.19, no. 6, page insert.  !

4 3

f mummarr.arrnr A-59 w 15. im l 3

i

! Bryant, W.A.,1985, Southern Newpon Inglewood fault zone, southern Los Angeles and i nonhern Orange counties, California: California Division of Mines and Geology Fault i Evaluation Repon FER-172.

j Bullard, T.F., and 12ttis, W.R.,1993, Quaternary fold deformation associated with blind thrust i

faulting, Los Angeles Basin, California: Journal of Geophysical Research, v. 98, no. B5, 1 p. 8349-8369. ,

! Clark, M.M., Harms, K.K., Lienkaemper, J.J., Harwood, D.S., Lajoie, K.R., Matti, J.C.,

{ Perkins, J.A., Rymer, MJ., Sarna-Wojcicki, A.M., Sharp, R.V., Sims, J.D., Tinsley, J.C.  ;

i HI, and Ziony, J.I.,1984, Preliminary slip-rate table and map of late-Quaternary faults of California: U.S. Geological Survey Open-File Report 84-106,12 p.,5 plates, map l scale 1:1,000,000.

l 4

l Clarke, S.H., Greene, H.G., and Kennedy, M.P.,1985, Identifying potennally active faults and L unstable slopes offshore, in Ziony, J.I. (ed.), Evaluating Earthquake Hazards in the Los i

Angeles Region - an Eanh-Science Perspective: U.S. Geological Survey Professional

{ Paper 1360, p. 347-374.

J l Clarke, S.H., Greene, H.G., Kennedy, M.P., and Vedder, J.G. (with contributions by Legg, M.R.),1987, Geologic map of the inner-southern California continental margin,

! California Continental Margin Geologic Map Series: California Division of Mines and l Geology Maps I A, IB, IC, ID.

I ,

j Coppersmith, K.J.,1991, Seismic source characterization for engineering seismic hazard j analyses: Proceedings, Founh Intemational Conference on Seismic Zonation, Stanford,  !

j California, August 26-29, v.1, p.1-60.  ;

I

! Cornell, C.A., and Van Marke, E.H.,1969, The major influences on seismic risk:

l Proceedings of the Third World Conference on Earthquake Engineering, Santiago l Chile, v. A-1, p. 69-93.

3 j Crook, R., Jr., Allen, C.R., Kamb, B., Payne, C.M., and Proctor, RJ.,1987, Q.:aternary

{ geology and seismic hazard of the Sierra Madre and associated faults, western San i Gabriel Mountains, San Bernardino County, in Morton, D.M., and Yerkes, R.F. (eds.),

j Recent Reverse Faulting in the Transverse Ranges, California: U.S. Geological Survey i Professional Paper 1339, p.

• 7 64.

! Crouch, J.K., and Bachman, S.B.,1989, Exploration potential of offshore Newport-Inglewood i fault zone (abs.): American Association of Petroleum Geologists Bulletin, v. 73, p. 536.

i

! Crouch, J.K., and Suppe, J.,1993, Late Cenozoic tectonic evolution of the Los Angeles basin and inner California borderland: a model for core complex-like crustal extension:

Geological Society of America Bulletin, v.105, p.1415-1434.

i axxasearr arrm A-60 wy is m 1

. ~ __ - . - - - - - - - - - - . - - - . -- . - -. ..

4 Dames and Moore,1984, Preliminary design repon: Pamo Dam and Reservoir: Prepared for the San Diego Water Authority.

Davis, T.L., and Namson, J.,1994, A balanced cross section analysis of the 1994 Nonhridge  :

canhquake and thmst fault seismic hazards in southern California (abs, and poster): l Seismological Society of America, 89"' Annual Meeting, Program for Northridge 1

Abstracts, Abstract No. 37.

j Davis, T. L, Namson, J., and Yerkes, R.F.,1989, A cross section of the los Angeles area:

seismically active fold and thmst belt, the 1987 Whittier Narrows canhquake, and j canhquake hazard: Joumal of Geophysical Research, v. 94, p. 9644-9664.

DeMets, C.R., Gordon, G., Argus, D.F., and Stein, S.,1990, Current plate motions

j Geophysical Journal International, v.101, p. 425-478.

! DeMets, C. R., Gordon, G., Stein, S., and Argus, D.F.,1987, A revised estimate of

Pacific-Nonh America motion and implications for western North America plate i boundary zone tectonics: Geophysical Research I.ctters, v.14, p. 911-914.

Dolan, J.F., and Sieh, K.E.,1992 Tectonic geomorphology of the nonhern Los Angeles Basin: ,

i seismic hazards and kinematics of young fault movement, in Engineering Geology Field i Trips, Orange County, Santa Monica Mountains, and Malibu: Association of )

3 Engineering Geologists, 35th Annual Meeting, Field Trip Guidebook, p. B20-B26.

l  ;

Dolan, J.F., Sieh, K., Guptil, P., Miller, G., Rockwell, T., and Smirnoff, T.,1993, Structural geology, fault kinematics, and preliminary palcoscismologic results from the Hollywood fault: new data .* rom continuously cored borings and geotechnical trenches, Hollywood, California (abs.): EOS, Transactions, American Geophysical Union, v. 74, no. 43, p. 427.

Feigl, K.L, Agnew, D.C., Bock, Y., Dong, D., Donnellan, A., Hager, B.H., Herring, T.A.,

Jackson, D.D., Jordan, T.H., King, R.W., Larsen, S., Larson, K. M., Murray, M.H., Shen, Z., and Webb, F.H.,1993, Space geodetic measurement of crustal deformation in central and southern California, 1984-1992: Journal of Geophysical Research, v. 98, no. B12,

p. 21,677-21,712.

Fischer, PJ., Gorsline, D.S., and Shlemon, R.J.,1992, Late Quaternary geology of the Dana Point-San Onofre-Carlsbad margin, California, in Heath, E.G., and Lewis, W.L. (eds.),

The Regressive Pleistocerie Shoreline Coastal Southern California: South Coast Geological Society, Inc.,1992 Annual Field Trip Guide Book No. 20, p.195-218.

Fischer, P.J., and Lee, C,1992, Evolution of Santa Monica - San Pedro Margin, California during the last 25,000 yr, in Heath, E.G., and Lewis, W.L. (eds.), The Regressive Pleistocene Shoreline Coastal Southern California: South Coast Geological Society, Inc.,

1992 Annual Field Trip Guide Book No. 20, p.161-194.

awas4mT arr.rrr A-6I w is, m4

_ ._ _ . - _ . _ _ _ _ . _ _ _ _ = _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ . _ _ . .

4 I

Fischer, PJ., Patterson, R.H., Rudat, J.H., and Simila, G.,1987, The Palos Verdes fault zone, 2

onshore to offshore, in Fischer, PJ. and Mesa Inc. (eds.), Geology of the Palos Verdes  ;

Peninsula and San Pedro Bay: SEPM Pacific Section Field Trip Guidebook, p.91-133. '

Fischer, P.J., and Mills, G.I.,1991, The offshore Newport Inglewood-Rose Canyon fault zone,  ;

California: Structure, segmentation and tectonics, in Abbott, P.L., and Elliott, WJ. (ed.),

Environmental Perils, The San Diego Region: San Diego Association of Geologists, San Diego, California, p.17-36.

Fischer, PJ., Kreutzer, P.A., Morrison, L.R., Rudat, J.H., Ticken, EJ., Webb, J.F., Woods, I M.M., Berry, R.W., Henry MJ., Hoyt, D.H., and Young, M.,1983, Study of Quarternary deposits (sand and gravel) of southern California: State Depanment of Boating and Waterways, p. 75.

Freeman, S.T., Heath, E.G., Guptill, P.D., and Waggoner, J.T.,1992, Seismic hazard  ;

assessment, Newpon-Inglewood fault zone, Pipkin, B.W., and Proctor, RJ. (eds.),

Engineering geology practice in southern California: Association of Engineering 4

Geologists Special Publication No. 4, p. 211-231.

Gath, E.M., Gonzales, T., and Rockwell, T.K.,1992, Slip rate of the Whittier fault based on 3-Darak trenching at Brea, Southern California: Geological Society of America Abstracts with Pmgrams, v. 24, no. 5, p. 26.  ;

Geomatrix Consultants,1992, Seismic ground motion study for West San Francisco Bay Bddge: unpublished repon prepared for CALTRANS Division of Structures, variously paginated.

Guptill, P.D., and Heath, E.G.,1981, Surface faulting along the Newpon-Inglewood zone of deformation: California Geology, p. 142-145.

Gutenberg, B., and Richter, C.F.,1954, Seismicity of the canh and associated phenomena,2nd (ed.): Princeton University Press, Princeton, New Jersey,310 p.

Harding, T.P.,1973, Newpon-Inglewood trend, California-an example of wrench style deformation: American Association of Petroleum Geologist Bulletin, v. 57, no.1, p.97-116.

Harris, R.A.,1992, A 3-d dynamic rupture model of fault segmentation with applications to the next Anza Gap canhquake (abs): EOS Transactions, American Geophysical Union, v.73,no.43,p.389.

Hauksson, E.,1987, Seismotectonics of the Newpon-Inglewood fault zone in the les Angeles basin, southern California: Bulletin of the Seismological Society of America, v. 77, no.

2, p. 539-561.

cxwearr arr.rrr A-62 u r u. w

Hauksson, E.,1990, Earthquakes, faulting, and stress in the Los Angeles Basin: Journal of Geophysical Research, v. 95, p.15,365-15,394.

Hauksson, E.,1992, Seismicity, faults, and earthquake potential in Los Angeles, southem California, in Pipkin, B.W., and Proctor, R.J. (ed), Engineering Geology Practice in Southern Califomia: Association of Engineering Geologists Special Publication No. 4, p.167-179.

Hauksson, E., Jones, L.M., Davis, T.L., Hutton, L.K., Brady, A.G., Reasenberg, P. A., Michael, A.J., Yerkes, R.F., Williams, P., Reagor, G., Stover, C.W., Bent, A.L., Shakal, A.K.,

Etheredge, E., Porcella, R.L., Bufe, C.G., Johnston, M.J.S., and Cranswick, E.,1988, The 1987 Whittier Narrows canhquake in the les Angeles metropolitan area, California:

Science, v. 239, p.1409-1412.

Hauksson, E., and Gross, S.,1991, Source parameters of the 1933 long Beach canhquake:

Bulletin of the Seismological Society of America, v. 81, no.1, p. 81-99.

Hauksson, E., and Jones, L.M.,1988, The July 1986 Oceanside (Mt = 5.3) canhquake sequence in the continental borderland, southern California: Bulletin of the Seismological Society of America, v. 78, no. 6, p.1885-1906.

~

Hauksson, E., and Jones, L.,1989, The 1987 Whittier Nanows canhquake sequence in Los Angeles, southern California: Seismological and tectonic analysis: Journal of Geophysical Research, v. 94, p. 9569-9590.

l Hauksson, E., and Saldivar, G.V.,1989, Seismicity and active compressional tectonics in Santa Monica Bay, southern California: Journal of Geophysical Research, v. 94, p. 9591-9606.

Hill, R.L., Sprotte, E.C., Bennett, J.H., Chapman, R.H., Chase, G.W., Real, C.R., and Borchardt, G.,1977, Seismicity and focal mechanisms of the study area, Pt. 2 of Santa Monica-Raymond Hill fault zone study, IAs Angeles County, California: California Division of Mines and Geology, Final Technical Repon, U.S. Geological Surve)

Contract 14-08-15858,35 p.

Hill, D.P., Eaton, J.P., and Jones, L.M.,1990, Seismicity, 1980-86, in Wallace, R.E. (ed.), 'Ihe San Andreas Fault System, California: U.S. Geological S'. rvey Professional Paper 1515, p.115-151.

Hummon, C., Schneider, C., Yeats, R.S., Dolan, J.F., Sieh, K., and Hutile, G.,1994, Wilshire fault: earthquakes in Hollywood 7: Geology, v. 22, p. 291-294.

Jennings, C.W.,1992, Preliminary fault activity map of California: California Division of Mines and Geology Open-File Report 92-03, scale 1:750,000.

toixasewr.wr.TxT A-63  % is, im 1

J

Jones, L.M., Sieh, K.E., Hauksson, E., and Hutton, L.K.,1990, The 3 December 1988 Pasadena, California earthquake: evidence for strike-slip motion on the Raymond fault:

Bulletin of the Seismological Society of America, v. 80, p. 474-482.

Kahle, J.E.,1988, A geomorphic analysis of the Rose Canyon, La Nacion, and related faults in the San Diego area, California: California Division of Mines and Geology Fault Evaluation Report FER-196.

Kennedy, M.P.,1975, Geology of the San Diego metropolitan area, California: California Division of Mines and Geology Special Repon 200, Section A., Western San Diego Metmpolitan Area, p. 7-39.

Kennedy, M.P., Bailey, K.A., Greene, H.G., and Clarke, S.H.,1978, Recency and character of faulting offshore from metmpolitan San Diego, California: Final Technical Repon FY 1977-1978, U.S. Geological Survey,47 p.

Khromovskikh, V.S.,1989, Determination of magnitudes of ancient canhquakes from dimensions of observed seismodislocations: Tectonophysics, v.166, p. 269 280.

King, G.C.P., and Stein, R.S.,1983, Surface folding, river terrace deformation rate and canhquake repeat time in a reverse faulting environment: the Coalinga, California, canhquake of May,1983, in Bennett, J.H., and Serburne, R.W. (eds.), The 1983 Coalinga, California, Eanhquake: California Division of Mines and Geology Special Publication 66, p.165-176.

)

l Knuepfer, P.L.K.,1989, Implications of the characteristics of end-points of historical surface fault ruptures for the nature of fault segmentation, in Schwanz, D.P., and Sibson, R.H.

(eds.), Fault Segmentation and Contmls of Rupture Initiation and Termination: U.S.

Geological Survey Open-File Repon 89-315, p.193-228.

Lajoie, K.R., Ponti, D.J., Powell, C.L., II, Mathieson, S.A., and Sarna-Wojcicki, A. M.,1992, Emergent marine strandlines and associated sediments, coastal California; a record of Quatemary sea-level fluctuations, venical tectonic movements, climatic changes, and coastal processes, in Heath, E.G., and Lewis, W.L. (eds.), The Regressive Pleistocene 1 Shoreline Coastal Southern California: South Coast Geological Society, Inc.,1992 l Annual Field Trip Guide Book No. 20, p.81-104. l l

Lamar, D.L., and Rockwell, T.K.,1986, An overview of the tectonics of the Elsinore fault zone, in Ehlig, P. (ed.), Guidebook and Volume on Neotectonics and Faulting in Southen California: Cordilleran Section, Geological Society of America, p. 149-158. I I

Larson, K.M.,1993, Application of the global positioning system to crustal defonnation measurements,3, result from the southern Califomia borderlands: Joumal of Geophysical Research, v. 98, no. B12, p. 21,727-21,740.

cot.432654.onrr.nrrxx r A-64 w is, w

legg, M.R.,1980, Seismicity and tectonics of the inner continental borderland of southem California and nonhern Baja California, Mexico: M.S. Thesis, University of California

, at Santa Diego,60 p.

Legg, M.R.,1985, Geologic structure and tectonics of the inner continental bonierland offshore northem Baja Califomia, Mexico: unpublished Ph.D. dissenation, University of California at Santa Barbara, Santa Barbara, California,410 p.

legg, M.R., and Kennedy, M.P,1991, Oblique divergence and convergence in the Califomia continental borderland, in Abbott, P.L., and Elliott, WJ. (eds.), Environmental Perils San Diego Region: San Diego Association of Geologists, San Diego, Califomia, p.1-16.

legg, M.R., Wong-Oretega, V., and Vidal, F.S.,1991, Geologic structure and tectonics of the inner continental borderland of northern Bajar ' "' mia, in Dauphin, P., Ness, C., and Simoneit, B.R.T. (eds.), The Gulf and Peninsula Provinces of the Califomias: American Association of Petroleum Geologists Memoir 47.

Iettis, W.R., and Hanson, K.L.,1991, Crustal strain panitioning: implications for seismic-hazard assessment in western California: Geology, v.19, no. 559-562.

Lin, J., and Stein, R.S.,1989, Coseismic folding, canhquake recurrence, and the 1987 source mechanism at Whittier Narrows, Los Angeles Basin, California: Journal of Geophysical Research, v. 94, p. %14-9632.

Lindh, A.G.,1983, Preliminary assessment of long-term probabilities for large canhquakes along selected segments of the San Andreas fault system in California: U.S. Geological Survey Open-File Repon 83-63,15 p.

Lindvall, S.C., Rockwell, T.K., and Lindvall, C.E.,1990, The seismic hazard of San Diego revised: new evidence for magnitude 6+ Holocene canhquakes on the Rose Canyon fault zone: Proceedings,4* U.S. Conference on Eanhquake Engineering.

Magistrale, H., and Rockwell, T.K.,1990, Seismic and geomorphic evidence for two active branches of the south-central Elsinore fault zone (abs.): EOS, v. 71, no. 43, p.1633.

Mark, R.K., and Bonilla, M.G.,1977, Regression analysis of canhquake magnitude and surface fault length using the 1970 data of Bonilla and Buchanon: U.S. Geological Survey Open-File Repon 77 614.

Matti, J.C., Monon, D.M., and Cox, B.F.,1992, The San Andreas fault system in the vicinity of the Central Transverse Ranges Province, southern California: U.S. Geological Survey Open-File Report 92-354,49 p.

ammzarr arrnr A-65 uru m

McGill, J.T.,1989, Geologic maps of the Pacific Palisades area, Los Angeles, California: U.S.

Geological Survey Miscellaneous Investigations Series Map I-1828, scale: 1:48,000.

Merifield, P.M., Rockwell, T.K., and I.oughman, C.C.,1987, Slip rate on the San Jacinto fault zone in the Anza seismic gap, southern California (abs): Geological Society of America Abstracts with Programs, v.19, p. 431-432.

Millman, D.E., and Rockwell, T.K.,1986, Neotectonics of the Elsinore fault in Temescal Valley, Califomia, in Ehlig, P. (ed.), Guidebook and volume on neotectonics and faulting in southern California: Cordilleran Section, Geological Society of America, p. 159-166.

Mills, G., and Fischer, P.J.,1991, Thrusting associated with the offshore Newpon Inglewood-Rose Canyon zone, Califomia (abs): Geological Society of America Abstracts with Progmms, v. 23, no. 5, p. A198.

Monon, D.M., and Matti, J.C.,1987, ne Cucamonga fault zone, geologic setting and history of Quaternary faulting, in recent reverse faulting in the Transverse Ranges, California:

U.S. Geological Survey Professional Paper 1339, p.179-203.

Morton, P.K., and Miller, R.V. (compilers),1981, Geologic map of Orange County, California, showing mines and mineral deposits: California Division of Mines and Geology, Department of Conservadon, Scale 1:48,000.

Moyle, W.R.,1973, Geologic map of western parts of Camp Pendleton, southern Califomia:

U.S. Geological Survey Open-File Map,2 plates, Scale 1:62,500.

Mueller, K.J., and Rockwell, T.K.,1984, Sense, recency and rates of faulting along the Laguna Salada and Canon Rojo faults, NE Baja California (abs.): Geological Society of America Abstracts with Programs, p. 602.

Muhs, D.R., Miller, G.H., Whelan, J.F., and Kennedy, G.L.,1992a, Aminostratigraphy and oxygen isotope stratigraphy of manne-terrace deposits, Palos Verdes Hills and San Pedro areas, Los Angeles County, California, in Fletcher, C., and Wehmiller, J.F. (eds.),

Quaternary Coasts in the United States: Marine and Lacustrine Systems: SEPM Special Publication No. 48, p. 763-376.

Muhs, D.R., Rockwell, T.K., and Kennedy, G.L.,1992b, Late Quaternary uplift rates of marine terraces on the Pacific Coast of North America, southern Oregon to Baja California Sur:

Quaternary International, v.15/16, p.121-133.

Nanlin, T.R., and Henyey, T.L.,1970, Pliocene-Pleistocene diastrophism of Santa Monica and San Pedro shelves, California continental borderland: American Association of Petroleum Geologists Bulletin, v. 62, no. 2, p. 247-272.

tou):654 cart-arr.rxT A-66  % u. im

__. _ ___._..m _____..____ _ ._ _ __ __ _ _._._ _.._____. _ _ __

i National Research Council,1988, Probablistic seismic hazard analysis: National Academy Press,_ Washington, D.C.,96 p.

Patterson, A.C., and Rockwell, T.K.,1993, Palcoseismology of the WFNr fault based on 3-dimensional trench *ng at Olinda Oil Field, Orange County, southem California:

Geological Society of America Abstracts with Programs, v. 25, no. 5, p.131-132.

Pinault, C.T., and Rockwell, T.,1984, Rates.and sense of Holocene faulting on the southem Elsinore fault: funher constraints on the distribution of dextral shear between the Pacific and Nonh American plates (abs.): Geological Society of America Abstracts with Programs, v.16, no. 6, p. 624.

Ponti, D.J., and 12 joie, K.R.,1992, Chronostratigraphic implications for tectonic deformation of Palos Verdes and Signal Hills, l.as Angeles Basin, Califomia, in Heath, E.G., and Lewis, W.L (eds.), The Regressive Pleistocene Shoreline Coastal Southem California:

South Coast Geological Society, Inc.,1992 Annual Field Trip Guide Book No. 20, p.157-160.

Proctor, RJ., Cmok, J.R., McKeown, M.H., and Morseco, R.L,1972, Relation of known faults to surface ruptures,1971 San Femando canhquake, southem California:

l Geological Society of America Bulietin, v. 83, p.1601-1618.

Raleigh, C.B., Sieh, K.E., Sykes, L.R., and Anderson, D.L.,1982, Forecasting southern l California canhquakes: Science, v. 217, p.1097-1104. I l  !

Real, C.R.,1987, Seismicity and tectonics of the Santa Monica-Raymond Hill fault zone and l l nonhern 1.os Angeles Basin, in Morton, D.M., and Yerkes, R.F., Recent Reverse l Faulting in the Transverse Ranges, California: U.S. Geological Survey Professional Paper 1339, p.113-124.

Rockwell, T.,1989, Behavior of individual fault segments along the Elsinore-Laguna Salada  !

! fault zone, southern Califomia and nonhern Baja Califomia: implications for the characteristic earthquake model, in Schwanz, D.P., and Sibson, R.H. (eds.), Fault segmentation and controls of mpture initiation and termination: U.S. Geological Survey Open-File Repon 89-315, p. 288-308.

l l Rockwell, T.K.,1990, Determination of slip rates and dating canhquakes for the San Jacinto l and Elsinore fault zones,in National Eanhquake Hazards Reduction Program, Summaries of Technical Repons Volume XXIX: U.S. Geological Survey Open-File Repon 90-54, p.144-146.

Rockwell, T.K., and Pinault, C.T.,1986, Holocene slip events on the southern Elsinore fault, Coyote Mountains, southern California, in Ehlig, P. (ed.), Guidebook and Volume on

l. -

1 au mu.raran nr A-67 wy is, im

Neotectonics and Faulting in Southern California: Cordilleran Section, Geological Society of America, p. 193-196.

Rockwell, T.K., McElwain, R.S., Millman, D.E., and Lamar, D.L.,1986, Recurrent late Holocene faulting on the Glen Ivy North strand of the Elsinore fault at Glen Ivy Marsh, in Ehlig, P. (ed.), Guidebook and Volume on Neotectonics and Faulting in Southern Califomia: Cordilleran Section, Geological Society of America, p. 167-176.

Rockwell, T.K., Wilson, S., and Hatch, M.E.,1987, Ages and deformation of marine terraces

{

within the Aqua Blanca fault zone, northem Baja California, Mexico (abs): Geological '

Society of America Abstracts with Programs, v.19, no. 6, p. 444.

Rockwell, T.K., Loughman, C., and Merifield, P.,1990, Late Quaternary rate of slip along the San Jacinto fault zone near Anza, southem Califomia: Joumal of Geophysical Research,

v. 95, p. 8593-8605.

l Rockwell, T.K., and Lindvall, S.C.,1990, Holocene activity of the Rose Canyon fault zone in San Diego, California, based on trench exposures and tectonic geomorphology: .

Geological Society of America Abstracts with Programs, v. 22, no. 3, p. 78.

I i

Rockwell, T.K., Lindvall, S.C., Haraden, C.C., Hirabayashi, C.K., and Baker, E.,1991,  ;

Minimum Holocene slip rate for the Rose Canyon fault in San Diego, Califomia, in l Abbott, P.L., and Elliott, W.J. (eds.), Environmental Perils San Diego Region: San Diego l Association of Geologists, San Diego, Califomia, p. 37-46.

2 Rockwell, T.K., Lindvall, S.C., Haraden C.C., Hirabayashi, C.K., and Baker, E.,1992, Minimum Holocene slip rate for the Rose Canyon fault in San Diego, California, in Heath, E.G., and Lewis, W.L. (eds.), The Regressive Pleistocene Shoreline Coastal Southern California: South Coast Geological Society, Inc.,1992 Annual Field Trip Guide Book No. 20, p. 55-64.

I Rogers, T.H.,1965, Geologic map of Califomia, Santa Ana sheet: California Division of Mines and Geology, scale 1:250,000.

Sanders, C.O.,1989, Fault segmentation and earthquake occurrence in the strike-slip San Jacinto fault zone, Califomia, in Schwartz, D.P., and Sibson, R.H. (eds.), Fault Segmentation and Controls of Rupture Initiation and Termination: U.S. Geological Survey Open-File Report 89-315, p. 324-349, i Sauber, J.,1989, Geodetic measurement of deformation in California: NASA Technical Memorandum 100732,211 p.

1 Saul, R.B.,1979, La Nacion/Sweetwater faults, San Diego County, California: California Division of Mines and Geology Fault Evaluation Report FER-82.

muousuarr nrrm A-68 w u. m

t j Schwartz, D.P.,1988, Geology and seismic hazards: moving into the 1990's,in Van Thun, J.L.

(ed.), Earthquake Engineering Soil Dynamics II--Recent Advances in Ground Motion

! Evaluation, American Society of Civil Engineers Geotechnical Special Publication 20, j p. 1-42.

i 1 a

Schwartz, D.P., and Coppersmith, K.J.,1984, Fault behavior and characteristic earthquakes--

examples from the Wasatch and San Andreas faults

Journal of Geophysical Research, l v. 89, p. 5681-5698.

i l Schwanz, D.P., and Coppersmith, K.J.,1986, Seismic hazards--new trends in analysis using geologic data, in Active Tectonics: National Academy Press, Washington, D.C., p. 215-j 230.

t i Sharp, R.V.,1981, Variable rates of late Quatemary strike slip on the San Jacinto fault zone, i Southern Califomia: Journal of Geophysical Research, v. 86, p.1754-1762.

! Shaw, J.H.,1993, Active blind thrust faulting and strike-slip fault-bend folding in Califomia:

] unpublished Ph.D. dissertation, Princeton University, Princeton, New Jersey,216 p.

Shaw, J.H., and Suppe, J.,1994, Active faulting and gmwth folding in the eastern Santa j Barbara Channel, California: Geologial Society of America Bulletin, v.106, p. 607-626.

l Shlemon, R.J.,1992, The Cristianitos fault and Quaternary geology, San Onofre State Beach, California,in Heath, E.G., and Lewis, W.L. (eds.), The Regressive Pleistocene Shoreline j

, Coastal Southern Califomia: South Coast Geological Society, Inc.,1992 Annual Field i Trip Guide Book No. 20, p. 9-12.

)

i Sibson, R.H.,1982, Fault zone models, heat flow, and the depth distribution of canhquakes ,

! in the continental crust of the United States: Bulletin of the Seismological Society of I l America, v. 72, p.151-163.

i Sibson, R.H.,1984, Roughness at the base of the seismogenic zone--contributing factors: l Journal of Geophysical Research, v. 89, p. 5791-5799.

! Sieh, K.E.,1986, Slip rate across the San Andreas fault and prehistoric earthquakes at Indio, j California (abs.): EOS American Geophysical Union, v. 67, p.1200.

Sieh, K.E., and Jahns, R.H., III,1984, Holocene activity of the San Andreas fault at Wallace Creek, California: Geological Society of America Bulletin, v. 95, no. 8, p. 883-896.

4 i Sieh, K.E., and Williams, P. L.,1990, Behavior of the southemmost San Andreas fault during the past 300 yr: Joumal of Geophysical Research, v. 95, p. 6629-6645.

J i

cowaserr.arr.rxr A-69  % is.i m i

a P - e- - -

T .

Slemmons, D.B.,1977, Faults and canhquake magnitude: U.S. Army Corps of Engineers, .

Waterways Experiment Station, Miscellaneous Papers 5-73-1, Repon 6, p.1-129.

Slemmons, D.B.,1982, Determination of design canhquake magnitudes for micruzonation:

Proceedings of the Thirti International Eanhquake Microzonation Conference, v.1, p.119-130.

Slemmons, D.B., Zhang, P., and Mao, P.,1989, Geometry and displacement of the surface rupture zone associated with the 1954 Fairview Peak, Nevada, canhquake:

Seismological Research Letters, v. 60, no.1, p. 29.

Southern Califomia Edison,1988, Geology, seismology and geotechnical engineering: Updated Final Safety Analysis Repon (document file category "K"), San Onofre Nuclear Generating Station Unit 1, v.1, section 2.5.

Stein, R.S., and Yeats, R.S.,1989, Hidden canhquakes: Scientitic American, v. 260, no. 6,

p. 48-57.

Stephenson, WJ., Rockwell, T.K., Odum, J.K., Shedlock, K.M, and Okaya, D.A., in review, Seismic-reflection and geomorphic characterization of the onshore Palos Verdes fault zone, Los Angeles, California, Southern California Eanhquake Center Publication 96, 11p.

Suppe, J.,1983, Geometry and kinematics of fault-bend folding: American Journal of Science,

v. 283, p. 684-721.

Suppe, J.,1985, Principles of structural geology: Prentice-Hall, Englewood Clifts, New Jersey, 537 p.

Suppe, J., Chou, G.T., and Hook, S.C.,1992, Rates of folding and faulting determined from growth strata, M. McClay (ed.), Thrust Tectonics: Unwin Hyaman, Publisher, p.105-121.

Sykes, L.R., and Nishenko, S.P.,1984, Probabilities of occurrence of large plate rupturing canhquakes for the San Andreas, San Jacinto, and Imperial faults, Califomia,1983-2003:

Journal of Geophysical Research, v. 89, p. 5905-5927.

Tapponnier, P., Armijo, R., and Lacassin, R.,1989, Fault bifurcation and panition of strike-slip and dip-slip of mechanical interfaces in the continental lithosphere (abs.): International Workshop on Active and Recent Strike-Slip Tectonics, Florence, Italy, April 18-20, Universitf degli Studi di Firenze,34 p.

Tocher, D.,1958, Eanhquake energy and ground breakage: Bulletin of the Seismological Society of America, v. 48, no. 2, p.147-477.

cowaucarr.arrTxt A-70 wy 15. :9a

4 i

i Vaughan, P.R.,1987., Alluvial stratigraphy and neotectonics along the Elsinore fault at Agua i Tibia Mountain, Califomia: unpublished Masters Thesis, University of Colorado, l Boulder,182 p.

F

! Vaughan, P., and Rockwell, T.K.,1986, Alluvial stratigraphy and neotectonics of the Elsinore i fault zone at Agua Tibia Mountain, southern California, in Ehlig, P. (ed.), Guidebook '

and Volume on Neotectonics and Faulting in Southern California: Cordilleran Section, Geological Society of America, p. 177-191.  !

Vedder, J.G., Yerkes, R.F., and Schoelhamer, J.E.,1957, Geologic map of San Joaquin San l i Juan Capistrano area, Orange County, California: U.S. Geological Survey Oil and Gas i Investigations Map, DM 193.

l l Vedder, J.G., Greene, H.G., Clarke, S.H., and Kennedy, M.P.,1986, Geologic map of the mid-

! southem California continental margin, in Greene, H.G., and Kennedy, M.P., (eds.),

California Division of Mines and Geology California Continental Margin Geologic Map  !

l Series, No. 2A, scale 1:250,000 l

l Ward, S.N.,1990, Pacific North America plate motions: new results from Very Long Baseline l Interferometry: Journal of Geophysical Research, v. 95, no. B13, p. 21,%5-21,981.

4 i Ward, S.N, and Valensise, G.,1994, The Palos Verdes terraces, California: bath ub rings form l a buried reverse fault: Journal of Geophysical Research, v. 99, no. B3, p. 4485-4494.

l Webb, J.F.,1984, Geology of the San Diego margin, Carlsbad to La Jolla, San Diego County, California: unpublished M.S. thesis, California State University, Northridge, California.

i, Weichert, D.H.1980, Estimation of the earthquake recurance parameters for unequal

observation periods for different magnitudes

Bulletin of the Seismological Society of

America, v. 70, no. 4, p.1337-1346.

I

] Weldon, R., and Humphreys, E.,1986, A kinematic model of southern California: Tectonics, j v. 5, p. 33-48.

4

Weldon, R.J., and Sieh, K.E.,1985, Holocene rate of slip and tentative recurrence interval for i large earthquakes on the San Andreas fault, Cajon Pass, southern California

Geological l Society of America Bulletin, v. 96, p. 793-812. -

4 l Wells, D.L., and Coppersmith, K.J.,1994, Updated empirical relationships among magnitude,

{ rupture length, rupture area, and surface displacement: Bulletin of the Seismological

Society of America, v. 84, in press.

1 j Wells, D.L., and Lettis, W. R.,1990, Empirical assessment of earthquakes on reverse thrust j faults and surface deformation (abs.): Seismological Research Letters, v. 61, p. 23.

j J

J l a w a54o ure r n r A-71 wy is. i9w s

b

+-4w

Wesnousky, S.G.,1986, Eanhquakes, Quaternary faults, and seismic hazards in Califomia:

Journal of Geophysical Research, v. 91, no. B12, p.12587-12631.

)

Wesnousky, S.G., Prentice, C.S., and Sieh, K.E.,1989,' Offset Holocene stream channel and I rate of slip along the northem reach of the San Jacinto fault, San Bemardino Valley, I California: EOS, v. 70, no. 43, p.1349  !

Williams, P.L.,1989, Aspects of the earthquake geology and seismotectonics of the southern San Andreas and related faults, Southern California: Ph.D. thesis, Columbia University, New York,276 p.

Woodward-Clyde Consultants,1979, Report of the evaluation of maximum earthquake and site ground motion parameters associated with the offshore zone of deformation, San Onofre Nuclear Generating Station: prepared for Southern California Edison.

Working Group on California Earthquake Probabilities (WGCEP),1988, Probabilities of large canhquakes occurring in California on the San Andreas fault: U.S. Geological Survey Open-File Report 88-398,62 p.

Working Group on Southem California Earthquake Pmbabilities (WGSCEP), in progress, Siesmic hazards in Southern Califomia, probable earthquakes, 1994-2024.

Wright, T.L.,1991, Structural geology and tectonic evolution of the Ims Angeles basin, California, in Biddle, K.T. (ed.), Active Margin Basins: American Association 'of Petroleum Geologists Memoir 52, p.35-134.

Wyss, M.,1979, Estimating maximum expectable magnitude of earthquakes from fault dimensions: Geology, v. 7, p. 336-340.

Yeats, R.S.,1986, Active faults related to folding,in Active Tectonics--studies in geophysics:

National Academy Press, Washington, D.C., p. 63-79.

Yeats, R.S.,1973, Newport-Inglewood fault zone, Los Angeles basin, Califomia: American I Association of Petroleum Geologists Bulletin, v. 57, p.117-135.  !

Yerkes, R.F., and Lee, W.H.K.,1987, Late Quaternary deformation in the Westem Transverse l Ranges: U.S. Geological Survey Professional Paper 1339, p. 71-82.

Yerkes, R.F., McCullock, T.H., and others,1965, Geology of the Los Angeles basin, California: An introduction: U.S. Geological Survey Professional Paper 420A,57 p.

Youngs, R.R., and Coppersmith, K.J.,1985a, Implications of fault slip rates and earthquake recurrence models to probabilistic seismic hazard estimates: Bulletin of the Seismological Society of America, v. 75, no. 4, p. 939-964.

axmaumarr.arrxrr A-72 wy u, i9w

Youngs, R.R., and Coppersmith, K.J.,1985b, Development of a fault-specific earthquake recurrence model (abs.): Earthquake Notes, v. 55, p.16.

l Youngs, R.R., Coppersmith, K.J., Taylor, C.L., Power, M.S., DiSilvestro, L.A., Angell, M.L., 1 Hall, N.T., Wesling, J.R., and Mualchin, L.,1992, A comprrNnsive seismic hazard  ;

model for the San Francisco Bay region, in Borchardt, Glenn, and others (eds.), l Proceedings of the Second Conference on Earthquake Hazards in the Eastern San '

Francisco Bay Area: California Department of Conservation, Division of Mines and Geology Special Publication 113, p. 431-441.

Ziony, J.I., and Yerkes, R.F.,1985, Evaluating earthquake and surface faulting potential, in Ziony, J.I. (ed.), Evaluating Earthquake Hazards in the Ims Angeles region - An Earth-Science Perspective: U.S. Geological Survey Professional Paper 1360, p. 43-91.

Zoback, M.L., Anderson, R.E., and Thompson, G.A.,1981, Cenozoic evolution of the state of stress and style of tectonism in the western United States: Philosophical Transactions of the Royal Society of landon, Ser. A., v. 300, p.'407-434.

i w

l l

i ax m orrwrnr A-73 wy is, i9w

TACLE A-1. ESTIMATED FAULT RUPTURE PARAMETERS AND WEIGHTS' PROSABILTTY TOTAL RUPTURE DOWNDIP DOWNDIP FAULT NAME-(Model (Weight)] OF ACTIVITY SLIP RATE LENGTH (km) LENGTHS (km) WIDTH (lan) GEOMETRY (mmlyr)

Newport-Inglewood-Offshore Zone of (1.0) 148 32 (0.3) 12 (0.6) 90* 0.8 (0.2)

Deformation 43 (0.4) 15 (0.4) 1.5(0.6) '

[Model A (0.5)] 75(0.2) 2.1 (0.2) 116 (0.I)

Rose Canyon (1.0) 52 18 (0.2) 10 (0.6) 90* 1.0 (0.2)

[Model A (0.5)]  ?

34 (0.5) 15 (0.4) 1.5 (0.6) .

52 (0.3) 3.0 (0.2)  !

Newport-Inglewood Onshore (1.0) 70 30 (0.4) 12 (0.6) 90* 0.1 (0.3)

[Model B (0.5)) 40 (0.5) 15 (0.4) 0.8 (0.5) 70 (0.I) 1.5(0.2) e Offshore Zone of Deformation-Rose (1.0) I15 32(0.3) 90*

10 (0.6) I.0 (0.2)

Canyon 43 (0.4) 15 (0.4) 1.5(0.6)

[Model B (0.5)] 52 (O.1)

> 2.1 (0.I) 75 (0.I) 3.0 (0.1) -

4 a 115 (0.1)

Palos Verdes (l.0) 115 27 (0.3) 12 (u.7) 90* 2.0 (0.2)

Eastern Trace 46 (0.4) 15 (0.3) 3.0 (0.6)

[Model A (0.5)] 85 (0.2) 4.0 (0.2) .

115 (0.I)

Palos Verdes (1.0) 95 27 (0.3) s 12 (0.7) 90* 2.0 (0.2)

Westem Trace 46 (0.4) 15 (0.3) 3.0 (0.6)

[Model B (0.5)] 70 (0.2) 4.0 (0.2) 95 (0.I)

Coronado Bank (1.0) 215 50 (0.6) 90*

12 (0.7) 2.0 (0.4)  ;

90 (0.3) 15 (0.3) 3.0 (0.4) 125 (0.I) '

4.0 (0.2)

San Diego Trough (1.0) 175 60 (0.6) 90*

12 (0.5) 0.5 (0.2)

I15 (0.3) 15 (0.5) 1.0 (0.6) 175 (O.1) 1.5 (0.2)

Santa Catalina Escarpment (0.5) 100 40 (0.7) 90*

12 (0.5) 0.5 (0.2) 60 (0.3) 15 (0.5) 1.0 (0.6) 1.5 (0.2)

(Dt A)2564:TABt.E-1.RFT

i TABLE A-1 (Cont'd) r FAULT FROBABILITY total RUFIURE DOWNDIF DOWNDIF SLIP RATE i

[Medel (Weight)] OF ACTIVITY LENGIII Gesi) LENGTHS (kas) WIIFIll(kas) GEOMEIRY (misslyr) l San Clemente-San Isidro (1.0) 340 54 (0.4) 12 (0.5) 90* 0.5 (0.2) '

, 76 (0.3) 15 (0.5) 1.5 (0.4) 130 (0.2) 3.0 (0.3) 184 (0.1) 4.0 (0.1) i Elsinore (1.0) 190 35 (0.3) 12 (0.4) 90* 3.0 (0.2)  ;

50 (0.3) 14 (0.4) 5.0 (0.6) 80 (0.3) 16 (0.2) 7.0 (0.2) 130 (0.I)

Whittier (1.0) 32 15 (0.5) 12 (0.6) 90' l.5(0.2) 32 (0.5) 14 (0.4) 2.5(0.6)  !

3.5 (0.2) y Aguana-Agua Tibia. Earthquake (1.0) 84 10 (0.3) 12 (0.6) 90' O.5 (0.3) 4 Valley 23 (0.5) 17 (0.4) 1.0 (0.4) r

u 42 (0.2) 2.0 (0.3)

San Jacinto (1.0) 211 40 (0.4) 13 (0.5) 90* 7.0 (0.1) 90 (0.5) 15 (0.4) 9.0 (0.5) 1 130 (0.1) 17 (O.1) 12 (0.4)

San Andreas - Southern California (1.0) 445 115 (0.1) 10 (0.4) 90* 25 (0.6) 200 (0.6) 12 (0.6) 30 (0.4) 330 (0.3) ,

Malibu Coast-Santa Monica (1.0) 54 25 (0.5) 16 (0.3) 50*N 0.04 (0.2) -

30 (0.5) 20 (0.4) 0.4 (0.4) 25 (0.3) 2.0 (0.4) l Hollywood.Raymond (1.0) 40 20 (0.6) 16 (0.3) 50*N 0.04 (0.2)  !

40 (0.4) 20 (0.4) 0.4 (0.4) [

25 (0.3) 2.0 (0.4)

Sierra Madre (1.0) 85 17 (0.3) 17 (0.3) 50*N 0.5 (0.3) 30 (0.4) 20 (0.4) 3.0 (0.4) 45 (0.2) 25 (0.3) 4.0 (0.3) 85 (0.1)

(Dt.AM5KTABt.E-l.ItPr -

July 14,1994 ;

TABLE A-1 (Cont'd)

FAULT PROBABILITY TOTAL RUFI'URE DOWNDEP DOWNDIP SLIP RATE

[Model (Weight)] OF AC!1VITY LENGTH (km) LENGTHS (km) WIDTH (lum) GEOMETRY (mmlyr)

Peralta Hills-Norwalk 0.8 25 10 (0.6) 60*E 12 (0.4) 0.1(0.I) 15 (0.2) 14 (0.4) 0.2 (0.3) 25 (0.2) 16 (0.2) 0.5 (0.5) s.5(0.I)

Cucamonga (1.0) 25 15 (0.6) 17 (0.3) 50*N 4.05 (0.2) 25 (0.4) 20 (0.4) 5.0 (0.6) 25 (0.3) 6.0 (0.2) i Temescal (0.2) 29 10 (0.7) 13 (0.7) 90' O.01(0.3) 29 (0.3) 17 (0.3) 0.05 (0.3) 0.1 (0.3) 0.5 (0.1) y la Nacion (0.1) 27 13 (0.7) 12 (0.5) 75'W 0.01 (0.4) 4 cx 27 (0.3) 15 (0.5) 0.05 (0.4) 0.1 (0.2)

(Dt.A)2%4 TABLE.1 RPT July 14,1994

TABLE A-1. (cont'd)

BLIND THRUST PROBABILITY TOTAL RUPTURE DOWNDir DOWNDIP SLIP RATE SOURCEZONE OF AC'I1VITY LENGTH (km) LENGTHS (km) WIDTH (km) GEOMETRY (mm/yr)  ;

LA Basin Source Zone - B (0.2) 85 20 (0.4) 8 (0.4) 25*SW 0.01 (0.2) 40(0.4) 12(0.4) [Model A(0.5)] 03(0.4) 60(0.2) 16(0.2) 25*NE - 0.8 (0.4)

(Model B (0.5)]

LA Basin Source Zone - A (1.0) 28 8 (0.3) 8 (0.4) 25*N 0.2 (0.5) 16(0.5) 12(0.4) 0.5 (0.3) 28 (0.2) 16(0.2) 1.5 (0.2)

REGIONAL AREALSOURCE PROBABILITY TOTAL MAXIMUM DOWNDIP DOWNDIP ZONE OF ACTIVITY LENGTH (km) MAGNITUDE WIDTSI(km) GEOMFTRY SLIP RATE O Central LA. Basin [Model A (0.8)] 1.0 45 5.5 (0.2) 12(0.5) 90* based on 6.0 (0.4) 15(0.5) seismicity 6.5 (0.4) ,

Peninsular Ranges [Model A (0.8)] 1.0 180 5.5 (0.2) 12(0.5) 90* based on 6.0 (0.6) 15(0.5) seismicity 6.5 (0.2)

LA. Basin and Penmsular Ranges 225 5.5 (0.2) 12(0.5) 90* based on

[Model B (0.2)] 6.0 (0.6) 15(0.5) seismicity 6.5 (0.2)

Offshore Basin 1.0 135 5.50.2) 12(0.5) 90* based on 6.0 (0.6) 15(0.5) seismicity 6.5 (0.2)

' Probability weights shown in parentheses after parameter values. Individual rupture scenarios are discussed in text..

(DLA)2564 TABLEIJtPT October 25,1994

-__ . . _ _ . . _ _ _ _ _ _ . . _ ._. _ ..___ _ _ __. m . _ . . _ _ _ . _ _ _ _ . _ . _ . _ . _ _ . - . . . _ . _ _ _ . .

l f

A I

Segmentation Geometry Downep Wide R o ve Lange Siip Rete i (km) (km) (mm4r) i f 0.8 1

@.2) j 12 32 1.5 _

j

' _ - , :1;r-;_ _. : Str* Hep (90')

/ (0.5) (0.3) \ (0.8)

SCoZD /

\

g g, l

I0)

\ 15 (0.5)

Wd) 75

@2)

.. M.2) -

(o.s) 118 (0.1)

't 10 j / (0.8) 18 1.0 1 Rose Canyon - SM (W) / 2 3

(02) (0.2) 1 (1.0) X l \ 15 34 1.5 _

4 (0.4) (0.5) (0.8) 1

52 _ 3.0 _

l (0.3) (0.2) i a

i 30 _ 0.1 (0.4) (0.3)

12 f 40 / 0.8

/

(0.8) (0.5) (0.5)

.'__ ; -;-;r-;_ . x-j Stnkreto (90*) /

(1.0) \N 15 (0.1)

^

(02)

(0.4)

Model 8 E _

(GE) (0.3) 1.0 _

a _

(0.21 (0.4) 1.5 _

~

l ,o g, (10.8)

$ SCOZD-Rose Canyon _

Str*>eep (W) > / (0.8)

T (0.1) 2.1 (0.1)

I (0,1) . 3.0 15 _

j (0.4) @' I 115 _

} (0.1)

Figure A 1. Logic trees showing ahomative models for the Newport-Inglewood SCOZD-Rose Canyon Fault Zones.

2722 A-78 1

EXPLANATION E San Onofre Nuclear Generating Statlan (SONGS) y ( l 100 keemeter redlue around SONGS site x

,, + /^ , N , I l **~-a'-" '

A1

\ [yA e e -e ,.,,-

g g At Centraf LA. boshi

/ \

A2 A3 Peninsider ranges onse ore boshi A2

> \ me . o. . . --

0 20 30 40 M teLES

+

to 6! r + ^3

+

\ 1, 0 .0 . - 40 = - m ----

\

N ,

x  % -

1 er + + + \ +

,,r

,,, ,,r ,,,

Figure A-2a. Line sources Indluded in seismic source model.

EXPLANABON E son onoa. wucse , c.neroimg station (soNos)

N f

• 17 100 kBometer rodlue around SONCS alle 16 e 14

/

\  % Lhe source Jr h '

4 y 1 Newport-Inglewood fault rone 20 2 58"th Co**l 0"'hore Zone of Deformotion (SC0ZO) g N 3 pose canyon fourt rene 4 Paloe verties fault

/ 32 5 6

coronado senke fouet San Diego Trough fault 7 Santo catosha Escorpment fault 8 San Comente - San leidro 5 NCS 11 2 g 9 Debore fault zone 7 10 Whittler fault k

• 11 Aesongo-Aguo Tble fault

> \ 5

,, ,2 S. ~ t.

@~+ + + j 13 u

San Andreos

- m ,o - f t rene 15 Hosywood-Royrnand fault j

3 / 1.

jy me t... f-t r e g,,,,,,,, ,,,,, ,,,,

8 19 / jg 7,,,,,,,,,,,, j 19 Le Nacine faJt

_ .20 Perotte Hme-Norwelk fault a

s seur: i ooi - no nowwas

-+ + +r s I

+ a e io a m so a rs t

"r nr '8r l o io no ao .o so so a so nowans j b

Figure A-2b. Areal sources included in seismic source model.

\ .v i

~

\

. ' fr 4

'\ l17 W

Qo,

\

SOUTH LoS ANGELES BASIN

, SEGMDrr l .'s , @

':0; , #

9 -

M:: % /,6 a M DANA PODrr I

r e e.ee aere  : '- ,p/ g SaGMENT l w m.mo -

\

A' r k <

1

\\ s Ivapent  : dwCd / j SONGS s a

SONGS San Onofre nuclear tw

',/ B g

\ ' /s'

\to OCEANSmE SEGMENr generating station I se chee w.mo  ;

5 6 \

'A' cross sections E '

shown in Figure 2 k \\

A l af.,.n, H-* Cross sections shown o'm s'\

l in Figure 6 /' sucMExt g ,~ f N

.U N W /

SAN D! EGO

( , p f' SEGMENT

- N- e%joca  :

':f *,.',

,,na, I ,' c andr sessasig &amt 1,., ,s FeuerLemesAlack SE,VER N seGMEKr 0 10 20 30 40 % , // \

U

/

UNITED STATES MEXICO Figure A 3. The Newport Inglewood - SCOZD Rose Canyon fault zone, showing major fault segments (modified from Fischer and others,1992)

A-81 l

t i

,. . . . _ . _ . _ . .___m . ._ . 2-- . . . - , . _ _ _ _ _ _ _ . _ _ _ . . _ _ _ . _ _ _ _ _ _ . . . _ . . . . .. . . . - . . _ _ . . _ _ _ - _ _ . .

i __ _-

a't 1

i.

1 g3 M4C k 1 4'

i

.u :aC-= --i __ _ .-

s 4 ** . . . .,.::::p. .- -

1

.*.." ,:r g.pp.

,,. (>):;d.t:: .,,.,%s:.;*=: **,

~~....,.).-

rr .

s, .. p .

?

=

,<..?,.. .

h '.~~~...***p=t".:

f . .:: t... .. .

• r. ,

) .t' Q. ; f **== ~~e .v..:::.:

'%# ======***

. . . . -u S

_ g.p f.k.r, i!. l.55).f.;. .. ' ' " ' ' ------ ~~~~

m -,a .: .::y" 1

• 0 [ ~~~

V

: . ..y.e::. :.1:.:

. . y ,

y ':m 1 l j .) h{jhhhkib.v..;b . . .::Ri:::i? -

j -

[ ' '

y y n

[ w .. g l

[ -

i j, .. .

I . 3'

-u N 1

2::: 4 ".":- .: t

~

-u

/

i /' 4 t .

l i

t

= = 3J

j. r Figure 2b Cross secnon "B" along JEBCO CDP prcae 49 I

"8' 104 (off San Mateo Pouw).

Figure 2a Crose secoon T along Westem CDP proflie i W71 141 (off Rest Pont. soum of Corons Del Mar).

1 1

I t i

a j f 38 eA 4W _ _ _ _

-44 N64C z

4

- _._. 1 1 w i .** __ _Y f 4

4 no

,,,- . Ii .-- ...

1 l ,

, .., A.1 *gi e. . l!.

,,a""# * ~.;.

• es. et

=- .1

- 130p= *

'k:J':',.* ~.h f

***,,e**#*

,[{.. Id . .. 9N.-vrfrf!?I. (t.-[*!I!* *)s'.').**$*,*-}. *ll. ?!

t Y.f.***:I' .[.~O, . ," !

l:,,,.M.e5.lg$.$$h.$y:- D, '

. F~'il -i F, - \ fI I/

2,p. :.v.&.  : : ..:.::.:. .9 ... .s 1  :: . .-

F '

', as 2389- -d j- i 3

3 v;i v.,:.)...:(.<:f

. .: 1 -'

I a

,,, 's '

i/

.A i...: i.iiF. ii.*- i /

m u g;.,.7.g .

. . 7 ." . 3 .a

,,, , ,, /

u 1

e

..o a .e N /o i.o m==$=

mes . O oJ Miles ,.

0 0.5 Miles agy.

N,  % **

O *g *SA F'gure 8c. Cross secDon "C" along JEBCO CDP profWe 49 Figure 2d Cross secoon "D" along Westem CDP proftie 110 (off San Onofre). W71 109 (off Leucedia).

Figure A-4. Cross sections of the Newport-Inglewood - Rose Canyon fault zone form digitaly processed seismic reflection data. Locations are shown on Figure 4. (from Fischer and Mills,1991)

A-82

l l

l l

y D F SQWs sawg, $OWS Saws

~

\ X X X x

. \ cs r c, c, N ce jl i \

g EN tu - EN

-N- \

'% ' }&tu

\

th N

\pu m te s

u .

s u

u MOORE'72 ZIONY andothers,'74 GREENE and others,'79 LEGG,'79

~

w w

\

% , \ w saws sows 50 % s k sows Nx x N x x/( hx cs Cs cs

' . \ ce cx , EN Ex Ex S h~ -l u

u

.. e u e V s CLARKE,'80 CLARKE and others,'87 SOUTHERN CALIFORNIA MESA2INC.,1985; EDISON,1988 FISCHER AND MILLS,1991 ,

Figure A-5. Previous inte9retations of the Newport-Inglewood fault zone from 1972 - 1988. Location abbreviations are as fwws: DP, Dana Point; SONGS, San Onofre Nuclear Generating Station; OC, Oceanside; CB, Carlsbad; EN, Encinitas: DM, Del Mar; LJ, La Jolla; CF, Cristianitos fault; NIFZ, Newport-Inglewood Fault Zone; SCZOD, South Coat Zone Of Deformation; RCFZ, Rose Canyon Fault Zone; SOOF, San Onofre-Oceanside Fault. (modified from Fischer and Mills,1991)

I A-83

emunu:

32WPORT. .to) y nii 3 I . . .una 22,ur"Nh M, ,,

....; (  ?

' 'l-

'.'uBEACH c- uun , ..

5,(v  :'u"un : uu Mitt:,w"i s'.

,,",, ,uu:, -i. 4"unuuibuu:,u:M<g m .-

m:,, :mem: n ,tt .

,  : ,. u -/ nunmamu , u.:

u,, u, ,

. . u m m , m u- .

. l .r . ' 'umutu  :.

.- '

,K,,

u N -

• i - gA ,a,a:n mu n:m - a'ub , y

,,a u -

s

'- numg e m u. - . i,; \ ,,i i-. .umm 'uu:t.ut, unu, una ., ' 3 uum ,, .uc. uv ..

m:n unnuu, m !. -'

m,r d. ,> s. . ,. s , s sI,s nuvan. .s. e: . --

u,u,:,

y cy. u. _. nm.

..u, me . < u. ,

mu-qi moAg ,:r e n ".s -

i.

ne, y.

u, ,e u- a, ' ./' l ,3, s ..\ r

• poi .,3- ' :, .. o.

4

. .. i -

u,

\s\

't'\f..s\'::,p'-. ( d '\' ,  %

s

- 33* ri . . ,,o , , 1

, .i ..

33. ,

,,", .f s s . ,v.,s .. ., ,,

M , ' '. . ,,

!!,!,!,i.

• ' /

N.  : .N I;. : San ono6.

',,,,,;,;i:,'iil:

i: '?iiii SONGS i .* :n nu, e )b .

\'i ,

n'm:n, umm, yr -l ,

um m ,a .

quutm 'u3 '

\ \ :',:' :,'

>:':nmu,"uu, "

a

')

, "'"uun #

u 344ceg,'

1, "n:, : .

4

un u"r ,

t ':n , ' : u .'.

s u  ?

p \q N\

' 'l ,i CCEANSIDE l

f t  :," . .'

i

- 33* 1 7 .. A

,4 3

\1W r4:-

'7. , ' CARLSBAD sv te 4

4

' \;. -,

. C*

1 ENCINITAS

+

i i

\

.a I '8 DEL MAR 0 10 20 Kilometers 0

Csafoo TORREY 5 10 Mil. s rys i.

JoLLA Figure A-6. Geologic setting of the inner margin from the Palos Verdes Peninsula to the Silver Strand (from Fischer and Mills,1991).

A-84

l l

l 1

Regional Local 23 to 6 km s3 to 6 km 6km 3 0 __ ,

l --

O

- Tertiary

\g \ _

O Region of low

~ l

. -moment release 2

~ Localstrain partitioning W ea % akes

~

Basement - "'

5- e w .-

' k. '

-Transition zone r

. Brittle RW Mn partW wn of large ,

10- \ -

-moment release during earthquakes QQ,TQ.*Q9Q s

^ '

' 7j' g; Base of seistmges A'j?O'1LM'N'+*$}

'>y '74 5 , zone or change in}

~

• Uk' I f' ..

g.,%'6$XP'*!BD 4 ' u 'q:+v.t', ' 73;v .,

fy73;g{ rtwological property g3 3 $$ of crust .

'$5 $,5 g 5

15- Ductile 5<5 3 Delormation

< 5g 55 UNN 3 5 5 Oblique strainin hthosphere 3 $g 5

- oT Displacement

. ,o. _ ,

4$s5

hb)s,5,),$,}h j g Displacement --

5))f$f5I$,$,g 5

93

,$ away from viewer i

Figure A-8. Idealized section of upper lithosphere, illustrating features of regional an Regionally partitioned structures originate at crustal depths where large-moment-r and should be charactertzed independently as separate seismic sources. Locally partitio necessarily separate seismic sources and should be characterized collectively to a sources. (from Lettis and Hanson,1991)

A-85

OUTER INNER NI-RC

_, THRUST. FOLD COMPLEX THRUST-FOLD COMPLEX D s

r * ..:

.s a :d'

.f ,

t

-- 4r*, -

y r- c y g s g g - ..* f @ r :p .

. l - Q y _e  ! ._l

---. . m, _ -

/ ~. . .a-

<- I . g- _[

s %..,, y,v -# m,

~

r- .

.s s

u..

t e

.- est pi. MM,.t pen

\l- .. . -

as Figure A-7. Cross section along JEBCO digitaNy processed seismic reflection Ene number 49-102 off San Mateo Point.

San Mateo Point. Location of cross section shown on Figure 4 as a dashed une (from Hscher and Muis.1991).

e

i l

f I l

! l l

l l

l APPENDIX B l Maximum Magnitude Distributions i

i Prepared by  !

l

) Geomatrix Consultants, Inc.

i San Francisco, California i

i 4

{

4 i

i

i J

APPENDIX B l MAXIMUM MAGNITUDE DISTRIBUTIONS FOR SEISMIC SOURCES i

I

The maximum magnitude distributions for each seismic soume are given in this appendix. The distributions are developed based on the fault rupture parameters for each soume (Table A-1).

Specifically, the values of rupture length and downdip width are used with empirical regressions

, between subsurface rupture length and ruture area (Wells and Coppersmith, in press) to derive magnitude values. The weights associated with each magnitude value reflects the weights given to various parameter values. These maximum magnitude distributions are used directly in the

, development of recurrence relationships for the hazard analysis.

1 1

1 1

i B-1

Magnitude (M) Probability Distribution t

, ., ,.I ' '

i '

i i '

i '

i ' '

i e 4 i '

6

.6 Newport-Inglewood SCOZD (A) -

Palos Verdes Western Trace (B) --

San Diego Trough _

l .5 -

-[ -

y .4 -

O O .3 h .2 -

1 - - - - -

e -

g .,

~

0 ' ' ' ' ' ' ' ' '

.7 ,

h .6 Rose Canyon (Model A) -

Palos Verdes Eastern Trace (A) --

San Clemente-San Isidro -

r ,

j .5 - - - - - -

c -

m y .4 - - - - -- -

OO O .3 -

a - -

h .2 -

8 g .1 - - - - -- - -

o

, t , I- , , t , t . . t , l. I M, i , , t , t . . .01 ,

t

.7 . , ,

h .6 Newport-Inglewood Onshore (B) _ Rose Canyon SCOZD (Model B) - Coronado Bank -

3 lc .5 - - - - - -

y .4 - - - - -

o -

O ,3 _ _ - _ _ -

a C -

g ,2 _ _ - _ -

O e '; - - - - - - -

CL.

0 ' ' ' ' ' ' ' ' ' ' ' ' ' M ' ' ' ' ' ' ' ' ' ' ' ' '

5.5 6 6.5 7 7.5 8 8.55.5 6 6.5 7 7.5 8 8.5 5.5 6 6.5 7 7.5 8 8.5 Maanitude (M) Mannitude (M) Macnitude (M)

.7 -

Magnitude (M) Probability Distribution i

h .6 Whittier _

_ San Andreas-So. Californio _ _ Hollywood-Raymond -

C j .5 - - - - - -

c y .4 - - - - - - -

o o .3 - - - - - - -

a _ -

C o .2 -

o ,

.1 - - - - - -

I -

0 ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' '

.7 .

3 i 3 h .6 Elsinore - San Jacinto -

Malibu Coast-Santa Monico -

r a

p .5 - - - - - -

'C g y .4 - -- - - - -

6 o o ,3 _ _. _ _ _ _ _

"c e ,2 -

o -

L g .1 -

o . . .

if, i .,n.i. . . . . . . n. i . . . . . . . . . . .

.7 .

3

, . , ,,.3 . .

, . , . [ . , .

h .6 Santa Catalino Escorpment _

Aguono-Aguo Tibio-Earthquake Valley _ _ Peralta Hills-Norwalk -

z .

D -

eu '5 y .4 - - - - - -

O o ,3 _ _ _ _ _ _ _

"c _

o .2 - - - - - --

o -

g ,i _ _

g i . i . . . . i . . .f i . . i . i . . . i m i , i , i .

5.5 6 6.5 7 7.5 8 8.5 5.5 6 6.5 7 7.5 8 8.5 5.5 6 6.5 7 7.5 8 8.5 Maanitude (M) Magnitude (M) Magnitude (M)

5 8 _

i 8

) 5 )

, B

( d , A i 7 M

(

, e e n

n e o o d

, Z , Z ,

e e 1 6 7 t

u c c - i r

u -

r u

n g

o o

, S ' , S - i 5 a n - n f 6 M

, i , i s '

s - .

a a

, B , B i 6 A. A.

, L '

, L - ,

n i

o - - - _ - - l 5

5 t

u _ - - _ -

5 8

b i

t r

i s , ' ,

i 8

D ,

y 5 )

t ,

i l

i i

7 M(

b ,

e a d b ,

i 7 u o

r t

i P

1

- n g

)

i 5 a n 6 M M o - ' ,

l a

( i c c a s e , N

- , e 6

d m a e t

u ,

L ' ,

T .

i - - - _ - - - - - 5 n - - -

5 g 5 a - - - - - - - - - _ - -

8 M ,

, ' , i 8

5 )

M 7 (

e d

7 ut f - i L

~ n g

, o _ , er i 5 a g

n d

a 6 M o '

M - ,

m a

, o ' , r i 6

c r u i e

C '

. S .

- - - - - - - - - - - - 5 7 6 5 4 3, 2 1 0 7 6 5 4 3, 2 1 g 5 h 3 l t y o O >. h o .

$r3tyOo1e20

="

i i

i I;

l l

l l

l l

APPENDIX C Earthquake Recurrence Relationships Prepared by i

j Geomatrix Consultants, Inc.

I San Francisco, California l

l i

l i

i b

i 5

l APPENDIX C EARTHQUAKE RECURRENCE RELATIONSHIPS FOR FAULT SOURCES i

I The earthquake recurrence relationships for each fault source (including blind fault sources) listed i

in Table A-1 are given in this appendix. The plots show the cumulative annual frequency of

' occurrence of various magnitudes (or greater). Shown with the solid lines are the mean,5th and 95th percentile recurrence relationships predicted from fault slip rate and using the characteristic j canhquake magnitude distribution model of Youngs and Coppersmith (1985). The characteristic j canhquake model is used in the hazard analysis. Shown with dashed lines are the mean,5th, and

! 95th percentile recurrence relationships predicted usir.g an exponential magnitude distribution l model and fault slip rate.

i l Also shown as circles is the observed seismicity occurring within a narrow corridor along each j fault (see main text for discussion of seismicity and correction for completeness, etc.). The i circles are mean estimates of the rate of occurrence; the error bars reflect the uncertainties in-i asssessing the rate of occunence and increases with increasing magnitude and decreasing numbers of events. As can be seen, in most cases the rates of observed seismicity are well-constrained for small-magnitude canhquakes, and is unconstrained for large-magnitude events.

i The canhquake recurrence relationships developed from fault slip rate data is totally independent i from the recurrence rates of observed seismicity. Seisimicity data are robust for the small-magnitude earthquakes, while the slip rate data are derived from the geologic expression of 1

multiple large-magnitude canhquake events. Therefore, the comparison of the two data types

!- provides only limited information, and is not a sufficient basis to modify the predicted recurrence

relationships (based on slip rates) for use in the hazard analysis. However, a useful aspect of the

! comparison is the constraint that it places on the appropriate magnitude distribution model for l

~ use with fault slip rate. For nearly every fault, the characteristic earthquake magnitude distribution provides a better match with observed seismicity than the exponential distribution.

This result is consistent with conclusions found for faults in the San Francisco Bay area (Youngs i and others,1993) and confirms that the characteristic canhquake model is appropriate for use in i the hazard analysis.

i The recurrence relationships for the Newpon Inglewood-SCOZD-Rose Canyon, which is the fault most important to the hazard analysis, are panicularly instructive. Assuming Model A (where

, the Newpon-Inglewood fault is assumed to be independent of the SCOZD and Rose Canyon i

faults), recurence relationships are shown for the Newpon-Inglewood and SCOZD-Rose Canyon i faults. For Model B, the Newpon-Inglewood-SCOZD is presented seperately from the Rose

Canyon fault. In order to compare the predicted recurrence for the NI-SCOZD-Rose Canyon i zone across both models A and B, a comparison is provided. As can be seen, the combined i relationship compares quite well with the observed seismicity within the zone.

j The two blind thrust sources (which are modeled as zones in the hazard analysis, but the l predicted recurrence comes from constraints on slip rate on the blind thrust ramps) are also given.

j The predicted recurrence relationship for the Los Angeles Basin source A, which includes the j Whittier Narrows and Elysian Park thrust area, compares favorably with observed seismicity.  ;

l Los Angeles Basin source B, which includes postulated blind thrusts in the Palos Verdes Hills

C-1 l

1

l 1

l l

and Torrance-Wilmington fold trend, is also shown. The predicted recurrence underestimates the I observed seismicity. However,in this area the Palos Verdes fault is also modeled in the hazani analysis and may account from some of the observed seismicity assumed to lie within Los Angeles Basin source B.

Finally, a comparison is also shown between the predicted recurrence rate for all of the seismic j

sources in the region (both faults and background source zones) and the total observed seismicity within the study region. The predicted recurrence curves shown are the mean, the 5th and the l l 95th percentile using fault slip rate (i.e. the characteristic model). The predicted recurrence rates l l compare favorably with the observed seismicity. The majority of the observed rates are due to  !

the contribution of the fault seismicity.

1 l

C-2 l

10 g , , i , , _

m lN Nswport-Inglewood SCOZD (A)l s Rose Canyon (A)

O .* s _ .s d _\\ Characteristic s \s Characteristic e

s s . s NN

~

~

3 Exponential __\, \ s ---- Exponential _

s s _: -

Seismicity Seismicity

{ *N, @ 5 s, g @ 5 u a \\ -

_, ,s  :

( NN _

N 's N ' _

-a 39-1 's ' 's _

~_

's 's d _

sN-  : -

N \sN i

)\ '  %

s T ~

i

( )

N\\ %  %

f 1

l Z -

5,\ -

\,s\ -

l 4

o 10 ~'

E 2

Nb) Y 1

N,\

\'s s r 1

- s

% s i

g i

(

i g

- _ i .

h4 L

1. 10-3

{

\ ',6i, Mi i t

q p

g *',

i i

'4 l

l i

y g

i

_ g _ _ q y _

i, ' . . ',' _

. 10-' '

l 10 _

N i Newport-Inglewood (B) i h Rose Canyon SCOZD (B) E 4

6  :  : i\

1. _\ Characteristic - _s  % Characteristic -

o .s

! # 1 (

A\ ---- Exponential _

__ '\,s N ---- Exponential _

& j\ \ @ Seismicity Ns \ Seismicity

< o _ s- s [  : s @  !

hN L _ s N

(

_ s s

(

) s

\s s%

D 10 -'

's \

(hh N \'s -

P i

_N s s .

E

)

's s

's\s s E s

s

_ ss *()s s

s ss -

s . _ s ss g '

N s No _

g s, _

o 10-2 _

's x 'N' \ o N 's 's _

P Ns ,

s E  : Ns, E

. e s

: ss  :

s

% s .i

s

\s $ - ~

-.d ' s

'c N

, o '

1. 10-3 r ,',  ;  ; i ',  :

J s

5

's s

i i

E j _

, _ s I _ _  %\ _

j h -

g 1 - -

I g g -

' ' ' 'J ' '

1 10-* - '

3 4 5 6 7 8 9 3 4 5 6 7 8 9 Magnitude Magnitude C-3

10 .

i ,

s i i > i .

s si i i e i  :

A a, N Palos Verdes (A)  :

-\\ San Clemente-San Isidro  :

o -

s s s s

s s

s -

Z -

s s Characteristic s s

s Characteristic e s s s s

1. 1 L NN ---- Exponential __ \ 's ---- Exponential _

& ( '

's \ @ Seismicity i 's, \ Seismicity 5 Q) -

s, N  :  :

s s s Ns

,)N s s s - -

-s s

'g (

(5 -

()\ ' 'g \ .

~

e 10-i y

7 ..\,\, , _

.s\

s

.s ,

\,\

,s E

s\

s s E

s g

\ - -

s - _

s O

s O ) *

.g o

10-2 N 's g\

s 5

\ ,',

\

N sd A

E y g - -

s, s, q -

~

d -

l ', -

s

s. s

1. 10-3 i i

, 7 g i* i, ,

i i,  :  :  :

$  : i i,  :

- ,i g - i 8, i - -

i i

.i -

Q -

i' -

j

• 10-' ' ' ' ' ' ' ' ' '

10 _q , , , ,

A ;s "s Ns Palos Verdes (B)  :

is sNs s Coronado Bank o  :

1. ~ 'ss\

Characteristic 's \ Characteristic 5 e s s

1. 1 \s ----

Exponential _ --

's N ---- Exponential

> ( 6 's\,'g Seismicity '\ , \

q) -

s 6 s

@  ! g 'N s s

@ Seismicity  !

s, \ s b -

N s s,\

( ,

)N s s -

,s n -

's'\ss -

' , s'y ss f r ss - - s% -

()

N s\

Y* s ss s,

i 10-2 7 \n" g\ ,

t)  :  :  : '

9 e  : s s,  :

'ss , . -

\ ss,

~

d -

I g i

g i

1. 10-3 p , i

', i ,

k  : i

'.  ! 2 ' '

i i  !

1. - i i - - -

O i

i, 10 -'

3 4 5 6 7 8 9 3 4 5 6 7 8 9 Magnitude Magnitude l

C-4 l l

k

'?

10 , , , , , .,,

7, , , , ,

9 Whittier  :  :

- Ns N Elsinore  :

6 lg't

~

s

{# 1 r

0 Chorocteristic Exponential _

6 's, N

\, N ----

Chorocteristic Exponential _

{ 'g D

e, @ Seismicity  ! EN,. s, 'N @ Seismicity

~

k r q . - s .

4 < ) )s, t g - -

's,s\ -

"D 10-1

()

\"

5 t

I g

3  :

\'s ss i

~

g _

N "tgq [ E () 's 'y i

- r - -

g e N,

\, .

9 j o -2 , N s 'g 's \

4 e -

s  :  :

1

, s  :

"P  : Ch  ! i \ ' .'s E U - '

\ 4, d -

s i

1. 10-3 r i g , 7 s ', ,

( E l

i E

i ',

E

. - s, .

_A l l I l I l i I l . t i 10 _ , , , , , _ , , , , , _

p)

~

Eg Son Diego Trough [  ! Santa Catalino Escorpment I O _s . >

$ i s \s Chorocteristic - \ Chorocteristic $

e s 1

bs\\ ---- Exponential _. r N ---- Exponential _.

N\ @ Seismicity  ! fs, \ @ Seismicity [

P 5

c

-s

. s, s s

g

's g -

k 's () s\

g

( ) 's \ -

% s s s s

e 10 3 m s, g $ ,

y 2 sI \N 2  : 's , \N E g -

% NN  :  : 3 s, 's  :

g _

N Nq . -

g\ _

9 10-2 ,

'g's 's

, N __

e  :

s\ E  : \s E P  : s \s s  :  : '

\  :

g _

g _ _

s _

~

e -

ss 's 1 's,,

's -

$ 10-3 jr \ ,'

_ $ s s k  !

\\s, i

\' i

' t,

(#) i i

' l ' ' ' ' ' ' ' ' '

10-'

3 4 5 6 7 8 9 3 4 5 6 7 8 9 Magnitude Magnitude c-5

4 d

10 :s rs , i i i i  : ,

's is i i i i  :

A 's N Ron Andreas  : A N N San Jacinto  :

O

~

\N

~

< b \\

g _

, NN Characteristic _ -

, Ng Chorocteristie

1. 1

's s 's N---- Exponential N NN ---- Exponential ss ss s

& E Ns ', N@ Seismicity  : "s s Nq @ Seismicity o E E

b _

\ss\ _

\ss\ _

4 - s \s _ _

s \s _

'[

]g 10-' 7 ( \ 'El, -

p h l, e

g

\ N,O",' s,  :  : Qv,  :

Y so '

q 's 10-2 ) s, , _

\,

t, g g e V.,  !  !  !

d -

i -

i ,

~3# 10- r i

r i

i i

i

E E

' i' E

$  :  :  : i i  :

d

: i  :

i 10 -' -

10 g , , , , , , , , , ,

2 }\ Aguono-Aguo Tibio-Earthquak:eWhgy Malibu-Santa Monico  :

0 _s N  : :s -

$ \, 's Chorocteristic

\ Chorocteristic e

-s , s y j _ \\ ----

Exponential . .. N N ---- Exponential _

g j,,s 's, 'N @ Seismicity  !  : N, 's @ Seismicity  !

g NN m  : -

N N  :

4 _ ,

s s, N s s s,

s Ns _

]9 10-' r 'c \\ --

r 's N -

N

\ E Es \\ i C

\, , f f\f \, \, f 9 10-2 Os s\ \s '

s s ,

e  : s s  : _-

s ss  :

w P

%e E

\ s\s ',

s E

\s \\ s i

i Ns i s

~

e -

i i i i s

ss -

si

1. 10-3 r

i i

r Ns i i -_

i i  : _

s i  :

i s y - i i

i N

i i

C) -

i N '

10-' '

3 4 5 6 7 8 9 3 4 5 6 7 8 9 Magnitude Magnitude c-6

._ . . - .-. . .. . .. . - - - . _ . .. .-.. - - _ ~ .- - .. -..

1 1

4 10 -

i i i e i i i i

s _ :s si i _

Hollywood-Roymond s, Sierra Madre  :

r3

N ~

N O  : N _ s, N  :

1 , s s

- Characteristic - -

s s

s s

Characteristic -

y j ;N N ---- Exponential _ \ \ ----

Exponential _

& i \, \ @ Seismicity A 's \ @ Seismicity i 1, s e -

s s s -

J 5  : Ns \s  :  : N \\  :

4 s s

N, \, s s

] 10-1 m N \ - -

N- \ --

i

< > g \ i o N \s 2

A s s ss -

, # ~ \O \  : ON ss  :

g s s, _ _

Ns ' ,'s _

s - s g 10-2 i

3 g gz _

3 y \  :

e s s s s ,,

p -

- Ns s s s s -

g g s 9 - -

ss -

s

%a s g i s s gn

~

G - - -

g i i i s , ,

1. 10-3 r N s s , 7 i 5, ', -

h 3

- 'd, 's, 's, 3 i

,', E h - g g g - . l I

O -

\ 5, i, - - ', i i i is ., , i i i ii ,

3 0_.

10 .

i i i i i

, :s .s i i  :  : i i i  :

2 :N N Cucamongo  :

~

~

Temescal  :

O  : NN q) s s s s Characteristic 5 Characteristic 5 s t y \\ ' '

Exponential _ _

Exponential _

\ s s,sNs @ Seismicity [  ! @ Seismicity l 6 g  % g . . -

} - - -

{  %  % g "d 10-1 m s \s \ --

A

s i #  : N s,,  :  : N  :

1. : O N gs,  : [, N  :

O

~

' 's 's

~

9 10-2 __ i, , , ._. s, s.s ,

Q) i i' ', i  : \s  :

P

~

_ s. _

s, \, s d 4 g d

• g

- ~

s N t

\

p 10 3 r i i

s g

s s

N  : ',  : 5 i ,

_ , . _ s ,

, o# . .

s

. s, ,

s 4 I l 1 4 L l l I l I I 3 4 5 6 7 8 9 3 4 5 6 7 8 9 Magnitude Magnitude C-7

10 _

3 , , , ,

l A t

,s L.A. Basin Source A  : -

L.A. Bosin Source B  :

O _  :  :

's Chorocteristic - -

Characteristic o , s '

y j y, N ---- Exponential _ <, .x\ ----

Exponential _

> 3 \s 's @ Seismicity

~

N @ Seismicity  !

o s 6 s N _

. 'O 'y.s s

4 <3 \ N' .

N .

]g 10-' N -

- \, -

3-s ()'N

\  :  : s, N  :

O s .

s s s -

s.

s -

- s - _ s s .

' , s s 10-2 , s , s _ __ s _;

's  ! \ l

,d -

\g ( -

(s N

g -

l d

~e s i, g

s i

g 10-3 7 '

',,g r \ --

e  : -

s i  :

R  : i i\  :  : ' '  :

9 _

(

• _ _

' i i

( ', g 4

, . _ g .

10 -' '

10 e i

i i i

: i i i i i  :

m O

Peralto Hills-Norwalk  :  : La Nacion  :

' Characteristic -

Chorocteristic -

1. 1 -\ ' ----

Exponential , _

---

Exponential _

-s , N @ Seismicity

~

j @ Seismicity

~

s N , . _ .

4 gN _ , _

g 10-' 's 'N , ,

\

s, g s s, N  : as  :

g s s,,s s

. ,ss ss  :

() s s s s ( s 10-2 s

g g 's _

e a  : N ',  : \s  :

d -

\

s s sg - s, s s -

~

e _

N s

,g . . ,

,s .

, , s

1. 10-3 7 \ 5, ,  ; ,

, N

($  ! 's  ! N 's i 3 .

i, .

s s, ,

o -

\ -

s

, g _, , i i i i iN ,. i 3 4 5 6 7 8 9 3 4 5 6 7 8 9 Afagnitude Afagnitude C-8

Faults and Background Sources Study Region to i .

! D Predicted Predicted

@ Observed Fault Seismicity @ Observed Regional Seismicity ll] Observed Background Seismicity 0 [] (.

e 1 1 - - -

en - .

e b I e io-'

e

\

\ -

o

• N

[] O O e

p []

=ts

~

10-2 _

[] - - -

c 3 0  !

i i

10-3 ' ' -

3 4 5 6 7 8 9 3 4 5 6 7 8 9 l i

Magnitude Magnitude i

I i

i APPENDIX D Attenuation Relationships Prepared by Woodward-Clyde Consultants Santa Ana, California 1

l

1 l

TABLE OF CONTENTS 1

1.0 ATTENUATION RELATIONSHIPS

SUMMARY

. . . . . . . . . . . . . . . . . . . . . . . D-1 1

l 2.0 SELECTION AND EVALUATION APPROACH 1

OF ATTENUATION RELATIONSHIPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-2 {

2.1 SELECTED ATTFNUATION RELATIONSHIPS . . . . . . . . . . . . . . . . . D-2 2.2 APPROACH TO THE EVALUATION OF ATTENUATION RELATIONSHIPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-4 2.2.1 Weighting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-4 2.2.2 Adjustment of Standard Errors . . . . . . . . . . . . . . . . . . . . . . . . . . D-5 3.0 EMPIRICAL DATABASE AND SPECTRAL SHAPES  !

USED IN THE EVALUATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-6 3.1 EMPIRICAL DATAB ASE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-6 3.2 SPECTRAL SHAPE FROM NUMERICAL SIMULATION . . . . . . . . . . D 7 4.0 WEIGHT EVALUATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-9 4.1 WEIGHTS BASED ON EMPIRICAL DATABASE . . . . . . . . . . . . . . . . D-9 4.2 COMPARISON WITH SITE-SPECIFIC SPECTRAL SHAPE . . . . . . . D-11 4.3 OTHER CONSIDERATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-12 4,4 COMBINED ATTENUATION WEIGHT . . . . . . . . . . . . . . . . . . . . . . . D-12 5.0 ADJUSTMENT OF STANDARD ERRORS . . . . . . . . . . . . . . . . . . . . . . . . . . . D-13 5.1 STANDARD ERRORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D- 13 5.2 STANDARD ERROR ADJUSTMENT . . . . . . . . . . . . . . . . . . . . . . . . . D-13 6.0 STANDARD ERROR TRUNCATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-15 6.1 MAGNITUDE-DEPENDENT STANDARD ERRORS . . . . . . . . . . . . . D-15 6.2 BEHAVIOR OF UPPER TAILS OF GROUND MOTION DISTRIBUTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D- 16

7.0 REFERENCES

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D- 18 l

i

___m _ _ . _ . _ _ _ . . . _ _ _ _ . _ _ . _ _ _ . . . . _ _ _ . _ . _

TABLE OF CONTENTS (Cont'd)

I _IRT OF TABI FR D-1 Horizontal attenuation relationships weight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 D-2 Horizontal attenuation relationship by Idriss (1985,1987,1994) for stiff soil site . 21 D-3 Preliminary horizontal setenuation relationship by Abrahamson (1994) for soil sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 D-4 Horizontal attenuation relationship by Sadigh (1994) for soil site . . . . . . . . . . . . . 30 D-5 Horizontal attenuation relationship by Boore, Joyner and Fumal (1994) for the random oriented horizontal component of ground motion for soil sites . . . 33 D-6 Horizontal attenuation relationship by Campbell (1993,1994) for soil site . . . . . . 37 D-7 Vertical aaenuation relationship by Campbell (1990) for soil site . . . . . . . . . . . . . 41 D-8 Empirical database for horizontal spectral accelerations . . . . . . . . . . . . . . . . . . . . 44 D-9 Parameters used in ground motion simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 D-10 Attenuation relationship weight using the empirical database . . . . . . . . . . . . . . . . 47 I _IRT OF FIGURFR i

)

D-1 Idealized native soil shear wave velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 l D-2a Predicted median peak ground accelerations - Mw=5 . . . . . . . . . . . . . . . . . . . . . . 49 !

D-2b Predicted median peak ground accelerations - Mw=6-1/2 . . . . . . . . . . . . . . . . . . . 50 D-2c Predicted median peak ground accelerations - Mw=7-1/2 . . . . . . . . . . . . . . . . . . . 51 D-2d Predicted median 25 Hz (T=0.04 s) spectral accelerations - Mw=5 . . . . . . . . . . . . 52 ,

D-2e Predicted median 25 Hz (T=0.04 s) spectral accelerations - Mw=6-1/2. . . . . . . . . 53 l D-2f Predicted median 25 Hz (T=0.04 s) spectral accelerations - Mw=7-1/2 . . . . . . . . . 54 D-2g Predicted median 10 Hz (T=0.1 s) spectral accelerations - Mw=5 . . . . . . . . . . . . . 55 1 D-2h Predicted median 10 Hz (T=0.l s) spectral accelerations - Mw=6-1/2 . . . . . . . . . 56 D-2i Predicted median 10 Hz (T=0.1 s) spectral accelerations - Mw=7-1/2 . . . . . . . . . 57 D-2j Predicted median 5 Hz (T=0.2 s) spectral accelerations - Mw=5 . . . . . . . . . . . . . . 58 D-2k Predicted median 5 Hz (T=0.2 s) spectral accelerations - Mw=61/2 . . . . . . . . . . 59 D-21 Predicted median 5 Hz (T=0.2 s) spectral accelerations - Mw=7-1/2 . . . . . . . . . . 60 D-2m Predicted median 2.5 Hz (T=0.4 s) spectral accelerations - Mw=5 . . . . . . . . . . . . 61 D-2n Predicted median 2.5 Hz (T=0.4 s) spectral accelerations - Mw=6-1/2 . . . . . . . . . 62 D-2o Predicted median 2.5 Hz (T=0.4 s) spectral accelerations - Mw=7-1/2 . . . . . . . . . 63 D-2p Predicted median 1 Hz (T=1 s) spectral accelerations - Mw=5 . . . . . . . . . . . . . . . 64 D-2q Predicted median 1 Hz (T=1 s) spectral accelerations - Mw=6-1/2 . . . . . . . . . . . . 65 D-2r Predicted median 1 Hz (T=1 s) spectral accelerations - Mw=7-1/2 . . . . . . . . . . . . 66 D-2s Predicted median 0.5 Hz (T=2 s) spectral accelerations - Mw=5 . . . . . . . . . . . . . . 67 D-2t Predicted median 0.5 Hz (T=2 s) spectral accelerations - Mw=6-1/2 . . . . . . . . . . 68 D-2u Predicted median 0.5 Hz (T=2 s) spectral accelerations - Mw=7-1/2 . . . . . . . . . . 69 D-3 Comparison ofidealized native soil shear wave velocity with empirical database shear wave velocity range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 D-4a Magnitude and distance distribution of spectra data . . . . . . . . . . . . . . . . . . . . . . . 71 ii

I  !

TABLE OF CONTENTS (Cont'd) l l

D-4b Magnitude and distance distribution of spectra data . . . . . . . . . . . . . . . . . . . . . . . 72 i D-5 Normalized site-specific horizontal response spectra at 5%

damping - Mw=6- 1/2 and Mw=7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 D-6a Comparison of quantile plots of residuals for PGA - Strike Slip and oblique everits 74 D-6b Comparison of quantile plots of residuals at T=0.04 s - Strike-Slip and oblique even ts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 5 D-6c Comparison of quantile plots of residuals at T=0.1 s - Strike-Slip and obliq ue events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 6 1 l D-6d Comparison of quantile plots of residuals at T=0.2 s - Strike-Slip and oblique even ts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 D-6e Comparison of quantile plots of residuals at T=0.4 s - Strike-Slip and oblique even ts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 8 D-6f Comparison of quantile plots of residuals at T=1 s - Strike-Slip and oblique e ve n u . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 9 l D-6g Comparison of quantile plots of residuals at T=2 s - Strike-Slip and oblique evenu ......................................................... 80 D-7 Comparison of median residuals at diffeaent spectral periods . . . . . . . . . . . . . . . . 81 D-8 Weighting of deviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 D-9 Weighting evaluation procedure for median residuals . . . . . . . . . . . . . . . . . . . . . . 83 i D-10a Comparison of normalized horizontal response spectra at 5%

damping - Mw=6-1/2 -Site specific spectrum and median of five attenuations . . . 84 1

D-10b Comparison of normalized horizontal response spectra at 5% damping - Mw=7-Site-specific spectrum and median of five attenuations . . . . . . . . . . . . . . . . . . . . . 85 D-11 Typical relationships between standard errors and magnitudes . . . . . . . . . . . . . . . 86 D-12 Comparison of computed and attenuation standard errors at different spectral peri ods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 D-13 Standard error reduction approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 j D-14a Comparison of attenuation relationship and revised standard errors - Idriss (stiff soil) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 D-14b Comparison of attenuation relationship and revised standard errors - Abrahamson (soil) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 D-14c Comparison of attenuation relationship and revised standard l

errors - S adigh (soil) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 i D-14d Comparison of attenuation relationship and revised standard ermrs - Boore et al. (average class B and C) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 D-14e Comparison of attenuation relationship and revised standard errors - Campbell (soil) . . . . . . . . . . . . . . . . . . . . . . . . . -. . . . . . . . . . . . . . . . . . . 93 D-15a Comparison of computed and attenuation standard errors at different spectral periods -Idriss (stiff soil) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 D-15b Comparison of computed and attenuation standard errors at different spectral periods -Abrahamson (soil) . . . . . . . . . . . . ...................... 95 iii 1

TABLE OF CONTENTS (Cont'd)

D-15c Comparison of computed and attenuation standard errors at different spectral periods -Sadigh (soil) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 D-15d Comparison of computed and attenuation standard errors at different spectral periods -Boore et al. (sverage class B and C) . . . . . . . . . . . . . . . . . . . . . 97 D-15e Comparison of computed and attenuation standard errors at different spectral periods -Campbell (soil) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 D-16 Relationship between spectral acceleration, magnitude, and standard error . . . . . . 99 D-17 Residuals normal probability plot for spectral acceleration at 5 Hz . . . . . . . . . . . 100 l

l 1

l iv

i 1.0 ATTENUATION RELATIONSHIPS

SUMMARY

l 1

A study was performed to select horizontal and vertical ground motion attenuation relationships for use in the probabilistic seismic hazard analysis (PSHA) and to provide weights to be used with the selected attenuation relationships as part of the IPEEE study of the Units 2 and 3 of the San Onofre Nuclear Generating Station (SONGS). Ihe main emphasis of the study was :o evaluate the selected attenuation relationships using a set of empirical ground motion data appropriate for the PSHA study of the SONGS site. As part of the results of this study, adjustments in standard errors assocutted with the selected attenuation relationships are recommended.

The horizontal attenuation relationships selected in the study are those by Idriss for stiff soil site (1985,1987,1994), Abrahamson for soil site (1994), Sadigh for soil site (1987,1994), Boore et almsponding to the geometric average of their class B and C sites (1994), and Campbell soil site (1993,1994); the vertical attenuation relationship selected in this study is that by Campbell (1990) for soil site. Table D-1 lists the selected horizontal attenuation relationships and the corresponding weights to be used in the PSHA. Tables D-2 through D-6 respectively describe these attenuation relationships along with the recommended adjustments to standard errors, and Table D-7 describes the vertical attenuation relationship.

I i

D-1

2.0 SELECTION AND EVALUATION APPROACH OF ATTENUATION RELATIONSHIPS This section presents the attenuation relationships selected and provides the general approach used to develop weighting of the selected attenuation relationships and adjustments to standard errors associated with the attenuation relationships.

2.1 SELECTED ATTENUATIONRELATIONSHIPS Based on previous geotechnical studies performed by Woodward-McNeill for the SONGS site (Woodward-McNeill,1971), the ideahzed native soil shear wave velocity was estimated as shown on Figure D-1. It is noted on Figure D-1 that the site classification boundaries for site classes B and C according to Boore et al. (1993) are also indicated. On the basis of the site-specific shear wave velocity information on Figure D-1 together with the fact that the site is located in Cahfornia, attenuation relationships were selected from the published literature. The following attenuation relationships, also listed in Table D-1, were selected for horizontal ground motions:

1. Idriss stiff soil site (1985,1987,1994)

2. Abrahamson soil site (1994)

3. Sadigh soil site (1987,1994)

4. Boore et al. average site class B and C (1994)

5. Campbell soil site (1993,1994).

These attenuation relationships are described in Tables D-2 through D-6, respectively.

Comparisons of these attenuation relationships are shown on Figures D-2a through D-2u for different earthquake magnitudes at peak ground accelerations and spectral periods. The <

attenuation relationship for soil sites by Campbell (1990) was selected fo: vertical ground motions.

Table D-7 describes this attenuation relationship.

All of these authors have developed a number of attenuation relationships in the past. In selecting these attenuation relationships, these authors (Idriss, Abrahamson, Sadigh, Joyner, and Campbell)

I were asked to identify their most recent attenuation relationships appropriate for use in this study.

As a resuh of this process, Idriss and Sadigh suggested changes in standard errors associated with D-2 i

. . - . . . _ - - .. ~ -.- -. _-.- . - - - -..-. - .- ---.-

l their most recent published relationships, and these suggestions are reflected in Tables D-2 and D-4. t Based on the subsurface shear wave velocity shown on Figure D-1, the average shear wave velocay in the upper 100 ft at the SONGS site is at the low end of site class B, close to site class C. For this reason (and the presence of relatively low shear wave velocity values in the upper 20  ;

ft of the subsurface soil at the site), a geometric averara of the Boore et al. site class B and C was selected in this study. It also should be noted that & randard errors provided in Table D-5 for the Boore et al. relationship do not include the "o," term, which accounts for the use of the geometric mean rather than selecting randomly a component to be used in the estimation of the I attenuation parameters. The elimination of this term from the Boore et al. relationship provides consistency among the standard errors associated with all the attenuation relationships selected in this study because the term is not included in the other four attenuation relationships.

The attenuation relationship for peak ground acceleration (PGA) by Campbell and Borzorgnia  !

i l (1994) reflects recent worldwide recordings (1957 to 1993). However, it does not include response spectral values. Therefore, the response spectral values for Campbell used in the study is based on the Campbell and Borzorgnia relationship and those from the 1993 Campbell i relationship (Campbell,1993), where he provided equations for both the peak and spectral accelerations. The spectral shape is obtained by subtracting the logarithm of the PGA term from that of the spectral acceleration.

'Ihe Campbell attenuation relationship requires the depth to basement rock, which affects the distance used in the attenuation relationship. At the time of this study, distances used in the Campbell attenuation relationship were not published nor available to this study. Therefore, the distances were estimated based on the definition of the distance by Campbell using an assumed depth to basement rock of 2 km.

All the attenuation relationships considered in this study provide the following two ground motion parameters: (1) the median value of the spectral acceleration, and (2) a dispersion or standard i

i D-3 -

error about the median value. The dispersions associated with all the attenuation mlationships selected in this study except for the Boore et al. relationship is magnitude-dependent.

2.2 APPROACH TO THE EVALUATION OF ATTENUATION RELATIONSHIPS 2.2.1 Weighting The weight for each of the five horizontal attenuation relationships was developed based on the following three criteria:

1. Appropriateness of the attenuation relationship with respect to selected empirical ground motion data, and

2. Appropriateness of the attenuation relationship with respect to spectral shapes l reflecting the site-specific information.

3. Additional considerations.

The first criterion was the main focus of weight development. The second criterion was used only in a qualitative way without significantly affecting the weights developed. (As discussed i i Section 4, this was in part due to relatively small differences among the five selected attenuation relationships with respect to the second criterion above.) The third criterion, which reflects considerations that are not easily quantifiable, was used to make minor adjustments to the weights developed using the first criterion.

l In evaluating the first criterion, the empirical ground motion data were selected based on their conditions (magnitude, distance, site conditions, etc...) being similar to those that control the hazard at the SONGS site. The closeness of fit of the attenuation relationships to the empirical database willprovide the basis for attenuation weighting. The closeness of fit of the attenuation relationships to the empirical database developed for the SONGS site is measured by the closeness of fit of the median residuals. Definition of the residual quantity is given in Section 3.0.

The empirical ground motion database for soil sites located close to large, strike-slip canhquake events is relatively limited. For this reason, it is useful to supplement the evaluation of the attenuation relationships using the first criterion above by that using the second criterion above.

Therefore, the Band-Limited-White-Noise source model combined with random vibration theory D-4

l l

1 (Silva et al.,1990) was used to generate site-specific response spectra. Appropriateness of spectral shapes indicated by the attenuation relationship was evaluated by comparing them to the site-specific spectral shape thus generated. l 2.2.2 Adjustment of Standard Errors Because residuals are used to evaluate the weighting of the attenuation relationships, standard enors based on the site-specific database are computed and compared with those provided by the attenuation relationships. Wherever appropriate, standard errors of the attenuation relationships were adjusted for the purposes of this study in a reasonably conservative way to reflect the trends indicated by the available empirical ground motion data that are most appropriate for the conditions contributing most to the ground motion hazards at the site.

The empirical database used in the present study, and the site-specific numerical spectral shapes obtained are presented in the Section 3.0.

l l

I D-5

l 3.0 EMPIRICAL DATABASE AND SPECTRAL SHAPES USED IN THE EVALUATIONS 3.1 EMPIRICAL DATAB ASE The empirical ground motion data used to evaluate the horizontal attenuation relationships were selected to reflect the conditions controlling the ground motion hazard at the site. Although all the canhquake magnitudes and distances from all the seismic sources used in the PSHA contribute to the hazard at the site, nearest canhquake sources usually dominate the hazard at the site. In this study, the hazard at the SONGS site is dominated by magnitude 6 to 7 events on the Newport-Inglewood-OZD-Rose Canyon fault system at relatively close range. Funhermore, the SONGS site is a soil site with the shear wave velocity variation shown on Figure D-1. Accordingly, the following criteria were used to select an empirical database among the published canhquake recordings within California:

1. Recordings should be made at soil site stations.

2. Approximate magnitude range should be 6 to 7-1/4.

3. Approximate distance range should be 0 to 25 km.

4. Main style of faulting should be strike-slip.

The records selected as the myirical database used in the study are listed in Table D-8. This table consists of the following information: eanhquake event; date; the style of faulting; moment magnitude; recording station; distances to seismic source according to the closest distance (as used in the relationships by Idriss, Sadigh, and Abrahamson), the Joyner and Boore definition, and the Campbell definition; site classification according to Boore et al. (1993); horizontal spectral accelerations at selected penods; and reference. As discussed in Section 2.0, Campbell distances were estimated based on the Campbell's definition. In addition to canhquake events involving strike-slip rupture, the empirical database also includes two earthquakes with oblique rupture (North Palm Springs and Iema Prieta). Nine earthquakes were considered in the empirical datahm providing 27 recording stations involving strike-slip rupture, and 14 recording stations involving oblique rupture.

D-6

As indrated in Table D-8, the geometric mean of the horizontal spectral accelerations at 5 percent damping included in the empirical database consist of those for peak ground acceleration and spectral periods of 0.04,0.1,0.2,0.4,1.0 and 2.0 seconds (25,10,5,2.5,1, and 0.5 Hz).

To evaluate the sinnlarity between subsurface conditions at the various recording stations and that l at the SONGS site, the i i standard deviation shear wave velocity range about the mean shear wave velocity of the empirical database is shown on Figure D-3. Also shown on Figure D-3 are the idealiwd SONGS site shear wave velocity profile (Figme D-1) and the average shear wave velocity of the upper 100 ft layer at the SONGS site. As can be seen on Figure D-3, the shear wave velocity range cus+raling to the empirical database is reasonably consistent with the site shear wave velocity profile, indicating the adequacy of the empirical database with respect to its subsurface conditions.

The magnitude and distance distribution of the database recordings are shown on Figures D-4a and D-4b for the different distance definitions. As can be seen on Figures D-4a and D-4b, the criteria 2 and 3 above are reasonably satisfied by the empirical database.

3.2 SPECTRAL SHAPE FROM NUMERICAL SIMULATION To supplement the empirical database, numerical ground motion simulations were performed by Dr. Silva of Pacife Engineering using the Band-Limited-White-Noise source model combined with l random vibration theory (Silva et al.,1990). These simulations reflected site-specific data l

including the site shear was velocity profile (Figure D-1) to generate many response spectra. The parameters used in the simulations are summarized in Table D-9. The computed median sindation response spectra is normalind by the spectral acceleration obtained by averaging the spectral values corresponding to the periods or frequencies of interest to the study (0.1 to 1 I seconds or 10 to 1 Hz). The average spectral acceleration used in the normalization process is given by:

fSa,. Sara . Sa, . . Sa7 "

(D 1) h,- 3 I

4 D-7 l

where Sar is the computed spectral acceleration at the period T.

Using this average spectral acceleration, the normalized response spectra, Nar , is given by the following equation:

Sa'

+

Na,- (D 2) h, he normahzed simulation response spectra corresponding to earthquake magnitude 6-1/2 and 7 are shown on Figure D-5. Rese earthquake magnitudes were selected to represent events that 3 contribute most to the hazard at the site.

l l

l l

i 4

I T

l i

i s

4 5

D-8

l l

4.0 WEIGHT EVALUATIONS 4.1 WEIGHTS BASED ON EMPIRICAL DATABASE Difference between spectral acceleration obtained from the empirical data and that obtained from 1

an attenuation relationship (at the same period and damping) is often evaluated using the residual l value, r, defined by the following equation:

r In(SaJ -In(Sa.) (D-3) where Sa , = Calculated spectral acceleration from an attenuation relationship Sa. = Calculated spectral acceleration obtained from an empirical database.

It is noted in equation (D-3) that r greater than 0 means that the attenuation relationship used underestimates the spectral value from the empirical database while r less than 0 means overestimation.

Using the spectral values from the empirical database conesponding to the strike-slip and oblique events (41 data points), residuals as defined above were calculated for horizontal ground motions for the following conditions: peak ground acceleration and spectral values (at 5 percent damping) conesponding to penods of 0.04,0.1,0.2,0.4,1, and 2 seconds (or frequencies of 25,10,5,2.5,

! 1, and 0.5 Hz).

The quantity ofinterest for evaluating weights was the " median residual." The following steps were used to compute the median residual for an attenuation I, at a period T:

l

1. Compute rj for all the specified empirical data,f, as given in equation (D-3).

l 2. Order the values of rj in ascending order.

3. Compute the cumulative frequency distribution of the trsiduals (Benjamin and Comell,1970) in a quantile (fraction) form.

4. Compute the median residuals corresponding to an 0.5 quantile (a cumulative frequency of 50%). -

D-9

The quantile plots (ordered calculated residuals versus their fractions) are shown on Figures D-6a through D-6g respectively for peak ground acceleration and spectral values at 0.04, 0.1, 0.2,0.4, 1, and 2 seconds. In these figures the residual r values corresponding to a 0.5 fraction is then the median of the residuals.

I 1

I

• Ihe resulting median residuals for the period range ofinterest at the SONGS site, expressed in  !

percentage " deviation" from zero (negative being the overestimation and positive being the underes*imation), are plotted as a function of period for each of the five selected attenuation relationships on Figure D-7. As shown on Figure D-7, the negative median residuals mean i

overestimation and most of the relationships overestimate the median residuals throughout the i entire period range ofinterest except the Boore et al. relationship, which somewhat underestimates l

the median residuals at all but onc period. The median residual deviations (expressed in l

percentage) shown on Figure D-7 were used to develop weights for the five attenuation  !

relationships.

l In developing a weight for each attenuation relationship, the median residual deviations shown on Figure D-7 for various periods need to be " integrated" or " summed" into a single number. Such a summation should account for the following:

1. A median residual deviation being closer to zero is better than that having increasing negative or positive value. I

2. A negative deviation should be favored over a positive deviation because the former, which implies overestimation, in general is conservative than the latter, which implies underestimation.

Based on the above considerations with inputs from Dr. Abrahamson, SCE consultant to the project, the curve used as part summation process as " deviation weights" was judgmentally developed as shown on Figure D 8. According to this curve, a deviation weight of 1 is given if deviation is zero and, as the absolute value of the deviations increases, the integration weight becomes smaller until it becomes zero when the deviations are sufficiently high. As shown on Figure D-8, the distribution of the deviation weight is unsymmetrical with overestimation given higher weights than underestimation.

D-10

Figure D-9 illustrates the approach used to evaluate the attenuation weight using the results shown on Figures D-7 and D-8. As can be seen on Figure D-9, the median residual deviation from Figure D-7 was used to obtain the corresponding deviation weight from Figure D-8. Deviation values I

and the corresponding deviation weights are listed in Table D-10.

In the actual weighting process, normaliveA deviation weights were used for computational convenience. The deviation weights, DWu, for each period Tof each attenuation considered are given in column 5 of Table D-9. 'Ihe normalired deviation weight is obtained using the following equation:

DW

{D 7 (D-4) where, NWe, is the normr.lized deviation weight for attenuation I, at a spectral period T, The values of NWuare given in column 6 of Table D-10. Finally the attenuation weight, W,, of an attenuation, I,is computed using the following equation:

Wtrai.' Wtr4a,

• Wtr4= ' W tr.to.

(D 5) 3 The weighting r.bove was selected based on the periods ofinterest for this study. The resulting total weight for each attenuation relationship is listed in column 7 of Tal.!e D-10.

4.2 COMPARISON WITH SITE-SPECIFIC SPECTRAL SH APE Figures D-10a and D-10b compare median spectral shapes obtained from the attenuation relationships with those obtained from the site-specific simulations for magnitudes 6-1/2 and 7, respectively. Figures D-10a and D-10b show that the spectral shapes associated with various attenuation relationships are relatively similar and somewhat shifted toward lower period range compared to the site-specific spectral shapes. On the basis of tho results shown on Figures D-10a t

D-11 1

1 l

and D-10b, it was decided not to make any adjustments of weights among various attenuation relationships.

l Although the results shown on Figures D-10a and D-10b did not affect the total weights assigned l to the attenuation relationships, the comparison was considered useful as an overall check for reasonableness. He main conclusion from this comparison is the similarity of the spectral shapes exhibited by all the attenuation relationships.

1 4

4.3 OTHER CONSIDERATIONS

! The unit of total weight assigned to each attenuation relationship was judged to be no less than  !

0.05. Herefore, the total weights listed in the last column of Table D-10 based on the median

! residual deviatens need to be adjusted somewhat. In making the final minor adjustment a number of considerations were used with inputs from Dr. Abrahamson.

i i

ne Idnss attenuation relationship reflects his prior knowledge of the site subsurface condition and l

hisjudgment regarding the appropriateness of the attenuation to be used for the SONGS site. The Abrahamson attenuation relationship reflects most of the recent earthquake events. The Sadigh attenuation relationship does not reflect most of the recent earthquake events, but it does reflect hisjudgment regarding the appropriateness of the attenuation for use at the SONGS site based on his prior involvement in the project. In addition, his standard errors do reflect more current database. The database used in the Boore et al. attenuation relationship is limited mainly to I

earthquake events of magnitude 6.5 to 7.3 for site class B and C, but these ranges generally coincide with the ranges ofinterest to this study. The Campbell attenuation relationship reflects the uncertainties involved in determining the distances to be used in this relationship and the depth to basement rock.

4.4 COMBINED ATTENUATION WEIGHT On the basis of all the considerations presented in this section, with an emphasis on the median residual deviations as discussed in Section 4.1, the total attenuation weights in the unit of 0.05 were obtained as listed in Table D-1. -

D-12

5.0 ADJUSTMENT OF STANDARD ERRORS l

l 5.1 STANDARD ERRORS l Figure D-11 schematically shows typical standard errors associated with attenuation relationships.

In the Idriss, Abrahamson (except period of 2 seconds), Sadigh, and Campbell (PGA only) l attenuation relationships, they are magnitude dependent similar to the solid line variation shown l

on Figure D-11. The Boore et al., Campbell (other than PGA), and Abrahamson (period of 2 l seconds) relationships have magnitude independent variation similar to the dashed line shown on Figure D-11. ,

1 5.2 STANDARD ERROR ADJUSTMENT Figure D-12 shows two sets of standard errors plotted versus period. The upper plot on Figure  !

l D-12 shows standard errors (natural log) computed using the empirical database and the median

]

values of the attenuation relationships; the lower plot on Figure D-12 shows standard errors I (naturallog) associated with each attenuation relationship for magnitude 6-1/2 (corresponding to the average magnitude of the empirical database). Comparison of these plots indicates that the i standard errors associated with the empirical database, which reflects available empirical data most appropriate for this study, appear to be somewhat lower than those associated with each attenuation relationship. It is noted that attenuation standard enors for each relationship reflect the totality of the database used by their respective authors as opposed to focusing only on the l database most pertinent to the SONGS site study.

On the basis of the results shown on Figure D-12, some downward adjustments in standard errors of the attenuation relationships were considered appropriate. With the exception of the Abrahamson attenuation relationship, Figure D-13 summarizes the procedure used to develop the adjustments. As can be seen on Figure D-13, downward adjustments were introduced in a conservative way by ensuring the reduced standard errors to be at least one standard deviation away from the median standard error computed using the empirical database and the reduction at l

j high magnitude is no more than 30 percent. The adjustments to the Abrahamson attenuation relationship was provided by Abrahamson (1994).

D-13

Figures D-14a through D-14e show the adjustments made for each attenuation relationship; Idriss, Abrahamson, Sadigh, Boore et al., anri Campbell, respectively. On Figures D-14a through D-14c, l; it should be noted that the reductions in standard errors were developed based on the results at i

magnitude 6-1/2 and checked for reasonableness at magnitude 7.

l Figures D 15a through D-15e show the original standard errors associated with each attenuation relationship (in dashed lines), the revised standard errors associated with each attenuation

. relationship (in light solid lines), and the standard errors computed using the empirical database j and the median attenuation relationship (in dark solid lines) respectively for the Idriss, j Abrahamson, Sadigh, Boore et al., and Campbell relationships. These figures show that the

revised standard errors fall between the original and the empirical database standani errors in a i l reasonable way.

k j The recommended revised standani errors for each attenuation relationship to be used in the I i- present PSHA for the SONGS site are included in Tables D-2 through D-6 for the five

! relationships considered in the present study. Because the standard error adjustments were psfum.d using a short distance range empirical database, the revised standard errors should be used only when the Offshore Zone of Deformation is involved. For other cases, the original I standard error provided by the attenuation relationship should be used.

D-14

l i

6.0 STANDARD ERROR TRUNCATION I

\

In developing enyLical attenuation relations for response spectral values, the variability is typically assumed to be log-normally distributed. If the upper tails of such distributions are not truncated, however, it becomes possible to calculate unreasonable high spectral values. In addition, with the l recent trend of magnitude dapa=iant standard errors to express this variability, it becomes possible at low probability levels to have small magnitude canhquakes producing larger spectral values than larger magnitude earthquakes.

To avoid these unrealistic condinons, the following relationship was used in this study to truncate the upper tails of the distributions:

Magnitude Number of Sigma for Truncation less than 6 2.5 6 to 7 2.5 + (M 6) greater than 7 3.5 Background infonnation for the above relationship is provided below.

6.1 MAGNITUDE-DEPENDENT STANDARD ERRORS Recent analyses of the vanabihty of ground motion has shown that the standard error is dependent on the magnitude of the earthquake (e.g. Youngs et al.,1994). The standard error is larger for small magnitude events than for large magnitude events. A consequence of this magnitude dependence is that the attenuation relations may not be monotonically increasing with magnitude  !

at all levels of probability (e.g., at a given number of standard deviations).

To illustrate this point, the 5 Hz spectral attenuation relation for Sadigh (1987,1994) is shown on Figure D-16 for a distance of 10 km as a function of magnitude for various number of standard i deviations. Below 2o, the curves are all monotonically increasing with magnitude; however, i above 2o, this is not the case. Similar trends are found for other periods less than 1 second. At long periods (1-2 seconds), the magnitude scaling differences are larger than standard error differences so the attenuation relation remains monotonically increasing with magnitude up to 30.

D-15

l l

This non monotonic behavior of the attenuation at low probabilities (large number of standard 1

l deviations) is controlled by the tails of the log-normal distribution. The tails of the distribution are l

not controlled by the regression analyses that were used to develop the ground motion models.

Beyond 2o, the behavior of the tails has been assumed to follow a log-normal distribution.

6.2 BEHAVIOR OF UPPER TAILS OF GROUND MOTION DISTRIBUTIONS I

To discuss this point further, it is noted that the variability of ground motion can be separated into I source, path, and site contributions. The source variability can be separated from the total variability by the regression methodology: a Joyner-Boore two-step appreach or a random effects approach. The path and site variability are more difficult to separate and are usually grouped together. 1 l

The standard assumption in developing empirical ground motion attenuation relations is to assume that all components of the variability are log-normally distributed. With the large number of strong motion recordings currently available, it is possible to evaluate the behavior of the upper l tails of the distribution of ground motions directly from the data to test the assumption of a log-normal distribution.

The site and path variability of peak acceleration was evaluated using recordings fmm soil and rock sites from earthquakes with magnitudes greater than 5.2 and rupture distances less than 40 knt Sadigh (1987) for soil sites and Sadigh et al. (1993) for rock sites are used to compute the residuals for soil and rock sites. For both soil and rock sites, the Sadigh 1993 standard error model is used. The residuals for spectral acceleration at 5 Hz are compared to the log-normal distribution by using a normal probability plot on Figure D-17. In a normal probability plot, the normalized residuals are ordered from smallest to largest and are plotted against the expected value based on the total number of data points. If the normahzed residuals were truly log-normally distributed, th< n the data points wouki follow the 45 degree line. This figure shows that between

-20 and +2c,, the residuals follow a log-normal distribution; however, at >20, the residuals begin to depart from the log-normal distribution and appear to saturate below 2.50. Similar truncations D-16

I of the log-normal distribution at 2.5o were found for all of the spectral periods considered in this study (0,0.1,0.2,0.4,1.0, and 2.0 seconds).

l l

The souxe variability can be evaluated in a similar manner. However, there is not a large enough ,

1 number of earthquakes in the data set to test the distribution above the 20 level. It is noted that '

for the source and path variability, the total number of recordings was used, but for the source variability, the total number of canhquakes needs to be used.

l D-17 l

7.0 REFERENCES

Abrahamson, N.A. (1994). Personal communications.

Benjamin, J.R., and C.A. Cornell (1970). Probability. Statistics. and Decisinn for Civil Encrineers, McGraw-Hill Book Company, New York, New York,684 p.

Boore, D.M., W.B. Joyner, and T.E. Fumal (1993). Estimation of Response spectra and Peak Accelerations from Westem Nonh American Eanhquakes: An Interim Repon. Open-File Repon 93-509, U.S. Geological Survey, Menlo Park, CA.

Boore, D.M., W.B. Joyner, and T.E. Fumal (1994). Estimation of Response spectra and Peak Accelerations from Western Nonh American Eanhquakes: An Interim Repon, Pan 2.

Open-File Repon 94-127, U.S. Geological Survey, Menlo Park, CA.

Campbell, K.W. (1990). Empirical Prediction of Near-Source Soil and Soft-Rock Ground Motion for the Diablo Canyon Power Plant Site, San Luis Obispo County, Califomia, Prepared for Lawrence Livermore National Laboratory by Dames & Moore, Evergreen, Colorado, September 1990.

Campbell, K. W. (1993). Empirical Prediction of Near-Source Ground Motion From Large Eanhquakes. Intemational Workshop on Eanhquake Hazard and Large Dams in the Himalaya, Sponsored by the Indian National Trust for Art and Cultural Heritage (INTACH), January 15-16,1993, New Delhi, India.

Campbell, K. W., and Y. Bozorgnia (1993). Near-Source Attenuation of Peak Horizontal Acceleration From Worldwide Accelerograms Reconied from 1957 to 1993. Submitted to Fifth U.S. National Conference on Eanhquake Engineering, Chicago, Illinois, July 10-14,1994.

Idriss, I.M. (1985). Evaluating Seismic Risk in Engineenng Practice. Proc. Eleventh International Conference on Soil Mech. and Foundation Eng., August 12-16,1985, San Francisco, Califomia, vol.1, pp. 255-320, A.A. Balkema, Rotterdam.

Idriss, I.M. (1987). Eanhquake Ground Motions. Lecture Notes, Course on Strong Ground Motion, EERI, Pasadena, California, April 10-11,1987.

i Idriss, I.M. (1991). Personal communications. '

Idriss, I.M. (1994). Personal communications. l l

D-18 l 1

.=.__-- - _ - - ._- - . - . . _ _ . _ . . -.

{ Joyner, W.B., and D.M. Boore (1988). Measurement, characterization, and prediction of strong ground motion. Proc. Conf. on Earthquake Engineering and Soil Dynamics H, GT Div/ASCE, Park City, Utah,27-30 June 1988, pp 43-102.

Sadigh, K. (1987). Written communication to W. Joyner published in Joyner and Boore (1988).

Sadigh, K., C.M. Chang, N.A. Abrahamson, S.J. Chiou, and M.S. Power (1993). Specification of Long Pe:iod Ground Motions: Updated Attenuation Relationships for Rock Site Conditions and Adjustment Factors for Near-Fault Effects, Technical Papers on Seismic Isolation, ATC-17-1, pp. 59-70.

Sadigh, K. (1994). Personal communications.

! Silva, W.J., R. Darragh, C. Stark, I.G. Wong, J.C. Stepp, J.F. Schneider, and S. Chiou (1990). A methodology to estimate design response spectra in the near-source region of large canhquakes using Band-Limited-White-Noise ground motion model, Proc. Fourth U.S.

Nat. Conf. on Earthquake Eng., vol.1, pp. 487-494.

Woodward-McNeill & Associates (1971). " Elastic and Damping Properties, Laydown Area, San Onofre Nuclear Generating Station," Report dated October 14,1971.

D-19

TABLE D-1 Horizontal Attenuation Relationships Weight Attenuation Relationship Weight m Idriss Stiff Soil Site 0.15 Abrahamson Soil Site 0.20 Sadigh Soil Site 0.20 l Boore, Joyner and Fumal Geometric Average of Site Class B & C 0.25 Campbell Soil Site 0.20 Note: (1) Weights are given in 0.05 increment.

l l

l l

l D-20

1 l

TABLE D 2 Horizontal Attenuation Relationship by Idriss (1985,1987,1994) l for Stiff Soil Site Idriss (1985,1987,1994) derrved the following equation for horizontal peak ground acceleration of stiff soil sites: l in(PGA ) In [a(M)] - b(M) in(r.20) + 0.2F + a o a (D-6) where:

PGA = horizontal peak ground acceleration (g)

M = earthquake magnitude (M t for Ms6, M, for M>6; thus M represents moment magnitude Mw) r = closest distance to the source (km); for small magnitude earthquakes (Ms6) hypocentral distance is used F = style of faulting factor F=0.0 for strike-slip / normal fault F=1.0 for reverse / thrust fault F=0.5 for oblique fault a = number of specified standard error terms om = standard error tenn (natural logarithm) l l

l Values of the coefficients a and b are given in the following table. )

l l

M a b l usammmmmmmmmmmmm mmmmmmmu I l

4.5 606.0 2.57 5.0 617.0 2.46 l 5.5 452.0 2.28 6.0 282.0 2.07 6.5 164.0 1.85 7.0 91.7 1.63 7.5 49.8 1.41 8.0 28.5 1.21 8.5 15.9 1.01 D-21

TABLE D-2 (Cont'd)

Values for magnitude other than those listed above should be obtained through linear

, interpolation. Idriss proposed that peak acceleration from equation (D-6) be used to scale the response spectral shapes for different site conditions with magnitude- and period-dependent correction factors, Su and Sr, respectively. The spectral magnification factors at 5 percent

, damping for a stiff soil site and normalized to a magnitude 6-3/4 are given below.

I 1

f Sr l Period Spectral Magnification Factor (sec) for Stiff Soil Site at M=6-3/4

, PGA 1.00 4

0.04 1.08 i

0.10 1.82 i

0.20 2.72 0.40 2.60 1.0 1.13 2.0 0.52 I

j a

e i

i 2

1 D-22

TABLE D 2 (Cont'd) l The dependence of spectra 1 ordinates on magnitude normalized to a magnitude 6-3/4 are given in the following table.

y M Su Su Su Su Su Su j T_-0.04 T-0.1 T=0.2 T=0.4 T=1 T=2 1 see see see see see sec

ummmmmmmme mmmmmmmmme mummmmmmmmu 4.5 1 1 0.660 0.415 0.210 0.120 5.0 1 1 0.740 0.520 0.310 0.205 5.5 1 1 0.810 0.640 0.450 0.340 6.0 1 1 0.900 0.780 0.630 0.540  !

1 6.5 1 1 0.960 0.920 0.860 0.820  !

l 6.75 1 1 1 1 1 1 7.0 1 1 1.050 1.09 1.14 1.200

! 7.5 1 1 1.130 1.26 1.45 1.600 8.0 1 1 1.220 1.44 1.74 1.960 8.5 1 1 1.220 1.44 1.74 1.960 l As in the first table, values for magnitude other than those listed above should be obtained through linear interpolation.

Finally the pseudospectral acceleration, S. (in g), is obtained as follows:

S, PGA S, S u (D-7) where PGA is obtained through equation (D-6).

D-23

l TABLE D 2 (Cont'd)

The standard error terms recommended by Idriss (1994) are given in the following table:

1 l

Period (sec) Standard Error o,.m  ;

M < 7-1/4 M a 7-1/4 {

mmmmmmmmmmmmmmmmmmimumm-PGA 1.29 - 0.12M 0.42 i 0.04 1.29 - 0.12M 0.42  :

0.1 1.32 - 0.12M 0.45 l

0.2 1.37 - 0.12M 0.50 i 0.4 1.41 - 0.12M 0.54 1.0 1.47 - 0.12M 0.60 2.0 1.47 - 0.12M 0.60 The following is the recommended standard errors for the Idriss horizontal attenuation relationship i to be used in the present PSHA for the SONGS site only when the Offshore Zone of Deformation isinvolved:

Standard Error o,.m

_ Period (sec) M,, M < M ,, M a M,,

PGA 7.25 1.380 - 0.138M 0.38 I l

0.04 7.25 1.290 - 0.120M 0.42 0.1 7.25 1.320 - 0.120M 0.45 0.2 7.25 1.506 - 0.147M 0.44 0.4 7.1 1.833 - 0.205M 0.38 1.0 6.8 2.092 - 0.245M 0.42 2.0 7.I 1.943 - 0.21SM 0.42 D 24 1

TABLE D-2 (Cont'd) l' Reference Idriss, I.M. (1985). "Evaluat'ng Seismic Risk in Engineering Practice", Proc. Eleventh International Conf. on Soil!!e:h. and Found. Eng., August 12-16,1985, San Francisco, California,1, pp. 255-320, A. A. Balkema, Rotterdam.

Idriss, I.M. (1987). Earthquake Ground Motions, Lecture notes, Course on Strong Motion, l Earthquake Engineering Research Institute, Pasadena, California, April 10-17, 1987.

Idriss, I.M. (1994). Personal communication.

l l

l l

l l

i 1

l 4

D-25

TABLE D 3 Preliminary Horizontal Attenuation Relationship by Abrahamson (1994) for Soil Sites Abrahamson (1994) derived the following equation for horizontal peak ground acceleration at a soil site:

Peak Ground Acceleration The peak acceleration at the surface of soil site is modeled by:

.533 -0.400 M. [-1.006 0375 (M-6))h(r.5).0.25 F. s o for M < 6

.008 0.819M.[-1,712 0J52(M-6))h(r.15).0.25F.so for M a 6 where:

PGA = horizontal peak ground acceleration (g)

M = earthquake moment magnitude r = closest distance to the fault rupture surface (km)

F = style of faulting factor, F = 1.0 for reverse / thrust fault F = 0.5 for oblique fault F = 0.0 fcr strike-slip fault a = number of specified standard enor terms o = o,3, standard error term (natural logarithm) for horizontal peak ground acceleration.

e D-26

TABLE D-3 (Cont'd) t Spectral Ordinates The normalind spectral shape of soil sites is given by:

S c*

i 1 1. for T * < 1

• ) ' s t 8

s >

(D 9) l c . ( c . c,(8 - M)2)[In(T*)] . c [In(T*))8 3

for T* a 1 i

I l where T* is defined as:

l T,,,

T*

! T, l In(T,) . -2.274 - 0.146 (8-M)" + 0.338 In(r.50 )

l PGA = horizontal peak gmund acceleration (g) determined from equation (D-6)

! M = earthquake moment magnitude r = closest distance to the fault rupture surface (km) c, = 0.934 l c2 = -0.657 l c, = 0.0462 c, = 1.16 I c3 = 2.32 (=2c,)

c, = -0.0505 N, =3 N2 = 6 (=2N,)

The standarti errors for S, at various periods are given in the following tables.

l 4

D-27 l

t

TABLE D-3 (Cont'd)

Standard Error For Abrahamson (1994) Average Horizontal Spectral Acceleration Period Standard Error o wu, (sec) M s 7.2 M > 7.2 mmmmmmmmmmmmmmmmmmmmnummmmmmmmmmmmmmmmmme PGA 1.43 - 0.14M 0.42 0.04 1.46 - 0.14M 0.45 0.1 1.47 - 0.14M 0.46 0.2 1.48 - 0.14M 0.47 0.4 1.49 - 0.14M 0.48 1.0 1.43 - 0.12M 0.57 2.0 0.95 - 0.05M 0.59

'Ihe following is the recommended standard errors for the Abrahamson horizontal attenuation relationship to be used in the present PSHA for the SONGS site only when the Offshore Zone of Deformation is involved:

Period Standard Error o,.m.,

(sec) M s 7.2 M > 7.2 PGA 1.662 - 0.186M 0.32 0.04 1.692 - 0.186M 0.35 1

0.1 1.679 - 0.182M 0.37 0.2 1.735 - 0.191M 0.36 0.4 1.813 - 0.205M 0.34 1.0 1.807 - 0.195M 0.40 2.0 1.086 - 0.077M 0.53 D-28

TABLE D-3 Cont'd)

Reference Abrahamson, N.A (1994). Personal communication.

D-29 l

l

TABLE D-4 l Horizontal Attenuation Relationship by Sadigh (1994) for Soil Site l

l Sadigh (1987) developed peak horizontal acceleration and 5% damping horizontal response spectrum attenuation relationship using earthquake data from the westem United States supplemented by significant recordings of earthquakes at depth less than 20km from other parts of the world. Both horizontal components were used in his derivation. The following equation describes this relationship:

Im(Y). a . bM . c (s.5 Mf2 . din [r . h icxp (h2 M)] . In F . e og 3 (D-10) where:

Y = horizontal peak ground acceleration (g) or horizontal pseudoacceleration (g)

M = earthquake moment magnitude r = closest distance to the fault rupture surface (km) a = number of specified standard error terms 4 i

oc = standard error term (natural logarithm) l 1

F = Sadigh's attenuadon was originally derived for strike-slip events, Sadigh recommended the following values of F:

F=1.0 for strike-slip events F=1.2 for reverse-slip events For the present study, the following value of F was assumed for oblique events:

F=1.1 for oblique events.

Values of coefficients a, b, ci, c2, d, hi, h 2 , and o are given in the following tables.

D-30

t

TABLE D 4 (Cont'd) l i

Coefficients for Spectral Attenuation Relationship ( 5% damping) by Sadigh (1987,1994) for Strike-Slip Fault Eventsm t

Period M < 6.5 M > 6.5 PGA -2.611 0.000 0.8217 0.4814 0.3157 0.6286 i 0.04m -2.611 0.000 0.8217 0.4814 0.3157 0.6286 0.1 -2.024 0.007 0.8217 0.4814 0.3157 0.6286 j 0.2 -1.696 0.000 0.8217 0.4814 0.3157 0.6286 j 0.4W -1.610 -0.016 0.8217 0.4814 0.3157 0.6286 i

j 1.0 -1.975 -0.060 0.8217 0.4814 0.3157 0.6286 4

j 2.0 -2.414 -0.105 0.8217 0.4814 0.3157 0.6286 c

Standard Errors for the Sadigh Attenuation Relationship i

Period Standard Error [omm]m (sec) M < 7.0 M > 7.0 j -

2 PGA 1.52 - 0.16M 0.40 0.04m 1.52 - 0.16M 0.40 1 0.1 1.54 - 0.16M 0.42 s

j 0.2 1.56 - 0.16M 0.44

0.4W 1.60 - 0.16M 0.49 i

j 1.0 1.66 - 0.16M 0.54 i 2.0 1.66 - 0.16M 0.54 i

)

D-31

TABLE D-4 (Cont'd)

The following is the recommended standard errors for the Sadigh horizontal attenuation relationship to be used in the present PSHA for the SONGS site only when the Offshore Zone of Deformation is involved:

Period Standard Error [omm]m (sec)

PGA 7.0 1.590 - 0.174M 0.37 0.04 7.0 1.520 - 0.160M 0.40 0.1 7.0 1.540 - 0.160M 0.42 l

0.2 7.0 1.560 - 0.160M 0.44 0.4 7.0 1.807 - 0.201M ,1 0.40  !

1.0 7.0 2.067 - 0.241M 0.38 2.0 7.0 1.930 - 0.214M 0.43 NOTES:

(1) For all periods, b=1.1, c2=2.5, and d=-1.75.

(2) Sadigh (1994) Personal communication.

(3) Assumed to be the same as PGA.

(4) Response spectra at T=0.4s wer- determined by non-linear interpolation. l l

l Reference Joyner, W.B., and D.M. Boore (1988). Measurement, characterization, and prediction of strong ground motion. Proc. Conf. on Earthquake Engineering and Soil Dynamics 11, GT Div/ASCE, Park City, Utah,27-30 June 1988, pp 43-102.

Sadigh, K. (1987). Written communication to W. Joyner published in Joyner and Boore (1988).

Sadigh, K. (1994). Personal communication.

D-32

TABLE D-5 Horizontal Attenuation Relationship by Boore, Joyner and Fumal (1994) for the j Random Oriented Horizontal Component of Ground Motion for Soil Sites i

Boore, Joyner and Fumal (1993) modified the Joyner and Boore (1988) attenuation to include l reconfings from the 1989 Imma Prieta, the 1992 Petrolia, and the 1992 Landers canhquakes. In 4

4 1994, Boore, Joyner and Fumal funher modified their attenuation relationship to incorporate fault mechanism as part of the predictors. 'Ihe result of their 1994 study is presented in the following j equation for the random oriented horizontal peak ground acceleration and spectral acceleration for periods up to 2 seconds:

In(Y)- {bi .bfM-6).bfM-6)2.b / 3 sor .b,Gj.bf}c) In(10) s a w

.b 1og (D-11) where:

Y =

ground motion parameter to be predicted; horizontal peak ground acceleration (g) or spectral acceleration (g)

M = canhquake moment magnitude (5.0sMs7.7) i r =

(d2 + h2)i/2  !

d =

shortest distance, in km, from the recording site to the venical projection of the canhquake fault nipture on the surface of the canh (ds 80km) a = number of specified standard error terms o wn = standard error term (natural logarithm)

G ', =

site classification (=0.5 for average of class B and C used in the present study)

=

G'c site classification (=0.5 for average of class B and C used in the present study) b, =

bu Gu + bn Gn For strike-slip events, Gu = 1 and Gu = 0.

For reverse-slip events, Gu = 0 and Gu = 1.

Strike-slip events are defined as those with a rake angle within 30 degrees of horizontal. Reverse-slip events include all events where rake angle are above 30 degrees from horizontal. According to Joyner (1994), reverse-slip events also include reverse, reverse-oblique, and thrust events. Although the data set used by Boore, Joyner and Fumal do not include normal-slip events, the strike-slip coefficients should be used for normal events in the present study.

Values of coefficients b2 , 3b , b,, b3 , b,, b7 , bu, bu, h, and o wnare given in the following tables for the randomly oriented horizontal component of ground motion.

D-33

TABLE D-5 (Cont'd)

Coefficients for Calculating Response Spectra at 5% Damping for Random Horizontal Componentmm Period bss bas b2 b3 b b3 b. b, (sec) muummmmmmmmmmmmmmmum ---mmmmmmmm--

PGA -0.136 -0.051 0.229 0.000 5.57 -0.778 0.162 0.251 0.04m -0.136 -0.051 0.229 0.000 5.57 -0.778 0.162 0.251 0.10 0.436 0.471 0.327 -0.098 6.27 -0.934 0.046 0.136 0.20 0.433 0.507 0.309 -0.090 7.02 -0.924 0.190 0.279 0

Q 0.40 0.091 0.183 0.361 -0.052 4.91 -0.867 0.264 0.405 1.0 -0.493 -0.439 0.450 -0.014 2.90 -0.798 0.314 0.517 2.0 -0.739 -0.783 0.471 -0.037 5.85 -0.812 0.360 0.537 Notes:

(1) b, = 0.00 for all periods.

(2) Spectral accelerations for oblique events are obtained as the geometric mean of the spectral accelerations of strike-slip and reverse events.

(3) Assumed to be the same as PGA.

i l 1

TABLE D 5 (Cont'd)

! Standard Error for the Boore et al. Attenuation Relationship i

Period om l (sec) mummmmmmesummmmmme PGA 0.467 0.04m 0.467 O.10 0.440 0.20 0.435 1

0.40 0.454 '

1.0 0.520 l 2.0 0.566

)

The following is the recommended standard errors for the Boore et al. attenuaion relationship to be used in the present PSHA for the SONGS site only when tl e Offshore Zone of Deformation isinvolved:

1 Period ohm (sec) summmmmmmema m mmmme PGA 0.40 0.04 0.43 0.10 0.44 0.20 0.44 0.40 0.43 1.0 0.43 2.0 0.54 Note:

(1) Assumed to be the same as PGA.

D-35

. . _ . - __= . . _ - - - . . _ . . . . _ - _ . _ - . . - _ . . . - - - , _ . ,

TABLE D 5 (Cont'd) l Reference Boore, D.M., W.B. Joyner, and T.E. Fumal (1993). Estimation of Response Spectra and Peak Accelerations From Western Nonh American Earthquakes: An Interim Report, U.S.

Geological Survey, Open File Report 91-509.

Boore, D.M., W.B. Joyner, and T.E. Fumal (1994). Estimation of Response Spectra and Peak Accelerations From Western North American Earthquakes: An Interim Report, Pan 2.

U.S. Geological Survey, Open File Report 94-127.

Joyner, W.B., and D.M. Boore (1988). Measurement, characterization, and prediction of strong ground motion. Proc. Conf. on Earthquake Engineering and Soil Dynamics 11, GT Div/ASCE, Park City, Utah,27-30 June 1988, pp 43-102.

Joyner, W.B. (1994). Personal communication.

i l

D-36

i j TABLE D 6

) Horizontal Attenuation Relationship by Campbell (1993,1994) for Soil Site Peak Ground Acceleration Carmpbell and Bororgnia (1994) presented the following equation for the horizontal peak ground I

accelerations: '

l' 1

h(PGA ) -3.512 0.904 M - 1.328 in[h,* . [0.149 sup (0.647 M)]']

i

. [1.125 -0.112 in(R,)- 0.0957 M]F s o g D lh l where:

PGA =

geometric mean of the two horizontal components of the peak ground acceleration (g) l M = canhquake moment magnitude R, =

closest distance to seismogenic rupture on the fault (km). Campbell recommended that this parameter should not be assigned a value less than 3 km.

F = fault rupture mechanism F=0 for strike-slip, and nonnal faulting  !

l F=1 for reverse, reverse-oblique, and thrust faulting l a = number of specified standard error terms o m = standard error term (natural logarithm)

Although the dispersion (standard error) in the predicted value of PGA was found to be better contlated to the amplitude of the motion (PGA) than the canhquake magnitude,in the present study, the value of the dispersion (standard error) used is the one related to the magnitude of the canhquake as expressed below:

P 0.889 - 0.0691 M for M < 7.4 0.38 for M a 7.4 l

[

D-37 1

TABLE D-6 (Cont'd)

'Ihe following is the recommended PGA standard errors for the Campbell horizontal attenuation relationship to be used in the present PSHA for the SONGS site only when the Offshore Zone of Deformation is involved:

0.935 - 0.078 M for M < 7.4 O.36 for M 2 7.4 Sne.ctral Ordinatee The normali*cd spectral shape used in the present study corresponds to the one presented by Campbell (1993). 'Ihe normahzed spectral shape is obtained by dividing the spectral values by the corresponding PGA (Campbell,1993) for periods up to 4 seconds. The following equation of the normatired spectral shape is given by:

f h 3* i

$ PGAj p; . pitanh [0.647 (M-4.7)) - (p; . p}M)R, (D-13)

. pis . p;tanh (o.620 D) where:

S, = horizontal spectral acceleration (g)

PGA = peak ground acceleration (g)

M = momeat magnitude (Mw) (or local magnitude (Mt ) for M < 6 and surface wave magnitude (Ms ) for M a 6.0)

R, =

shortest distance between the recording site and the seismogenic rupture zone on the fault (km)

D = depth to basement rock (km). A recommended value of D of 2 km should be used in the present study S = local site coefficient (S=0 for Quaternary deposits (soil)).

i Although the normalired spectral shape developed in equadon (D-13) was based on the peak ground acceleration derived by Campbell (1993), in the present study, the spectral acceleration was obtained by multiplying the normahzed spectral shape in equation (D-13) by the peak ground i acceleration from equation (D-12).

D-38

b

'I i

l TABLE D 6(Cont'd)

Values of coefficients pi, pi, pi, pl, p!, pi, and standard enor for different periods are given in the following tables.

J Period p/ pi pi p/ p/ p/

f (se)

]

- ummmmmmme mummmmmmmmmmmmmmmmmmmmm i 0.04 0.01 0.00 0.22 0.00 0.00080 -0.000055 0.10 0.54 0.00 0.08 0.00 0.00240 0.000007 0.20 0.83 0.00 -0.21 0.00 -0.00110 0.000525 0.40 0.13 0.60 -0.46 0.12 -0.00470 0.000783 1.0 -1.58 1.37 -0.41 0.57 -0.00845 0.000995 2.0 -3.09 1.96 -0.32 0.83 -0.01004 0.000995 Standard Error for the Campbell Attenuation Relationship Standard Period Error M

0.04 0.53 0.10 0.58 0.20 0.64 0.40 0.65 i

1.0 0.72 2.0 0.52 D-39

TABLE D 6 (Cont'd)

The following is the recommended standard errors for the Campbell horizontal attenuation relationship to be used in the present PSHA for the SONGS site only when the Offshore Zone of Deformation isinvolved:

Standard Period Error mummmm&%mmme 0.04 0.47 0.10 0.51 0.20 0.48 0.40 0.46 1.0 0.50 2.0 0.52 i

Reference Campbell, K.W. (1993). Empirical Prediction of Near Source Ground Motion from Large Earthquakes, International Workshop on Eanhquake Hazard and Large Dams in the Himalaya, January 15-16,1993, New Delhi, India, Indian National Trust for Art and Cultural Heritage (INTACH).

Campbell, K.W, and Y. Bozorgnia (1994). Near-Source Attenuation of Peak Horizontal Acceleration From Worldwide Accelerograms, Submitted to the Fifth U.S. National Conference on Earthquake Engineering, Chicago, Illinois, July 10-14,1994.

1 D-40

- -. - ._ . _ _ . . =- . .. . . . - . _.

TABLE D-7 Vertical Attenuation Relationship by Campbell (1990) for Soil Site J

Campbell (1990) derived the following relationship to model the near-source attenuation of strong ground motion:

e in(Y) a + bM + d in[R, + ic exp(cy)] . eF

+ f itanh [f,(M +f)] + gitanh (g2D) + a o g (D-14) 4 where:

Y =

vertical components of the peak ground acceleration (g) or pseudospectral j acceleration (g)

M = moment magnitude (Mw) (or local magnitude (MJ for M < 6 and surface wave

magnitude (Ms ) for M 2 6.0)

R, = shortest distance between a recording site and the assumed zone of seismogenic mpture (km).

j F = fault mpture mechanism l F=0 2,r strike-slip, l F=1 for reverse, reverse-oblique, thrust, and thrust-oblique

, faulting D = depth to basement rock (km). A recommendeu vhiue of D of 2 km should i

j be used for the present study.

j a = number of specified standard error terms

, oc = standard error term (natural logarithm) 4 i

I D-41

\

TABLE D-7 (Cont'd) l l

Values of coefficients a, b, ..., g,, and standard error for different periods are given in the following table.

l l

Period a fi f2 f3 gi g2 (sec)

- mummmmemanusmann-PGA -3.829 - - - - -

0.04 -3.733 - - - - -

0.10 -3.053 - - - -

0.20 -3.179 - - - - -

0.40 -3.845 0.181 0.711 -4.7 - -

1.0 -5.471 1.13 0.711 -4.7 0.177 0.513 2.0 -6.985 1.65 0.711 -4.7 0.613 0.513 For all periods, b=0.991, ci=0.0790, c2=0.661, d=-1.50, and e=0.111.

The standarti error terms are given by the following table. ,

l l

Magnitude Range 4.7 - 6.1 6.2 - 7.8 Period ,

PGA 0.668 0.476 0.04 0.957 0.685 0.10 0.905 0.464 0.20 0.761 0.498 0.40 0.739 0.519 1.0 0.832 0.545 2.0 0.764 0.666 D-42

- _ ~ . _ - ._- . - . .- -- . .- . - . . _ . - , . . _ . _ _ _ . - . - -

j TABLE D-7 (Cont'd) l 1

Reference 4

Campbell, K.W. (1990). Empirical Predrtion of Near-Source Soil and Soft-Rock Ground Motion

) for the Diablo Canyon Power Plant Site, San Luis Obispo County, California, Prepared for Lawrence Livermore National Laboratory by Dames & Moore, Evergreen, Colorado, l September 1990.

4 1

D-43

TABLE D-8 Empirical Database For Horizontal Spectral Accelerations DIsenere (kap Slee Hesteement Spedral - gp - Geomeerte 88 sum Seyte of Joyeer A Chess T-U 04 s I-0Is T-02s T-U 4 s I-I s T-2s Fuehgaste Deee _ Feeleseg(*) his Seestem Closest Beere Camphet (**) pOA f-25 Hr f-10 Hr f=5 Hr f=2 5 Hr f=1 Hr f-0 5 Hr Re4erence Impenef Valley 5/19v40 SS 72 I V hmguaien thserict 10 0 12 0 83 U 0 273 0 271 0 428 0566 0 587 03M 019F i l' art field 6/27/66 SS 6i Choleme 85 (G*fG) 53 93 63 C 0392 0 394 0 571 0 726 0 948 0165 0 071 2 SS 61 Cholome st(CDMG) 92 13 0 10 1 C 0255 0 255 0424 0 711 0 352 0 150 0 047 2 SS 68 Chisme st 2 (CDMG) 14 7 17 3 15 2 B 0 058 0 058 0 116 0 121 0 096 0 058 0 090 2 Impenal Valley 1915/79 $$ 65 D Centro s7(USGS) 06 06 55 C 0 392 0 402 0 478 0 628 0 578 0662 0 309 3

$$ 65 BCentro s6(USGS) to 13 56 C 0 388 0 453 0 898 0 655 0 652 0 515 0407 3 SS 65 B Centro s5(USGS) 10 40 56 C 0444 0469 0 791 0 913 1 997 0 530 0 287 3 SS 65 Bonds Corner (USGS) 25 26 6i C 0679 0 733 1085 I640 1 933 04M 0 205 3 SS 65 U Centro es (USGS) 38 38 67 C 0 533 0590 1 230 0 902 0 746 0 354 0204 3 SS 65 8 Censo 84(USGS) 42 68 69 C 0 419 0 437 0622 0 877 0 651 0 589 0 307 3 SS 65 a Centro 89 Diff Array (GS) 53 5I 76 C 0 413 0426 0 719 0959 0 887 0 144 0 237 3 SS 65 Holrnfle Post Off(USGS) 75 75 93 C 0 233 02M 0 414 0665 0552 0 295 0 til 3 SS 65 bep Cnty Caer BlWCDMG) 76 76 94 C 0224 0 235 0379 0 539 0 198 0387 0 208 4 SS 65 Breeley Amport(USGS) 85 85 75 C 0 191 0 206 0 309 0467 0290 0247 0 147 3 SS 65 a Centro fl0(USGS) 86 85 to I C 0197 0200 0 328 0 445 0 ell 0 241 0 243 3 C

SS 65 H Censro #1(USGS) 93 12 7 10 9 C 0 241 0 245 0 689 0766 0 484 0 189 0 150 3 SS 65 B Censo #2(USGS) 10 4 16 0 12 3 C 0 MI O3D 0 101 0 864 0 Sol 0 150 0 135 3 SS 65 Calence Fue $ts(USGS) 10 6 10 6 11 9 C 0 235 0 241 0 544 0 553 0477 0 169 0 071 3 SS 65 0 Centro all(USGS) 12 6 12 6 13 8 C 0 172 0 380 0580 1 095 0 808 0 235 0 174 3 SS 65 Pareduee Test See(USGS) 14 2 14 0 15 B 0 140 0 152 0 215 0 379 0250 0 123 0 081 3 Memceh Vly 6/9/80 SS 64 Chshuahus 14 6 I46 84 9 C 0 lit 0 121 0 088 0249 0 288 0 174 0 147 5 Morgan Hill 4/24/84 $$ 62 Hells Vly (CDMG) 34 34 5i B 0 220 0 221 0 379 0 461 0445 0 263 0 055 6

$$ 62 Ostroy s4 (CDMG) 12 8 12 8 14 3 C 0 272 0 274 0 461 0 680 0 473 0 269 0 055 6 SS 62 Sn 3ame Fwy I01480(USGS) 14 4 14 4 15 9 U 0 142 0 155 0 148 0273 0 317 0226 0 089 7 SS 62 Gilroy al(CDMG) 14 6 14 6 16 C 0187 0187 0 122 0 374 0 230 0 166 0 076 6 N Paha Spr 7/8/86 06 6I Whosewater Cyn (USGS) 73 00 73 U 0 543 0 689 0964 4504 1 193 0297 0 078 9 OB 61 Dement Hot Spr (CDMG) 50 20 8 U 0 284 0 285 0 648 0 851 0 781 0 278 0 099 8 OB 6I N Pebe Spr (USGS) 52 00 82 U 0678 0 712 1 169 1455 0 865 0 491 0 160 9 OB 61 Dewre (SCE) 97 30 97 0 0 797 1 232 1 517 1 231 0 875 0 520 0 112 to OB 6i Morcego Vly(USGS) 10 I 40 10 I U 0 219 0 230 0 272 0444 0437 0437 0 llo 9 sup Hills (H) I t/24'87 SS 67 bep Cray Cner BlqCDMG) 14 8 14 8 15 1 C 0 295 0 295 0 389 0 659 0562 0276 0246 II 1 ome Pncte lort 719 OB 69 Canaistos (CDMG) 51 00 15 i B 0 549 0 547 0669 0996 I I42 0 460 0 144 12 OB 69 Carnole(CDMC) 14 5 86 14 5 B 0 433 04M 0 743 1039 1 421 0 374 0 133 12 OB 69 Gewilan Coll (CDMG) Il 6 10 9 18 6 B 0 335 0 339 0 729 0 925 0 841 0 164 0 078 12 OB 69 Seret, Alohn(G)MG) l30 Il 7 13 B 0 403 0 401 0 648 0 977 0557 0 393 0 203 12 OB 69 Saimege W Vly Coll (CDMG) I37 12 0 13 7 B 0 289 0 288 0 146 0497 0 568 0 555 0 231 12 OB 69 Gilroy Old Feehme(CDMG) 12 7 12 3 12 7 B 0264 0 264 0328 0420 0 754 0 353 0 112 12 OB 69 Seres Cruz(CDMG) 17 9 12 5 179 B 0425 0424 0 746 1 001 0 806 0 177 0 041 12 OB 69 Gilroy #2(CDMG) 12 7 12 1 12 7 C 03M 0 338 0490 0 846 I 088 0 348 0 184 12 OB 69 Gilroy #1(CDMG) 14 4 14 0_ 14 4 C 0 447 0 446 0 924 3 579 0 715 0 317 0149 12

? andres W28/92 SS 71 Joshua Tree iI 6 II 1 12 B 0 279 0 279 0292 0447 0 593 0 461 0 158 11 NOTES (*) SS-Stra e-$ lip. OB-ONeque

(**) Sete class U owbcsees uranoen sne classancetion accordeg to Beore et el (1993)

i TABLE D 8 (Cont'd)

(1) National Geophysical Data Center, NOA A, Boulder, CO, US ACA02.

(2) National Geophysical Data Center, NOAA, Boulder, CO, US ACA01.

(3) National Geophysical Data Center, NOAA, Boulder, CO, USACA23 and USACA24.

(4) California Division of Mines and Geology (1980). " Compilation of strong-motion records l 1

and preliminary data from the Imperial Valley Eanhquake of 15 October 1979,"  !

Preliminary Repon 26,53 pp.

(5) National Geophysical Data Center, NOA A, Boulder, CO, MEX01.

(6) National Geophysical Data Center, NOAA, Boulder, CO, USACA36.

(7) National Geophysical Data Center, NOAA, Boulder, CO, US ACA35.

(8) Califomia Deptuinent of Conservation, Division of Mines and Geology, Office of Strong Motion Studies (1987). " Processed strong motion data from the Palm Springs Eanhquake of 8 July 1986, Part I, Ground response records," Repon OSMS 87-01,256 pp.

(9) Porcella, R.L, E.C. Etheredge, R.P. Maley, and J.C. Switzer (1987). " Strong-motion data from the July 8,1986 Nonh Palm Springs Eanhquake and Aftershocks," USGS OFR 87-155,37 pp.

(10) Hanzell, S. (1990). Personal communication. USGS, Denver, CO.

(11) Califomia Depanment of Conservation, Division of Mines and Geology, Office of Strong Motion Studies (1987). "CSMIP strong motion records for the Superstition Hills, Imperial County, Califomia Eanhquakes of 23 and 24 November 1987," Repon OSMS 87-06, 42 pp.

(12) California Depanment of Conservation, Division of Mines and Geclogy, Office of Strong Motion Studies (1989). "CSMIP strong motion records for the Santa Cruz Mountains (loma Prieta) , California Eanhquake of 17 October 1989," Repon OSMS 89 06,196 pp.

(13) California Depanment of Conservation, Division of Mines and Geology, Office of Strong Motion Studies (1992). "CSMIP strong motion records for the Landers , Califomia Eanhquakes of June 28,1992," Report OSMS 92-09,330 pp.

D-45

TABLE D 9 Parameters Used in Ground Motion Simulations Q is a lognormal distribution with mean of 150 and standard deviation (in) of 0.18 ti is a normal distribution with mean of 0.6 and standard deviation of 0.05 l x is a lognormal distribution with mean of 0.04 s and standard deviation On) of 0.30 Shear wave velocity profile varied with c.o.v. of 0.3 Vadable slip on rupture surface l Variable focus location within a zone extending from center to bottom along dip and out to 10% of the length at each end along strike I

l D-46

l l

l TABLE D 10 Attenuation Relationship Weight Using the Empirical Database l

Normalized l Attenuation Period Sample Median Deviation Deviation Attenuation i T Size Residual Weight Weight Weight (s) (*A) (D War) (NWt r) (W ) 4

, (1) (2) (3) (4) (5) (6) (7) l Idriss 0.1 41 -17.2% 0.79 0.17 0.13 i Stiff Soil Site 0.2 41 -30.8% 0.43 0.10 l (1994) 0.4 41 -36.6% 0.29 0.09 1.0 41 -13.5% 0.88 0.22 '

Abrahamson 0.1 41 -8.2% 0.99 0.22 0.19 ,

Soil Site 0.2 41 -18.2% 0.76 0.18 (1994) 0.4 41 -24.7% 0.59 0.18 1.0 41 -12.1% 0.92 0.23 Sadigh 0.1 41 -4.7% 1.00 0.22 0.20 Soil Site 0.2 41 -10.0% 0.97 0.23 (1987,1994) 0.4 41 -24.0% 0.61 0.18 1.0 41 -27.1% 0.53 0.13 Boore et al. 0.1 41 7.8% 0.82 0.18 0.24 Avg. Site Class 0.2 41 -2.1% l.00 0.24 B and C 0.4 41 2.8% 0.99 0.30 (1994) 1.0 41 8.1% 0.80 0.20 Campbell 0.1 41 -1.7% l.00 0.22 0.24 Soil Site 0.2 41 -3.4% l.00 0.24 (1993,1994) 0.4 41 -14.5% 0.86 0.26 1.0 41 -15.1% 0.84 0.21 I

l D-47 t

0 , , , ,, , ,

i i  :

t  ! s-g 1  !

20 t i 1 1 i  !

t i i

] I i  !

, 40 , , ,

1 1 1 1 i F i I t

?

l 60

$ t t

!  ! 1,

# $ t i 1 .

6a 80 e f, 4

^

i i

$ ~

SITE 3 SITE i 'CLASE C' A 1 C ASSB 1

l 100

~

\

Idealizec Shear f .

Wave V elocity

! V8'i^ti' " " '

120 SONGS Site l

l 1

~

140 1

160 O 500 1000 1500 2000 2500 l

Shear Wave Velocity, Vs - fps IDEALIZED NATIVE SOIL SHEAR WAVE VELOCITY Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-1

] _ _ , , , p.48 Woodward-Clyde Consultants l

! Mw=5 Idriac Abrahamson

Depth =5 km Sadigh Boore, Joyner & Fumal Campbell 10 _. , , , , ,,,,, , , , , , , , , , __

O 1 _ _

C Z Z

,o . _ _

, e - -

2 - -

e w =___r o .

• Q -

k 10'1 :  :

, =

C .

=

% 'ss

=

s N o

w

, s N

V s _

O i

N N s

m ,

$ 10-2 ; N' , =

i  : -

4 ' ' ' ' ' ' ' ' ' ' ' ' '

10 i ' ' ' ' 8 1 10 10 2

Horizontal Distance - km J

e q PREDICTED MEDIAN PEAK GROUND ACCELERATIONS -Mws6 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig D-2a

-w D-49 Woodward Clyde Consultants

I Mw=6-1/2 Idrina Abrahamson Depth =5 km Sadigh 1

Boore, Joyner & Fumal  !

Campbell 10 , , , ,,,,,

= , , , , , , , , , , ._

i _

l

\ -

l

\ -

1

1. 1 -

i  :  :

c  :

O . -

5 #-*% . ~

O  %

1 3u  : -

Q _

l' o  %

( 10'1 = =

g -

N Z C

s .

\)s s _

o w -

s s _

O -

es 8

Q. 10'2  :

I 4 ' ' ' ' ' ' ' '

10 1 10 2 '

10 Horizontal Distance - km PREDICTED MEDIAN PEAK GROUND ACCELERATIONS - Mw=G-1/2 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2b u.w p.50 Woodward-Clyde Consultants

Mw=7-1/2 Idrina Abrahamson D.epth=5 km Sadigh I 1 Doore, Joyner & Fumal Campbell

, 10 _ , , , , , , , , , , , , , ,,,,, _ j l  :  : 1 l _

l l _ _

l 08 1 - _

e -r-  :

I

--;s ,~~. y o

e N _

m 2

u -

NN _

l , _

N _

l l

8 u

N 4 10 E 3 E ,

E . 5 5 m _

m - _

, o , - _

l ,

t a l e i

CL 10'2 : _

\

l 1

l I I I I Iff1I i i I I i l Ii1 l 10-3 1 10 10 2

l Horizontal Distance - km l

l PREDICTED MEDIAN PEAK GROUND ACCELERATIONS - Mw=7-il2 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2c D-51 Woodward-Clyde Consultants l

Mw=5 Idriaa Abrahamson Depth =5 km Sadigh Boore, Joyner & Fumal i

Campbell 10 _ , , , , ,,,,, , , , , , , ,,, __

to a 1  : =

c  :  : l o

.g .

- i m

w i

G --

l e  : -

~q _

o - -- -. n

( 10'I - -

3 5 N' 5 y . - -

e -

N -

a

_ s N _

V) _ s N

s _

Z LO N 10-2 ; _

h 's, _

y  :

4 ' ' ' ' ' ' ' ' ' '

10

' ' *

• 8 1 10 go 2 Horizontal Distance - km PREDICTED MEDIAN 26 Hz (T=0.04 s) SPECTRAL ACCELERATIONS - Mwn6 Project No. 934E361 A Date:17 OCT 94 Project: SONGS IPEEE Fig. D-2d

. . a ,o ,, D-52 Woodward-Clyde Consultants

i Mw=6-1/2 idrina l

i Abrahamson Depth =5 km Sadigh Boore, Joyner & Fumal

- Campbell 10 , , , ,,,,,, , , , , ,,,,, g

=

\ s - -

l tz i 1  : =

c  :

0 s

.g g ": =-mw -

2 e

u

- %s g _

o

! 4 10'I :  :

5w  : I g .

\N Ns j

g - s _

l M  : -

\ _

Z l

' W 10'* = =

1 "  :  :

s - _ \

l 4 ' ' ' ' ' ' ' '

10

• * *

• 8 l

1 10 10 2

Horizontal Distance - km

PREDICTED MEDIAN 26 Hz (T=0.04 s) SPECTRAL ACCELERATIONS - Mw=6-112 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2e

,, D-53 Woodward-Clyde Consultants

Mw=7-1/2 Idrino Abrahamson Depth =5 km Sadigh Boore, Joyner & Fumal Campbell 10 i i

i i i i i ii i i i i i i iii -

cn

. 1  :  :

8 _--_ E cm '

~ ~ ~ _~ r_=. ~

m -

3 -

%N -

= -

\ -

4 10'I : \

3 5

'\'- E y . -

a. -

M = -

y _

z W 10-2 ;

m 4 ' ' ' ' ' ' ' ' ' ' ' ' '

10 1 10 2 8 10 Horizontal Distance - km PREDICTED MEDIAN 26 Hz (T=0.04 s) SPECTRAL ACCELERATIONS - Mw=7-1/2 Proje.ct No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2f

. .. 2,. , D-54 Woodward Clyde consultants

Mw=5 Idriac Abrahamson Depth =5 km Sadigh Boore, Joyner & Fumal  ;

i Campbell  !

10 i i i illi a i i i iiisi

1 _

e - _

l CD i 1  : _

c  :  :

0 - -

's e

T" _

e r-u a

-s~% _

u 's ' %

I 10'I - -

NkN ,

$ =

Yst, Z g

a -

NL -

I ' -

NN. -

10-2 ;

Y _

e _

=

e - _

4 ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' '

10 1 10 2 10 Horizontal Distance - km l 1

l l

i i

PREDICTED MEDIAN 10 Hz (Tm0.1 s) SPECTRAL ACCELERATIONS - Mwa6 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2g

- . = = D-55 Woodward-Clyde Consultants l 1

Mw=6-1/2 Idrim Abrahamson Depth =5 km Sadigh Boore, Joyner & Fumal Campbell 10 _. , , , , ,,,,, , , , , , , , , , __

tn

, 1  : =

g --

O

.g .

> ~ ~ ~~~'* ' ~ .

~

g w

,N -

2  : _

u o R 5 10 E %N E e

y .

p.  :

g, N, -

&)  : _

Z C 10'E :  :

w -

I 4 ' ' ' ' ' ' ' ' ' ' ' ' '

10 1 10 2 8 10 Horizontal Distance - km PREDICTED MEDIAN 10 Hz (T=0.1 s) SPECTRAL ACCELERATIONS - Mw=6-1/2 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2h

- , , , _ , D-56 Woodward-Clyde Consultants

Mw=7-1/2 Idrico I ---

Abrahamson i Depth =5 km j ---

Sadigh a

Boore, Joyner & Fumal Campbell l 10 _ , , , ,,,,,, , , , ,,,,,, _

i - -

1 cn

, 1 F.  :

!_ e  :---.%  :

-% -Vx

.O .

:

i 11 4

w N -

s -

s

e -

s -

< o

o -

4 4

10- =

N x

. E  :  :

{ g . - -

e - -

et - -

v)

{  : - -

g

Z i o

,- 1o.2 ;  ;

1 4 ' ' ' ' ' ' ' '

I 10 l 1 10 10 2 '

Horizontal Distance - km PREDICTED MEDIAN 10 Hz (Ts0.1 s) SPECTRAL ACCELERATIONS - Mws7-1/2 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2i

_, D-57 Woodward-Clyde Consultants

Mw=5 Idri2a Abrahamson Depth =5 km Sadigh Boore, Joyner & Fumal Campbell 10 __ , ,,,,,ii, , , , , ,,,,, _

l m 1 _ _

e  :  :

C  :  :

o . =-- _

m ___

%_ . _c> -

E - - - - - _ - _ _ -

e -

g a _

~N _

(

8 10'1 :

2 .

_ NA s _-

g g

N'KsN -

M _

N Z

in 10 2 4 ' ' ' ' ' ' ' '

10 1 10 10 2

Horizontal Distance - km PREDICTED MEDIAN 6 Hz (T=0.2 s) SPECTRAL ACCELERATIONS - Mwn6 Project No,934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2j D-58 Woodward-Clyde Consultants

, Mw=6-1/2 Idrico Abrahamson Depth =5 km Sadigh Boore, Joyner & Fumal Campbell 10 , , , , ,,,,, , , ,,,,,

= , ,

=_

i - _

I E N P-- %-s -

E g _ _

O . - _

'n - _

2 _ _

e , - _

e \s s o

o '% N 4 10'I - -

- E NN E g .

u N' _

g _ _

m , _ _

N Z

in 10'I :  :

l 4 ' ' ' ' ' ' ' ' ' ' ' ' '

10

' ' *

• 2 '

1 10 10 j 1

Horizontal Distance - km '

{ PREDICTED MEDIAN 5 Hz (T=0.2 s) SPECTRAL ACCELERATICNS - Mw=6-112 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2k D-59 Woodward-Clyde Consultants l

l _

Mw=7-1/2 idriza Abrahamson Depth =5 km Sadigh )

l Boore, Joyner & Fumal I i i l' Campbell 10 _ , , ,,,,,i, , , , , ,,,,, __

- _ l D8

""' ~ "= - h 1

e  :  :

i c  : N  :

,o

%t -

N u - .

%gk 1

f  :

f - - -

8  %\

, <C 10' - -

l

1 2

. o - -

e - -

, CL to  : - _

N I 4 2 l

so 10-2  ;

e - -

] _ _

4 ' ' ' ' ' ' ' ' ' ' ' ' '

10 1 10 10 2

Horizontal Distance - km PREDICTED MEDIAN 6 Hz (T=0.2 s) SPECTRAL ACCELERATIONS - Mwn7-1/2 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-21

_ . , , , D-60 Woodward-Clyde Consultants 1

I l

Mw=5 Idriac Abrahamson Depth =5 km Sadigh Boore, Joyner & Fumal Campbell 10 _ , , , ,3,,,, , , , , , , , , , _

e -

i -

m s 1 -

c  : -

! O -

i g e -

i f5 -

g - -

_____'~s e  : -

o - - - - -

._ sq

< u -

's '

10

\h

-- g

' w _

a 5e e -

a -

N,N _

M \'s N

y -

gN _

. ;z- .

9 10-2 _ \s '

N E E e -

4 ' ' ' ' ' ' ' ' ' ' ' ' '

10

• ' *

• 8 1 10 2 10 Horizontal Distance - km PREDICTED MEDIAN 2.6 Hz (T=0.4 s) SPECTRAL ACCELERATIONS - Mwn6 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2m

-.mo. ,,, D-61 Woodward-Clyde Consultants

Mw=6-1/2 Idrino Abrahamson Depth =5 km l Sadigh Boore, Joyner & Fumal Campbell 10 _

i , , , ,i,,, , , , , ,,,,, __

cn i 1 __ -

p 5  :

sN

%@g g -

s -

2 Ng e _

l o ~

4 10'I -

\' -

W E NN E x  :

cL _ _

W) g - -

Z 9

N 10-2  ;

4 ' ' ' ' ' ' ' ' ' ' ' ' '

10 1 10 2 10 Horizontal Distance - km PREDICTED MEDIAN 2.6 Hz (T=0.4 s) SPECTRAL ACCELERATIONS - Mw=6-1/2 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2n

_ , , , , D-62 Woodward-Clyde Consultants

l Mw=7-1/2 Idriss l

Abrahamson Depth =5 km 1

--- Sadigh i Boore, Joyner a Fumal Campbell 10 , , , , , , , , , , , , ,,,,,

= , _

l  : _ _

l ~ ~ -

l t:n -- . y . __- 7 x c

o b

Ds b p . -

s _

m - N -

6 N N -

_e e  : - - _

0 l

4 10-1 m  :

g

• o cL - _

t/)

g _ _

Z l 9

N 10-2 ; = j l - _

i 4 ' ' ' ' ' ' ' '

10 1 10 10 2

Horizontal Distance - km l

l l

i

, PREDICTED MEDIAN 2.6 Hz (T=0.4 s) SPECTRAL ACCELERATIONS - Mw=7-1/2 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-20 ,

1

. . , , , , , , D-63 Woodward Clyde Consultants l

Mw=5 idriss Abrahamson

] Depth =5 km Sadigh Boore, Joyner & Fumul Campbell i 10 j

i i i iiiits i i i i iiiii -

0, _

i 1 _

d b

a

1. 1

e

-  :

! C  : 2 O . -

l *@ -

1 2 -

4 j g  : -

?* u u

} (

10'I E ~ -%

_% N

]

2 '

__~

o -

e -

n _

, s 4

4 M  : - s N

N -

4 N ' N

! Z N

,- 10-2

\ _

l

! . b \ [  !

.\ -

l 4 -

x- _

4 ' ' ' ' ' ' ' '

10 1 10 '

2 10 Horizontal Distance - km PREDICTED MEDIAN 1 Hz (T=1 s) SPECTRAL ACCELERATIONS - Mw=6 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2p

- .w. , D-64 Woodward-Clyde Consultants

Mw=6-1/2 idri2c Abrahamson l D.epth=5 km Sadigh l l

Boore, Joyner & Fumal f

i  !

l Campbell 10 _ , , , ,,,,,, , , , , ,,,,, __

\

t i 1 e - -

\

l l

l \

1. 1 -

i  :-  :

C

• D '--% n-

.)m

- 2 N %.

e' N s, g  : -

N g _

u  % ~~

u N

\ Ns

<C 10'  : s  :

-  : s N, 3 .

u - -

e g -

W - -

N Z

v 10-2 ;  ;

l 4 ' ' ' ' ' ' ' '

10 I 1 10 10 2

l Horizontal Distance - km  !

l I

i PREDICTED MEDIAN 1 Hz (T=1 s) SPECTRAL ACCELERATIONS - Mw=6-1/2 i

Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2q l

l

_, D-65 Woodward Clyde Consultants l

i l

Mw=7-1/2 Idriss Abrahamson Depth =5 km Sadigh Boore, Joyner & Fumal

- Campbell 10 _ , , , , , , , , , , , , , , , , , , _

i _ _

1 l

~-

f 1 r_ _ _ _--

s

, ~ .

=

g

.O . - ~~ %  %

~

g -

Ng -

g s _

g -

N\ _

o \'-

(

10-1 :

\ g g .

u - -

G _ _

CL M _ _

N Z

e 10'* = =

3 l l l l 1llll l l 1 1 I I III

• * *

• 8 1 10 2 10 Horizontal Distance - km 1

1 PREDICTED MEDIAN 1 Hz (T=1 s) SPECTRAL ACCELERATIONS - Mw=7-112 Project No. 934E361 A Date:17 OCT 94 Project: SONGS IPEEE Fig. D-2r

, , , , D-66 Woodward-Clyde Consultants l

Mw=5 Idriac Abrahamson 1 Depth =5 km Sadigh j ---

Boore, Joyner & Fumal Campbell 10 i , , ,,,,ii , , , , , , , , ,

]

{ _ _

s - _

O I ' 1  :  :

c  :  :

p o .

m w

e s - _

o

, o 4 , 10-1 m

g

• o - -

e

=--__---s

. CL -

' 'N N C/)

] N * :- - ---

\ 'A s 1 2- - - - - - -

. s.\

N s

s ss to. 10-2 o -

- N, , N -

s _

N

)

'% s g s

! ,s \

= -

NA - '

4 N l

10-3 f I I I I f I 11 1 I I I h 114

• * *

• 8

1 10 10 2

Horizontal Distance - km PREDICTED MEDIAN 0.5 Hz (T=2 s) SPECTRAL ACCELERATIONS - Mwn6 Project No. 934E361 A Date; 17 OCT 94 Project: SONGS IPEEE Fig. D 2s

_ ,, D-67 Woodward-Clyde Consultants

Mw=6-1/2 idriso Abrahamson Depth =5 km Sadigh Boore, Joyner & Fumal Campbell 10 i i i i iiiii i i i i iiiii  :

cn i 1  :  :

e  :  :

o ,

.a.

w g ,

o _

4 o m e,Ns 10-1 :

_ N- _

m _ 4  :

2g s N% _-

a. -

en g _

z '

9 o

1o-2 ;  ;

4 ' ' ' ' ' ' ' ' ' ' ' ' '

10 1 10 2 10 Horizontal Distance - km PREDICTED MEDIAN 0.6 Hz (T=2 s) SPECTRAL ACCELERATIONS - Mw=6-il2 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-2t

. . , , D-68 Woodward-Clyde Consultants

l Mw=7-1/2 Idrina Abrahamson Sadigh Boore, Joyner & Fumal I

Campbell 10 i i i i iiiii i i i i iiiii _

)

- _ \

j l

4 -

I cn

' . 1  :  :

c  : -

o -  :

, cm . =_- ,

s

! Ei --_'d 5- -

i

!o * -

hs ge i 4

__ io-1 _

Ns  :

i 2 I

$ 6 _ \-

y . _

N  :

o. _ _

i M y . - _

1 z l

"? 1o-2 ; _

o  : E l . - _

4 ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' '

i 10 1 io 2

} 10 i

Horizontal Distance - km i

4 i

j PREDICTED MEDIAN 0.6 Hz (T=2 s) SPECTRAL ACCELERATIONS - Mw=7-112 Project No. 934E361 A Date: 17 OCT 94 Proje:t: SONGS IPEEE Fig. D-2u 1

. . .., D-69 Woodward Clyde Consultants

1 Shocr Wcvo Volocity, Vo - fpa 0 500 1000 1500 2000 2500 0

m g e b y ? d eg -

20 -

WN9"'.

3;;y ;.-~; , pygm

v^P*M,,,, - l

. -B%i+?&a.

- e a;., .

. o..;.:Vl  : s, .:, ,

A l$h% -

1

'A- 9,< s ,, _- . I' ,'

40 - *

• o+; ' n,:e -

l 1

I.'...  ;. < m i .

-  ;; n ;

/q

. . . ; v,

-  :, .- - ' ',~ .i ;'.

, ; g,: '

, ' , ','4 'z j'...- ';

...7l: -

z, 60 -

ai -

' ;%J' f L.

. . a r, . . .

" .'7:0%:f - ? ,, ,

me;;qz .,;;s+.

. .. '97t.> < m 13 g ;a,, j

& ~gg'

[;;:;; ' ;" :. f, i' : l ; .%,

. . .,' j.; '.[

e

:f L'_.')[ ':I f Q 4:1f f:l{

fcL 80 -

AskM.Mrd HWyd M ~t -

• -1 .<',ve N yw ~ o.  : Tn9:? "i W ) +1 o O ~

4 m e g m,;:r;;p:c(';; 93 '

QJ. T ;" >l . l-W:li i; , .;;f'@ \

mynxn:. c n:: p ; ~ :q 100 -

b' g

- EMPIRICAL  % \

DATABASE \ -

\

\

120 - 1 -

\

Idealized Shear I I

Wave Velocity g Variation at g SONGS Site g 140 -

1 -

I

. l .

I i

160 O 500 1000 1500 2000 2500 Shear Wave Velocity, Vs - fps COMPARISON OF IDEALIZED NATIVE SOIL SHEAR WAVE VELOCITY WITH EMPIRICAL DATABASE SHEAR WAVE VELOCITY RANGE Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-3

,,, D-70 Woodward-Clyde Consultants

- . - . . . . _ . . - . . - - - . - - . . . . . - - - . . - . . . . . . . = - - . . . . . . - - . . .

l I

(

+ Strike-Slip a Obilque 8 .

y , + .

E 7 _

i

. a a ma, a e *

+ -

3

+ +M+ ++++ +

++

'h -

+

+ a m +M

+

-H+-

I a g _

m _

E .

6 0 6 10 16 20 26 Closest Distance (Idriss, Abrahamson, Sadigh) - km i

8 ,

g .

+, .

E 7 -

e a a om a 0 + -

+++++ +++ +

+ ++ +

• 5

+ -

M 6 "-

^^^ + ++ + +

m -

3 . .

6 O 6 10 16 20 26 Closest Surface Distance (Boore et al.) - km l MAGNITUDE AND DISTANCE DISTRIBUTION OF SPECTRA DATA

( Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-4a

. . , . , , , D-71 Woodward-Clyde Consultants l

l l

1 i

i

1 l

l

+ - Strike Slip A Oblique l I

8 , , , , , , , , ,

g -

, + _

E 7 - _

e a m A AA A g -

4 D l

g -

4+-H-+ + + + -f+ -

+++

~

• E M 6 -

++^^ +++ -

843 E

6 ' ' ' '''''''''''''''' j 0 6 10 16 20 26 1 j Closest Distance (Campbell) - km l

MAGNITUDE AND DISTANCE DISTRIBUTION OF SPECTRA DATA Project No. 934E361 A Date: 17 OCT 94 Project SONGS IPEEE Fig. D-4b

_w D-72 Woodward-Clyde Consultants

i Frequency (Hz) 2 l 10 . , 10 . , 1 . , 10-1 1Iil i i i i i iii i i i i i I Iiii 1 ii i i i 2.6

, n i CD o* 2.0 l

2 Tlll3h_

. l

! G* 1 tn

n. .&

2 O 4a O i CO 1.5

)

O4 N It

g. W Mw=G-1/2 Al

/ t Mws7

T CL 1.0  : '-

s em NW J

/ 3 V,

s

*E

. / \$

W / $

h _.E.

OO 0.6

. s k s \,g ZW s 4 c. \

i u) *

\

~_.

0.0 '

j g o-2 g o.1 1 1o Period (sec) l t

s i

4 NORMALIZED SITE-SPECIFIC HORIZONTAL RESPONSE SPECTRA AT 6% DAMPING Mw=G-112 AND Mww7 Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-5

.. . . e D-73 Woodward-Clyde Consultants

i i

PEAK GROUND ACCELERATION 1.oo . . . . , . . . . .....,.... . . . .

,,_m__- ., . .

j .-

Idriss (Stift Soil Site) ffg'" _

--- Abrahamson (Soil Ste) 'fp - -

Sadgh (SoM Site) -

o.7s - / '9 ),

- 1 Boore et al. (Average Site Class B and C) g,8/.

3n ,/

~

---

campba (Sa sie) _

c3 -

I -

~

o - > ] .- -

0.so p<j c j ,

0 .O

,y ,. . - t 2 o -

j -

2 -

')I /

L -

o.2s -

g' -

g, y ,. -  ;

bI -

A' .

.- ,m

__ _ J. -I' ' -

t i

a,aa . . . . i . . r. . i.... . . . . i....i....i....

2.o -1.s -1.o -o.s o.o o.s 1.o 1.s 2.o overestimation Residuals underesumeiron ,

l NOTE: Number of data points =41 t

COMPARISON OF QUANTILE PLOTS OF RESHMJALS FOR PGA - STRIME-SUP AND OBLIQUE EVENTS  ;

Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fg. D-6a i

~ = . - - Woodward-Clyde Consultants  !

T = 0.04 sec FREQUENCY = 25 Hz 1.00

_ , . c '- - _ , _ _ - ,

_ Idriss (Stiff Soil Site) , .

• o

--- Abrahamson (Soil Site) *

[ -

---

Sadgh (Soil Site) [/9 ,

0.76 ~

((

Boore et al. (Average Site class B and C)

[ ----- campbeit (Sail Site)

/ ,[ [

Q -

J'y ) -

f -

• / .' -

c 0.50 .I )s f

9 .9 -

,h, '

d t; .

m u

e

/!V .

u.

llln 0.26 -

f

- 7'ly .

) l'j .

- /- -

2 ,,.T.

0.00 ' ' ' ' ' 'Y ' ''' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' '

-2.0 -1.6 -1.0 -0.6 0.0 0.6 1.0 1.6 2.0 Overestimatim Residuals underestimation  ;

i NOTE: Number of data points-41 COMPARISON OF QUANTILE PLOTS OF RESIDUALS AT T=0.04 s - STRIKE-SLIP AND OBLIQUE EVENTS Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-6b

~~- e Woodward-Clyde Consultants

T = 0.1 sec FREQUENCY = 10 Hz 1.00 . . . ,

. . . , ._j. ,_ m . . , . . . .

j '-

. --- Idriss (Stiff Soil Site) <- _

r

--- Abrahamson (Soil Site) *

/, -

0.M

---

Sadgh (soit site)

Boore et al. (Average Site Class B and C)

[/

• 8

~

~

---

Campben (Soil site) *'

f O -

./ w$ -

O

~

c 0.60 }' l2

), '.k['

y .e -

- J i l -

t 15 - -

(U w

./ - _

U. -

/,s' s -

0.26 -

[',A f

.4.t -

/Y -

// ' ' ' ' ' '''''''''''''

0.00 '

-2.0 -1.5 -1.0 -0'.6 0.0 0.6 1.0 1.6 2.0 Overestimatim Residuals underestimation NOTE: Number of data points =41 COMPARISON OF QUANTILE PLOTS OF RESIDUALS AT T=0.1 s - STRIKE-SLIP AND OBLIQUE EVENTS Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-6c

---r Woodward-Clyde Consultants

T = 0.2 sec FREQUENCY = 5 Hz  !

1.00 . . . ,

Idriss (Stiff Soil Site) 8,- "

/

- - - Abrahamson(SoilSte) f " # ,:

^ '

---

Sadgh (Soil Site) /' -

~ ~

~

gg -

Boore et al. (Average site Class B and C)

/ ,f -

W tu -

---

Campbell (Soil Ste)

• s' -

o . ./ g. /,s O

0.50

( f! .l.,b c -

9 O -

lI /

d s o -

s'r ..

0 -

,/ /,)

u. . -

0.25 -

, /,' -

/ l', - '

/

j:<t,l' 0.00 ' ' ' ' ' ' ' ' '''' ' ' ''''''''''''

-2.0 -1.6 -1.0 -0.6 0.0 0.6 1.0 1.6 2.0 Overestimatim Residuals underestimation NOTE: Number of data points =41 COMPARISON OF QUANTILE PLOTS OF RESIDUALS AT T=0.2 s - STRIKE-SLIP AND OBLIQUE EVENTS Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-6d

~ ~ - < Woodward-Clyde Consultants  ;

4 T = 0.4 sec FREQUENCY = 2.5 Hz  ;

1.00 . . . . , , . . .

_ --- Idriss (Stiff Soil Site) , _

l

--- Abrahamson (Soil Site) ,

---

Sadgh (Soil Site) _

4 e' 0.76 -

Boore et al. (Average Site Qass B and C) [' ',e 6

---

Campbell (Sail Site) o , II -  :

~

fg -

O ,f g 'l i

c 0.60 y ',  ;

O O ) ' -

} l' U -

it  : ,/[  :

O.25 -

[ e J .)1'#*

i

(

jj j -

- ) .x ' / .

0.00 ' ' ' ' ' J ' 'J f 1.6 2.0

-2.0 -1.6 , -1.0 -0.6 0.0 0.6 1.0 ovmessmason Residuals underessmason i

NOTE: Number of data points =41 COMPARISON OF QUANTILE PLOTS OF RESIDUALS AT T=0.4 s - STRIKE-SLIP AND OBLIQUE EVENTS Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-6e

- - - - < Woodward-Clyde Consultants

T = 1 sec FREQUENCY = 1 Hz 1.00 . . . . , . . . . , . . ..,.... . . ,, , , , m, , . . . ,

, . .I Idriss (Sts Soil Site) , , _

--- Abrahamson (Soil Site) e j'

~.gE(w'/ /

~

- - - - Sadgh (So8 Site) j [

~

Boore et al. (Average Site Cfo B and C) /

[ ----- campben (Sail Site)

,/ d _'

a

<f I O

^ ~

j j g

e y

5 .

('f{

.j { -

2 -

[ s'fl

u. _

.I -

0.26 - i is/

>'I -

. . . -W -

.; 'pf  :

C#

~

0.00 ''''f'''' ''''''''''''''

-2.0 -1.5 -1.0 -0.5 0.0 0.6 1.0 1.5 2.0 Overestimatim Residuals underestimation NOTE: Number of data points =41 COMPARISON OF QUANTILE PLOTS OF RESIDUALS AT T=1 s - STRIKE-SUP AND OBLIQUE EVENTS Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-6f

-m- r Woodward-Clyde Consultants

_ _ _ _ _ _ _ _ _ . _ _ - _ _ _ _ - _ _ _ _ _ _ _ = _ _ _ _ _ _ __

T = 2 sec FREQUENCY = 0.5 Hz L e s s s--

a s g s s s e g s e e s g u a s a a a z s, "' ' ' ' ' ' '

r l Idriss (Stiff Soil Sete)

/ /

[ -

- - - Abrahamson (Soil Site) ~_j

---

Sadgh (Soil Site) / -

l 0.75 - - / -

Boote et al. (Average Site Cass B and C)

) / r,"

campben (Soil Site) I/

p' ,f' _

O -

Ij '*)

l / e o -

/ -

i O

g::

O 0.50 l ll,AC f- -

l

@ j -

,f' e'/ ,

E e , .

Li- -

l la',/ -

0.25 - * -

f _[, g,s o -

/ -

/gf] -

-A -'M 0.00 ' ' ' ' I.-+ '''''''''''''' i

-2.0 -1.5 '''''''.0

-1.0 -0.5 0 0.5 1.0 1.5 2.0 omestimatim Residuals underesinnation i

l NOTE: Number of data points =41 fI i

i COIWPARISON OF QUANTILE PLOTS OF RESIDUALS AT T=2 s - STRIKE-SUP AND OBUQUE EVENTS l Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-6g i

-aw Woodward.Clyde Consultants

_ _ . - - ....- - . - . . - . . - - . . - . - ~ . - - - - ~ _ ~ - . . _ . - . ~ . . . . - - . . - .

--@- Idnes (Stiff Soil)

- A- - Abrahamson(Soil)

-Q- Sedgh(Soil)

--)(-- Boore et al. (Avg. B&C)

-@- Campbell (Soil) 100 ,, , ,

l l

76 -

_ _ w E

t

- . E

- - . 3 50 w

g

,g I e  :

- e i= 26 - - 3 l  :  :

'O l 'e o e - E.-- , .

g A

, ._______________ ,_ p.

l _

_ s l g;;

-26 N , ~ . , ~a + __ __ ~

g g - -- . y ,y:- ---.._.._ 9 e

,- - . .y

.  :  : g 3 40 -

3 e

1 1

e .

i' l = =

l 75 -

1

' ' ' ' ' ' ' '''''''' ~

, -100 l I 0.0 0.1 0.2 0.3 0.4 0.6 0.6 0.7 0.8 0.9 1.0 l Period, T (s)

I t

COMPARISON OF MEDIAN RESIDUALS AT DIFFERENT SPECTRAL PERIODS Project No. 934E361 A Date

17 OCT 94 Project: SONGS IPEEE Fig. D-7

,, D-81 Woodward-Clyde Consultants

._. _ , . .-_ . . . __ _ . __- . . _ _ . _ _ _ _ . . . . = _ - .

l l

l l

l 1.0 ,,, ,,, ,,, ,,, ,, ,,, ,,, ,,, ,,, ,,,

. . I 0.8 M - .

C O - .

'D g . .

> 0.6 G

Q . .

O g . .

.6 0.4

,,C . .

.G 0.2 0.0 ' '

-100 -80 -60 40 -20 0 20 40 60 80 100 Deviation (%)

Overestimation Underestimation WElGHTING OF DEVIATIONS Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-8

. -- _ ; m_,, D-82 Woodward Clyde Consultants

-200 -100 0 100 200 1.00 .........

T=0.10 EU - f J -

3 ,/

3 -

to G

. ./

Ik f

g 0.76 -

//

% - i .

O .

}I' .

C ~

/0 ~

l .$.

+ -

1 # -

4 t

l

=

,m_ 0.60 ljf l

g,

,g

,/[/ 17% Overestimation of Median Error G

[#

is / /

/ -

--- Idriss (Strff Soil) 0.26 Abrahamson (Soil)

[,ef/ -

45 - es .

Sedgh (Soil)

V

[ -

Boore et al. (Avg. B&C)

I a

/ -

- - - - - Campo.u (Soii) o -

f I f f f I I t 4t I I f f I t t t i I f f f f 1 f f f i t i e t I t i1 2 1.00 , . . . .

Residuals (%)

,, , 1.00 l p .,, . . . . , , . . . . , , . .

l _ _

Weight for Idriss l

Attenuation Based on Deviation of i to -

Median Residual l l @ 0.76 - -

0.75 l g . .

l (U - . l i >

l e - -

O l

l $ 0.50 - -

0.60 g . .

C - .

• C g . .

m . .

8 0.25 - -

0.26 l

l .

0.00 ' ' ' ' '

O.00

-200 -100 0 100 200 l

Deviation (%)

WElGHTING EVALUATION PROCEDURE FOR MEDIAN RESIDUALS Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-9

- .e D-83 Woodward-Clyde Consultants

Frequency (Hz) 2 10 , , 3, , , , , ,

, ,-1 lili i i I i i liliI i i i i litI i i i i I l 2.6 l l l l l ll!

+ Idriss (Stff Soil)

]

v

-dr - Abrahamson(Soil)

-O-O Sade(Soa) ,

V E 2.0 X Boore et al. (Avg. B&C) _

-@ Campbell (Soil) m Site-speerfic Spectrum

n. @ .

N O E8 1.6 o<

N

• g G V Q. 1.0

'p i

=

k@

=N-a:

/

/ ,' .

2

/ '

(E O 0.s '

's

' ZW M

O, lL

'. \

\

hl%

0.0 4 * * *

• 10 10 1 10 Period (sec) i NOTE: Attenuaton response spectra computed for R=8 km. Stnke-Slip, Mw4-1/2 4

~

COMPARISON OF NORMALIZED HC R 713IITAL RESPONSE SPECTRA AT 6% DAMPING - Mwas-1/2 l SITE-SPECIFIC SPECTRUM AND MEDIAN OF FIVE ATTENUATIONS Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-10a

__ D-84 Woodward-Clyde Consultants

. . - . ._ - - = _ - _ - _ . - - -.- _ _- - - --- .- . - . -

i i

. Frequency (Hz) 2

10 . . 10 , , 1 , , io-

! Iii i i i l I i iii i i i i I i iii i i i i i i I l 2.6 l l l l l lll

+ Idriss (Stiff Soil)

) Q w

- dr- - Abrahamson (Soil)

-O- Sadigh (Ssi) l C i V 2.0 X Boore et al. (Avg. B&C) _

-@- Campbell (Soil) 1 m Site-Specific Spectrum

o. $ e

%e e U gg 1.5 O(

N

=g e p Q. 1.0 eW Ne

/

k 4

= cr:

N-4f

[

s l rl hh l b [ # I OU 0.5 -

l/ k -,

i 2e  ; V N

C. '

,D.

4 (f) i Y4 N

! a

! 0.0 I ---

s * '

j 1 10 10 2 101 l Period (sec) l

{ NOTE: Attenuaton response spectra computed for R=8 km. Stnke-Stip, Mw=7

}

l 1

i COMPARISON OF NORMALIZED HORIZONTAL RESPONSE SPECTRA AT 6% DAMPING - Mws7 j SITE-SPECIFIC SPECTRUM AND MEDIAN OF FIVE ATTENUATIONS Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-10b I

. ,,,,,,,, D-85 Woodward-Clyde Consultants

.i

- - -.- . . . - . . _ = . - _ . - - - _ . . . . _ - _ - . . - .. ... .- . . . -

1.00 ,

m (l73 0 0.80 -

.J G -

lll3 0.60 .. _ ___ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __

6 -

O D

W 0.40 -

V6 g -

V c

3 0.20 -

v>

0.00 ' ' ' ' ' ' ' ' '

6.0 6.6 6.0 6.6 7.0 7.6 Magnitude, Mw TYPICAL RELATIONSHIPS BETWEEN STANDARD ERRORS AND MAGNITUDES Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-11

-e D-86 Woodward.Clyde Consultants

. - . . - = . - . . - - . - . . . - __ _. - . - . - . - . - . - . -

l i

1.00 . , . ,

l 4

4- Idnss (Stff Soil)

^ - . l

- Abrahamson(Soil)

o - -

d 0.76 -O- Sadgb (Soil) l _

W -

)( Boore et al. (Avg B&C) -

3N -

-G- Campbell (Soil) -

Sz o- - -

O' O.50 -

l C

a f .

D2hy.v95 ..

w n

e - -

, j 0.26 - -

l 3 - -

. m - -

l 0.00 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 l Period, T (s) 1.00 . . . , . , . . . . , . .

i , , i . ,

l Mw=G-112 .

-) g . .

j 4

g o - -

- 0.76 - -

6 -

~,.G. e ' w ,. _ . _ _ _ _ _ . _ . e s ,,..,. -

e* g. - - -3. -

~~ _

• 6-

_. % ~_ _ _ _ _ " #

o2

-~

~~

' ~

~~

~~~~"

j g. [' $' ,,, h 36 0.60 0# T 3

1 c@ wo ,e h 6- -

i 3: W

<C , . .

6 - .

S

'O 0.26 - -

c g - -

, w ~ ~

l M g - .

i 0.00 l 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 l COMPARISON OF COMPUTED AND ATTENUATION STANDARD ERRORS AT DIFFERENT SPECTRAL PERIODS Project No. 934E361 A Date: 17 OCT 94 Project SONGS IPEEE Fig. D 12

---. e D-87 Woodward-Clyde Consultants

l Leg:nd~'

NOTE s'is the standard deviation of the computed standard crror. ----- Att nuaton Standard Error Revised Standard Error Q Computed Standard Error i . .

i ~~

6 '.

O 6

6 - CASE 1:

W N

%.A If A'islargerthan A the standard error of the V6 +s* '*A - attenuaton relatonshipis not revised.

1 > 's%

.g , ,, %s________ g C

ein 1

N M

. I i I i i

6 0 '

6 - CASE 2:

W N

, If A'is less than A, the standard error of the g ,

attenuaton relatonship is reduced to A* at 6 A magrutude 6.5 while mamtaining the same M P standard error at magnitude 5. The location D 'QN , of magrvtude B is unchanged because o, is

+3.{ '

k i >

" 4. - ,- -

N reduced by less than 30%.

g ..

i i 1 i B

i 6 .

O ,

6 .

6 .' CASE 3:

W -

' If A'is less than A, the standard error of the

=g A attenuston relatonship is reduced to A' at 6 ', '

- magrutude 6.5 wtvle mantaining the same M "

, standard error at magnrtude 5. This resutted

• D '-------- % in a reducton of standard error greater than 30%

f'" at magrutude B. Therefore, the reducton of w

h +s*} = 30% the constant porton is fixed at 30% o, by 1 >

o, h' , ,

moving B to B' to sabsfy the reduction at magrvtude 6.5.

B'B 5 6 7 8 Magnitude t

STANDARD ERROR REDUCTION APPROACH Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-13

. D-88 Woodward Clyde Consultants

1.0  ;

w PDA T=0.04 o e.)

w I ,

u l

, W ' l i D  ; l

. g 0.5 ,_ ,

L 1

y ..

3- T '

c ,,  ;

1 M

O.0 1.0 g Tm0.1 s T=0.2 s O

w m

~~

0.6 -

x ~ x ~ .

V __

w c

m l M J

1 0.0 1.0

, T=0.4 s T=1 s i

8 m

, uJ sT. .'-

g -

N --- -----

- -@ AN A c

g M l l

0.0 1.0 6 6 7 8 6

T 2, Magnitude b

s

- ----- Attenuation Standard Error W _...' -

Revised Standard Error Etti 0.6 O Computed Std. Err. (all database) 1 hL g ..

$ Computed Std. Err. (Mwit6.9; avg. mag.=')

3 NOTES: 1) Vertical error bar shows M 11 standard deviation.

0.0 2) Horizontal bar shows database magnitude range, 6 8 7 8 average magnitude = 6.5 Magnitude COMPARISON OF ATTENUATION RELATIONSHIP AND REVISED STANDARD ERRORS IDRISS (STIFF SolL)

Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-14a D 89 Woodward-Clyde Consultants

1.0 6

POA T=0.04 s l 8 6

l l - I l -;

'l ' -d ,-

i I

W c '*

l  ; l y

g 0.6 Q3 p---

j w

A

,. _g____

l c

, i o o W

0.0 -

1.0

, T=0.1 s T=0.2 s O

w - w

%  %~ a.. ~ -

w 0.6 -

___p____ y , ___ ____

g ..

N W

l 0.0 1.0 6

T= 0.4 s T=1 s 2 m  %. ._ i y M' .

.,'- Xt .

.o , , ' N . , ' -

g0.6 w _ __ _____

Ng#

D e

g A

N& ,

T a o

• l M

l 0.0 l

1,o 6 6 7 8 l

6 T=2 s Magnitude O

6 -----

Attenuation Standard Error 6

W

-__,..,,.-___7____; Revised Standard Error 6 0.6 T C Computed Std. Err. (all database) l N L o l ag C

e Computed Std. Err. (Mwr6.9; avg. mag.=7) l 3 NOTES: 1) Vertical error bar shows M i 1 standard deviation.

0.0 2) Horizontal bar shows database magnitude range, 6 6 7 8 average magnitude = 6.5 Magnitude COMPARISON OF ATTENUATION RELATIONSHIP AND REVISED STANDARD ERRORS ABRAHAMSON (SolL)

Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-14b

_. . D-90 Woodward-Clyde Consultants

I An-1.0 Tso.04 o PCA I l s

l O >

j t '

l W -

Qa -

l -

A

^

0 mg o

l g

w M

l 0.0 1.0 Tao.2 s Tso.1 s I

w I o _____ _____ _____ j t __________ _____

m _____ _____ _____ _____ ___________ -

,,w o.5 .

1 __

A I

$e __

e m

M -'

' l 0.0 1.0 T=1 s Tso.4 s w i I o _______________._____._____..____ -

t _____ _____/_____

m _____ _____

,5 vw 0.5 b

m 2 o m __

c  ! __

m w

M 0.0 7 8 5 6 ilti 1.0 Ts2 ,

Magnitude idar o ---- Attenuation Standard Error rdE g Revised Standard Error W Computed Std. Err. (all database)

'~

O  :"'

E 0.5 .=

$ Computed Std. Err. (Mwt6 9, avg mag.=7) a g

,- *f c NOTES: 1) Vertical error bar shcws ird 3 + 1 standard deviation. ba M 2) Horizontal bar shows na database magnitude range, ag 0.0 average magnitude = 6.5 6 7 8 5 __

Magnitude 'A COMPARISON OF ATTENUATION CAMPBELL (SolL)

RELATIONSHIP AND REVISED STANDARD -

Fig D-14e _

Date: 17 OCT 94 Project SONGS IPEEE Project No. 934E361 A 'd D-93 Woodward-Clyde 1 asultants

1.0 w l P2A T-0.04 s pj I w i I w

m V

6 0.5 m ----r----*--------------r----

~

g l -,i-1 l r (Mw=

m -,-

m f. I r (Mw=

0.0 1.0 g T=0.1 s T=0.2 s -.

O ,

w w

l w

V w 0a6 g

1 i

T g s.

A g .". __

w (U

m 0.0 1.0 T=0.4 s T=1 s O

t-m -

t 60.5 ----- ----- ----- -----

gg _____-_____.______________._____ .

y ,

. o g

-; I __

m 0.0 1.0 5 6 7 8 w T=2 , Magnitude O

w

---

Attenuation Standard Error m '

y Revised Standard Error '

60.6 '

O Computed Std. Err. (all database) 1.8 (U .L t --

$ Computed Std. Err. (Mwif>.9, avg. mag.=7)

C to w

M NOTES: 1) Vertical error bar shows i 1 standard deviation.

0.0 2) Horizontal bar shows 5 6 database magnitude range, 7 8 average magnitude = 6.5 Magnitude COMPARISON OF ATTENUATION RELATIONSHIP AND REVISED STANDARD ERRORS BOORE et al. (AVERAGE SITE CLASS B AND C)

Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE ~

Fig. D-14d RROR[

-- D-92 Woodward-Clyde Consultants

__ _  ?

1.0 POA T=0.04 e N

2m0*6 N* -N- N-y 2 x - ___[____ ] x.

g __ ..

$ l. '

I.'

m 0.0 1.0 T=0.1 s T=0.2 s 2

a N x '

N '

l

]s0*6

~~

. .Ni ..

~

D --

C 1 5

M 0.0 1.0 T=0.4 s T=1 s 2 s-N.-

$  %' Q__ N ,

p ,,, N'- _____ _____

t m 2 N :--- r---- , N.

" 4 C ".. 1 g .. .

m 0.0 1.0 6 6 7 8 T 2: Magnitude

---

Attenuation Standard Error L1J Revised Standard Error E(15 0.6 O Computed Std. Err. (all database)

.L  %

g $ Computed Std. Err. (Mwr6.9; avg mag.=7) g ..

3 NOTES: 1) Vertical error bar shows e 1 i standard deviation.

2) Horizontal bar shows 0.0 database magnitude range, 6 6 7 8 average magnitude = 6.5 Magnitude COMPARISON OF ATTENUATION RELATIONSHIP AND REVISED STANDARD ERRORS SADIGH (SolL)

Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D 14c D-91 Woodward-Clyde Consultants

1.0 P2A T=0.04 s 4

w l

W

g 0.6 _____g___________________ ____ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ , , , _ _ _

D -

s c ^

g l +

i

-I-w 1 o g __

0.0

1.0

, T=0.1 s T=0.2 s o

w 6

)

3

=mm- 0.6 -

T 4

g .'. ..

w eg i 0.0 1.0 T=0.4 s Tai s )

o '

t W

=6g 0.6 __________L____ ____m___

4 V  ;, __

+ "

e ,

i o . --

Q ..

m 0.0 1.0 6 6 7 8 m T=2 , Magnitude o

g ---- Attenuaton Standard Error l

W __ Revised Standard Error

.g ____________________ __________

6 0.6 M

O Computed Std. Err. (all database) t --

O Computed Std. Err. (W.9; avg. mag.=7)

C N

NOTES: 1) Vertical error bar shows M i 1 standard deviation.

0.0 2) Horizontal bar shows database rnagnitude range, 6 6 7 8 average magnitude = 6.5 Magnitude COMPARISON OF ATTENUATION RELATIONSHIP AND REVISED STANDARD ERRORS BOORE et al. (AVERAGE SITE CLASS B AND C)

Project No. 934E361 A Date: 17 OCT 94 Project: SONGS iPEEE Fig. D-14d

_ , , , D-92 Woodward-Clyde Consultants

i 1.0 1

, P2A T=0.04 s i

2 6

I

W E8 0.6 ' ~

- - - - - ~ - - - - ~ - - - - ~ - - - - ' * - - - - - - - - -

I 3

qll3 T

A --

~

c g

0 0

3 w

! U3 0.0 i

J 1.0 ;

, T=0.1 s T=0.2 s O

b b ,

uJ _____ _____ _____ _____ _____ _____

1

.n-, 0.6

$c ' __

1 --

w N

&)

0.0 1.0 T=0.4 s T=1 s o I t:

W _____ _____ _____ _____ _____ _____

w b 0.6 m i e

1 -_

= s o

^

e ._ _"_

w m

0.0 1.0 6 8 7 8 n- T 2s Magnitude O

a

n. ---- Attenuation Standard Error W Revised Standard Error E5 0.6 O Computed Std. Err. (all database)

A "

=g $ Computed Std. Err. (MveS 9. avg. mag.=7) g __

3 NOTES: 1) Vertical error bar shows e 11 standard deviation.

0.0 2) Horizontal bar shows database magnitude range, 6 8 7 8 average magnitude = 6.5 Magnitude COMPARISON OF ATTENUATION RELATIONSHIP AND REVISED STANDARD ERRORS CAMPBELL (SOIL)

Project No. 934E361 A Date: 17 OCT 94 Project: SONGS lPEEE Fig. D 14e D-93 Woodward Clyde Consultants

. - _ . . - . - - . . - .. . . . . - - . . - - ~ . - - . . . . - .. - . - - - . - . - - - .

1

- Q - Original Attenuation Standard Error (Mve-6-1/2)

? Revised Attenuation Standard Error (Mw=6-1/2)

O Empirical database 1.00 ,

_ 0.80 - -

tlD O

d -

,, _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _.o e ___

o '

g 0.60 -

y' -

k \, "

h H

o p

6. q ( .. .. .. . .

un. 0.40.7 ~~

m W .

c -

n CO 0.20 - -

0.00

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Period, T (s) a j NOTE

Vertcal error bar shows f.1 standard deviation.

i a

i

COMPARISON OF COMPUTED AND ATTENUATION STANDARD ERRORS j AT DIFFERENT SP ECTRAL PERIODS -IDRISS (STIFF SolL) i Project No. 934E361 A Date: 17 OCT 94 Project SONGS IPEEE Fig. D 15a

-w D-94 Woodward Clyde Consultants

! - Q - Original Attenuation Standard Error (Mw=6-1/2) 4

? Revised Attenuation Standard Error (Mw=6-1/2)

C Empirical database 1.00 ,

5 I

_ 0.80 - -

en o

J - -

a u < -----

o ..-

0.60 -

a z

! k

,e

?...%- . .. ..

0.40 i . ..

Qa ..

T . .

c

. ~

a 1

m 0.20 - -

'I 9

i

~

0.00 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Period, T (s) 1 l

i i.

j NOTE Verbcal error bar shows + 1 standard deviaton.

i i

i COMPARISON OF COMPUTED AND ATTENUATION STANDARD ERRORS l AT D FFERENT_ SPECTRAL PERIODS . ABRAHAMSON (STIFF SOIL)

.; Project No. 934E361 A Date: 17 OCT 94 Prc}ect: SONGS IPEEE Fig. D-15b

,_, D-95 Woodward.Clyde Consultants i

- Q - Original Attenuation Standard Error (Mw=6-1/2)

? Revised Attenuation Standard Error (Mw=6-1/2)

$ Ernpirical database i 1.00 ,

j

_ 0.80 -

en O

J .

E 3m ,,,, _ _,, _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ,,

• ~..

Z. .. .. ,

A w

i,t s

~

g y .. .. .. ..

0.40 =- --

p ..

'O .

g .

m se C0 0.20 -

0.00 ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' '

0.0 0.2 0.4 0.8 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Period, T (s)

NOTE Verbcal error bar shows 11 standard deviaton.

1 1

COMPARISON OF COMPUTED AND ATTENUATION STANDARD ERRORS AT DIFFERENT BPECTRAL PERIODS - SADIGH (SolL)

Project No. 934E361 A Date: 17 OCT 94 Project SONGS IPEEE Fig. D-15c

-, D-96 Woodward.Clyde Consultants

- Q - Original Attenuation Standard Error (Mw=6-1/2)

$ Revised Attenuation Standard Error (Mw=6-1/2)

O Empirical database l

j 1.00 , , , , , ,

)

1 l

l l

, m 0.80 -

J CD C

J -

l i

(U w

ll3

• 0.60 -

i 85 Z

w ____

u . . _

_ i 0 __ - .p i b 7 216 . *~~~

0.40 -

6 .-

m ..

'O - ..

g .

(

.U .

CO 0.20 -

t i

O.00

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Period, T (s)

J i

5 NOTE: Vertcal error bar shows + 1 standard deviaton.

i COMPARISON OF COMPUTED AND ATTENUATION STANDARD ERRORS AT DIFFERENT SPECTRAL PliRIODS - BOORE et al. (AVERAGE CLASS B AND C)

Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-15d

. - ,, D-97 Woodward.Clyde Consultants

i___ . _.

j

- Q - Original Attenuation Standard Error (Mw=6-1/2)

$ Revised Attenuation Stanerd Error (Mw=61/2) 1

{ O Empirical database i

i 1.00 ,

l , , ,

i j -

t l _ 0.80 -

cn i O i

I _

_ +. .

3 2

g #___

1

• 0.s0 - ' '

1 a <

z .

1 M J/

p l 2 w k_

! W /%

A -

w 0.40) --

u -

e ..

w -

1 c -

e

! w a

i en i 0.20 -

i

]

! 0.00 ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' '

1 j 0.0 0.2 0.4 0.s 0.8 1.0 1.2 1.4 1.s 1.s 2.0 l Period, T (s)

E i

i 1

1 i NOTE Vertcal error bar shows 11 standard deviaton.

i i

4 i

l .

COMPARISON OF COMPUTED AND ATTENUATION STANDARD ERRORS AT DlFFERENT SPECTRAL PERIODS - CAMPBELL (SOIL) f Project No. 934E361 A Date: 17 OCT 94 Project: SONGS IPEEE Fig. D 15e

-- D.98 Woodward.Clyde Consultants i

a

_._. . . _ . . _ . ____ . _ . _ .. . . _ . . _ _ . . _ _ _ . _ _ . _ _ _ . ~ . _ . . _ . _ _ _ _ _ . _ _ _ . _

t i

l l  :

I 5 Hz, d=10 km, Sadigh 1

5 i i i i ,

gI Median

- - 1 sigma 2 sigma ,

4 _

- - - - 2.5 s!gma

~

............. 3 sigma ....*

2  :

" " " " " " " ~ . . . . . , ., - . . ~ . . .

c3 -

8p ...............

, s

)

e - -

- s 2 - - -

? p en"* .nn**.

=-

D.

1 _

p

. .===

~

' ' i i i 0

5 5.5 6 6.5 7 7.5 8 Magnitude RELATIONSHIP BETWEEN SPECTRAL ACCELERATION, MAGNITUDE, AND STANDARD ERROR Project No. 934E361 A. Date: 17 OCT 94 Project: SONGS IPEEE Fig. D-16

"~~ '-v D-99 Woodward-Clyde Consultants

e l

I 5 Hz 4 i , , , , , ,

3 -

2 -

e a

e4 e 1 -

c e 0 -

k e

$o -1 -

-2 -

-3 -

-4 ' ' ' ' '

-4 -3 -2 -1 0 1 2 3 4 Expected Value RESIDUALS NORMAL PROBABILITY PLOT FOR SPECTRAL ACCELERATION AT 6 Hz Project No. 934E361 A Date: 17 OCT 94 Project SONGS IPEEE Fig. D-17 w w ...-<e D-100 Woodward-Clyde Consultants

l l

l '

l APPENDIX E l TIME HISTORIES FOR FRAGILITY ANALYSIS Prepared by

Woodward-Clyde Consultants

! Santa Ana, California

1 i

TABLE OF CONTENTS l 1.0 INTRODU CHON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E- 1 2.0 TIME HISTORIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-2 2.1 TIME HISTORIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-2 l 2.2 DURATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-4 2.3 APPLICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-4

3.0 REFERENCES

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-5 i I_IRT OF TABLES E-1 Empirical time histories selected for the fragility analysis . . . . . . . . . . . . . . . . . . . E-6 E-2 Duration (seconds) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-7 I E-3 Median d urations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-8 I TST OF FIGURFR E-1 Example of scale factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-9 E-2 Comparison of ensemble average to uniform hazard shape, horizontal i spectral accelerations at 5% damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-10 E-3a Ensemble median and 84th percentile response spectra, horizontal spectral accelerations at 5 % damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-11 E-3b Ensemble median and 84th percentile response spectra, vertical spectral accelerations at 5% damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-12 E-4a Acceleration time histories, horizontal component (x) . . . . . . . . . . . . . . . . . . . . E-13 E-4b Acceleration time histories, horizontal component (x) . . . . . . . . . . . . . . . . . . . . E-14 E-5a Acceleration time histories, horizontal component (y) . . . . . . . . . . . . . . . . . . . . . E-15 E-5b Acceleration time histories, horizontal component (y) . . . . . . . . . . . . . . . . . . . . . E-16 E-6a Acceleration time histories, vertical component (z) . . . . . . . . . . . . . . . . . . . . . . . E-17 E-6b Acceleration time histories, venical component (z) . . . . . . . . . . . . . . . . . . . . . . . E-18

1.0 INTRODUCTION

This appendix summarizes the development of a suite of normalized time histories for fragility analysis that are consistent with the normalized spectral shape presented in Section 5. The average of the nonnalized spectral shapes with peak ground accelerations of 0.67g and 1.34g is used to derme the normalized uniform hazard spectral shape for developing the fragility time histories. The initial empirical time histories are selected to be consistent with the SONGS MCE (M=7.0, d=8 km, strike-slip) for a soil site condition. A minimum of 15 3-component time histories are needed for the fragility analysis.

E-1

2.0 TIME HISTORIES 2.1 TIME HISTORIES The strong motion data sets used by Sadigh et al. (1993) and Boore et al. (1993) were searched to find appropriate empirical time histories. The initial selection criteria used to select the appropriate records is given by the following: magnitude in the range 6.5 to 7.5, distance in the range of 0 to 15 km, and soil site condition. Fifteen records that met this criteria are listed in Table E-1. (To insure a wide range of time histories, similar time histories from closely spaced stations from single events were not included.) Additional records were also considered by relaxing the selection criteria to include rock site (3 records) and soil sites with magnitudes down to 6.1 (8 records). The additional recottis are also listed in Table E-1.

Each component of each record listed in Table E-1 is modified to account for the expected difference in spectral content for the magnitude, distance, and site condition of the selected records and the spectral content of a preliminary uniform hazard shape. The resulting spectra are then compared to the final uniform hazard shape to justify their use. Specifically, a scale factor, a,(T), for the i* record at period T was computed by:

i Sa,(T) / pga ,(T) a,(T)

• Sa(M,, d,, Site ,, T) / pga (M,, d ,i Site ,) (E-1) where Sa,(T) and pga, were determined from a preliminary normalized uniform hazard shape.

Different scale factors were computed for the horizontal and vertical components. An example of the scale factor is plotted on Figure E-1. The spectral accelerations in the denominator of Equation (E-1) were computed using the median empirical attenuation relations by Idriss (1991) and Sadigh et al. (1993) for rock sites and by Sadigh (1987) and Campbell (1990) for soil sites. l The empirical time histories were modified using a time domain spectral modification method (Abrahamson,1991). This procedure is similar to the method of Lilhanand and Tseng (1988) but it uses a different functional form for the adjustments. Following the spectral modifications, a baseline correction was applied to the modified time histories using a high-order polynomial for the baseline model.

E-2

_ _ _ _ ___ _ _ _ _ _ . _ . ~ . _ _ _ . _ . _ _ _ _ _ _ _ _ _ _ . . _ _ _ .

! i l

f i

Finally, for each 3-component time history, the horizontal components were normalized such that i

the median spectral acceleration of the average of the two horizontal components over the '

frequency band of 1 to 10 Hz is unity (even sampling along the log frequency axis). The vertical ,

component was normahzed by the median vertical spectral acceleration over the frequency band of 1 to 10 Hz and then further scaled by the average ratio of the vertical to horizontal uniform hazard spectral shape over 1-10Hz.

j l

De ensemble average for horizontal spectral acceleration at 5% damping for the three data sets are wW to the final uniform hazard shape in Figure E-2. In terms of the ensemble average,

! the full data set (2f.,tations) gives as good agreement to the uniform hazard shape as the smallest l

\

i '

set (15 stations). Therefore, we recommend using the full set of 26 stations in the fragility analysis.

l The median and 84th percentile spectral accelerations of the suite of normalized time histories was computed in two ways: first with no assumption about the distribution of the data and second with an assumption of a log normal distribution. The ense.nble median and 84th percentile response spectra are plotted in Figures E-3a and E-3b for the horizontal and vertical components, respectively. Note that the vertical spectra have been normalized to the 1-10 Hz spectral acceleration of the horizontal component. This was done so that a single scale factor could be used for all three components. The acceleration time histories are plotted to the same time and amplitude scales in Figures E-4, E-5, and E-6.

1 The 1-10 Hz normalization produced unusual results for the 79_agr_z recording (Table E-1).

The unnormaliW spectrum for this recording has a peak at about 20 Hz and relatively low l

spectral accelerations in the frequency band of 1-10 Hz. The normalization over 1-10 Hz required scaling the time history by a factor of 5.6 which produced very large spectral accelerations at frequencies above 10 Hz. Although the large vertical spectral accelerations from  !

the scaled record appear to be unrealistic, this record was nonetheless used in this study.

E-3 l

2.2 DURATION He definition of duration used here is given below: '

Dration .

T (75%)- T (5%)

0.7 (E 2) l where T (x) is the time at which the arias intensity of the accelerogram reaches level x. He dumtions of the modified time histories are listed in Table E-2. The median durations for the three subsets are listed in Table E-3. He durations on the horizontal component for the large magnitude soil sites have a standard error of plus or minus a factor of 2. The durations on the vertical component are slightly less variable (standard error of about a factor of 1.65), he durations for the smaller magnitude soil sites are not significantly different from the durations from the larger magnitude soil sites. He durations for the three rock sites are smaller than the duranons for the soil sites, but they are within one standard error of the median duration for the soil sites.

2.3 APPLICATION For use in the fragility analysis, each time history should be scaled by the average horizontal spectral acceleration over the frequency b'and of 1-10Hz. The same scale factor should be used for all three components.

E-4

3.0 REFERENCE Abrahamson, N.A. (1991). Non-stationary spectral matching, Seismological Research Letters, Vol. 62, No.1 (Abstract).

('m@H. K.W. (1990). Enpirical Prediction of Near Source Soil and Soft-Rock Ground Motion for the Diablo Canyon Power Plant Site, San Luis Obispo County, California, Prepared forlawrence Livemmre Nationallaboratory by Dames & Moore, Evergreen, Colorado, September 1990.

Idnss, I.M (1991). " Selection of Earthquake Ground Motions at Rock Sites," Report prepared for the Structures Division, Building and Fire Research laboratory, National Institute of Standards and Technology, Department of Civil Engineering, University of California, Davis, September.

Lilhanand, K. and Tseng, W.S. (1988). Development and application of realistic earthquake time .

histories compatible with multiple damping response spectra, Ninth World Conf. Eanh.

Eng., Tokyo, Japan, Vol. II, pp. 819-824.

Joyner, W.B., and D.M. Boore (1988). Measurement, characterization, and prediction of strong ground motion. Froc. Conf. on Earthquake Engineering and Soll Dynamics 11, GT Div/ASCE, Park City, Utah,27-30 June 1988, pp 43-102.

Sadigh, K. (1987). Written communication to W. Joyner published in Joyner and Boore (1988).

Sadigh, K., Chang, C.Y., Abrahamson, N.A., Chiou, S.J., and Power M.S. (1993). Specification of Long-Period Ground Motions: Updated Attenuation Relationships for Rock Site Conditions and Adjustment Factors for Near-Fault Effects, Technical Papers on Seismic Isolation, ATC-17-1, pp. 59-70.

E-5

I L

TABLE E.1. EMPIRICAL TIME HISTORIES SELECTED FOR THE FRAGILITY ANALYSIS File Name -

Prefix Earthquake Mag Mech Dist (km) Station Site ,

40_icb 1940 Imperial Valley 7.1 Strike-Slip 8.3 ICSB Soil 78_ tab 1978 Tabas 7.4 Reverse 3.0 Tabas Soil 79_aep 1979 Imperical Valley 6.5 Strike-Slip 5.2 Aempuerto Soil i 79_agr 1979 Imperical Valley 6.5 Strike-Slip 5.5 Agrarias Soil j 79_ bra 1979 Imperical Valley 6.5 Strike-Slip 8.5 Brawley Soil  ;

79_e10 1979 Imperical Valley 6.5 Strike-Slip 8.6 El Centm #10 Soil 79_hyp 1979 Imperical Valley 6.5 Strike-Slip 7.5 Holtville PO Soil 83_pvp 1983 Coalinga 6.5 Reverse 8.5 PVPP Soil 87_wsm 1987 Superstition Hills (B) 6.7 Strike-Slip 13.4 Westmoreland Soil 89_ cap 1989 Inna Prieta 7.0 Oblique 14.5 Capitola Soil 89_g02 1989 Inna Prieta 7.0 Oblique 12.7 Gilmy #2 Soil 89_gav 1989 Inna Prieta 7.0 Oblique 11.6 Gavilan Col. Soil  :

92_jos 1992 Landers 7.3 Strike-Slip 12.0 Joshua Tree Soil i 92_ pet 1992 Petmlia 7.1 Reverse 10.0 Petmlia FS Soil  !

$ 92 rio 1992 Petmlia 7.1 Reverse 14.7 Rio Dell Soil M6.S-7.S, Rock Sites 71_ pac 1971 San Fernando 6.6 Reverse 2.8 Pacoima Dam Rock 76_gaz 1976 Gazli 6.8 Reverse 3.0 Gazli Rock 89_cis 1989 Loma Prieta 7.0 Oblique 5.1 Corralitos Rock M6.1-6.4, Soil Sites 66_c05 1966 Parkfield 6.1 Strike-Slip 5.3 Chalome #5 Soil 66_c08 1966 Parkfield 6.1 Strike-Slip 9.2 Chalome #8 Soil 72_eso 1972 Managua 6.2 Strike-Slip 5.0 Esso Soil i 80_ chi 1980 Mexicali 6.4 Strike-Slip 14.6 Chihuahua Soil  ;

80_cvk 1980 Mammoth Lakes (A) 6.2 Strike-Slip 15.0 Convict Creek Soil 80_ mis 1980 Mammoth Lakes (A) 6.2 Strike-Slip 14.0 Mammoth H.S. Soil 84_g04 1984 Morgan Hill 6.2 Strike-Slip 12.8 Gilmy #4 Soil i 84_hvy 1984 Morgan Hill 6.2 Strike-Slip 3.4 Halls Valley Soil i i

TABLE E-2 l DURATION (Seconds)

File Name Component 4

Prefix x y z Soil sites (M6.5 - 7.5) 40_icb 16.0 24.2 13.4 4

78 tab 11.6 11.1 15 3 79 aep 8.6 8.1 8.8 79_agr 9.0 93 7.8 79_ bra 6.4 4.6 7.1 79_e10 6.8 5.9 16.0 79_hyp 63 6.7 7.9 83_pvp 5.5 6.5 7.5

, 87 wsm 16.7 13.9 16.0 89[ cap 7.6 7.5 8.8 89_g02 2.7 3.5 5.8 89Jav 2.0 1.9 43 92jos 30.9 30.0 323 92_ pet 3.4 9.0 8.2 92_rio 5.9 2.5 8.9 Rock sites (M6.5 - 7.5) 71_ pac 73 8.1 6.1 i 76_gaz 6.6 6.9 6.2 89_cis 5.8 5.2 4.7 Soil sites (M6.1 - 6.4) 66_c05 3.7 2.6 73 66_c08 5.6 9.6 9.4 l 2 72 eso 7.0 5.9 6.1 l 80_ chi 7.6 14.0 8.2 l

- 80 cyk 9.4 10.2 8.0 I 80 mis 7.0 7.2 7.9 l l 84_g04 12.2 11.6 3.0 84_hvy 13.7 12.6 13.2 1

E-7

TABLE E 3 MEDIAN DURATIONS ,

Duration (sec) Std Error (In sec)

Horiz Vert an a, N M6.5 - 7.5, Soil sites 7.4 9.8 0.73 0.50 15 M6.5 - 7.5, Rick sites 6.6 5.6 0.17 0.16 3 M6.1 - 6.4, Soil sites 8.0 7.4 0.48 0.43 8 U

4 4

E-8

1 Brawley Recording of the 1979 Imperial Valley Earthquake (M=6.5, soil) 1.4 1.2 . ,

- . a

1. an #

1 1 ...

g 0.8 )

U 1 S

l g

2 0.6 1

- Horizontal 0.4 .. Vertical 0.2 0

0.01 0.1 1 10 Period (sec)

EXAMPLE OF SCALE FACTOR Project No. 934E361 A Date: 18 OCT 94 Project: SONGS IPEEE Fig. E-1

% ,, E-9 Woodward-Clyde Consultants

1 2.5 ,

Uniform Hazard Shape - H

- - - median (15 stations)

- - - - Median (18 stations) 2

.............. Median (26 stations)

T b

E h 1.5 E

o E

.5 e

l

%s.

$1 If \

1. %S j

3

/

. / j' M

en +

0.5

/ .

'T

\

0 %-

0.01 0.1 1 10 Period (sec) l COMPARISON OF ENSEMBLE AVERAGE TO UNIFORM HAZARD SHAPE '

HORIZONTAL SPECTRAL ACCELERATIONS AT 6% DAMPING _

Project No. 934E361 A Date: 18 OCT 94 Project: SONGS IPEEE Fig. E-2

,, E-10 Woodward-Clyde Consultants 1

4 i

1 1

1 l 2.5 Uniform Hazard Shape - H j - - - 50th (no distribution) -

- - - - 84th (no distribution)

2-- ""** 50th (log normal) 4

R - - 84th (log normal) i E '

e e l E ,Y

! $1.5

g />'

h(

b $ _

f.h' A  %.

V

! i y u.

6 e .-  % k.

f  ? 41 h .

g1 r -

p. .

i '(

< / I. t 5  %

/ f

{s

.l -//

0.5 - \ t -

/ *

\.

• s.\s

\, s .

N s,

\g 'N O

0.01 0.1 1 10 Period (sec)

ENSEMBLE MEDIAN AND 84th PERCENTILE RESPONSE SPECTRA HORIZONTAL SPECTRAL ACCELERATIONS AT 6% DAMPING Project No. 934E361 A Date: 18 OCT 94 Project: SONGS IPEEE Fig. E-3a

% ,, E-11 Woodward Clyde Consultants

2.5 Uniform Hazard Shape - Z

- - - 50th (no distribution)

- - - - 84th (no distribution) 22 50th (log normal) y - - 84th (log normal)  ;

1 I '

e l

? I,

$'. $'{\

• \ l "1 '5 E /' 'l t

j I ,' t E /'8 h O \

E Ie 14 .! ' \

.8 /,' -

1 a 1

,, j .

$o . I, .1 k.T t S.

• h-j ),9

. 4.

\r\ 'g

\

,,y m

0.5 **'l 4.

\ 3 w

%e s*k N asa..g

'N ,%

0 0.01 0.1 1 10 Period (sec)

ENSEMBLE MEDIAN AND 84th PERCENTILE RESPONSE SPECTRA VERTICAL SPECTR AL ACCFt FRATIONS AT 6% DAMPING Project No. 934E361 A Date: 18 OCT 94 Project: SONGS IPEEE Fig. E-3b

% ,, E-12 Woodward-Clyde Consultants

x Cuiw a s 80_rnls

==e=0) f>fh%b==n 80_cw fr1 % $ y/ M $/f W --- - _ -

80_d 79_hvp

^

Jh __ m_ _ ._____ _

-2 g%kWh_ -_- _ _ _ _ _ - 79_e10 79 h Sl' 79_agr 7d h h%a- . _ _ _ _ , _ _

79 aap Mh%btA%e -

78

• v W /y//$w m e_= - -

' ^ '

76_gaz

^- 4 r 72_eso M'h[ h e m I - __ _

71_ pac

~

4% !{fp -

66_c05 hh -

0.7 -

@ M*/4 W 4 p .u._= w m- W-

,i,,I ,i,,I . , i , l . i , .  ! , , i , Ii,i,! , i , i 1,,,, l 0 5 10 15 20 25 30 35 40 TIME (SEC) I ACCELERATION TIME HISTORIES i HORIZONTAL COMPONENT (x)  !

Project No. 934E361 A Date: 18 OCT 94 Project: SONGS IPEEE Fig.E4a a-,e E-13 Woodward-Clyde Consultants l

XM i

i l

l

& & w : ~ : -_ ,_= --- : Rt10 1

W ::-- ::_-

92 Pet m (-:: -__-

_ gr_ =._ .-

aoon Q ds QQ . :: .

_v -

umyqw= -

=n

-Wm%wIpm - - T 84Jw up,. .m . _-%-- _ - _

1. MWN--- -- -_-

84_g04 0.7 -

83.P9 i: & W _: -

-0.7 -

iiliii,Ii,,,I,, , , I , , , , t , ,,,l , , , , l , , ,

0 5 10 15 20 25 30 35 40 TIME (SEC)

ACCELERATION TIME HISTORIES HORIP.ONTAL COMPONENT (x)

Project No. 934E361 A Date: 18 OCT 94 Project SONGS IPEEE Fig. E-4b

• * ' E-14 Woodward Clyde Consultants

-J

y CuTw-6

-4 1$s , y _ _.. - 80 mis hNi j kk W - i 80 cvk 80 m s .

-_ w- - -

p W:: :~ . - - -  :

h _ - = - ---

44_ ""

, 797/hf IM^+~ n ano

~

m k Q y.-!;- m --- ~-

n us-

-::.fMhMs #

4hl$ v

^~

~

n Yp:,

4 g __- _-

se a

__- n :_- _- es a 0.7 -

myw w- 4o m

~

, , ,,!,,,,I , , , i li iiiliii>I i ie i li ' ' 'I' ' ' '

O 5 10 15 20 25 30 35 40 TIME (SEC)

ACCELERATION TIME HISTORIES Project No. 934E361 A HORIZONTAL COMPONENT (v)

Date: 18 OCT 94 Project: SONGS IPEEE Fig. E-5a

. E-15 Woodward-Clyde Consultants

yCmwws 1

M

&; - : .= , = :::_ - - = ~

~

&gby,- = _ -y;.

% - 1 4pm -

• 9

_ s p W -:

t .

etw_=

87 ***

ggw,Ag _-

' 8'

• 4 = .

wg g w p _- _%--- - --

~

• • ' ~

mn

p. -

-0.7 -

,,,!,,,il iiI,,i li , , ,

,,,iI i,,iI,,,,I, i , i 10 15 20 25 30 35 40 0 5 TIME (SEC)

ACCELERATION TIME HISTORIES HORIZONTAL COMPONENT fv)

Project No. 934E361 A Date: 18 OCT 94 Project: SONGS IPEEE Fig. E-5b

. ,e E-16 Woodward-Clyde Consultants

zCu w s 1

1

' Y , .'. . f,' = 80_%

80_cW

.z' .*

/

] W L',k %',* ^ _ -

= _ _ - _ -

_n__ _ _Y w ut , .,m _ .. . __

7s %

_ mi, c v 1 x, .c y r _ -

.t m io ph t , 79 e10 t

. u , 9. .' r "' ''& ' O --

' - - - ~

i i _ Lt Ml-maJa bl . - 7Q g g, p r, P __ _

'8: '

sni A , :Id e d u- a u . _ _ _ , _ 79 g i

qg r,o p y yg n r - -, - - - -

.ai.L.Lt. .t1 i ha 1 . .u m 79 ,

i -- ,'i'rr 'P f 'rl' r -ir "

7B_2

-- M f%p b.1v'-A&.pptm -

- t

..Li t a.i t [ .1 a 6 76 p

. g

r L- t .L A4- _

. 72_eso en rvi si 71_ pac pl. z,, j .

M c0B M.$.Yn

.- - anL. -'---

66_c05 Tyr 7 Y 0.9 -

a- i a um o .,s,__--..

40_ich

--r.7,,__..,..._,.-

j

-0.9 -

i i i i li, i! , , , i I,, , , Ii,,,l , ,,,l , , , , l , , , ,

0 5 10 15 20 25 30 35 40 I

TIME (SEC)

ACCELERATION TIME HISTORIES VERTICAL COMPONENT fr)

Project No. 934E361 A Date: 18 OCT 94 Project: SONGS IPEEE Fig. E-6a

'-*' E-17 Woodward Clyde Consultants

z Cvii w .- .i.

l eV _% _ - - -

-- _ __._ 92_rio 92_pst Mg _ - ._

1

- l ',', ^cMWI ",, - 14pec_  :-;^et__*n.2 h -

'.7; ,hp. _

89_ gov 1

2 ,-^^4 ^

_ ~ - - - - - -

I w_ --

89_ds

- .22h1J Latini, auu i . ,u __ . _ _ _ _ . 89_ cap i,

n p v , ,, y ,, ,, . , . .. -

i k.l 811L11 Ad. , 8, .! Llg ,L {J,l j , lla.i n g __ { hM n vmTvrpy m rnvv "v ,"-- - -

j 84_hvy Khby, 42 r u -- -- -____ ___ _

84_go4 0.9 -

83_pg O"fiict' 7 h . = _-

-0.9 -

, , , i  !,i,,! , , , , I,, , ,t, ,,,I , ,,,I , , , , I,,,,

0 5 10 15 20 25 30 35 40 TIME (SEC)

ACCELERATION TIME HISTORIES VERTICAL COMPONENT (r)

Project No. 934E361 A Date: 18 OCT 94 Project: SONGS IPEEE Fig. E-6b wm E-18 Woodward-Clyde Consultants

l .

l OVERSIZE l DOCUMENT ~

l PAGE PULLED i

l SEE APERTURE CARDS NUMBER OF OVERSIZE PAGES FILMED ON APERTURE CARDS 4

APERTURE CARD /HARD COPY AVAILABLE FROM RECORDS AND REPORTS MANAGEMENT BRANCH

i OVERSIZE DOCUMENT l

l PAGE(S) PULLED SEE APERTURE CARD FILES i

l APERTURE CARDIPAPER COPY AVAILABLE THROUGH NRC FILE CENTER NUMBER OF OVERSIZE PAGES FILMED ON APERTURE CARD (S)

ACCESSION NUMBERS OF OVERSIZE PAGES:

i j l'ft //G d o 8 5 - O l f

i

!