ML20027A438
| ML20027A438 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 11/15/1978 |
| From: | Stepp J Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20027A437 | List: |
| References | |
| NUDOCS 7812040324 | |
| Download: ML20027A438 (41) | |
Text
i b
I gTY,;.,
,(,'
--m
.f s
/.
s,'
W -g
. m
- g. ?,
t..
V
/
.,/
TESTIMONY OF DR. CARL J. STEPP
. /
n
pa DIABLO CANYON TESTIMONY GEOLOGY AND SEISM 0 LOGY CONTENTS.
I.
INTRODUCTION (history of OL Review - Summary of staff's conclusions)
A.
Conclusions Following CP Review B.
New Data Following CP Review'.
C.
Concerns of USGS and NRC During OL Review D.
Investigations by Applicant to Resolve Concerns E.
Intervenors Concerns F.
Summary of Staff's Conclusions II.
BASES FOR THE STAFF'S CONCLUSIONS A.
Length of the Hosgri Fault Zone 1.
Southern extent - relation to Transverse Range Structures 2.
Northern Extent of the Hosgri and Relationship to San Simeon Fault Zone 3.
Relationship of the Hosgri and San Gregorio Fault Zones B.
Amount of Strike Slip Movement on the Hosgri Fault Zone C.
Location of the 1927 7.3 Magnitude Earthquake D.
Maximum Earthquakes on the Hosgri Fault Zone E.
Ground Acceleration for Magnitude 7.5 Earthquake on the Hosgri III.
References 7
fN DIABLO~ CANYON TESTIMONY GE0 LOGY AND SEISM 0 LOGY I.
Introduction (history of Operating License (0L) review and statement of staff's conclusions)
A.
AEC,'USGS and NOAA's Conclusions Following Construction Permit (CP) Review This geology and seismology evaluation reflects our review of investigations conducted since 1969. These investigations are described in the Final Safety Analysis Report (FSAR) for the Diablo Canyon Nuclear Elant site.
The geology and seismology of the Diablo Canyon site as presented in the Preliminary Safety Analysis Report was reviewed by the AEC staff and its geological and seismological advisors, the U. S. Geological Survey (USGS) and the U. S.-
Coast and Geodetic Survey, during the construction permit review.
The findings of that review were published on November 18, 1969,.as part of the Safety Evaluation Report (SER) for.
Unit 2.
With respect to the seismic design input, the staff concluded:
(1)
"There are no identifiable major faults or other geologic structures in the area that could be expected to localize seismicity in the immediate vicinity of the site. The nearest seismically active major fault is the Nacimiento fault, a northwest-trending fault zone that approaches to-within about 18-20 miles of the site to the northeast," and
E 4
,- x 2
(2)
..the Coast and Geodetic Survey agrees with the applicant's statement of 0.20g at the site and on rock for the predicted maximum ground accelerations of the design earthquake and twice that value, 0.40g on the rock for the safe shut-down conditions."
B.
New Data Following CP Review Since publication of the SER in 1969 studies of the geologic structure offshore from the site have been published (Hoskins and Griffiths,1971; Wagner,1974). These studies revealed significant geologic structure offshore from the Diablo Canyon site. To determine the detailed structural relationships in the offshore region the applicant conducted extensive high resolution geophysical investigations along that reach of the structure.
Profiles obtained by the applicant were made available to the USGS and those obtained early in the investigations were included in the independent interpretation of the offshore structure by Wagner (1974).
The applicant's interpretation together with a summary of the f
results presented by Hoskins and Griffiths (1971) and Wagner l
(1974) are included in the FSAR for the Diablo Canyon site.
The Hoskins and Griffiths (1971) paper gives the results of an interpretation of extensive deep penetration seismic reflection surveys along the California Coast.
The survey: revealed a i
l
,x 3
structural basin offshore of the southern Coast Ranges which they called the Santa Maria basin.
It is described as being a shallow, synclinorium about 140 miles long and 25 to 30 miles wide. Structural grain within the basin trends nortn-west parallel to the trend of the basin. Major faults bound the basin on both the east and west. The eastern border fault as identified by Hoskins and Griffiths passes within about 5 miles of the Diablo Canyon site.
It is about 90 miles in total length.
Wagner (1974) utilized both deep penetration ::eismic reflection methods and high resolution seismic surveys.
The configuration of the sea floor was obtained using precision bathymetric measurements and, locally by side-scan sonar. These techniques provided a considerable refinement of the structure along the eastern boundary of the Santa Maria basin in the region between Cape San Martin and Point Sal. The basin is l
indicated to have formed in Middle-to Post-Miocene (post 26 m.y.)
l time.
It contains from 2,000 to 5,000 ft of Miocene sediments unconformably overlain by up to 3,500 ft of Pliocene (7 m.y.)
section. An erosion surface is indicated to have formed on these Tertiary beds during Pleistocene time.
Post-Wisconsinan age sediments, deposited during the past 20,000 years, overlie much of the Tertiary erosion surface.
[
p-4 Wagner (1974) concurred with the interpretation of Hoskins and i
Griffiths (1971) that a major fault zone forms the eastern boundary _of the offshore Santa Maria basin. He called it the Hosgri fault. The Hosgri fault is a zone containing from 2 to 5 subparallel fault splays which locally offset Tertiary and Pre-Tertiary rocks with apparent vertical displacements ranging between 1,500 ft and 6,000 ft. The Hosgri fault is discontinuous and segmented in the late Tertiary and Quaternary section.
The applicant interprets the East Boundary Zone (the Hosgri fault zone of Wagner,1974) as being the boundary between synclinal downwarping of the offshore Santa Maria basin and regional uplift of the southern Coast Ranges. The style of faulting in the zone is extensional as shown by its localization along the flank of a regional upwarp and by its pattern of basin down normal faults and crested faults along the flank of local structural highs at Point San Luis and Point Piedras Blancas.
Reverse drag downfolding characteristic of _ extensional defomation is also shown in the strata adjacent to the normal faults. Normal l
faults with east-facing scarps have also been identified and are interpreted as being antithetic faults of the overall extensional system. The applicant states that,due to the lack of evidence for compressional deformation in the Pliocene and Pleistocene sediments i
x 5
and the presence of the positive evidence for extensional deformation, the Santa Maria basin is in a region that has probably been characterized by extensional strain during much of the time since initial deposition in the basin during the Miocene.
While the movement on the fault zone was predominantly vertical during Tertiary, Wagner (1974) cites evidence of
. lateral (strike-slip) movement in the upper section.
Earthquake focal. mechanisms for this zone determined by the applicant support a strike-slip component of movement. Thus vertical movement on the fault may currently be subordinate to the strike-slip.
Evidence of recency of movement on the Hosgri fault zone is found in offsets of the sea floor together with offsets of the Post-Wisconsinan sediments. Wagner (1974) found those offsets on three of his profile crossings of the zone.
On other high resolution seismic profiles, offsets of the base of the Post-Wisconsinan sediments are observed but with no offset of the sea floor. Still other profiles show no offset
~~
of the Post-Wisconsinan sediments. This pattern of offset is largely supported by the applicant's investigations. We, therefore, cor.;1ude that the Hosgri fault zone must be considered capable within the meaning of 10 CFR Part 100 Appendix A, Section III (1).
1
6 The applicant places the Hosgri fault in his seismic potential category of Level III which is defined as " Potential for earthquakes resulting chiefly from mcvement at depth with no surface faulting, but at least with some possibility of surface faulting of as much as a few miles strike length and a few feet of slip." Although current movement on the Hosgri fault appears to be limited to local fault segments, we assume for the purpose 9f establishing the safe shutdown earthquake (SSE), that the fault is continuous over its 90 mile length.
In its geological input to the SafetyTveluation Report dated 28 January,1975 (Supplement No.1, Appendix D) the USGS concluded that the Hosgri fault and the Santa Lucia Bank zone (the fault zone that bounds the western side of the Santa Maria Basin) "should be considered inextricably involved with the strike-slip fault mechanics of plate boundary motions that are currently concentrated along the San Andreas fault." The USGS further concluded that earthquakes along the Hosgri fault.should not be expected to be as large as those expected along the San Andreas, but that, based on the nited information on the Santa Lucia Bank fault, "the occurrence of an earthquake as large as events characteristic of subparallel strike slip faults, which bound basins, such as the Santa Maria..." could not be precluded.
p-7 In the Seismology Section of that report the USGS concluded that "with the limit of the present information as to the interpretation, of the relationship of the East Boundary fault to the Sante Lucia Bank fault, an earthquake similar to the November 4,1927, event but occurring along the East Boundary Zone or the Santa Lucia Bank fault zone represents the maximum earthquake that is likely to occur near to the site."*
In its review of Amendments 31, 32, 34, 37 and 40 to the FSAR, which was transmitted by letter dated 29 April,1976 to the NRC from the Acting Director for the U. S. Geological Survey, the USGS reaffirmed its conclusion transmitted on 28 January, 1975 and that was quoted above.
In addition to reaffirming that conclusion, the USGS recommended that a magnitude of 7.5 on the Hosgri fault be used for the design basis earthquake.
The NRC accepted the USGS recommendations.
C.
Concerns of the USGS and NRC during OL Review Following our review of the data regarding the Hosgri fault zone, the Staff requested information to address the following:
1.
The nature of the intersection between the Hosgri fault zone and the Transverse Ranges faults.
East Boundary zone is equivalent to Hosgri fault zone.
i- ' %
8 2.
The~ northern extent of the Hosgri fault zone and relation to the San Simeon Fault Zone.
3.
The location and probable mechanism of the 1927 earth-quake, and the relationship of this event to geologic 1
structure.
4.
The determination of mar.imum earthquakes that can be expected on faults of various ranks within the San Andreas Fault System, and the relationship of each to historic seismicity.
5.
The maximum earthquake on the Hosgri fault done, and the potential effects on the site if such an event occurred on the segment of the Hosgri fault nearest the site.
D.
Investigations performed by the Applicant In response to NRC concerns regarding the relationship between the southern segment of the Hosgri fault and its intersection with the Transverse Ranges structure, the Applicant obtained new field data and synthesized it with the data which had been previously compiled and analyzed in 1974. The post 1974 data consisted of:
l.
Widely spaced seismic reflection lines in the region 1;
between the latitudes of Pt. Sal 'nd Pt. Conception.
2.
Four lines from the southernmost part of the USGS Kelez survey.
3.
Data open filed by the USGS in 1975 from the 1972 R/V
9 Bard ett cruises -(Silver and Von Huene, 1975) and the R/V Polaris cruises (Wolf, 1975).
4.
Proprietary data Regarding the northern reach of the Hosgri fa>1t zone the applicant utilized interpr,etations given in earlier reports by Hoskins and Griffiths (1971), Wagner (1974) and Appendix 2.5D to the Diablo Canyon FSAR (ESA, 1974).
Data obtained subsequently included sparker records from the USGS Bartlett cruise (Silver and Von Huene, 1975) and the sparker records from a survey of the offshore region north of Point Lopez.
Additional studies were conducted by PGLE to better define the location of the 1927 earthquake and to attempt to determine
~
i its relationship to structure. These additional studies include:
1.
Re-analysis of the data reported in the International Seismological Summary to locate the 1927 event and to estimate standard errors and joint confidence regions for-theepicentralcoordibates.
2.
A study of the aftershock sequence of the 1927 darthquake.
3.
A reevaluation of Byerly's~(1930) intensity data.
4.
Onshore geologic mapping.
5.
Offshore seismic reflection profiling.
g.
10 The results of these studies indicated that the 1927 earthquake was closer to shore than Byerly. (1930) had initially located it; it experienced a large component of dip-slip movement; and a probable structural candidate for the source of this earthquake was determined to be the offshore Lompoc fault which shows evidence of considerable Late Pleistocene or Holocene displacement.
E.
Intervenor's Concerns The Intervenor transmitted to the Diablo Canyon ASLB by letter dated April 24, 1978 the following contentions in the areas of geology and seismology:
The seismic design for the Category I structures, systems and comporents of the Diablo Canyon Nuclear Power Plant (Unit 1) fails to provide the margin of safety required by 10 CFR Part 50 and 10 CFR Part 100 in that:
1.
The Applicant has failed to conduct investigations of the Hosgri fault system to determine adequately (i) the length of the fault; (ii) the relationship
(
of the fault to regional tectonic structures; and
~~
(iii) the natui e, amount, and geologic history of displacements along the fault, including particularly the estimated amount of the maximum Quaternary displacement related to any one earthquake along the fault.
I l
l
n-11 2.
A 7.5 Magnitude earthquake is not an appropriate value for the safe shutdown earthquake.
3.
A.759 acceleration assigned to the safe shutdown earthquake is not an appropriate value for the maximum vibratory acceleration that could occur at the site.
E. Surrinary of the Staff's Conclusions The staff has reviewcd the results of the investigations mentioned above and other data. The following is a summation of our conclusions. The bases for these conclusions is presented in Section II.
1.
It is the NRC staff's conclusion that the Applicant has conducted an adequate investigation, which, when synthesized with data by other investigators, provides a basis for making a reasonable and conservative interpretation as to the length of the Hosgri fault zone, its relationship to other regional tectonic structures, and the nature, amounts, and geol% ic history of displacements on the fault.
2.
The Hosgri fault, although possibly belonging to the same fault system, does not appear to be directly linked to the San Simeon fault.
3.
The Hosgri fault may have experienced strike slip movement up to a few kilometers.
It has not, in our view, experienced strike slip movement on the order of 80 to 115 kms, as suggested by Hall (197'7).
s, 12 4.
The 1927 earthquake could have occurred on either the Hosgri fault zone or faults of the Transverge Ranges structure based on error in location. The totality of the data supports an association of this event with the Transverse Ranges structures.
5.
Itisconservativetoassume37.5magnitudeonthe Hosgri fault.
6.
Anaccelerationvaiueof0.75gisaconservativevalue for scaling the response spectra which describe the horizontal ground motion for seismic design at the site.
II. Bases for the Staff's Conclusions A.
Length of the Hosgri Fault Zone 1.
Soethern Extent and Relationship to Transverse Range Structures In Appendix 2.5E of the FSAR the Applicant presents seismic profiling data that led to an interpretation that the Hosgri fault bends to the east a few miles south of Point Sal and l
l eventually dies out. However, based on an independent review of the seismic profiles, the U. S. Geological Survey interprets the data to indicate that the Hosgri fault extends l
at leave five miles south of Point Arguello (April 29, 1976 letter to B. C. Rusche, NRC from H. W. Coulter, Acting i
Director,USGS). Although the precise location of the l
l termination or merging of the Hosgri fault with the Transverse 1
l
O' 13 Range structure is not known, we conclude that the preponderance of the geological data support such an occurrence within the offshore zone of intersection of the Coast Range and the Transverse Range structures.
During the review of the FSAR, the NRC staff and the USGS reviewers were concerned about the southern extent of the Hosgri fault zone and its relation to the Transverse Ranges structure. Because of that concern we requested that the Applicant " provide additional documentation including seismic reflection profiles on the intersection of the Hosgri-fault 2cne with the Tr.ansverse Range faults " It was further-requested that the Applicant " include geologic maps southward-of those provided in the FSAR showing the structural relationships of the Transverse Range faults and structure to the faults r
and structures having a northwest trend."
Prior the request for additional information in 1974 the geological consultant for the Applicant, Earth Sciences Associates (ESA) had analyuzed data from the literature by Von Huene, (1969 and 1971), H skins and Griffiths-(1971), Jennings (1973), and Beyer, et. al (1974).
ESA had also evaluated new seismic reflection data by Bolt, Beranek and Newman, Inc. and Aquatronics, Inc. Subsequently additional data became available,. including USGS open file data from the 1972 Bartlett (Silver, 1975) and Polaris (Wolf, 1975) cruises. Proprietcry data were also examined by ESA. These new data were synthesized with the old data mentioned t.
.m
.. ~,. -. -
. __.~,
p-~%
14 above and interpretations made by ESA concerning the southern extent of the Hosgri fault. This analysis is presented in Appendix 2.5E to the Diablo Canyon FSAR.
Hoskins and Griffiths (1971) described the Santa Maria Basin l
as being 140 miles long, 25 to 30 miles wide and extending southward to the latitude of Point Conception. The recent seismic reflection work appears to support that description.
Evidence presented in Appendix 2.5E of the Diablo Canyon FSAR suggests that:
1.
South of Point Sal, the eastern portion of the Santa Maria Basin begins to show development of fold and fault structures more characteristic of the Transverse Ranges than are seen further to the north. The Hosgri fault, which appears to be the eastern boundary of the basin, loses its identity as a major dislocation through the Pliocene section and becomes a complex disturbed zone with apparently less vertical separation.
- 2. The Santa Lucia Bank structural trend appears to swing to the east at a point about 25 miles west of Point Conception and projects toward an east-west fault, which apparently cuts across a projected extension of the Hosgri fault.
- 3. Between Point Arguello and Point Sal the dcminant strt ctural grain is N30*W in the area where the Transverse Ranges structure has veered from east-west to a more northerly trend, merging with structural trends of the Santa Maria Basin.
'3
- a.
Structural trends of the Transverse Range province are aligned in an east-west direction, but west of Point Arguello the structural trend is more northward.
East-west oriented folds are interpreted to overprint or to be overprinted by north-south folds, b.
The east-west synformal structure underlying the Santa Ynez River Valley turns northward just offshore and does not touch the Hosgri, c.
The Lions Head fault extends to sea in a north-sweeping are to a distance of about 2 miles from the Hosgri.
4.
A north-to-south plot of similar disturbed zones appearing on seismic profiles, which are interpreted by the applicart to be the southern extension of the Hosgri fault, suggests an eastward bending of the zone south of Point Sal toward the Lions Head and Santa Ynez faults. (Polaris Line 1-7,'
4-9, and 2-17A, Figure 10, 12b and 12c respectively, Appendix 2.5E, Diablo Canyon FSAR).
5.
An interpretation that the West Hosgri fault which is located west of the main trace of the Hosgri fault, bends eastward south of Point Sal.
6.
The Purisima fault, which is three miles west of the Hosgri fault, subparallels the southern termination of the Hosgri as defined by the applicant, and extends five miles further, arcs shoreward, and becomes a series of well-defined compres-sional folds west of Purisima Point.
m 16 Dames and Moore (1977) in its analysis of the Central California coastal area for a potential LNG site, using all available offshore and onshore geologic data regarding the southern extent of the Hosgri fault zone, which is summarized by Dibblee (1978), concluded that the bulk of available evidence indicated that the Hosgri veers toward the coastline between Point Sal and Point Arguello and does not extend southward to or beyond the latitude of Point Arguello. Dibblee concurred T
with this conclusion based on.his familiarity with the work of the USGS and Earth Science Associates investigators.
We consider the above data to support the interpretation that the Hosgri fault either terminates or passes into the Transverse Ranges structure. We regard this interpretation to be consistent with mapped Coast Ranges structures in the region where they intersectTransherseRangestructures.
2.
Northern Extent of the Hosgri Faul-t Zone and Its Relationship to the San Simeon Fault Zone The staff's review of the Diablo Canyon FSAR considered the relationship between the Hosgri and San Simeon fault zones.
By NRC letter dated 12 February,1975 the staff requested that PG&E " provide additional documentation, including seismic profiles, on the northern reaches of the Hosgri fault zone." We further requestedPG&Etoincludeafullerdehelopmentofits hiewsconcerningthestructuralrelationshipbetween the Hosgri fault and the San Simeon fault. The Appli-cant responded to this request in Appendix 2.5 E T
rww' y
v v
17 dated August,1975, wi.:h summarized the results of previous studies in that area and the results of additional investigations conducted in direct response to the Staff request.
The area within which the northern reaches of the Hosgri fault zone extends had previously been described by Hoskins and Griffiths (1971), Wagner (1974) and by Earth Sciences Associates (ESA) in Appendix 2.50 to the Diablo Canyon FSAR. Subsaquently, data from the USGS Bartlett cruise (Silver, et. al.1975) became available.
In addition, data from J. sparker Survey conducted by BBN, Inc. in 1974 was provided.
Based on our review of these data, it is the staff's concicsion that the Hosgri and San Simeon fault zones belong to the same coastal zone of deformation. The style of tectonism within the coastal deformation zone is one of anastomosing and en echelon faults, which is typical of otherfault systems within the Coast Range that are subsidiary to the San Andreas.
Data presented in Appendix 2.5E of the FSAR indicate that these two faults approach as close to each other tas 21/2 miles north of Estero Bay.
However, the weight of the data leads us to conclude that they are not directly linked. Thc following paragraphs describe the evidence on which we rely in reaching that conclusion.
s 18 1.
Both Hoskins and Griffiths (1971) and Wagner (1974) indicate that the Hosgri fault, or branches of it, extend north of any postulated intersection with the San Simeon fault.
From these data the two faults appear to form an en-echelon or anastomosing pattern rather than that of a single fault.
2.
Seismic reflection lines presented in Appendix 2.5E to the FSAR which cross a northern projection of the Hosgri fault zone suggest that it terminates in a series of discontinuous branches and folds north of Point P.iedras Blancas. These branches do not appear to veer toward the San Simeon fault but rather to. gradually die out along strikes that are subparallel to the San Simeon' fault.
3.
Seismic reflection lines presented in Appendix 2.5E to the FSAR which cross the Hosgri fault between Point Estero and San Simeon do not show any major branches of the Hosgri fault extending toward the projected southerly extension of the Sam Simeon fault.
4 4.
These reflection lines suggest that the contact between acoustic Units A2 and A3 approximately parallels the shoreline and there is no indication that it is offset by major vertical or lateral faulting.'.
7 19 5.
The straight coast line between Cambria and Point Estero is strongly suggestive of a continuation of the San Simeon fault immediately offshore in that area.
This is well southeast of the northern reach of the Hosgri fault zone. At Point San Simeon which is the most southerly onshore exposure of the San Simeon fault, the Monterey cherty shale lies along the southwest side of the San Simeon fault. This unit can be traced 4 miles to the southeast in seismic reflection records, suggesting that there is no major faulting in that reach'..
3.
Relationship of the Hosgri.and the San Gregorio Fault Zones We consider the data to indicate that the Hosgri and San Gregorio are not linked to form one fault.
1.
The San Simeon and Hosgri faults form the eastern boundary of the Santa Maria Basin.
Hoskins and Griffiths (1971) map thc: northern boundary of the Santa Maria Basin as being the west-northwest trending Point Sur antiform and the Pfeiffer fault. Data presented by the applicant in Appendix 2.5E of the FSAR (Plate 1(N) and Figures 6 (N))
indicate that the northwest trending offshore structures
~~
(including the San Simeon fault) turn into a more westerly trending structural grain at the Point Sur antiform. This suggests that the San Simeon fault either veers to the west-northwest or continues as the Point Sur fault. The Point Sur fault is mapped as a thrust fault while the San Simeon displays predominantly normal movement.
r
s 20 2.
In its review of the geologic and seismologic data i
relevant to the Diablo Canynn Nuclear Power Station, Units 1 and 2, transmitted by letter dated 29 April 1976 to Ben Rusche from H. Coulter, the U. S.
Geological Survey concluded that offshore fualts north of Point Piedro Blancas (an area of possible
~
linkage between the San Simeon and San Gregorio faults) do not form a single continuous fault.
3.
The USGS states that the San Simeon fault if projected northwest immediately offshore is truncated by the Sur Nacimiento fault zone, (USGS 1976).
4.
Greene et al ('9/3) interpret the more southeasterly trend-ing Palo Colorado as being the southerly continuation of the San Gregorio fault.
5.
Jennings (1977) maps the San Gregorio and Palo Colorado as part of the same fault zone.
6.
Data presented in USGS Open-File Report 77-79 (1973) by McCulloch and Chapman show a change in trend of magnetic anomalies from NW tr WNW between Lopez Point and Point Sur.
We interpret the data to indicate that the Hosgri fault terminates in folding in this region or trends more westerly.
l
3 21 B.
AmountofStrikeSlipMohementontheHosgriFault Zone Hall (1975) postulhted that more than 80 km of right-lateralmohementhasoccurredontheHosgrifaultsince Miocene time. He based his conclusions primarily on stratigraphic sections west of the San Simeon fault and eastoftheHosgrifaultwhichheconsiderstohahebeen once contiguous, but which are now separated by more than 80 km.
l Thestaffconsiderstheahailablegeologicinformation to indicate that some strike-slip motion has taken place on the Hosgri fault. We consider it likely that the cur-rentmodeofmohementispredominantlystrikeslip. Al-though the possibility of large lateral displacement can-notbecompletelyruledout,webeliehetheavailable data support the conclusion that no more than a few kilometersofstrike-slipmohementhasoccurredsince the Miocene (20 m.y.).
This conclusion is based on:
1.
An analysis of the southern extremity of the Hosgri fault zone indicates that the fault zone merges with the Transverse Ranges structures.
l.
22 2.
Thepreponderanceofahailablegeologicevidence supports the conclusion that the relationship between the Hosgri, San Simeon, and San Gregorio fault zones is one of an echelon or anastomosing series of faults, 1
which is typical of fault systems in the Coast Ranges, and not acontinuous plate margin master break like the San Andreas.
3.
Data from a test well (Standard Oil "Oceano No. 1")
which encountered the Middle-Miocene Obispo formation, which is of restricted extent, at a point directly op-posite its main onshore occurrence indicate a lack of large lateral offset of rocks east and west of the Hosgri fault.
4.
Submarine geomorphic features, such as the buried extension of the Sur Canyon, which is filled with probable Plio-Pleistocene sediments, and the Monterey submarine-1 canyon, which cross the projection of the Hosgri and San Gregoriofaultzonesrespectihelyappeartopreclude more than a few kilometers of strike ~ slip displace-ment.
The staff considers the available geologic information to indicate that some strike-slip motion has taken place on the Hosgri fault. We consider it likely that the current 4
1 v
,m r
.e e
-- ~,,, -. - - -
23 mode of mohement is predominantly strike slip. How-eher,webeliehetheavailabledatasupporttheconclusion thatnomorethanafewkilometersofstrike-slipmohement.
has occurred since the Miocene (20 m.y.)
C.
Location of the 1927,7.3 Magnitude Earthquake An accurate location for the earthquake of Novebmer 4, 1927 can not be obtained because of the lack of precise recordingsofthatehent. Several sets of geological and seismological data have been evaluated in considering the location of this earthquake. Although each data set containsuncertainties,theNRCStaffbeliehesthatthe data taken together support an association of this earth-quakewithtransherseRangesstructure. The uncertainty of the dataissuchthat-wecannotconclusihelyidentifythefault on which this earthquake occurred. Specific data and arguments considered are discussed below.
l l
l.
Byerly (1930): western margin of the Santa Lucia Bank fault zone.
2.
Gawthrop (1975): Near the Hosgri and Lion's Head faults.
3.
Hamilton (1975): The Lompoc fault.
i 4.
Engdahl (1975): A few miles south of Purisima Point on the coast, a location that falls on a shoreward j
extrapolation of offshore-Purisima fault.
l l
l
.{
~
.24 5.
Smith (1975): The 1927 aftershocks identify an area of permissible sources for the main earthquake.
Smith prefers a location between the Lompoc and-Santa Lucia Bank faults.
6.
Hanks,etal.(1975): Prefer a location 30 km due west of Point Arhuello at 341/2 N,121 WI 7.
International Seismological Summary (1931): Gihes a location at 34.9N,121.b W.
J 8.
USC & GS Japort (1927): Gihesalocationat34.5N, 120.7 W.
Items which we considered are:
1.
Docketed report-by Smith (1975):-
Smith concludes that the 1927 earthquake probably occurred closer to shore than originally thought'by Byerly in--192b. His conclusion is based on P-and S-wahe arrihals at two nearby stations and the area of maximum damage.
2.
The relocation of 1927 earthquake based on seismograms collected by Gawthrop (1975): Most data are clustered in Europe. Theinstrumentationgenerallyahailableat
[
that time had peak responses between 4 and 6 seconds.
l Seismic waves emerging from bac'kground noise on record-inngs from these instruments are difficult to measure ondistantrecordswithanaccuracybetterthanseheral seconds.
l-
4 25 -
The errors in the data are such that a convergent solution for the earthquake epicenter could not be obtained when the entire data set was used. Gaw-throp, therefore, used several selected subsets of recordsfromtheentiresetofrecordingsahailable for the earthquake. Each subset resulted in a sep-arate location being obtained. Thelocationsharied within a 50 mile diameter circle which included the Hosgri fault and several others. The selected group of stations preferred by Gawthrop resulted in con-hergence on the Hesgri Tault.
3.
Engdahl (1975): This study was similar to that of Gawthrop.
It,too,showedthatconhergencecould not be obtained when the entire data set was used.
Selected subsets of data each converged on a different epicenter. Theseepicentersfellwithinacircleof I
about 40 km diameter. Error ellipses were calculated for each epicenter. These error ellipses could be j
enhelopedbyanellipticalareaabout50x80 km.
l l
This area encloses many known faults, including part i
of the Santa Lucia Bank fault zone. Many of the error ellipses did not touch each other, suggesting large errors in the data. Two of the nearest stations l
w
~
26 in 1927, Berkeley and Mount Hamilton, clearly differ by6secondsfromtheothergoodstationsahailable in 1927. These stations were controlled by the same time source. Thus, there is a suspicion that a tim-ing error of 6 seconds existed on the day of the 1927 earthquake. These good, proximate stations with sharp high amplitude arrivals would normally be used to test other less precise more distantly recorded data.
Withtheuncertaintimecorrection,howeher,nogreater reliance can be placed on these stations than on others.
BeceusemoststationsahailablewereinEuropeat essentially the same azimuth from the earthquake, only 4 or 5 recordings actually control the eoicenter.
Twooftheseappeartohaheaserioustimingerror.
We consider Engdahl's analyses to demonstrate that therecordingsahailablefromthe1927earthquakeare notadequatetodeterminealocationforthisehent 1
~~
with enough precision to associate it with a particular I
fault based on these data alone, The results obtained by 5:ith,(1975)
, appear to indicate that the earthquake was probably closer to shore than the original location (Byerly, 1930). The center of the circles generated by Gawthrop and Engdahl
s 27 are also closer to shore, but the error ellipses on their locations embrace part of the Santa Lucia Bank fault zone. Bylerly's original epicenter, the location of most offshore earthquake activity in this region since instruments were installed in Southern California (about 1934), also is within what is now known as the Santa Lucia fault zone, but is not em-
~
braced within the error ellipses.
i ii i
i
~.
28 4.
Determination of seismic moment by Thatcher and Hanks (1973), determination of stress relationships by Hanks et al (1975), and a 30 bar estimate for average stress drop on strike-slip faults by Kanamori and Anderson (1975):
Hanks et al (1975) computed seismic moments for many southern California earthquakes. Using their calculated seismic moment for the 1927 earthquake and Thatcher and Hanks (1973) relationship between moment, magnitude and stress drop, a stress drop of about 100 bars is obtained for the 1927 event.
If the area inclosed by the MMI=VI isoseismal, (assuming the-Hosgri fault as the' source of the 1927 earthquake) is used to estimate stress drop, about 1000 bar results. This estimate is higher by more than a factor of 30 than Kanamori and Anderson (1975) find to be the average for strike-slip faults of the San Andreas system. We consider this added evidence that the November 4, 1927 earthquake was most likely centered on Trans arse Ranges structures.
5.
Byerly(1930)andUSC&GS(1972):
Both of these sources report the occurrence of a tsunami along the California coast near Point Arguello. This would imply that a vertical 1
F.
29 component of movement.resulted from the earthquakes.
Faults for which geologic evidence of vertical movement has been found are the Santa Lucia Bank fault and Lompoc fault.
6.
Wagner (1974) and Hamilton et al (FSAR): Wagner I
found that offshore seismic profiling indicates j
only small local vertical displacement of the
~ ~ "
ocean floor along the Hosgri zone. Hamiltonetal(FSAR) state that within the resolution of the offshore seismic profiles no vertical displacement of the ocean bottom can be verified. While we cannot conclude that no displacement of the ocean floor has occurred, we do find that the areas g
where such displacements may have occurred are small in extent and that the vertival displacement, if any, is also small.
A large earthquake similar to the-1927 event show show a i
substantial area of vertical movement.if the fault-generating
~
the' earthquake had a significant component of vertical movement.
Hence, since no significant vertical movement has been-found on the Hosgri fault, the movement must have been strike slip, if the 1927 earthquake did. occur on the Hosgri i-fault. Moreover, earthquakes similar to the 1927 event would not be expected to occur as an isolated event in an area which otherwise shows little or no evidence of tectonic f
activity in the recent past. Gawthrop has argued that a i
1y
=-w a1,-r
,n-y7, r-+,--
,w3
.,fe
__.f,
+--.-,.
_,--.r.-en.ev,%,.,,.-_,,..+m-y
%~y,-.--,,c--
-s r
30 vertical scarp could have been formed which subsequently was eroded away by offshore currents. However, older scarps, former shorelines which cross the Hosgri fault obliquely, have not been removed by erosion. Thus, the evidence supports the conclusion that the tsunami generated by the 1927 earthquake was not caused by movement on the Hosgri fault and further supports a location for this event on the Transverse Ranges structure.
7.
Fault plane solutions obtained by Smith (FSAR): The distribution of stations about the 1927 epicenter and the poor quality of first arrivals at many stations do not permit a reliable fault plane solution to be obtained.
A few good records (e.g., Uppsala), however, indicate a substantial amount of vertical movement at the earthquake source. Yeh (1975), using surface wave data, also found that some vertical movement at the source is i
indicated. Combined with the evidence for a tsunami, l
(
this supports the conclusion that the 1927 earthquake l
source had a substantial vertical component of motion.
The data are, however, sufficiently ambiguous that the possibility of strike-slip motion cannot be ruled out.
8.
Seismicity of the California offshore area, Hileman et al l
(1973) between 1932 and 1972 shows some activity which may possibly be associated with the Hosgri structure. The l
l 1
31 extremely low level of activity in the area of the Hosgri fault zone, however, suggests that it has a low earthquake potential.
9.
Fault ranking is discussed by the applicant in the FSAR.
The applicant ranks California faults in order of charac-teristics and tectonic setting and concludes that a magnitude in the 6 range is the maximum that could be expected from the Hosgri fault. The staff concurs that faults of the San Andreas System subparallel to but not connected with the San Andreas fault have not produced earthquakes of magnitude in excess of 6.5.
Although the available data permit the conclusion that the 1927 earthquake could have occurred on the Hosgri fault, they do not favor a location for this event on the Hosgri fault. The NRC staff considers the weight of the available evidence to support the conclusion that the 1927 earthquake was not centered on the Hosgri fault and most likely occurred on structures in the Transverse Ranges.
However, for the purpose of reevaluating the seismic safety of the Diablo Canyon Units, we have assumed a magnitude 7.5 earthquake on the Hosgri fault, consistent with recommendations of USGS.
D.
Maximum Earthquake on.the Hosgri Fault Zone The staff concludes that the assumption of a maximum earth-quake of magnitude 7.5 on the Hosgri is very conservative.
The
q.
32 United States Geological Survey (USGS), the staff's advisor on the geology and seismology aspects of the Diablo Canyon review, recomended the assumption of a magnitude 7.5 earth-
. quake on the Hosgri fault for use in evaluating the seismic safety of the DCNGS. The staff has adopted this recommendation.
The re-evaluation of the seismic safety of the Diablo Canyon units has been based on an assumed magnitude 7.5 earthquake centered on the sector of the Hosgri fault nearest the plant site (See Section 2.5, SER Supplement 4). The rationale and supporting data for the USGS recomendation are contained in the USGS report, Appendix 'C to SER Supplement 4.
We consider this to be a very conservative assumption. We base this on the following:
- a. The fault has experienced, at most, minor movement during post-Wisconsinan time (10,000 years) and, possibly, during l
a much longer interval of geologic time,
- b. Seismicity which may be associated with the fault is very l
low.
I E.
Ground Acceleration for a Magnitude 7.5 Earthquake on the.
Hosgri Fault Dr. N. M. Newmark, the staff's consultant on seismic design, has reconmended a design response spectrum scaled to 0.75g for re-evaluating the seismic safety of the Diablo Canyon l
l l
1, - uni ts. This ground motion is based on an assumed magnitude 7.5 earthquake on the Hosgri fault (See Sections 2.5 and 3.7, SER Supplements 4 & 5). The rationale and supporting data for Dr. Newmark's recommendation are contained in his report, which is Appendix C to SER Supplement 5.
This reconsnendation has been adopted by the staff, and we consider it to be conservative.
The staff considers several other observations to support the conclusion that the ground motion proposed by Dr. Newmark is edequately conservative.
- 1. Peak ground acceleration near earthquake sources appears to be only weakly dependent on magnitude for earthquakes larger than magnitude 41/2 (Hanks and Johnson,1976) and may be more directly a function of stress conditions at the source.
- 2. Available observations support a value of several tens of bars for the average stress drop earthquakes on strike slip faults of the San Andreas fault system (Hanks,1978; Kanamori and Anderson, 1975).
- 3. Boore (1972) and Bouchon (1976) have shown that the ground motions recorded-at Pacoima Dam from the San Fernando earthquake of 1971 were significantly amplified by topo-i
s 34 graphic effects. This amplification effect should be considered in using the Pacoima Dam accelerogram to extrapolate to larger events.
- 4. Observed damage to structures near earthquake sources generally is not commensurate with large accelerations.
For example, at a distance of 31/2 miles from the 1906 San Francisco earthquake source, the intensities were generally no greater than Rossi Forel IX (at or close to the VIII-IX isoseismal except in areas of saturated soil).
The equivalent to Modified Mercalli Intensity is VIII.
These observations suggest that large peak accelerations if present near the 1906 earthquake source were not effective in causino damage to structures.
5.
Empirical relationships among peak acceleration, distance from source and source magnitude suffer from lack of observational data control at distances closer than about 20 kilometers to the source and for magnitudes larger than about 6.5.
Some investigators have extrapolated limited existing near source data to higher magnitudes (Page et al 1972). Others have developed attenuation relationships from distant data which have been extrapolated to distances near the source by incorporating consideration of the finite source dimension (Donovan 1973; Schnabel and Seed, 1973; Seed, et al, 1976; Blume, 1977a, 1977b). The latter relatinships support 0.75g as a conservative value of peak horizontal acceleration at the DCNGS for a magnitude 7.5 earthquake on th'e Hosgri fault.
,z.
35 The staff recognizes that there is uncertainty in all of the above points. Taken together, however, we believe them to support 0.75g as being a proper conservative effective acceleration at the DCflGS for a magnitude 7.5 earthquake on the Hosgri fault.
i i
L REFERENCES Cernreuter, D.
L., 1977, A Stu'dy of Some of the Important Factors that IM1uence the Design Response Spectra, Lawrence Livermore i
Laboratory, University of California /Livermore.
Blume, J. A., 1977a, Diablo Canyon Nuclear Power Plant, The Effect of Arbitrary Variations in Peak Ground Acceleration on Spectral Response,~ PG&E Seismic Evaluation for Postulated 7.5 Hosgri Earthquake, Volume IV, Section 30.
Blume, J. A.,.1977b, The SAM Prccedure for Site-Acceleration-Magnitude-Relationships, Preprints, Sixtn World Conference on Earthquake Engineering New Delhi Jan. 10-14, 1977, p. 2-87, 2-92.
Boore, D. M., 1972, A Note on the Effect of Simple Topography on l
Seismic SH Waves, Bull. Seis. Soc. Amer. Vol. 62, p. 275, t
Bouchon, M., 1976, Discrete Wavenumber Representation of Seismic Wave Fields with Application to Various Scattering Problems, Massachusetts lnstitute of Tachnology, Submitted in partial fulfillment of the requirements for the Degree of Doctor of Philosophy.
f Byerly, P.,1930, The California Earthquake of November 4. -1927, Bull. Seismolog. Soc. Am. V. 20, 1930, pp. 53-66.
i Dibblee, T. W. Jr., 1978, Analysis of Geologic-Seismic Hazards-to Point Conception LNG Terminal Site, County of Santa Barbara,-
Calif. March, 1978.
Donovan, N. C., 1973, A Statistical Analysis of_ Strong Motion Data Including the February 9, 1971 San Fernando Earthquake Proceeding Fifth World Conference, Earthquake Engineering, j
Vol. 1, pp. 1252-1261.
Earth Sciences Associates, 1975, The. geology of the region of.
intersection and merging of the offshore Santa Maria Basin and Hosgri fault trends with the Transverse Ranges trends, in Appendix 2.5E, Additional Geologic Studies, 1975, to Units 1 and 2, Diablo Canyen Site FSAR (Amendment 31).
t i
i L. ' -
2 Engdahl, E. R., 1975, Teleseismic Location of the 1927 Lompac Earthquake, TERA Technical Report, TERA Corporation.
Gawthrop, W., 1975, Seismicity of the Central California Coastal Region, B. S. Thesis, California Polytechnic University (USGS Open File Repoer 75-134).
Greene, H. G., Lee, W. H.
K., McCulloch, D. S., and Brabb, E. E.,
1973, Fault Map of the Monterey Bay Region:
U. S. Geological Survey M sc. Map MF-518.
Hall, C. A.,
Jr., 1975.
" San Simeon-Hosgri fault system, Coastal California, economic and environmental implications,: V. 190, No. 4221, p. 1291-1294.
Hamilton, D., 1976, "A discussion of the proposal by C. A. Hall, Jr. that 80 km of right lateral slip has occurred along the combined San Simeon-Hosgri fault system." Unpublished manuscript 18 pp.
Hamilton, D. (1975), Appendix 2.5E, Diablo Canyon, FSAR.
Hanks, T. C., J. A., Hileman, W. Thatcher, 1975, Seismic Moments of Larger Earthquakes of the Southern California Region, Geol.
Soc. of Amer. Bull. V. 86, pp. 1131-1139.
Hanks, T. C., 1978, Earthquake Stress Drops, Ambient Tectonic Stresses and Stresses that Drive Plate Motions, in press.
Hanks, T. C. and D. A. Johnson, 1976, Geophysical Assessment of Peak Acceleration Bulletin of Seismological Society of America, Vol. 66, pp. 958-968.
Hileman, J. A., C. R = Allen, J. M. Nordquist, 1973, Seismicity of the Southern California Region, 1 January, 1932 to 31 December, 1972, Seismological Laboratcry, California Institute of Technology.
Hoskins E. G. and J. R. Gr'tffiths, 1971, Hydrocarbon Potential of Northern and Central California Offshore, in Cram, I. H. (editor), Future Petroleum Pgovinces of the U. S. - Their Geology and Potential; Amer. Assoc. Petrol.
Geol. Mem. 15, V. 1, p. 212-218.
3 -
International Seismological Summary, 1931.
Jennings, C. W., et al. (1977), Fruit Map of California, scale 1:750,000, Calif. Div. of Mine; and Geology.
Kanamori, Hi roo and D. L. Anderson,1975, Theoretica! Basis for Some Empirical Relations in Seismology, Bull. Seis.
Soc. Am. Vol. 65 pp.1073-1096.
McCulloch, D.S. and R.H. Chapman, 1972, Maps showing residual magnetic intensity along the Calf tornia coast, Lat. 37* 30' N to 34' 30' N, USGS Open File Report /7-79.
Newmark, N. M., 1976, A Rationale for Development of Design Spectra for.Diablo Canyon Reactor Facility, Report to the U. S. Nuclear Regulatory Commission, SER Supple 5,
Appendix, pp. C-1-C-41.
Page, R. A., D. M. Boore, W. B. Joyner, & H. W. Coulter, 1972, Ground Motion values for Use in the S6ismic Design of.the Trans-Alaska Pipeline System, U. S. Geological Circular 672.
Richter, C.
F.,
1958, Elementary Seismology, W. H. Freeman and Co., San Francisco, Calif. 768 pp.
Schnabel, P. B. and Seed, H. B., (1973) " Acceleration in Rock for Earthquakes in the Western United States, Bulletin of Srismological Society of America, Vol. 63, No. 2.
Seed, H. B., R. Muraraka, J. Lysmer and I. M. Idris, 1976, Relationships of Maxir;m Acceleration, Maximum Velocity, Distance from Source and Local Site Conditions for Moderately Strong Earthquakes, Bull. Seis. Soc. Amer. Vol. 66, p. 1323.
Silver, Eli, and R. Von Huene (1975).
Seismic Reflection Profiles, USNS Bartlett Nov. 197E Cruise, offshore central and southern California (Legs 2 and 3), USGS Open-File Report 75-121.
Smith, S. W.,1975, An Evaluation cf themaximum credible earthquake on the Hosgri fault zone and Associated Site Response Spectrum Comparisons (with John A. Blume and Associates), in Appendix 2.5E, Additional Geological and S6ismologic Studies, 1975, to Units 1 and 2, Diablo Canyon Site FSAR (Amendment 31).
Thatcher, W. and Hanks, T., C., 1973, Source parameters of Southern California earthquakes, Jour. Geophys. Research,
- v. 78, pp. 8546-8576.
i
.m -
4 Thatcher, W., J. A. Hileman, and T. C. Hanks, 1975, Seismic slip distribution along the San Jacinto fault zone, Southern California, and its implications; Geol. Soc.
Amer. Bull, V. 86, pp. 1140-1146.
Trifunac, M. D. and Anderson, J.
G., 1977, Preliminary Empirical Models for Scaling Absolute Acceleration Spectra, U. of Southern California Rrsearch Report 77-03, 141 pgs.
Trifunac, M.
D., and Brady, A. G., "On Correlation of Peak Acceleration of Strong Motion and Earthquake Magnitude Epicentral Distance and Site Conditions, U. S. National Conf. on Earthquake Enginering, 1975.
U. S. Coast and Geodetic Survey, 1927, United States Earthquakes, U. S. Department of Commerce.
U. S. Geological Survey, 1976, Review of the geologic and seismologic data relevant to the Diablo Canyon Nuclear Power Station, Units 1 and 2, report to the NRC transmitted by letter to Benard C. Rusche from H. W. Coulter, April 29, 1976.
Von Huene, R., (1969), Geologic structure between the Murray fracture zone and the Transverse Ranges: Marine Geology, v. 7, p 475-499.
, (1971), A Possible Relation Between the Transverse Ranges of California and the Murray Fracture Zone; expanded abstract, Symposium on Transverse Range Province, Geol. Soc. American Cordilleran Section Meeting.
Wagner, H. C., 1974, Marine Geology between Cape San Martin and Pt. Sal, South-Central California Offshore; a Preliminary Report, August 1974; U. S. Geological Survey Open-File Report 74-252.
Wolf, S. C., (1975), Seismic R flection Profiles R/V Polaris March 1972 Offshore Southern California, Point Conception Cruise, U.S.G.S.
Open-File report 75-166.
Yeh, Huey-Ching, 1975, Mechanism of the 1927 Lampoc earthquake from surface and Analysis, Thesis, 91 pgs, University of Washington, Seattle.
L