ML20031B708
| ML20031B708 | |
| Person / Time | |
|---|---|
| Site: | Midland |
| Issue date: | 07/31/1981 |
| From: | EECWESTG, WESTON GEOPHYSICAL CORP. |
| To: | |
| Shared Package | |
| ML20031B698 | List: |
| References | |
| ISSUANCES-OL, ISSUANCES-OM, NUDOCS 8110050455 | |
| Download: ML20031B708 (26) | |
Text
.
I
-~'gx ItEllTED C(JRHFSM)M)ENm
' r.g.,
DC CliET NUr,tc2n@Ni
'2-
\\ '"* *' ' U -('500$dE:S.h lcl cccymn c:nne
.tp OCT 1198t * :-
i lE T ::' n am,
}/l E
L
.. - a a: a '
i
-G E:: :::
A k
i BASIS FOR THE REJECTION OF THE 1966 PARKFIELD EARTHOUAKE l
l ACCELEROGRAMS FOR USE IN MIDLAND PLANT SITE l
SPECIFIC SPECTRA I
Prepared for CONSUMERS POWER COMPANY July 1981 I
<(c ' 6 >, d
/7 ?'
7,w', 0 fil
[
Wes on Geophysical CORPORATION 5
811005045S 810929 5
PDR ADOCK 050003po Holt Exhibit 7 G
\\ ~P
E E
i l
l BASIS FOR THE REJECTION OF THE 1966 PARKFIELD EARTHQUAKE i
l ACCELEROGRAMS FOR USE IN MIDLAND PLANT SITE l
SPECIFIC SPECTRA i
Prepared for CONSUMERS POWER COMPANY
!E i
!E l
July 1981
!,E
- E Y
pj' e.m
- E
!j Weston Geophysical CORPORATION
!E i
I l
l dhf,l Weston Geophysical coanoauios 7
l
(!
II July 21, 1981 (I
Consumers Power Company 1945 West Parnall Road I
Jackson, Michigan 49201 Attention:
Dr. Thiru Thiruvengadam
Subject:
Basis for the Rejection of the 1966 Parkfield Earthquake Accelerograms for Use in Midland Plant Site Specific Spectra.
Gentlemen:
The enclosed report together with comments by Dr. Otto W.
Nuttli address the above subject matter.
Sincerely, WESTON GEOPHYSICAL CORPORATION l
%Lfj%d Richard J.
Holt RJH:eag Enclosure B
I P
l
E E
Table of Contents 5
1 1
E Page
1.0 INTRODUCTION
1 1.1 Objective 2
l
1.2 Background
2 2.0 THE PARKFIELD EARTHQUAKE 3
2.1 General Background 3
2.2 Specific Arguments Against the Inclusion of the Parkfield Nearfield Strong Motion Recordings in the Midland Site Specific Response Spectra 5
2.2.1 Strface Rupture 5
2.2.2 Supersonic Incoherent Rupture 5
2.2.3 Sources of Conservatism 8
2.2.3.1 Large Mangitude Range 8
3.0 CONCLUDING REMARKS 9
REFERENCES 11 APPENDIX I FIGURES E
E 5
5 5
g
_m e_ce
E E
Table of Contents E
E Page
1.0 INTRODUCTION
1 i
1.1 Objective 2
1.2 Background
2 1
l 2.0 THE PARKFIELD EARTHQUAKE 3
{
2.1 General Background 3
1 2.2 Specific Arguments Against the Inclusion of the Parkfield Nearfield Strong Motion
+
Recordings in the Midland Site Specific Response Spectra 5
l 2.2.1 Surface Rupture 5
2.2.2 Supersonic Incoherent Rupture 5
i i
2.2.3 Sources of Conservatism 8
i 2.2.3.1 Large Mangitude Range 8
j i
3.0 CONCLUDING REMARK 3 9
REFERENCES 11 Il ApeENDIx I FIGURES E
IE I
4
!E IB Weston Geophysical
B 1
E Table of Contents j
B I
I Page
1.0 INTRODUCTION
1 1.1 Objective 2
1.2 Background
2 2.0 THE PARKFIELD EARTHQUAKE 3
2.1 General Background 3
2.2 Specific Arguments Against the Inclusion of the Parkfield Nearfield Strong Motion Recordings in the Midland Site Specific Response Spectra 5
2.2.1 Surface Rupt'are 5
2.2.2 Supersonic Ir..:oherent Rupture 5
2.2.3 Sources of Conservatism 8
E 2.2.3.1 Large Mangitude Range 8
3.0 CONCLUDING REMARKS 9
REFERENCES 11 APPENDIX I FIGURES l
I Weston Geophysical j
(
{
E l
E List of Figures ll f
(I Figure No.
Title 1
Response Spectra for the Original Ground Surface at the Midland Nuclear Plant With and Without I
Parkfield (5% of Critical Damping).
2 Map of the Fault Trace and I
Aftershock Epicenters of the Parkfield Earthquake of 1966 (Modified from Papageorgiou, 1981).
3 Response Spectra for the Original Ground Surface at the Midland Nuclear Plant Compared to Parkfield I
Accelerogram B034 (5% of Critical Damping).
I I
I I
I I
I I
E g
o
1.0 INTRODUCTION
This report is in response to discussions with the NRC staff concerning the use of accelerograms from the Parkfield California Earthquake of 1966.
The discussions were concerned with the appropriateness (or non-appropriateness) of these accelerograms for use in the site specific spectra developed for the Midland site.
Included with this report is a letter by Dr. Otto Nuttli which addresses the same issue (see Appendix 1).
Site specific spectra are developed by selecting accele-rograms from the world-wide data set that most closely match the safe-shutdown earthquake magnitude, distance (from the site), and local geological conditions.
In the case of the Parkfield earthquake, the distances to the accelerometer stations from the epicenter appear to fall in the correct range.
The local geologic conditions, as determined by shear-wave velocities, are acceptable.
Some published magnitudes for this earthquake appear to be appropriate for use at Midland; however, as it will be discussed, individual station magnitudes determined at close distances would make it inappropriate.
In addition, the earthquake effects, charac-terized in previous reports and in this one as "near field,"
make it inappropriate for use at Midland.
As previously mentioned in the addendum to Part 1, Response Spectra - Original Ground Surface, in applying
" Appendix A",
one or more of three geologic conditions govern E
E g
--_ce
E the selection of a Safe Shutdown Earthquake: a " capable fault",
a " tectonic structure", and/or a " tectonic province".
The safe shutdown earthquake for Midland was based on a tectonic province only.
This means that a maximum observed earthquake B
(or higher) in the province, is assumed to occur at the site where no capable fault or tectonic structure has been identified.
It is believed that an accelerogram showing characteristics reiated to capable faulting or the effects of proximity to a tectonic structure should not be used in the 5
development of site specific response spectra within the context of a " tectonic province" approach, unless unusual tectonic circumstances exist at/near the site.
1.1 Objective The objective of this report is to demonstrate that I
accelerograms resulting from the Parkfield California earthquake of June 28, 1966 should not be included in a data set used to model a 5.3 m earthquake for the b
Midland Nuclear Power Plant.
1.2 Background
In the original Midland submissions Parts '. and II the Parkfield earthquake accelerograms that met 'he magnitude-distance-geologic foundation conditions criteria were not included because of the anomalous nature of the event.
E E
g
E E E In an addendum to Part I, at the request of NRC, sensitivity tests were performed including these records.
The results showed that the anomalous Parkfield records are much higher than the 44 components originally l
considered appropriate to model a magnitude 5.3 mb earthquake, and that their inclusion raises the 84th l
percentile by a significant amount (see Figure 1).
In th'.
next section (2.0), reasons for not including the near-l field Parkfield records to model the Midland site earthquake potential are documented in detail.
2.0 THE PARKFIELD EARTHQUAKE 2.1 General Background l
A voluminous amount of literature has been published concerning the Parkfield event because of the extensive nearfield instrumental coverage and its unusual features, st 'h as its dislocation and surf ace rupture.
Some of the major references are: McEvilly et al. (1967), Eaton E
(1967), Filson and McEvilly (1967), Aki (1968), Haskell (1969), Tsai and Aki (1969), Scholtz et al. (1969), Eaton et al. (1970), Aki (1972), Tsai and Patton (1973), Murray (1973), Trifunac and Udwadia (1974), Anderson (1974),
Lindh and Boore (1974), Kawasaki (1975), Levy and Mal (1976), Archuleta and Day (1977), Wiggins et al, (1977),
Hartzell et al. (1978), Kanamori and Jennings (1978), Aki (1979), Wu (1968), Housner and Trifunac (1967), and Papageorgiou (1981).
I ms ~ sm x
E Some of the important parameters determined by various authors and agencies exhibit a range of values:
E Magnitude:
5.3 (USGS - Filson and McEvilly, 1967),
E 5.8 mb (WU, 1968), 5.5 ML (BRK), 5.6 ML (PAS),
6.4 M3 (WU, 1968).
I Rupture Velocity:
2.2 to 2.5 km/sec most often quoted (Aki, 1979), some estimates above 3.0 km/sec (Anderson, 1974).
Rupture Length / Segmentation:
Main fracture zone about 37 km (Brown and Wedder, 1967).
Segmentation lengths vary with interpretation.
5 Fault Dislocation:
60 cm (Aki, 1968); surface rupture actually smaller.
Depth:
Quotes vary from about 3 km (Aki, 1968) to 15 km (based on af tershocks) depending upon the branch of the rupture considered.
This earthquake occurred in one of the most seismically active zones of the San Andreas Fault (McEvilly et al., 1967).
It caused surface rupture and very high recorded accelerations in the nearfield.
Many studies have shown that the dislocation was characterized by incoherent starting, stopping, and jumping over barriers.
2.2 Specific Arguments Against the Inclusion of the Parkfield Nearfield Strong Motion Recordings in the Midland Site Specific Response Spectra 2.2.1 Surface Rupture Extensive surface cracking was observed after the Pa field earthquake.
Aki (1968) stated that the surface as decoupled by a thin layer of about 100 m B
B below which a calculated 60 cm dislocation probably took place.
This is larger than the one observed at the surface.
In the central United States, surface rupture is not expected to result from magnitudes less than 6.5 m (Nuttli, personal communication b
1981); undoubtedly, the hypothetical 5.3 magnitude assumed to occur "at the site" would not be accompanied by surface rupture.
Thus, the Parkfield earthquake is not characteristic of a central United States scenario.
2.2.2 Supersonic Incoherent Rupture Murray (1973) and many other authors (Aki,1968; Haskell, 1969; and Papageorgiou, 1981) have addressed the affect of incoherent rupture or dislocation on I
seismic radiation.
When a fault rupture occurs, the crack propagation can be smooth, or it stops or starts, or actually jumps over a barrier.
Barriers may be classified into two types:
(Aki, 1979) geometrical barriers such as a bend in a fault; barriers made by inhomogeneities such as evidenced by velocity anomalies.
Quoting Aki (1979), these barriers (both types) can act "...not only as a stopper of rupture but also as an initiator of rupture, as well as stress concentrator".
Weston Geophysical
E For Parkfield, such a jump (discontinuity) did occur.
The fault trace jumped from the east side of the Cholame Valley to the west side, very close to accelerograph Station 2 (See Figure 2).
The vertical 5
component of the seismogram at Station 2 shows strong i
l higa frequency waves which have been generated by the l
jump (Papageorgiou, 1981).
Anderson also suggested that the rupture propagated through a region with larger than average irregularities and possibly even I
stopped, and restarted on a separate plane, resulting in higher amplitudes of the high frequency P-waves.
Anderson (1974) showed that, although it could possibly be the arrival from the S-wave out of the epicentral region, the high frequency phase arrival at Station 5 also corresponds to the time P-waves are arriving from the rupture front in the area of the jump.
A consistent model of crack propagation to explain the strong motion radiation from Parkfield was developed by Murray (1973).
He found that the rupture velocity for this event was supersonic, that is, the crack propagation velocity was greater than the shear velocity of the medium.
From his complex history of the rupture he concludes that the acceleration maxima were caused by the formation of Mach waves generated by stopping and starting.
He 5
WeJon Geophysical
E futhermore nates that
...We may then expect that l
sufficiently close to the fault, the attenuation will approximate that due to an infinite source.
At a distance x from the fault that is sufficiently large E
l compared to the vertical extent of the source, the
-1 amplitude should decay approximately as x In f
fact, the transverse acceleration " jump" amplitudes (when the accelerations attain maximum values) at Stations 5 and Temblor (5.20 and 6.45 km from the 5
fault) are the same as at Station 2.
At Station 8 (9.23 km), the amplitude has fallen by 33 percent, and at Station 12 (14.7 km) by 87 percent."
From the above discussion it is clear that the high values of acceleration recorded at the close E
stations were caused by an extremely unusual element in the dislocation history and that Station 5 and Temblor recorded nearly unattenuated Mach waves.
Consequently, the usage of these abnormal accelerograms would introduce additional conservatism E
for the modeling of a 5.3 m event at Midland.
b 2.2.3 Sources of Conservatism 2.2.3.1 Large Magnitude Range Imposing a large magnitude range, 5.3 m g0.5, or 5.4 M g0.5, already y
5 skews the average and 84th percentile of a site specific response spectrum towards the 3
l 5
l l level of the higher magnitude.
This l
influence was discussed in '. he Midland Addendum to Part I.
If we add anomalously 1
high recordings to a set of appropriate data, the resulting response spectra will not correspond to the targeted magnitude, 5.3 m.
As it can be seen in Figure 3, b
some of the Parkfield records are much higher than the 84th percentile spectrum.
5 In fact, these records are higher than any of the data set collected to represent the 5.3 m spectrum.
b Further substantiation is found in a paper by Kanamori and Jennings (1978),
where the acceleration time histories were run through a Wood Anderson simulator and Mg's (local magnitudes) were directly calculated from the strong motion records.
This was done for two distances:
- 1) the I
closest approach of the fault, and 2) the distance to the zones of aftershocks.
Their results for Parkfield are reproduced below:
E E
-,- e~
L Station Ref Mg Cholame #2 BO33 6.75, 6.35 Cholame 35 BO34 6.35, 5.95 I
6.25, 5.90 Cholame #8 B035 5.75, 5.35 6.1, 5.7 I
Cholame #12 BO36 5.75, 5.35 5.9, 5.55 Temblor BO37 6.4, 5.7 6.7, 6.0 It should be emphasized that these values are meant to be Mg's.
It is clear from the above that for an m =5.3, b
M =5.3-5.4 data set, only B035 and B036 g
should be given marginal consideration for inclusion.
The inclusion of Cholame 5, with an estimated M =6.35 or 5.95 would g
not characterize the m =5.3 (Mg:4.9 b
I to 5.5) at Midland.
3.0 CONCLUDING REMARKS The Parkfield earthquake caused surface rupture and experienced supersonic dislocation velocities complicated by the presence of barriers generating Mach waves.
These I
phenomena combined to generate extremely high accelera-tions in the nearfield.
These high accelerations have been shown to be typical of magnitudes greater than 6.0.
Indeed, this earthquake has been modeled as a series of multiple shocks (Wu-tSf 8).
It is not correct to bias the 84th percentile of the 5.3 magnitude data set, carefully developed for Midland, by adding these anomalous records.
If they are included and the 84th percentile spectrum is used, the results will substantially deviate from the defined potential at the site.
Since this defined potential was determined using the tectonic province approach, which of itself introduces I
substantial conservatism, the level of the resulting i
1 spectrum would be unreasonable for the tectonic environment of the Midland plant.
l 5
t I
k E
E l
weston seopnysica
E REFERENCES
- Aki, K.,
1968, Seismic Displacement Near A Fault, Journal of E
Geophysical Research, V. 73, p. 5359-5376.
- Aki, K.,
1979, Characterization Of Barriers On An Earthquake Fault, Journal of Geophysical Research, V.
84, p.
6140-6148
- Anderson, J.G.,
1974, A Dislocation Model For The Parkfield Earthquake, Bulletin of the Seismological Society of i
America, V.
64,
- p. 671-686.
Archuleta, R.,
and Day, S.M.,
1977, Near-Field Particle Motion
'E Resulting From A Propagating Stress-Relaxation Over A Fault Embedded Within A Layered Medium (Abstract), American Geophysical Union Transaction, V.
58, p. 445.
- Bouchon, M.,
1979a, Predictability Of Ground Displacement And Velocity At Proximity Of An Earthquake Fault:
An Example: The Parkfield Earthquake Of 1966, Journal of E
Geophysical Research, V.
84, p.
6149-6156.
- Das, S.,
and Aki, K.,
1977, Fault Planes With Barriers:
A 5
Versatile Earthquake Mouel, Journal of Geophysical Research, V.
82, p. 5648-5670.
E
- Eaton, J.P.,
O'Neill, M.E.,
- Murdock, J.N.,
1970, Aftershocks Of The 1966 Parkfield-Cholame, California Earthquake:
A Detailed Study, Bulletin of the Seismological Society of America, V.
60, p.
1151-1197.
l
- Filson, J.,
McEnvilly, T.V.,
1967, Love Wav Spectra And The Mechanism Of The 1966 Parkfield Sequence, Bulletin of the E
Seismological Society of America, V.
57, p. 1245-1257.
l
- Haskell, N.A., 1969, Elastic Displacements In The Near-Field l
E Of A Propagating Fault, Bulletin of the Seismological Society of America, V.
59, p. 865-908.
l
- Housner, G.W.,
and Trifunac, M.D.,
1967, Analysis Of Accelero-5 grams-Parkfield Earthquake, Bulletin of the Seismological l
Society of America, V.
57, No.
6, p.
1193-1220.
- Kanamori, H.,
and Jennings, P.C.,
1978, Determination On Local From Strong-Motion Accelerograms, Bulletin Magnitude, ML l
of the Seismological Society of America, V.
68, No.
2,
- p. 471-485.
l
- Kawasaki, I., 1975, On The Dynamical Process Of The Parkfield Earthquake Of June 28, 1966, Journal of Physics of the E
Earth, V.
23, p. 127-144.
l E
weston econnyscor 1
E E
- Levy, N.A.,
- Mal, A.K.,
1976, Calculation Of Ground Motion In A Three-Dimensional Model Of The 1966 Parkfield E
Earthquake, Bulletin of the Seismological Society of America, V.
66, p. 405-423.
- Lindh, A.,
- Boore, D.M.,
1974, The Relation Of The Parkfield Foreshocks To The Initiation And Extent Of Rupture (Abstract), Earthquake Notes, V.
45, p.
54.
McEnvilly, T.V.,
- Bakun, W.H.,
and Casaday, K.B.,
1967, The Parkfield, California Earthquakes Of 1966, Bulletin of the Seismological Society of America, V.
57, p.
1221-1244.
Papageorgiou, A.S.,
1981, On An Earthquake Source Model Of Inhomogeneous Faulting And Its Applications To Earthquake Engineering, MIT Research Report R81-10, No. 696, for National Science Foundation Grant No. PFR-7827068.
l
- Trifunac, M.D.,
and Udwadia, F.E.,
1974, Parkfield, California, l
Earthquake Of June 27, 1966:
A Three-Dimensional Moving Dislocation, Bulletin of the Seismological Society of America, V.
64, p. 511-533.
Weston Geophysical Research, 1981, Part II, Response Spectra Applicable For The Top Of Fill Material At The Plant Site, In Site Specific Response Spectra Midland Plant - Units 1
~
and 2: Prepared for Consumers Power Company.
'E Weston Geophysical Research, 1981, Addendum to Part I, Response m
Spectra - Original Ground Surface, In Site Specific Response Spectra Midland Plant - Units 1 and 2: Prepared for Consumers Power Company.
- Wiggins, R.A.,
- Sweet, J.,
and Frazier, G.A.,
1977, The Parkfield Earthquake Of 1966 - A Study Of Dislocation Parameters Based On A Complete Modeling Of Elastic Wave Propagation (Abstract), American Geophysical Union Transaction, V.
58, p. 1193.
E E
E E
E
-~~
i E
E
!E E
E E
E 5
APPENDIX I E
E E
E
- E i
5 E
E
~B
E E
otto W. NUTTLt PHOFESSOR OF GEOPH Y SIC S P. O. BOX 0099. LACLEDE ST A ST. LO U I S.
MISSOURI 63156
'3m /)*/3/P 658-3124 5
July 13, 1981 Mr. Richard J.
Holt, President Weston Geophysical Corporation P.O.
Box 550 Westboro, MA 01581
Dear Mr. Holt:
I am replying to your request to comment upon the appropriateness of using response spectra from the 1966 Parkfield earthquake in establishing a set of Site Specific Response Spectra for the Midland (Michigan) plant.
The mainshock of the Parkfield earthquake sequence was anomalous I
in a number of ways, which you pointed out in your response to,NRC, dated July 1981.
Principal among them are the surface rupture and the large am11itude of high-frequency waves associated with incoherent supersonic rupture across seismic barriers.
These are reflected in the I
high M values obtained by Kanamori and Jennings (BSSA, pp. 471-485, g
1978) when they used the acceleration time histories to produce simulated Wood-Anderson seismograms, and obtained M values greater than 6 from the g
I Cholane no.
2, Cholane no. 5 and Temblor strong-motion time histories.
From more distant Wood-Anderson seismograph records the ML value assigned to the earthquake was 5.3.
Thus the Kanamori and Jennings values
.I show both that the large acceleration observed at the above-mentioned three near-field stations results from localized features, and that it attenuated much more rapidly than the motion produced by the principal part of the rupture process.
The only characteristic of central United States earthquakes that is known to produce large amplitude, rapidly attenuating high frequency waves is very shallow focal depth.
Table 23 of NUREG/CR-1577 lists all the known central United States earthquakes of this type.
Of the 59 events listed, the maximum mb value was 4.3.
(The 1966 and 1967 Attica, 5
earthquakes had mb values of 4.6 and 4.4 and focal depths of 2 and N.Y.
3 km, respectively.)
These mb values are a measure of the far-field ground motion, as the Mt= 5.3 value was for the 1966 Parkfield earth-quake.
I believe we have a sufficient sample of very shallow central United States earthquakes to conclude conservatively that their mb value will not exceed 4. 8.
Bob Herrmann and I (manuscript in preparation) found that mb(EUS) = M.(WUS), using waves of approximately 1-Hz frequency, j
i.e. the amplitudes of"l-Hz waves excited by earthquakes in the two m
regions are the same when the magnitudes are numerically equal.
There-fore, I would not expect a very shallow central United States earthquake E
- E
-~
E E
Mr. Richard J.
Holt July 13, 1981 pg. 2 l
to produce near-field ground motion greater than that of an Mg = 4.9 California earthquake, which would be log-1 0.5, or 0.32 times that of the 1966 Parkfield earthquake.
It is generally accepted that central and eastern United States earthquakes have not produced surface rupture, with the exception of the great earthquakes of the 1811-1812 New Madrid series.
For the various reasons given above, it is my personal opinion that the Phrkfield earthquake caccelerograms at near-field stations I
should not be included in the set of spectra used to obtain a site-apecific response spectrum for Midland, Michigan.
Sincerely yours, E
Gs La T'.wl-i Otto W. Nuttli E
!E
- E
!E E
E E
Weston Geophysical
l-_
l5 I
l l
5 E
E E
i B
FIGURES I
l5 t
E
- E v
E l
B i
i E
E 1
Weston Geophysical 4
- - _. ~ _ - - - _ -, -, -, -,, - -, - - -,, - - - - - - - -
./,
9
'o +y G
0, q.
9 E
i/
- N/
\\/
r
-x
/
A
/N
)841,h PERCENTILE
\\~
~
~
/ /,
s
/
p'
/
g
>MEAN l /
y
/
NF hMEDIAN 5
f
/
}
f E {$
>l' '/
\\
\\
/
5 ENi
,/K
///
' /
\\
10-'
ORIGINAL SUBMITTAL I
\\
!E
- -- ORIGINAL PLUS PARK FIELD N
i l
l 10-2 10-'
10o 10' PERIOD (SEC) l RESPONSE SPECTRA FOR THE ORIGINAL GROUND ig SURFACE AT THE MIDLAND NUCLEAR PLANT l5 WITH AND WITHOUT PARKFIELD (5% OF CRITICAL DAMPING)
FIGURE 1 g
-e_
5 E
36'00' N.
b e...
E
'"4.e S
E5a
-N' E
13 4
--;s,
. E4'N PARKFIELD STATION 12 m')
w se
\\,
Meo of AM
,,, gg g) lO c
,E 2 a
t%
STATlON 2,;'~ 't STATION g,
, Q
, e
^
9 h
- k'g KILOMETERS CHOLAME h
E
.R STATION 5, 3
b 6,
+ 35 40' g4 7,
5 Map of the fault trace and aftershock epicenters of the Parkfield earthquake of 1966, reproduced from Eaton et. al.
(1970).
Both the fault trace and the fault plane at depth, identified by the aftershock zone, jumps from one side of the Cholame Valley to the other. Two lines were drawn by Aki (1979a), fitting the two zones of aftershock epicenters.
(Modified from Papageorgiou,1981).
FIGURE 2 5
E g
, ~,,,ce l
L
a s
'o,
- ~
o
+y a
o, o*
a a
i/
\\/
\\/
s-s
/
o l
k/\\\\
c
/
\\\\
w
//
\\(
/ B034 N85E
/
N i
s-
//
34 N05W D
/
js/ /
84th PERCENTILE
//
J"
/s
_ M EAN
//
MEDIAN
~o
/i
//
E 9
{
N/
N
[
ORIGJNAL SUBMIT TAL j
---PARKFIELD ACCELEROGRAM B 0 3,4 s
PERtOD (SEC)
RESPONSE SPECTRA FOR THE ORIGIML GROUND SURFACE E
PARKFIELD ACCELEROGRAM B034 AT THE MIDLAND NUCLEfa PLANT COMPARED TO (5% OF CRITICAL DAFPING)
FIGURE 3 a
- * ~ ~
-