ML19309G228

From kanterella
Jump to navigation Jump to search
Seismological Assessment of Probable Expectation of Strong Ground Motion at Getr Site, Prepared for GE
ML19309G228
Person / Time
Site: Vallecitos File:GEH Hitachi icon.png
Issue date: 04/28/1980
From: Kovach R
ENGINEERING DECISION ANALYSIS CO., INC.
To:
Shared Package
ML19309G227 List:
References
NUDOCS 8005050219
Download: ML19309G228 (15)
v • d • e


Text

_-

O seasosoziq P

A SEISMOLOGICAL ASSESSMENT OF THE PROBABLE EXPECTATION OF STRONG GROUND MOTION AT THE GETR SITE By Robert L. Kovach Report to ENGINEERING DECISION ANALYSIS COMPANY

)

for GENERAL ELECTRIC COMPANY April 28, 1980

- +. _ + - -.

.t 3(,

OFFICIAL b~ 'L ex J /'lT,. VIRGINIA C. CA50'BC '(

{;'Q(jff Jh M 1 NOTARY PUBUC - CAUTORN'A AtMEDA COUNTY Neer uf comm. ext cs !!An c.1 ti,y i

._e...

I Ground motion can be characterized in a number of ways.

The time history of particle motion would be the most complete specification but prediction of this unknown in the near-field depends on a complete a priori specification of details of the source time function, the rupture velocity, together with other properties of the media (Israel and Kovach, 1977).

.ww,

For many purposes of seismic design the motion is often og p amd

~

specified by parameters such as peak acceleration, velocity

< 8$ g {..

, w yg gt and displacement together with some measure of shaking dura-

!a u. v p

$ 08 <n u

=Ot tion.

Peak acceleration is the most widely referenced para-i t $2 3 d

$ kh E E

>s -

meter in the seismic engineering literature.

Of particular

-1 e

(

z interest to the engineering community has been the question 1 y,ccsgh;

]Qf($[jf {

f whether peak acceleration depends on earthquake magnitude.

u An examination of much of the near-field instrumental acceleration data was carried out by Hanks and Johnson (1976).

Further data were obtained during the 1975 Oroville earthquake aftershocks (Seekins and Hanks, 1978; Hanks, 1979).

The main conclusion of Hanks (1979) was that "at least above magnitude 5, peak acceleration data at a fixed, close distance (R1 10 km) only weakly depend on the magnitude of the earthquake.

That is, peak accelerations at R$ 10 km ' saturate'.

Boore et al. (1980) present a summary of peak acceleration data most useful for esti-mation in the distance range of 15 to 100 km from a ' causative' fault.

Since preparation of the above reports near-field data from two additional earthquakes have been recorded, the M

= 5.7 3

Coyote Lake earthquake or August 6, 1979, and the M

= 6.4-6.6 g

Imperial Valley earthquake of October 15, 1979.

These data i

l significantly expand the data base for the prediction of cear-field strong ground motion.

All of the pertinent horizontal near-field peak acceleration data for distances with 10 km of a ' causative' fault are given in Table 1.

These data are plotted against magnitude in Figure 1.

The curves of Schnabel and Seed (1973) represent smoothed data at distances of 10 km taken from their paper " Ranges of maximum accelerations in rock".

Also shown is the magnitude

- + ~'d-os Fic dependence of average acceleration given by Donovan (1973).

7g ?[

c5 pg 6

" kn

\\

Overall the data show a ' weak' dependence on magnitude for M a u. 5 ',1

!d

.O s,

j [j u g g. ;

magnitudes >5.5.

The observed average values for the 1979

& g; M 26 }j

/. i fy ~ s Imperial Valley and 1979 Coyote Lake earthquakes fall within S E3 - [

?

I ;'

the bounds given by Schnabel and Seed suggesting that these l

-E

)

/

data are of some use in extrapolation to higher magnitudes.

a i

The mean of the peak horizontal accelerations for the l,

,./j 1979 Imperial Valley shock is plotted as a function of the nearest distance to the fault surface in Figure 2.

Also shown in Figure 2 by the solid triangles are the computed fit to the Imperial Valley data utilizing a peak acceleration vs. attenu-ation dependency of the form a = a (R+C) a relation in common use in earthquake engineering practice (Esteva, 1974; Idriss, 1978).

The exponent 6 controls the decay of the curve for distances R>>C.

It is well known that seismograms contain more dominant longer periods with an increase in magnitude and in fact the ampli-tude of the long-period spectral level is found to scale with the seismic moment.

Therefore, S is probably at most only weakly

TABLE 1.

Corrected Peak Acceleration Data for Earthquakes < 10 km Earthquake Date H

R(km)

H i

Z Reference

,jj 2)

Beverly liills Aug. 27,72 3.2

~10 100 80 40 llanks & Johnson (1976)

~10 150 60 20

~10 150 120 40

~10 120 90 20 Imperial Valley Jan. 27,75 3.5 6

170 140 80 3.7 6

150 60 60 4.3 6

250 240 150 3.6 5

160 120 90 3.4 5

130 80 70 3.6 6

120 80 60 Bear Valley Sept. 4,72 4.8 9

602 486 144 Horrill & Matthiesen (1972) l Ancona, Italy June 14,72 4.9

~7 403 381 220 Hanks & Johnson (1976)

La' 4.7

~7 314 199 123 June 21,72 4.4

~7 153 72 109 Oroville Aug. 5,75 3.3 8

190 110 60 3.2 9

190 100 60 Aug. 11,75 4.3 5

420 340 280 Port !!ueneme Har. 18,57 4.7

~10 164 87 25 Parkfield Jan. 28,66 5.8 0.1 480 202 5

425 348 117 9

270 233 78 Santa Barbara July 1,40 6.0

~10 234 172 69 San Fernando Feb. 9,71 6.3 3

1150 1050 696 Koyna, India Dec. 10,67 6.5

'10 6 30 520 320 Krishna et al. (1969) y =--

^,.3 j,

.r f'

i

'..,[7;,

(

j ;. Q w;

' Il l

~ -.---s: a. m --o-o w-

s TABLE 1 (cont.)

Earthquake Date M

R(km) 11 11 Z

Reference 2

(cm/sec )

Imperial Valley May 18,40 6.5 7

342 210 206 Hanks & Johnson (1976)

Coyote Lake Aug. 6,79 5.7 0

225 157 98 USGS Open File Report 79-385 1

333 412 167 3

255 235 431 (values uncorrected) 5 265 255 147 7

255 196 176 8

127 98 78 Imperial Valley Oct. 15,79 6.4 1

454 327 504 USGS Open File Report 79-1654 1

1 424 341 1490 8

3 770 576 348 4

457 598 4

367 517 432 5

345 477 646 7

350 484 199 7

162 217 150 8

246 213 224

-9 222 103 11 270 197 179

(..

i rzr'

?

r

.,a

(

bb;U.

'" k

/

I E

e i

O l

v O'

6 MAXIMUM VALUE -BONDS CORNER 3KM 6 3.0

0.8g--

q

- - - - - - - - - - O.6 g - - - - - - +- - - - - - - - - - - - - - - - - _ V E

+

AVERAGE VALUE o

e o

X MAXIMUM y ell

'4'-

v JJ.L c

VALUE s

.e

, 9

.@ 2.5 X

e AVERAGE A /-

DONOVAN (1973) ig VALUEp

/ScHNABEL a SEED (1973) 8 oo 4

h a BEVERLY HILLS AUG. 72 5

e X

/-

e IMPERIALVALLEY JAN.75

-O O

+ BEAR VALLEY

+-

h x ANCONA. ITALY N 2.0 O OROVILLE

  • C
  • PORT HUENEME O

$ PARKFIELD 1

A SANTA BARBARA a SAN FERNANDO 2c O

V KOYNA O

O IMPERIAL VALLEY 1940 O_

O IMPERIAL VALLEY 1979 O'

Y COYOTE LAKE 1979 3

!.5 I

I I

I 6

7 8

3 4

5 WiAGNITUDE Figure 1.

)

Gi

{,,

' l.\\ '!!P ^:NI?. C. Cs

'j fiOTI.W PUOLC. CA _

,( s;

,/

r f U*E0 i CO';.

LAUSMi..-

.. j

+.

dependent on earthquake magnitude.

The solid triangles in

['.,-

Figure 2 represent a fit to the mean values of the Imperial

'N w '.2 0 g v; q[::-

shock.

A constant S = -1.75 and C = 17 produce an excellent ij U2 !!i~

fit to the observed data.

. TZ $ 2 3 c Ns E

$ g, ; r Comparison of the observed data for the Coyote Lake earth-quake and the Imperial Valley shock reveals that in the near-

,S

,[

S

.i Q @, 'q #

!' -p field the peak hors.zontal accelerations ' flatten' for distances

/

1 s

' ~ ~ ~

close to the fault.

This flattening trend requires that C increase with increasing magnitude.

Assuming that this relation is valid for higher magnitudes we can empirically fit the observed data for the M

= 7.2 Kern g

County shock, the M

= 7.6 Sitka, Alaska shock and the M

= 7.1 g

g Puget Sound shock (Figure 2) and extrapolate to a distance of 4 km from a %M

= 7.2 causative fault.

This extrapolation for 3

a value of C = 31 yields a mean value of 0.65 g.

Also shown for comparison in Figure 2 is the attenuation vs. distance rela-tion developed by Seed (personal communication, 1980) for a M

= 7.5 earthquake.

This relation also predicts a value of g

0.65 g at a distance of 4 km.

Therefore, it is believed that for the peak horizontal acceleration 0.65 g is a reasonable expect-ed value, based on available near-field data, for the GETR site for a M

= 7.2 shock on the nearby Calaveras fault.

3 The question of vertical peak accelerations can also be add-ressed.

Figure 3 shows the observed peak instrumental accelera-tions for distances of less than 10 km from a causative fault plotted against magnitude.

The data do suggest a slight depend-ence with an increase in earthquake magnitude.

However, the mean of the vertical acceleration data (with the exclusion of the

.,b e s 7, o n.;.q

}i ij os 51 Z ;i2 $ 4 '

b s 7. 2

. o if $ '

PRE 0lCTED VALUE O.65g AT 4KM FOR Mg

(

i% i f

1.0 J

c,ac ~_

rg 3 }

c = 31

==, g=%*

iw t

~. -- ~

(

g g%,

~~

4 h

+

h SEED (1980) M3 = 7. 5 4/

z9

+

\\.

\\

Q O.1

+

s i

CE tu

+

d C =12

++

o u<

.+-

C = 17

_.s A a = cr(R<)a

+#- +

s z

2 C = 17 E O.01 r s =-l.75

+

0 I

+ 1979 COYOTE LAKE M = 5.7 t

M 9 KERN COUNTY 1952 M = 7.2 MEAN OF 1979 g

y

_ E SITKA ALASKA 1972 M = 76 IMPERIAL VALLEY s

$ PUGET SOUND 1949 M = 7.1 M = 6. 6, M = 6.8 g

g 3

I I

I 0.0 01 1

3 10 30 10 0 300 CLOSEST DISTANCE (Km)

Figure 2.

l l

l

g MAXIMUM VALUE BRAWLEY-lMPERIAL Z

FAULT INTERSECTION 9

3.0 =---------------_____1.og________._______________

"4[E o

td *

._l

- - - - - - - - - - - - 0. 5 6 g - - - - - - - - - - - - - - - -

Ld -

-- - - - -- O.4 5 g - - -- - - - - - -

M A XIM UM r

VALUE y ---- - kM E A N - - - - - - - - - - -

O vi OZ 4 32.5 eqv

+

o MEAN WITHOUT MAX. VALUE 4 "^

Oo

(

O e BEVERLY HILLS X

MEAM e IMPERIAL JAN 75 L

pe gT g

9 x BEAR VALLEY tu E t ANCONA X

O OROVlLLE

( PARKFIELD 2.0 8

a SANTA BARBARA M

  • e 4

a SAN FERNANDO oo e

ld A

v KOYNA o IMPERIAL VALLEY 1940 (r)

O IMPERIAL VALLEY 1979 O

Y COYOTE LAKE 1979

-I a

I I

I I

1.5 3

4 5

6 7

M AGNITUDE Figure 3.

u OFF'CIAL :

[

v.) -

"'MHI A C. CA 7 0 '

~O f;O T/ RY FUEtt;. CAUr C R:; A

'/

/lA*/E3A COU';TY f4 Com et? tes WAR 8, MCI h.

i

~

~ -.,

observed maximum value of 1.5 g recorded during the Imperial Valley shock) are significantly less than the horizontal compo-nent.

(The high value of 1.5 g recorded during the 1979 Imperial.~...

,y U Valley shock was recorded near the intersection of the Brawley-s h p # ;t Imperial fault wedge where the rupture direction apparently J d 9 5f O:<bhb) shifted to a more vertical direction.

Such a condition would 9a Fj g d Ei :

not be applicable to rupture on the Calaveras fault in the l h ((

vicinity of the GETR site].

) $$ E ([

1o 53 i

w

(

i M '$

Extrapolation of these observed vertical acceleration data A p. >

(excluding the value described above) would suggest a value of j h'['A J'

' i:L

%0.45 g for a %M

= 7 earthquake at distances of less than 10 km'.

3 Figure 4 shows a regression analysis for the peak vertical accelerations of the 1979 Imperial Valley shock recorded as a function of distance.

In this regression analysis two values were excluded - the 1.5 g value previously described above and the value of 0.66 g recorded at Station 5165 at Dogwood Road at a distance of 5 km.

[The vertical accelerogram for Dogwood Road exhibits a singular, isolated spike of acceleration not repre-sentative of the total accelerogram].

With an adopted value of C = 5 to allow for a slight curvature of the data at distances of less than 10 km the regression analysis yields the expression

-1.15 A = 3.71 (R+5)

Figure 4 shows an extrapolation of the far-field vertical acceleration data for the M

= 7.2 Kern County shock of 1952, 3

the M

= 7.6 Sitka, Alaska shock of 1972 and the M

= 7.1 Puget g

3 Sound shock of 1949 into near-field distances using the same functional dependence as utilized for the horizontal component acceleration data.

An extrapolation of these data, using the

\\

-1[ b]&

a.- -

-1.0-

c:

x<

.j b9 o f f

~

1, os t _ t

. f? d 2 ; ' (

)

jEb5 'E i gt; c 4

l%

i) e1.5g i

1, j

~

e 3

O.42g e

~..,

Z e

'N Q(N 9

e N

p e

' O, 4

e N.

\\

CE O.1 e

W e

J e

W e

O g

O 3

+

+ bo J

ee O

, t = 6. 6 e1979 IMPERIAL VALLEY SHOCK M

h O.01 W

O KERN COUNTY 1952 M =72 t

a SITKA ALASKA 1972 Ms = 7.6 y

+ PUGET SOUND 1949 M = 7.1 t

W Q_

4 10 30 10 0 300 DISTANCE (km)

Figure 4.

EA !

a 9,0 g:

j {j Mj values shown in Figure 4, yield a value of %0.4 g at 4 km for a gu. be postulated M = 7.2 earthquake from the nearby Calaveras fault. g og g+g The regression analysis of the peak vertical acceleration 4( E S224 zg g q kO 55 E

data from the 1979 Imperial Valley earthquake require a lower 53

./

i value of C than that exhibited by the horizontal acceleration J%(dth

.W h data.

Therefore, in extrapolating the peak vertical accelera-

.m tions for the higher magnitude earthquake data shown in Figure 4 a value of C = 25 was used, a value less than that used for the extrapolation of the peak horizontal accelerations from the same data base (Figure 2).

As pointed out earlier, through comparison of the acceleration data from the M

= 5.7 Coyote Lake earthquake 3

and the M_ = 6.6 Imperial Valley earthquake the value of C increases with an increase in the magnitude of the earthquake.

It has been postulated that a magnitude 5 to certainly no more than a magnitude 6 shock might occur on the suggested Verona thrust fault close to the GETR site.

The recent near-field data used in extrapolating from a magnitude 6.4 shock to a magnitude 7 shock were recorded from predominantly strike-slip earthquakes.

Little data exists that would p'ermit a rigor-aus differentiation between the near-field accelerations produced by a thrust fault earthquake as compared to a strike-slip earth-quake.

Bouchon (1980a,b) has completed a theoretical study of

^

the ground motions produced by shallow strike-slip and dip-slip faults near the source a'nd concluded that the ground displacement can exhibit a high degree of complexity.

On August 1, 1975, a magnitude 5.7 earthquake occurred in the Sierra Nevada foothills southeast of Oroville, California l

(Bufe et al., 1976).

Focal mechanism studies of the aftershocks indicate normal faulting with the western, Great Valley side down-dropped relative to the Sierra Nevada block.

The dip of the fault plane is believed to be dipping 60 to the west.

A strong motion recording of the main shock recording at Oroville dam on bedrock at a distance of 11 km from the epicenter yielded a peak accelera-tion of 0.4 g (Maley et al., (1975).

The data of S,eekins and Hanks (1978) from analyses of the strong-motion accelerograms of the Oroville aftershocks consistently show that higher peak accelerations were recorded on bedrock sites as compared to sedi-mentary sites.

If we adopt the data from the'Oroville aftershocks as'being representative of what might occur on the postulated Verona fault then it appears that a reasonable value of the peak accelerations that might be anticipated from a magnitude 5.5 to 6.0 shock in the near-field from the Verona fault would be about 0.4 g.

-.-m~,.

(!

1y$o h,. '

U

.o e

t p:

1 b

J*#u se os 61 m zu Q

) O FT 1

"o E

i

>E 1

s

'it

(

1 I

RcJ G

____ References Aki, K.

(1967)

Scaling law of seismic spectrum, J. Geophys.

--72, 1217-1231.

Res.

f E'

Boore, D.M., W.B. Joyner, A. Oliver III, and R.A. Page (1980)

.]

[. '

Peak acceleration, velocity, and displacement from strong-

?9; a

_,0.u@

- <3 motion records, Bull. Seism. Soc. Am. 70, 305-321.

e

.o Bouchon, M.

(1980a)

The motion of the ground during an earth-g ug g.y quake 1.

The case of a strike-slip fault, J. Geophys. Res.

$t g2 g

85, 356-366.

$ge Bouchon, M.

(1980b)

The motion of the ground during an earth-i

>O E

quake 2.

The case of a dip-slip fault, J. Geophys. Res.

d? FEN

f%l 8_5,,

365-375.

Hanks'j c h./:-

o normal faulting in the Sierra ~~~ '"{

Lahr, L.C. Seekings, and T.C.

Bufe, C.G.,

F.W.

Lester, K.M.

(1976)

Oroville earthquakes:

Nevada foothills, Science 192, 72-74.

Donovan, N.C.

(1973)

Earthquake hazards for buildings, in Buildine Practices for Disaster Mitigation, Building Science Series 46, U.S.

Dept. of Commerce, National Bureau of Standards.

Esteva, L.,

(1974)

C' ology and predictability in the assessment of seismic risk:

2nd International Conference Association of Engineers and Geologists, Proceedings, Sao Paolo, Brazil.

ifanks, T.C.

(1979)

Seismological aspects of strong motion seis-mology, Second U.S. National Conference on Earthquake Engineering, Aug. 2-24, Proceedings, 898-912.

Hanks, T.C.,

D.A. Johnson (1976)

Geophysical Assessment of peak accelerations, Bull. Seism. Soc. Am. 6 6,,

959-968.

Idriss, I.M..

(1978)

Characteristics of earthquake ground motions Specialty Conference on Earthquake Engineering and Soil Dynamics, ASCE, Pasadena, California, p. 115.

Israel, M.

and R.L. Kovach (1977)

Near-field motions from a propagating strike-slip fault in an elastic half-space, Bull. Seism. Soc. Am. 67, 977-994.

Krishna, J.,

A.E. Chandrasekaran, and S.S.

Saini (1969)

Analysis of Koyna accelerogram of December 11, 1967, Bull. Seism. Soc.

Am. 69, 1719-1731.

Maley, R.P.,

V. Perez, and B.J. Morrill (1975)

Stror.; motion seis-mograph results from the Oroville earthquake of 1 August 1975, in Oroville, California, earthquake 1 August 1975, California Division Mines Geol. Spec. nacer, 124, 115-122.

l l

l l

I

l I Morrill, B.J. and R.B. Matthiesen (1972)

Strong-motion accelero-graph records from 4 September 1972 Stone Canyon earthquake, Earthquake Notes H, 17-20.

Porcella, R.L.,

R.B. Matthiesen, R.D. McJunkin and J.T., Ragsdale (1979)

Compilation of strong-motion records from the August 6, 1979, Coyote Lake earthquake, U.S.G.S. Open File' Report 79-385.

Schnabel, P.B. and H.B. Seed (1973)

Accelerations in rock for earthquakes in the western United States, Bull. Seism. Soc.

Am. 63, 501-516.

Seekins, L.D.

and T.C. Hanks (1978)

Strong-motion acceleration data, Bull. Seism. Soc. Am. 6 8,,

677-689.

O FE M ai-fg

)

ymniwA c. c/.

e (g D[.f: DOTARY PUSUC - CAdiGb!A o

3. 4 e, d gh c kes 8I e

a w-

Retrieved from "https://kanterella.com/w/index.php?title=ML19309G228&oldid=4801763"