ML20151H122
| ML20151H122 | |
| Person / Time | |
|---|---|
| Site: | Trojan File:Portland General Electric icon.png |
| Issue date: | 04/12/1988 |
| From: | Tinsley J INTERIOR, DEPT. OF, GEOLOGICAL SURVEY |
| To: | |
| Shared Package | |
| ML20151H012 | List:
|
| References | |
| NUDOCS 8808010150 | |
| Download: ML20151H122 (10) | |
Text
-
- n. g g m es m3 ~ g.yys rj. , . s. . - a.n,y:q.a..y . . - < 3
(' GEOLOGIC FACTORS AND THE REGIONAL EVALUATION OF SITE RESPONSE FOR URBAN SEISMIC HAZARDS STUDIES John C. Tinsley U.S. Geological Survey 345 Middlefield Road m/s 975 Menlo Park, CA 94025 415-329-4928 Several factors influence the character of earthquake-generated ground shaking at a point on the earth's surface, including distance from the causative fault or seismic source, characteristics of the earthquake source, and geologic conditions within the earth through which the vibratory anergy propagates. Certain frequencies of strong shaking may be amplified considerably owing to thin, low-velocity surface layers; the overall spectral level of ground motion may increase as the seismic velocities of near-surface materials decrease and/or as the thickness of the sediments increases (Murphy and Hewlett, 1975; Borcherdt and Gibbs,1976; Rogers and others, 1979, 1985). Ground shaking wherein the earth is not apparently permanently deformed has caused the greatest losses historically during earthquakes, because the seismic energy radiates over large areas where it encounters numerous works of man. In comparison, other mechanisms by which earthquakes cause damage to structures include direct displacement of the ground surface by a fault and damage that arises from ground failure. If surface faulting occurs immediately beneath a structure, the damage makes for spectacular T.V.
l footage, but the zone of damage tends to be restricted to the fault zone; hence, surface faulting damages a relatively small number of structures and consequently causes relatively small losses compared to losses caused by ground shaking. Losses from earthquake-generated ground failure (landslide, liquefaction, rockfall.) tend to be of intermediate magnituds in terms of the number of structures that they affect; ground failure tends to be localized, and can be predicted either deterministically or probabilistically by a careful analysis of the earthquake potential of a region and an analysis of the earth materials which comprise the surface and near-subsurface deposits, the topographic relief, and the association between cohesionless deposits and shallow ground water. The topics of surface faulting and ground failure are addressed in concurrent sessions of this workshop; thus, while important elements to consider, they will not be discussed herein. In this paper, I will discuss ongoing efforts to identify and understand the geologic factors correlated with attenustion or amplification of ground motion and which lead towards preparing predictive maps describing the ground shaking hazard in the Pacific northwest and the Puget Sound region.
llazards Assessment Program Evaluating the hazseds posed by moderate and large earthquakes in the Puget Sound-Portlant areas requires a concerted effort by numerous geologists, geophysicists, engineers, urban planners and elected officials who labor on dosens of related projects. The program properly includes at least 5 interrelated elements:
k
- 1. Collection and synthesis of earth science data;
o .
- 7. ; - ;, a; .-
~ ~,y; w, .
- 3. Preparation ~of loss estimation models;
[
11 . Incorporation of models into infonnation systems;
- 5. Selection and implementation of hazard mitigation measures. >
l This paper emphasizes selected aspects of the first two of these 5 i
elements. Once ground motion models describing the vibratory motions of large earthquakes in terms of the local geology are prepared, predictions addressing i how types and classes of man-made structures will respond to those motions are possible. A predictive ground response map also enhances evaluations of ground failure, including landslides, rockfall, and liquefaction-induced soil failures (see, for example, a discussion of liquefacticn hazards in the Los Angeles region by Tinsley and others, 1985).
t i
Predicting Ground Response '
The least model-dependent ground response maps are produced by combining !
measurements of ground motions with characteristics of the substrate beneath the recording instruments. In this way, the data reflect site conditions ;
typical of the map area and the attributes of ground motion in terms of spectral effects can be related to the regional geology. There are several phases to this effort which are described below.
Initially, one must cellect measurements of vibratory ground motion.
Typically these seismic recordings are made at many sites including at least one bedrock (reference) site and record ground motions caused by a variety of seismic sources. The recording sites should be selected to sample a variety :
, of subsurface conditions representative of the range of geologic conditions
( within a region. The seismic sources used in this part of the analysis include actual earthquakes, microtremors (see Kagami and others, 1986), local quarry blasts, and underground nuclear tests occurring at the Nevada Test Site. Each of these sources can be used separately or mutually for measuring relative ground response. However, naturally-occurring earthquakes are ,
difficult to use because their schedules are not advertised in advance and their geographic location will vary in place and time--azimuth-dependent effects are a variable which we desire to eliminate or control in the analysis, so a nuclear test or quarry blast which emanates from the same location are preferable for this and are most commonly used. Once obtained, the seismic records are each compared to the bedrock or reference site's ,
record of the same seismic event and the components of the ground motion l j
spectra which are attenuated or amplified are determined for each site. An l example from the Los Angeles region showing the time-histories of ground ;
motion generated by Nevada Test Site nuclear testing and recorded at eight sites in the Los Angeles region are shown in figure 1, after Rogers and others, (1995). The amplitudes at locations underlain by various types of i alluvium are significantly greater than the amplitudes at sites underlain by i rock. The reference station is CIT (Old Seismological Lab, Californis j
j Institute of Technology).
1 Tne second step is to collect geologic and geotechnical engineering dats !
l Trom the sites where the, recordings of groand motion were made. Those data j i
include measurements of the thickness, type, and physical properties such as
}
density, strength parameters, compressional-wave and shear-wave velocity of ;
the deposits benetth the site, as well as depth to ground water and depth to I( bedrock. In most esses, these data are not available because efforts are made t
i !
4 I
. ALLUVIUM
'"' 7 HO!
- I N
~
f Ir '
I ' ' ' ' ' '
- 0 031 -
03'7 MIL
'" ] l l iIN'NiU ;^ ! Ir^'^ ^ ' ^ '- 'l^^'="-"' '
-omi-~
0 3 ' --
O ATH oe h ll ,i.~4id'if,!ril#i-Il-- -" '^
W :
g -o031 -
w 8 C3 ' ~-
GMB
- o. .
I oom "
g .
f h,',7, 'i .M",7 ^4 - ~ ^ ^ " ^/ ' l - -
W 5 _ o03'7 c3t - FS4 H .
2 -
w com -
--.m .a .a i su . , . o g _ 7 . _ , u, n c i n ... ... ..._..- - - -- -
z
~~ ~
- c esi -
] p C 3'} 3838 SEDIMENTARY ROCK .
ou ^{,"$'!,2!t t '"c4" ^ ^ee 'c f
^ - -
1
-0031-
' 7 GOC CRYSTALLINE ROCK hii< -
000 -!' r-- ~ ^ N^ I k
^ ^i ' - ^ V -# ^
~
C011 J i
o cai 7 C:T M
oae f = =';yk%kff(W,':::. ;.v.-: ^- v :.v:- ::: -
_ l
~ 0 C3' 'i i i i I i i i i i i i i i t i t i o .'o es w w sw sn in ,w sw n m w n m u an TIME, IN SECONDS FIGUR E 1 -Radial component time-histora s of ground moton from a distant undergruurd nuclear esplosen kn Nesada rworded simuhanmusly at eight sites in the los An64es regWn and gruuled ectordmg to the t)1w of smlogic materals 1
immnha'nly le-neath cuh rei ordmg statum The amphiude levels et 1xations urderlain by alluvium dearly are greater j f than thmc at law atuis urviertain l,3 rock The devem of emphlkatbn alw appnars to t= ts44 ed t to the thukness of l
i underlurig allunum IlOI. IhMay Inn, > s) m. Mil M JLhan I.ibrary,372 m AT H, Athenaeum,372 m, CMll, Glendale M unn ipal linJ,hn.; 120 m F54. Fire Stat e m 4,15 m Q V'. Cl r, arul le M ers ria: k sites i
. . : .4. y, ,.
q..
g to locate the instruments away from the interfering effects of nearby
( buildings so that the so-called "free-field" ground motion is recorded.
Consequently, geological and geophysical experiments aust be conducted (such as seismic refraction and reflection profiles, exploratory drilling and sampling) to determine the properties of the materials.
The third step is to identify which attributes of the geologic units are associated with the seismic response data by forming statistically-determined clusters which group suites of geotechnical parameters with similar spectral characteristics measured at the recording sites. If the clusters are chosen so that the geologic properties associated with the various degrses of ground motion amplification or attenuation are mappable, maps predicting ground motion can then be prepared and used in loss estimation studies or compared to observed actual earthquake damage or Modified Mercalli or Rossi-Forel intensity levels to check the accuracy of the predictions. In any event, a :
matrix is easily prepared in which suites of geotechnical properties can bo l related to given attributes of ground motion spectra, even if a regional map ,
is not formally prepared. An example of this approach is that taken by Rogers and others (1985) in downtown Los Angeles, where maps of a s=all area were prepared in 3 period-bands in the range of 0.2 to 10 seconds. Although the importance of local geologic conditions on the relative severity of ground (
shaking has long been recognized, the quantitative prediction of the influence l of these conditions on ground shaking employing either empirical or ;
theoretical models is still in a developmental stage. The procedure is shown l conceptually in figure 2, after Rogers tnd others (1985).
Amplification of Seismic Ground Motion i
Studies of earthquakes in California, Japan, and Mexico have shown amplification of seismic waves at sites where thick sequences of uncon-solidated or semiconsulidated sediments and soil overlie more competent bedrock. Such sort and weak sedimentary deposits, such as San Francisco Bay i Hud (an estuarine deposit initially containing more than 50% water) or the recent lacustrine deposits of Lake Bonneville (Great Salt Lake) in Utah typically are characterized by low densities, high void ratios, and low shear- .
wave velocities. Some examples may be instructive. The 1933 Long Beach, California, earthquake caused more damage in Compton than in Long Beach, a finding that Wood (1933) ascribed to local geologic effects and which Campbell i (1976) studied and showed that for a given distance from the Newport-Inglewood '
fault zone, damage at sites underlain by unconsolidated soils (including the -
Compton area) was greater than at sites underlain by consolidated middle !
Pleistocene and late Pleistocens deposits, which underlie much of the City of !
Long Beach, California. The September 19, 1985 Mexic earthquake did j
extensive, localized damage hundreds of kilometers fros the epicenter in parts ,
of Mexico City underlain by 50+ meters of soft, water-saturated lake i deposits. The sedimentary soction had a fundamental vibratory period of about j 2 seconds; buildings in the range of 15 storeys had a similar period and i suffered severe structural damage and collapse, owing to resonance effects. ;
Accelerations were amplified from bedrock levels of 0.014 g on bedrock to 0.2 !
g, a factor of 5 on the soft soil sites.
\
The Puget Sound area does not have quite so severe site conditions as Mexico City, however, the effect of site geology on ground motions is expected )
( to have a considerable influence on levels of damage. The Seattle area, for
. . - - *' , ;v . . ,
DETERMINE SITE RESPONSE CLUSTERS COMPILE MAPS RELATING OEOLOGIC," l 4
AND THEIR MEAN SPECTRAL RATIOC FACTORS THAT INFLUENCE SHAKING f IN THE STUOY REGION TO 08 SERVE SPECTRAL RATIOS.
i COMPARE GEOLOGIC OATA MAPS TO
. /
i 10ENTIFY AREAS OF COINCIDENT ATTRIBUTES ;
PERTINENT TO SITE RESPONSE CLUSTERS *
. .i: ',: ;. 1:r
. i t. ec .f . . a:: MAP SHOWING DISTRIBUTION ,
. ,. , OF BEDROCK AND SEDIMENT I
i MAP SHOWING DEPTH j
g TO BASEMENT ROCKS ,
I i
- MAPS SHOWING THICKNESS OF
] 8-UNCONSOLIDATED HOLOCONE i 'Us E. AND PLEISTOCENE SEDI AENT l
i .
- w. . e ie. t
, 3' , ' ,'e ' MAP SHOWING VOID 8
s , , 7s e e RATIOS OF SEDIMENTS
$ ', '. ! t ' ,*
. i i
l u
4 l DEUNEATE RELATIVE RESPONSE UNITS AND ASSIGN THEIR MEAN SPECTRAL F.ATIO '
)
COMPARED TO CRYSTALUNE BEDROCK g
~
" N 33 g "/
),
W si I ,3 i
/
1 i
i 3 FIGURE 2 l
- I Procedure used to construct a predictive relative ground response map
- after Rogers and Others, 1985. i
! l i
3 I
t 1
!(.
i
__ _ _ _ - _ _ - _ - - - - __- _ - - -. _ _ - _ _ _ _ _ -__l
., c_
' %.f l .. i.
- .. f*i . ,
- w .- '. i l
example, is underlain by glacial till (ice-deposited sesioonsolidated mixtures I
(\ of clay, silt, sand, gravel, oobbles and boulders), post-glacial fluvial deposits, marine and estuarine muds, and fill materials such as sawdust and hydraulically jetted till. Thicknesses cf sedimentary deposits above bedrock range from 0 to more than 1 km near Seattle (Yount, 1983).
Historical Earthquake Effects in the Puget Sound Area The occurrence of two moderately large earthquakes in the Puget Sound I j
region during the past 40 years, the 1949 Olympia earthquake (M 7.1, Nuttli, i
- 1952) and the 1965 Seattle earthquake (Ms6.5, Algermissen and others, 1965), :
' illustrates several complex aspects of ground shaking and the geographic ;
distribution of the damage relative to some of the geologic deposits of ;
area. The prospect of a truly great earthquake occurring on the Cascadia subduction zone (Heaton and Hartzell, 1987; Heaton and Kanamori, 1984; !
Atwater,1987) is sobering. The 1949 and 1965 seismic events had rather deep j foci of 70 and 60 kilometers, respectively, and, in consequence, seem to have caused shaking damage that was slight (slight by California standards, where l
focal depths seldom exceed 15 km) compared to the magnitudes of the {
4 earthquakes (Algermissen and others,1965; Mullineaux and others,1967). The rather great focal distance and the relatively thick sedimentary sections -
through which the seismic waves were propagated and presu= ably attenuated are i thought to be responsible for the relatively low levels of damage (Langston,
- 1981; Shakal and Tokso , 1990). Yet damage to structures wa observed to vary greatly over short distances even if the buildings were sited on what were ;
- j ostensibly lithologically similar geologic deposits, especially for the 1965 event (Algermissen and others, 1965; Yount, 1983). ,
i
' Yount (1983, p. 263) reviewed the da= age patterns of the 1965 earthquake and noted that relatively heavy da= age occurred "in the lower Duwa=ish River .
area and southern downtown region of Seattle where unconsolidated Holocene l
alluvius and artificial fill make cost of the substrate; but damage was ;
relatively light in the upper Duwa=ish River Valley just a few kilometers to ,
the south, where si=11ar geologic c:aterials make up the substrate". The most !
l severe residential da: age involved brickwork and chimneys and seemed to be concentrated in the West Seattle area, a sector underlain by co= pact glacial
' sands and silts; yet the Beacon Hill and Magnolia areas, underlain by similar Pleistocene deposits, suffered little damage. Yount concludes that subsurface geologic conditions are of paramount importance in understanding the ground ;
response characteristics of the Puget Sound region. Initial efforts by the '
U.S. Geological Survey to study geologic aspects or ground response focus on areas of West Seattle and Olympia (see papers by Kenneth W. King and Arthur C.
1 Tarr, thir volu e, and figure 3) and will endeavor to discover if heretofore
] unrecognized differences in site geology can be discerned and used to i= prove 2
predictions of earthquake-generated strong ground motion.
i Systematic efforts to record ground motion at sites da: aged by the i
1 historical earthquakes are already underway under the direction of Kenneth King (USGS, Golden, Colorado) and include ground motion recordings, reflection l profiles to determine the depth to prominent reflectors in the subsurface and depth to rock.
- A program of exploratory drilling anc geotecnnical samplins !
intended to determine the nature of the substrates and assess the degrees of '
similarity or differences among the subsurface stratigraphic units of the
( region, will co .tence in early July, 1939, under the direction of John Tinsley
- l
- l l l 1
I
,( ,,Y?,'.l*,$
- _Elliott Ge ,
~
[ i,, , W Se ttle I I
A' F w'e $tation A $0ig # ' *;l- 'lIL' ANO
.A Hlewathe 4 t,iJii.eae 17 V3'k i y" ^I'3'i#Tahd y-
} ' ' f80 Genesee - ,s
- (O", '
n ,. ,s_
A Pek '
3-Genesee Hill I A ,$g"" l' school W. fonttle .
Gott Course
- Seward Perk Re f erenceA Wright o b' * '
(
4 8: '
k 0 1 i
l MILES =
i
-Budd in:a r' l g Rosseve:t School ;
OLYMPIA A
l Accelerograph $ite /
Capitol IOowntow n) /
l E8
8 A State Capitol Q '
Twmwat er A
\'
nere ence s.ie
- TUMWATER L
1 4
O. ?.
a woes
! l
- TIGURE 3
- {
Locations of seisnic stations comprising part of the array i for initial studies in the Seattle and Olympia areas, af ter l
! Tarr and King,1937. i i l I J J
l 4
l
)
i
7 1
l (USGS, Menlo Park). These initial' exploratory studies will include down-hole P-wave and S-wave geophysical studies and will commenee at 14 sites from
[. Genessee Park and downtown Seattle to the West Seattle and Magnolia areas, and additionally at 6 sites in the 0,1ympia area; the results of these and subse-quent studies are expected to contribute to understanding of geologic aspects
! of ground shaking hazards in the Puget Sound region during the next three years. The properties and character of Holocene alluvium and man-made fill, the presence of and depth to poorly-consolidated sand deposits which are
, situated within otherwise well-consolidated last-glacial and pre-last-glacial i
! deposits, and the configuration of the bedrock beneath the metro areas are '
) expected to be key geological aspects in the analysis.
' f The geologic exploration phase is commencing this year, so data are !
y rather sparse. I anticipate making full use of exploratory data obtained for l d
purposes other than studies of ground motion. Initial contacts with geotech- '
nical engineering firms have been encouraging and helpful. Future activities ;
4 will see expansion of the database to include studies of sites in the Port-land, Oregon, area as well as establishing additional recording sites distri-buted throughout the Puget Sound region. It would be desirable from a statis-l
-; tical standpoint to have as many seismic stations and detailed site studies as ;
possible. I will endeavor to use in the analysis generalized stratigraphic '
j systems such as those derived during water resour;;s investigations in the region. The pertinent geotechnical data will be incorporated into a database using the software modified by Art Tarr, hence will be available during later stages of this study for future reference and use by interested researchers.
i Coc:ents concerning the approach and methodology and the local geology are
{ respectfully solicited.
l 2
l 1
d i
I i
d ,
I l i !
2 l
J f
I 4
4
}
i
t t i REFERENCES CITED .
4 Algermissen, S. T. and Harding, ;3. T. , 1965 The Puget Sound, Washington, h earthquake of April 29, 1965: U.S. Coast and Geodetic Survey Preliminary [
Seismic Report, 26 p. l l Borcherdt, R. D. and Gibbs, J. F., 1976, Effects of local geologic conditions l in the San Francisco Bay Region on ground motions and the intensities of !
l the 1906 earthquake: Bulletin of the Seismological Society of America, [
- v. 66, p. 467-500. !
I Campbell, K. W., 1976, A note on the distribution of earthquake damage in Long [
j Beach, 1933: Bulletin of the Seismological Society of America, v. 66, p. ;
1001-1006.
Heaton, T. H. and Kanamori, H., 1984, Seismic potential associated with l
i subduction in the northwestern United States: Bulletin of the j i Seismological Society of America, v. 74, p. 933-941. L I
l Heaton, T. H. and Hartzell, S. H., 1987, Earthquake hazards on the Cascadia !
subduction zone: Science, v. 236, p. 162-168. l Kagami, H. , Okada, S. , Shiono, K. , Oner, M. , Dravinski, M. , and Mal, A. K.,
1936, observation of 1- to 5-second microtremors and their application to 4
earthquake engineering, Part III, A two-dimensional study of site effects [
- in the San Fernando Valley: Bulletin of the Seismological Society of ;
li America, v. 76, no. 6, p. 1801-1812. L i
i
' Langston, C. A. ,1931, A study of Puget Sound strong ground motion: Bulletin of the Seismological Society of America, v. 71, no. 3. p. 833-904.
Mullineaux, D. R. , Bonilla, M. G. , and Schlocker, J. ,1967, Relation of f building damage to geology in Seattle, Washington, during the April 1965 t earthquake: U.S. Geological Survey Professional Paper 575-D, p. 183-191. '
Murphy, J. R. and Hewlett, R. A., 1975, Analysis of seismic response in the L city of Las Vegas, Nevada: A preliminary microzonation: Bulletin of the l Seismological Society of America, v. 65, p. 1575-1597. !
i i Hutt11, O. W., 1952. The western Washington earthquake of April 14, 1949: !
Bulletin of the Seismological Society of America, v. 42, p. 2123. l I
Rogers, A. M. , Tinsley, J. C. , Hays, W. W. , and King, K. W. , 1979. Evalustion of the relation between near-surface geological units and ground response !
. in the vicinity of Long Beach, California: Bulletin of the Seismological I j Society of Ameries, v. 61, p. 1603-1622. i 1 ;
Rogers, A. M. , Tinsley, J. C. , and Sorchardt, R. D. ,1985 Predicting relativo !
ground responses in Ziony, J. I., ed., Evaluating earthquske hazards in !
4 the Los Angeles region: U.S. Coological Survey Professional Paper 1360, ;
} p. 221-247. j i
1 t
i b ,
t d
i Shakal, A. F. and Toksoz, M. N.,1980, Amplification and attenuation of site - '
-[
structure: Puget Sound strong ground motion, abstract Earthquake ;
Notes, no. 50, p. 20. ~
Tarr, Arthur C. and King, Kenneth W., 1987, USGS begins seismic ground response experiments in Washington: Washington Geologic Newsletter, v.
- 15, no. 2, (May), p. 11-18. ,
! l Tinsley, J. C. , Yeud , T. L. , Perkins, D. M. , and Chen, A.T.F. , 1935, ;
Evaluating liquefaction potential irt Ziony, J. I. (ed), Evaluating earthquake hazards in the Los Angeles region--An earth science perspective: U.S. Geological Survey Professional Paper 1360, p. 263-316.
a i Wood, H. O., 1933, Preliminary report on the Long Beach earthquake: Bulletin :
of the Seismological Society of America, v. 23, p. 42-56.
l fount, J. C., 1993. Geologic units that likely control seismic ground shakinc l
j in the greater Seattle area: U.S. Geological Survey Open-File Report 83- l j
19, p. 268-273 L I (
i !
4 i
5 e
l i
I i i !
i c >
i i i 1
i I l I
1<
l' i
I 2
I I
4