ML19248D355
| ML19248D355 | |
| Person / Time | |
|---|---|
| Site: | Framatome ANP Richland |
| Issue date: | 09/29/1978 |
| From: | TERA CORP. |
| To: | |
| Shared Package | |
| ML19248D353 | List: |
| References | |
| REF-PROJ-M-3 NUDOCS 7908150652 | |
| Download: ML19248D355 (58) | |
Text
.
SEISMIC RISK ANALYSIS FOR THE EXXON NUCLEAR PLUTONIUM FACILITY, RICHLAND, WASHINGTON s.e.:ne:: :
LAWRENCE LIVERMORE LABORATORY P.O. Box 808 Livermore, Californic 94550 Attention: Mr. Don Bernreuter Project Moncger TERA CORPORATION
_ Teknekron Energy Resource Analysts 2150 shcituck Avenue Berkeley California 94704 415 845 5200 September 29,1978 I 0$3:?.,'.';)
7008150b{]c
~~
J q
TABLE OF CONTENTS Section Poae
1.0 INTRODUCTION
AND SUMM ARY...............................
I-!
2.0 SEISMIC RISK METHODO LOGY.................................
2-1 Theory........................................................
2-2 3.0 G EO LO G Y....................................................
3-1 Summcry......................................................
3-1 R eg i ona l P hy si og ra phy..........................................
3-I S i t e G eo l ogy..................................................
3-5 G eo l og i c H a z cr d s..............................................
3-9 4.0 SE I SM O LO G Y..................................................
4-I S e is m i c i t y.....................................................
4-2 S i t e E f f ec t s...................................................
4-2 Se is m i c Sou rc e R eg ion s........................................
4-3 Ecr th quck e S ta t i st i cs...........................................
4-5 5.0 CALCU LATIONS AND R ESULTS................................
5-1 input.........................................................
5-1 A t t e n u a t i on...................................................
5-3 Results.......................................................
5-5 R es p ons e Sp ec t ru m.............................................
5-9 6.0 B I BL I O G R AP H Y................................................
6-
.., : I_ c 3
o O ;.>.
TERA CORDCiRATION
s
1.0 INTRODUCTION
AND
SUMMARY
In this report, TERA Corporation presents the results of a detailed seismic risk onclysis of the Exxon Nuclear Plutonium Facility at Richland, Wcshington.
This report is one port of c Icrger effort being directed by the U.S. Nuclear Regulctory Commission.
Tt e NRC's objective in commissioning the overall report is to essess and improve, to the extent practicable, the cbility of this facility to withstand adverse naturcl phenomenc without loss of ecpocity. This report focuses on earthquakes; the other natural hazards, being addressed in sepcrate reports, cre severe weather (strong winds and tornados) cnd floods. The overall analysis will provide on assessment of the consequences of an occident resulting from any of these natural phenomenc. Ti e essessment will express c quantitative probcbilistic mecsure of the potential structural dcmage and the relecse function. It will also provide o probcbilistic estimate of the resulting dose of radioactivity to the public.
This study was performed under contract to the Lawrence Livermore Laborctory (LLL). The study was directed by Don Bernreuter of the LLL Nuclect Test Engineering Division. At TERA, the study was managed by Lawrence Wight.
To ensure credible results, very sophisticated but well-accepted techniques were employed in this component of the project, on cnclysis, of the seismic risk. The calculationcl method we used, which is based on Cornell's work (1968), has been previously cpplied to safety evoluctions of major projects.
The historical seismic record used in this onelysis is the most current and complete ovcilable. The record, which is a synthesis of dato from the University of Wcshington Puget Sound Array, the Eastern Wcshington Array. cnd the NEIS dcto bcses, was provided by Battelle Northwest, R:chicnd.
The resulting seismic record, covering the period 1833-1977, was used to identify all possible sources of seismicity that could offect the site. Inodequccies and incompleteness in this record were explicitly considered in the definition of llUU.@
y TERA CORPORATION
source regions md their activity rctes. Where 1here were uncertainties, we cssigned subjective probabiliiies to the span of uncertainty.
The ecceleration attenuation relation used in the analysis wcs developed based on a careful evaluation of all relevent strong motion dato.
To guarantee use of the best available calculational methods and input, the entire project wcs reviewed by Professor Stewcrt W. Smith, a University of Washington seismologist, well known for his resecrch on west cocst seismicity and tectonics. Dr. Smith is responsible for several seismic crrays in Wcshington, Oregon, and Californic, one of which covers the Richlcrd crec.
The results of our malysis, which include estimates of the uncertainty, cre presented in Figure 1-1.
Our best estimate curve indicates that the Exxon fccility will experience 10 percent g with a return period of 600 years and 15 percent g w.ith a return period of 2,500 years. The curves on either side of our best estimates represent roughly the one standard deviation confidence limits about this best estimate.
These. curves provide the seismic cesign bcsis for the Exxon facility in term; of peck ground acceleration.
For those structures cna equipment that could experience structural amplification, we judge that the mean olluvium response spectrum prescribe 1 WASH 1255 (AEC,1973) adequately chcracterizes the frequency content.
bh3UN) l-2 TERA CCRDORATION
10,000 f
/
/'
/
\\
s l
/
1
/
/
/
I s
/
r I /
I
/'
I I
I
{
l I /
l[
/
l
/
j l/
)
/
t r
/
c
/
I e
l
/;
/
\\
/
/
/
1,000 1 !
/
J
/
/
/
'/
/
/
/
/
f f
/
/
/,
l 2
/ /
/ l/
E
- /
,b J
/
/
y y
/
5
/
9 100
/
/
1 C
iIJ
//
C f//
i
///
i if i 11l
- /'
/
4 I
I 10 I
I i
e i
l i
l l
l l
t i
I I
l I
i 1
t I
O 5
10 15 20 25 30 PEAK HORIZONTAL ACCELERATION (%g)
FIGURE l-1 n..3.
G.: Q. '.. b EARTHGUAKE RISK AT THE EXXON FACILITY TERA CORPORATION
2.0 SEISMIC RISK METHODOLOGY A seismic risk cnolysis is only as credible as the risk analysis methodology and the input to it. This section presents the basis for our selection of a probcba-fistic Poisson model for the risk cssessment at the Exxon facility.
There are generally two distinctly different cpproaches to seismic risk cnolysis; probcbclistic and deterministic.
Using the deterministic approach, the cnolyst judgmentally decides that an ecrthquake of c given magnitude or intensity occurs at a specific location. He then ottenuates the ground motion from the ecrthqucke source to the site and determines the effects of that quake. The problem in using this approach is that it is difficult to define the margin of safety or the degree of conservatism in the resulting design parcmeters. Analysts cre of ten asked to provide infarmation on the " maximum possible" or "most probable" earthquakes for design purposes, but the deterministic cpprocch does not easily provide those cnswers.
A probabolistic coproach, on the other hand, qucntifies the uncertainty in the number, size, cnd locction of possible future ecrthquakes and allows on analyst to present the trade-off between more costly designs or retrofits and the economic or social impoet of a failure. Because the product of a probabolistic approach is a measure of the seismic risk expressed in terms of return period, this trade-off con easily be quantified.
Although the probabilistic approoch requires significantly more effort than the deterministic cpproach, it has the following advantages:
it quantifies the risk in terms of return period; e
it rigorously incorporates the complete historical seismic e
record; it con incorporate the judgment and experience of the e
analyst;
- ;c. L 1 - a w.
- a T
2-1 TERA CORPCRAilCN
e it cecounts for incomplete knowledge regcrding the loca-tion of faults; e
it has the flexibility to assess the risk at the site in terms of spectral acceleration, velocity, displacement, or earth-quake intensity.
The credibility of the probabilistic cpproach has been estcblished through de-toiled technical review of its applicatior to several important projects and crecs.
Recent applications include assessments of the seismic risk in Boston (Cornell, 1974), in the San Francisco Boy Area (Vugliente,1973), in the Puget Sound Area (Stepp, 1974) and continental United States (A!germissen and Perkins,1976).
Results of these studies have been applied to, among other crecs:
e Development of long-range earthqucke engineering re-search goals; Plcnning decisions for urbcn development; e
e Environmental hczards associated with the milling of uranium; and Design considerations for radioactive waste repositories.
e This diversity of cpplication demonstrates the inherent flexibility of the risk assessment approcch.
THEORY The risk calculations con be fundamentally represented by the total probcbility theorem P [ A ] = (( P [A/m and r] f (*} I (r) dmdr M
R where P indicates probability, A is the event whose probcbility is sought, cnd M cnd R cre continuous, independent random variables which influence A.
The probability thct A will occur con be calculated by multiplying the conditional probability of A, given events m and r, times the probabilities of m and r, and integrating over all possible values of m and r.
- d3 E.I.28 2-2 TERA CORPORATION
6 4
in our assessment of the Exxon fccility, A will be taken as maximum accelerotion and therefore P [ A/m and r 3 will be derived from data relating peak acceleration to epicentrol distance end earthquake magnitude. Often krown as attenuation date, these dato are usucity lognormally distributed around a mean relationship of the form (McGuire,1977a).
M 2
A = C; e (R.r,)
The distribution on earthquake magnitude, fg(m), ccn readily be derived from an actual or postulated frequency relationship of the form log N = a-bM where N is the number of ecrthquakes having magnitude grecter then M, cnd a and b cre conste.its chcrocteristic of the particular source region under con-sideration. It follows (Corne!!,1968) that f can be derived from the cumulative M
distribution function, FM, which has the form, FM = k 0 -e M) where k is a normalizing constant and S = binl0.
The distribution on distence, f (r), depends on the geometry of the problem R
under consideration.
For simple geometries, the distributions can of ten be integiated analytically. Realistic geometries however, require numerical eval-uction of the integral. A very versatile comycter program has been developed (McGuire,1976b) that incorporates the theory presented cbove with a numerical integration scheme that allows for evoluction of vey complex source-site geometries.
YdU1,20 1.:
TERA CORPORAilC.N
The overall cpproach to performing seismic risk assessment by this theory is svalitiurized below. First, the hisioricci ecrthquake record and local attenuation dato are combined with the experience of the onclyst to produce the functional relationships cppliccble to the crea under consideration. The source regions cre divided into circular sectors cnd proportional seismicity is allocated to each sector. The total expected number of events cousing maximum accelerations at the site greater than a porticular test acceleration is obtained by summing the events from each sector within each source region. The risk associated with this test o:celeration is then calculated using the conventional cssurnption thct earthquckes have o Poisson distribution in time. It then follows that the return period is simply the reciprocal of the risk.
In recognition of the uncertainties in any seismological analysis, we test the sensitivity of results to variations, corresponding to the uncertainty, in key pcrameters. The results are then combined by cssigning subjective probcbilities to the range of variations and weighting the results accordingly.
The final synthesized estimate of the seismic risk con be considered a Boyesion estimcte (Cornell and Merz,1974).
6 % 130 TERA CORPORATION
3.0 GEOLOGY
SUMMARY
The Exxon Nuclear site is in the Pcsco Basin which lies in south-centrol Washing-ton. It is within the Columbia River Boscit Pictecu that extends over 50,000 square miles. The site is under:cin by thick, widespread boscific lava flows at a depth starting at 151 feet and centinuing downward over 10,000 feet.
The surface of the site is coverec by c Icyer of Winchester send of thickness two to seven feet. Site borings and necrby execvotions indicate that the superficici few feet of sand is underlain by 20 feet or :,o of the Pcsco gravels. The sand and grovel rest upon the deeply eroded surface of the Ringold Formation which, where not eroded, consists of a thick series of silts, sands and gravels. The Ringold Formation rests on the bascit, the top of which is estimated to be at c depth of 151. feet below the site surfcce.
REGIONAL PHYSIOGRAPHY Regiorp!!y, the study crec is situated in the Columbic Bcsin. This geologic and structural bcsin is physiogrcphicc!!y defined as the Columbia Plateau (Hunt, 1974).
The site is situated near the Columbia River which is the major water course in the region. A locci physiogrcphic and structural depression in the plcteau, the Pesco Bcsin, contains the site cnd comprises approximately 1,600 square miles of undulating semicrid plain with low lying hills, dunes and intermittent streams.
The northern and southern boundaries of the Pesco Bcsin are defined by the Saddle Mountains and Rattlesnake Hills, respectively.
The easterly ends of Umtonum and Yakimo Ridges mark the western boundary of the basin, whereas to the east the basin merges into a vast expense of dunes, dissected fictlands and coulees northwest of the Snake River.
- m.r ;
- t.e m. 1 > - s.
3-1 TERA CORPORATION
Recional Geolocv The structure of the region was produced by a combination of volcanic construc-tion and of moderate tectonism during Cenozoic time. The geologic provinces correspond to the physiogrcphic provinces.
Recional Straticrcohv The Pasco Basin crea is undericin and pcrticily surrounded by en extensive and thick sequence of bescits. The Rattlesncke Hills deep test well west of the project site penetrated more then 10,000 feet of bcscit composed of nearly 100 flows and interflow sediments (Raymond and Tillson,1968). Bescits of Miocene and Pliocene age exposed in the Columbic Bosin cre defined collectively as the Columbia River Group.
Recional Structure The principal structural elements of the region surrounding the Exxon Nuclear site include the stratification of volcanic and sedimentary units, major folds and faults, joints and clostic dikes. On a regional scale, stratification of lithologic units is horizontal or dipping slightly to the west-southwest (Brown,1968).
Folds - The folds in the Columbia River basalts are the most prominent structural fectures in the Columbia Basin. The folds con be classified according to three general orientations: ecst-west trending, west-northwest trending, and northeast trending.
Faults - The relatively few faults that have been mapped in the Columbia Basin cre directly relcted to the folds within the basalt. This relationship suggests that many of the fcults represent local adjustment to the folding, in generci, the faults found within the basin are (1) of limited length, discontinuously less then 30 miles; (2) of moderate offset, usually less than 500 feet; and (3) chorocterized by the obsence of structural features associated with recent fault activity such as offset streams and truncoted alluvial fans (Brown,1968).
.f.%7 &,>. w '.
, o 3-2 TERA CCRPORATION
The faults of significance to the proposed site are cssociated with anticlinct folds:
Wallula Gap fouit and Rattlesnake-Wallula alignment Yakima Ridge enticlines and a minor fault Umtonum fault cnd anticline Gcble Brutte end Gcble Mountain cnd minor faults Scddle Mountain fault and anticline Fold /Foult Relationships - Folding hcs been the primary response of the stra-tigraphic units in the Columbia Bosin to tectonic deformation, cnd fculting hcs been secondary (Jones and Deacon,1966 ond Brown,1968). The known faults hcve short to moderate length, have small cumulctive displacement, and lack evidence of recent movement, even though subsidence of the bcsin may have continued to the present (Binghcm, et al.,1970 and Brown,1968).
In general, faulting hcs been contemporcneous with folding. Faults were formed when joint surfaces could not relieve accumulated stress developed during folding. Mony of the supposed fcults that have been mapped in the Pcsco Besin crea could be interpreted as monoclinal flexures that have been eroded along their exes.
This interpretation may explain the steepening of dips and the presence of bcscit breccios cs erosional remnants of the monoclines (Brown,1968 and Grolier,1965).
Recional Ground Water - The ground water in the Hcnford region exists as confined water in the permeable ports of the layered boscit bedrock cnd cs unconfined water in the overlying sedimentary deposits. The rubbly tops of some of the basalt flows and the sedimentary interbeds are permecble.
These
~
permeable parts transmit ground water in o confined condition and may yield water to wells in the ccpocity of up to 1,500 gallons per minute. The pressure levels of the ground water in the bcscit cre controlled by: (1) local structure, d.>.!_33 3-3 TERA CORPORATION
wnich either impedes movement of ground water down the pressure gradient or facilitctes its leckage from confinement, and by (2) the meager recharge, which gains entry to the equifers locally. The ground water level in the basalt in this crec varies from o few feet to several tens of feet cbove the level of the Columbia River. Permeable zones in the stratigrcphic succession of Icyc flows occur irregulcrly ct unequal intervals of several hundreds of feet throughout the bescit sequence.
The occurrence of ground water in the bascit hos been described in several publications (Newcomb,1959 cnd Newcomb, et cl.,1973).
The structural controls on trie ground wcter for the Honford Reservation are described in Newcomb et al. (1973) and for type crecs in Newcomb (1961). The quality of the ground water has been described for the region in Newcomb (1972).
The sedimentary materials above the boscit contain a regional body of unconfined ground water that dischcrges to the Columbio River. The. egionc!
water table in these materials slopes gently to the river or to the zone of bank storage, which varies from a few hundred to severci thousands of feet wide, along the river (Newcomb and Brown,1961).
Over much of the Pasco Bcsin, the unconfined water table lies within the Ringold i crmation, which consists largely of silt, sond, gravel and cicy. Locclly along the river cnd benecth the most recent cbondoned channels, the sand cnd gravel of the glacio-fluvic! deposits extend below the water tcble and below higher stages of the river.
The unconfined water table beneath the Honford Reservation has a relatively steep ecstward slope outward from the lower ends of the mountain valleys and then c more gentle slope north, ecst and south beneath the terroin lands. Arti-ficial recharge hos citered the shape of the general water tcble of the reserva-tion; however, in the cree of the site there are no present significant sources of crtificici recharge.
bEUl@l TERA CORPORATION
SITE GEOLOGY General in general, the crec s >rrounding the plont is open, semi-crid desert land and is very spcrsely settled with the exception of the "Tri-Cities" of Richicnd, Kenne-wick on3 Pasco. Most developed land within a five-mile radius of the site is used for agriculture in raising irrigated crops and for residences. The undeveloped crea north and ecst of the site is in the DOE Hanford Reservation.
Site Geoloav Tooocrcohv - The local cree of the site is flat but covered with a series of wind-formed ridges ranging from five to cbout 30 feet high; the ridges are parallel and extend in the NE-SW directions. The generci trend of the surrounding terrain is a gentle upward slope toward the north and northwest (clong the Columbia River and Cold Creek), cnd downwcrd towcrd the south and southwest. The general slope of the land has been crected by the scouring and deposition action of water and wind.
The dominent topographic feature to the northeast and ceross the Columbia River is c nearly continuous outcropping, known as White Bluffs. The bluff clong the ecst bank of the Columbic River varies in cititude between 670 and 930 feet MSL.
There are no topographic obstructions to the southeast of the site, cs the bluff terminates abruptly. The land surface on both sides of the river is essenticily the scme cititude (!00 to 200 feet above the river) cnd gently rolling.
The confining topography to the west and south are the Rattlesnake Hills and the Horse Heaven Hills. The Rattlesnake Hills are cut in three places. The lowe::t cut is occupied by the Yakimo River (near Benton City). The continuation of the Rattlesncke Hills gradually merges into the Horse Heaven Hills south of Kennewick.
t:;..3.62 3-5 TERA CORPCRAT'ON
Straticroohv - The surface of the site is covered by c Icyer of Winchester sand of thickness two to seven feet. Site borings and necrby excavations indicate that the superficial few feet of scnd is undericin by 20 feet or so of the Pasco grovels.
The sand and gravel rest upon the deeply eroded surface of the Ringold Formation which, where not eroded, consists of a thick series of silts, sands and gravels.
The Ringold Formation rests on the bcscit, the top of which is estimated to be et a depth of 151 feet below the site surface. A geologic column is shown in Table 3-1.
Observations by R.E. Brown and others (1958,1960) show that the total thickness of these deposits wcs at tecst 1200 feet. The thickness of the Ringold matericts remaining under the building site is only cbout 150 feet.
it is probcble, therefore, that almost 1,100 feet were removed by the scouring action of glacici flood streams before the present 22 feet of send cnd gravel were deposited by the Icst (or Ictter few) f kx>d streams.
Therefore, the Ringold matericts underlying the building site have been subjected to o consolidating load of at least 33 tons per squcre foot.
In addition to the preconsolidation loading discussed cbove, major portions of the Ringold materials have been partially herdened (indurated) by cementation with calcium carbonate in the form of caliche. The average seismic velocity of saturated Ringold materials is reported to be cbout 10,000 feet per second (Domes and Moore,1970).
The most detailed necrby information on the besof t bedrock was provided by the logging of an oil well which penetrated over 10,000 feet of bcscit flows in the southern portion of Rattlesnake Hills. Although this well did not completely penetrate the entire series of bosof t flows, it did pass through over 100 recognizcble flows cnd their osxiated "interbeds." Raymond and Tillison (1968) and Brown (1958) have obtained evidence to indiccte that the entire thickness of the boscit flows is about three miles and that the flows individually developed at long intervals, i.e., cbout one flow per 500,000 years. The minimum compressed seismic velocity of these flows is reported to be 14,000 feet per second, with on b!!b13U 3-6 TERA CORPORAilCN
Table 3-1 GEOLOGIC COLUMN, EXXON NUCLEAR SITE Death and Ace Thickness varies from Winchester Sand.
Fine to medium 2 to 7 feet sand with occasional gravel. Loose to Recent Deposit medium density Lme windblown.
Thickness varies from Pasco Sand cnd Gravel with probably 15 to 20 feet some cocoles and coulders.
This Deposited cbout 10,000 meterial was deposited by the lost, or years ago Ictter few, glacici flood streams.
22 to 170 feet Ringold Formation.
Composed of Deposited during interval silts cnd fine sands of Ickebed depos-between about 20,000 and its and scnd, gravels and cobbles of 500,000 years ago glacial streams. Silty matericts pre-domincntly in the bottom portion of the formation. The materials have been pcrtially cemented with calcium carbonate.
These materials have been preconsolidcted by the sub-merged weight of cbout I,100 feet of material that was eroded by glacial strecms prior to the deposition of the overlying send cnd grovel.
151 feet to well over 10,000 Bescit.
This is a portion of the feet of bcscit flows, overaging C51umbia River bescits which are the 90 feet in thickness, with seccnd most extensive in the world.
interbeds often present The boscit flows have on unusually between the flows. It is uniform thickness. In the main the estimated that the flows rock is hard, high density, brittle occurred at intervals of about
" trop rock" which is sepcrcted by 500,000 years prior to 7,000,000 pentagonci vertical columncr jointing years cgo and horizontal interbeds. The inter-vol between flows was unusually long so that even some thin cool deposits are present in some of the interbeds between flows. Raymond and Tillison estimated that the total thickness is about 3 miles including the interbeds.
If this be so, then the deposit consists of about 170 flows with associated interbeds.
- 2 "
- p^?f ')'
- Adapted from Domes and Moore,1970.
3-7 TERA CORPORATICN
overage of 16,400. During the overage long-term period between the individuct flows, semi-continuous "interbeds" often developed and were at least in pcrt buried by the next lava flow. Attention is invited to the fact that these lavos had cn unusually low viscosity, which of ten resulted in basalt flows extending several or more tens of miles, with a fairly uniform thickness. The entrapped interbeds consist of materials with a low seismic velocity, which occumulated during the average 500,000 years between flows. These materials consist of residuct soils (soprolites), Icke bed and swamp deposits, as well as more isolated stream deposits and thin coal beds. The large number of individuct flows, which are of ten sepcrated by interbeds of sof ter materic!, lends cdditional credence to R.E. Brown's (1968) opinion that stresses resulting from tectonic forces were primarily relieved by slippage between the individual boscit flows rather then by faulting which would tend to produce ecrthquake forces.
The grect majority of the bcsalt fic us mostly occurred during the 14,000,000 years of the Miocene which ended about i 1,000,000 years ago. Some few flows are believed to have occurred during the following Pliocene Time and ceased about 7,000,000 years cgo. The composition and structural characteristics of the older rocks underlying the bcsalts hcve not been determined.
Structure - The fault necrest the site is the Rattlesacke-Wallulo Fouh which trends southeostward from the Rattlesnake Hills and through the Wallula " water gcp" for a distence of 50 miles.
The structure pcsses about seven miles southwest of the site. Recent U.S.G.S. (1966) field work showed that the fault, where exposed in basalt, had formed a crushed zone from 100-200 feet wide.
Just northwest of the Wallulo water gcp, almost vertical offsets ranging from 200 to 300 feet were found in the bcsolt. The fcult has been traced for 27 miles to the southeast from the Yckimo River; its trace north of the Yakima River has not been found. Evidence of " geological recent" movement was located by the U.S.G.S. cbout 35 miles southeast of the site.
4-b J u e; rz 3-8 TERA CORPORAilCN
Ground Water - The ground water table at the site is related to both the Yakimo River to the west and the Columbia River to the east. The site lies in the wye formed by the junction of these two streams so that fundamentally only events occurring in these strecms or in the site crea will significantly of feet the water table.
Borings drilled ct the site revealed that the lower two feet of the Pcsco gravels cre saturated. This puts the around water tcble cbout 19 feet below the surfcce.
This is consistent with the regional water table contours which are cecurate to 10 feet in the vicinity of the site (ERDA,1976).
Construction of the proposed Ben Frcnklin Dom is expected to cffect the water tcble only nominctly.
The construction of further upstream storage dams (Canada) is expected to reduce flood levels, and thus maintain the water table at the site at its present elevation or lower.
GEOLOGIC HAZARDS General Non-tectonic geologic hazards for the Exxon Nc.ecr facility have been evaE cted. This has been done on a preliminary bcsis with emphcsis placed upon a thorough search of the literature and review of existing maps of the crec.
Personal communications have been estcblished with experts knowledgecble with this particular crec. In addition, a brief site visit was made including both me facility proper and the immediate surroundings. The facility site was examined in context within the crect extent of the Pasco Basin.
Geologic Hazards Slooe Failure - Existing natural slopes have been determined to be stable; only those slopes on the enclosing cnticlinct ridges, Goble Mountcin, Wahluke Slope and the White Bluffs are steep enough for concern. Of these, the White Bluffs pose the greatest concern because of I) the clay-rich nature of some beds about b %.[ 3 3 3-9 TERA CORPORATION
river level and within the Ringold Formation,2) discharge of large quantities of irrigation water to ground atop the bluffs, 3) gentle dips of the Ringold beds toward the Columbia River, and 4) the eastward shif ting of the Columbia River and its uncercutting of the bluffs. Slides of a million or more cubic yards hcve occurred within the Icst cbout 12,000 years; consequently, more are expected.
Not likely to be impounded, the river would more likely be diverted to a more westward channel in the slides crecs.
The site proper is relatively flot and totally locking in slopes of a size sufficient to cause problems. Surficial matericls cre stcble.
Many Icndslides are present in the region cnd have been attributed tn ecrthoucke activity, as ecrlier noted. However, they cre only occcsionally directly caused by earthqvckes and more commonly result from excess water in the ground. A cicssic ccse is that of the slides clong the north face of the Scddle Mountains, attributed by some persons to ecrthquckes, especially the Corfu qucke. In view of the findings in recent years, the indications are that most of the features cre related to the glacio! Lcke Missoulo flood of about 12,000 years ago, not historiccIly recent quckes.
Geolocically Reicted Problems - Actual or potentic! problems relcted to surfcce or subsurface subsidence, uplift or collapse are not known to exist at the site.
Karst terrains, covernous conditions, tectonic depressions and related fectures have not been identified either in the region or in the vicinity including the nuclear facility.
Borings show that surficial deposits consist of from two to seven feet of loose
" blow sond" cnd grovel undericin by denser sand and gravel. Conditions are essentially the same throughout the site crec.
Pesco sends and gravels with a thickness of slightly less than 20 feet underlie the surfcce material. The penetration resistance blow counts indicate that the material is dense. Conditions are generally uniform and no fault is known within or adjacent to the crec.
W.SE0 3-10 TERA CORPORATICN
Differential compaction is co nmon in unconsolidated alluvium cnd was a major o ince William Sound (Anchorege) ecrth-cause of domcge in Alcska during the r
quake of 1964. The sediments at Honford have a high degree of natural com-paction. The Ringold Formation sediments at one time filled the Basin to en altitude of about 1,000 feet (the crest of the White Bluffs), but have been eroded in port from the Honford Reservction.
They have been irregularly covered subsequently by later sediments, the Pasco Gravels. Hence, et one time 200 to 600 feet more sediments existed as a static load than now are present over the topmost Ringold Formation beds, o factor in their compaction. The beds, too, are low in permeability, owing to their age, the resulting cementction and opportunity for compaction subsequent to deposition. Those foetors cnd the relatively high seismic (P) wave velocity of cbout 6,000 to 12,000 f t/see confirm the compaction.
The Pesco Gravels cnd their fine-grained equivalent, the Touchet Beds, were laid down by the glacial Lcke Missoulo floods. The grovels that underlie cil but the bcsin margins commonly cre open-work or semiopen-work gravels with high permecbilities. Seismic (P) wave velocities cre consistently low, cbout 2,000 ft/sec. However, their food-bearing capacity without undue settlement is high, 2
generally in excess of 6,000 lb/f t even for materials directly ct the ground surface. Commonly 10,000 to 12,000 lb/f t are measured. That high iood-becring ccpacity of the flood deposits is attributed to the point-to-point contact between cobbles and pebbles and the subsequent interstitial filling by finer-grained sediments es the floMwater velocities slockened.
Thus, loads cre supported on columns of cobbles and pebbles.
The Ringold Formation generally is soturated and the Pesco Gravels commonly cre dry except locally where tSey are normally saturated for only a few tens of feet of depth at the bcse of the gravel. Consequently the opportunity for reworking and compaction is minimal. Even where large qucntities of water pcss to ground and where vibrational and differential loading occur on undisturbed ground, settlement hos been negligible. Differentici compaction on undisturbed ground is of minimo! concern.
b80hl1 3-1I TERA CORPCRATION
Due to the prevailing dry environment and cs long cs the potentially liquefiable sediments remain confined, the likelihood of liquefcetion cf Honford sediments is very remote. The sediments lying at and necr the ground surface (the Pcsco Gravels and uppermost Ringold Formation beds) to depths of I10 feet have been compared to the rcnge of gradotion of liqueficble soils. Most materials had a high relative density (were compcct), o cocrse grain size cnd good size gradativas. The silts cnd clays that lie below the ground water table were of high plasticity but "cre also insensitive and exhibit high shear strength in excese of 3 to S tons per square foot." This is, in part, the result of deposition in calcium-rich environment, some resulting cementation, compaction under high load, and maintencnce of the calcium-rich environment for millions of years.
The deposited clays were not changed to forms licble to liquefaction in a chemically changed environment (WPPSS,1973).
Some sends in the P,ingold Formatien are elecn, uncemented and well sorted.
Upon penetretion by wells (a face toward which to flow), the scnd rises in the wells. This is not liquefaction in the sense of recrrangement of particles to occupy a lesser volume; it is controlled by oppropriate confinement of the sand.
The extraction and recharge of ground wcter in the region and site vicinity is not expected to hcve any mecsurable effect upon stcbility conditions. The lower two feet of the Pcsco gravels are saturated. The Pasco gravel was deposited by the last of a series of great floods resulting from the rcpid draining of large glacici lakes. The materials hcve a dry density of obout 130 pounds per cubic foot, and exhibit high shear strength and low compressibility characteristics.
Actual or potential problems related to surface or subsurface subsidence uplif t or collapse have been assessed. The cbandoned Rattlesnake Hills gas field is located on the northecst slope of the Rattlesnake Hills, Benton County, Wcshing-ton, opproximately 12 miles west-southwest of the proposed site. The rature of this withdrawc! cnd the distance from the proposed site indicate that it presents no problem to stability of the site.
3U3Ub10 3-12 TERA CORPCRATlON
Studies by R.E. Brown (1969) and Tillson (1970) suggest that the Pasco Basin is continuing to subside. Brown indicates rates of one foot in 5,000 yecrs to one foot in 1,000 years and Tillson determined an average rate of one mm per year from ovailable leveling dato. Although these analyses are not positive indicators of continued basining, they do provide dato which p.eclude regional worping as a problem at the proposed site.
la9l)l_175 3-13 IERA CORPCRAIiCN
4.0 SEISMOLOGY While the detailed elements of the seismic risk cssessment cre discussed in Section 5.0, the historical seismic record is of such significance thct it is dis-cussed separately below.
A complete evoluction of the historical record is the keystone to the risk assess-ment because of the importent time and sp9ic! distribution informction it con-tains. With regard to time, the record provides detailed historical earthquake frequency information that con best be represented by the relationship, log N =
o-bM. Further, the spatic! distribution of ecrthquckes cround the site con of ten be used to delineate seismic source regions within which earthquakes have common characteristics.
The historical record for the Pacific Northwest is based predominently upon ecrthqucke intensity reports.
Instrumental ecrthqucke information began to cccumulate beginning in 1906 when a seismograph wcs installed in Seattle.
Although additional instrumentation wcs instclled in the intervening yecrs, signi-ficant improvement in coverage was not achieved until 1962, when instrumenta-tion for the 'Norld-Wide Network was introduced. Finclly, severci local seismic or microseismic arrays have been deployed within the test decade, resulting in excellent coverage of the nearby Hcnford Reservation and the Puget Sound crec.
The earthquake data that have been collected for statistical anolysis cnd incorporation into the seismic risk cssessment include all reported or recorded ecrthquakes from 1833-1977. Data from the World-Wide Network, the University of Washington, Puget Sound and Hanford Networks, and the Dominien Observa-tory in British Columbia were collected and evaluated, os were the historicci intensity reports (Rasmussen,1967). Much of this data has been compiled and mode publicly ovcilcble by the Wcshington Public Power Supply System.
Because of the spatial cnd temporal varicbility of coverage, we have directed particular attention to the stability of earthquake statistics. This effort, which is described in detail below, results in a most occurate description of earthqucke recurrence.
.o<
i::.9.s U,10 4-l TERA CCRPORATION
SEISMICITY The spatici variability of earthqvckes around the site is illustrated in cn earth-qvcke epicenter map, Figure 4-l.
Note that there cre obvious zones of seismicity that are clearly unique in their occurrence of earthqvckes. Most dramatic among these zones is the Puget Sound crec. This crea is unique not only because of the frequency of earthquckes, but also their magnitude distribution (there have been several in the 6-7 rcnge) and their depth (roughly 60 km compared to less then 20 km elsewhere). The crea is unique in other geological cnd geophysical respects. For example, the Puget Sound crea is a significent gravity anomaly, with Seattle being at the center with cn isostatic cnomaly of -93 mgal.
There is, therefore, a firm gecphysical basis for identification of the Puget Sound crec as a unique ecrthquake source region.
The seismicity mcp further suggests that a similar portioning of source regions is cppropriate elsewhere. For example, the identification of a Cascede Mountcin Rcnge source region is not only intuitive!y cppealing, but is further supported by the physiography, surface geology, and the historical seismic record.
South of the site is another concentration of seismic activity in the Milton, Oregon-Walla Walla, Washington crec.
As with the other concentrations of seismicity, there is a geophysical explanation for this localized seismicity. The known surface structure in this crec, and elsewhere in the northwest, is generally a manifestation of shcIlow stresses (irrigation potterns, erosional crusto!
unloading, etc.) while the more significant deep-sected tectonic stresses have no surface mcnifestation. For these reasons we emphasize the indirect results of geophysical gravity cnd aeromagnetic reconnaisscnce.
SITE EFFECTS Because of the sporce seismic instrumentation coverage discussed earlier, the historical record of significent site effects ccn only be inferred from isoseismals.
The seismic network that covers the Richland creo has operated for only eight years and surrounding instrumentation is very spcrse resulting in very little VdbbD 4-2 TERA CORPORAilCN
i V,
w" s.
p-
~
e J
e t.
p p
\\
(..'r s;
\\
L e
t s
s
~~
s
((
.a.
\\.
A I
7-T 7
ur
. t /
- N,./
w
.~m,
( -
\\
L,j '
c
.c h
\\
~
( Wd/ s
-w i v ei. u e.
. s s, w
~
sx e,,
./.
7 m
.N '-
, ~g
%. ).
A
- n,... &,.. c..
~r
....n................._.......... -...,... - -. - - -. -,
/
-n
,. g g
- /.g Wm j
s o.
< &>.m p v h. o=,.
J.,.
.a N
r.;g.
s.
s... ~.... ?...
m
~5 % A t[f %.,,g.*.* * * *
. ' ^ ^
?.
...%;4. l 3.
9' N
~
.c de.
is.
ty. c..
..,6 3
p&
O ca *' g &;.
j.
v 4
c.3_.
- i.
4 p
P t...
i o
s
s s,(
....s g
A.s ~'.
e
~,
'(
y
.e
.~
.-. f.
? g y 7..-
, q~
~
9 W
s s
e.
1 f
g
~
ew
}'
{
._ }
(
r,,
I'
-r unt=snv w.anux
' I1 17 FIGURE 4-1 k_ '
PACIFIC NORTHWEST SElSMICITY 2 v.,,
(FROM WPPSS,1977)
.cc,-
2 it
- >3 t.: n <>J..* t tj TERA CORPCRATION
~
- "~.': 0 0.
instrumental data.
Examination of isoseismal mcps in the historical record indicates (Tcble 4-l) that c er c 150 year intervel, the site has experienced one MM site intensity VI and perhcps cs mcny cs nine MM site intensity IV earth-quckes. While this data must be used qualitatively since both the historical record end isoseismcl maps may be incomplete, the results serve to suggest a relatively low risk at the site. This is because first, the Icrge historical ecrth-quckes have occurred in distent source regions and second, the seismicity of the nearby source regions is such that large earthouckes infrecuentiv, if ever. occur.
In the following sections we analyze the hivarical seismic record stctistically to provide a quantitative bcsis for these statements.
SEISMIC SOURCE REGIONS As discussed ecrlier, various geologic, geophysical cnd seismological data suggest on intuitively cppealing seismic zonction of the crea cround the site. Figure h-2 presents the source zones considered to be cppropriate for this cnclysis; the superposed seismicity demonstrates the relationship between the historicci seismic record and the selected source regions, but gravity, aeromagnetic cnd geologie data were also employed in their selection.
For exemple, Figure 4-3 presents the physiographic zonction of the region and Figure 4-4 shows how surface geology results in a similar zonction. As discussed in Section 4, these are important data in our essessment of the seismic zonction of the region. However, because major earthquakes are a result of deep-seated stresses, we rely most heavily on the results of geophysical grcvity surveys.
Because a gravity enomaly is indicative of a stress differential, the anomalous crea con, therefore, be readily associated with a ecpobility for ecrthquakes. In Figure 4-5 we present a compcrison between the best ovcilable Bouquer Gravity mcp and the seismicity. Note first that there is a broad relationship between the gravity anomalies and the geologic and physiogrcphic zones, thus recsoncbly suggesting a correlation between deep-seated and surface structure. Also note that the concentrations of seismicity occur clong gravity gradients.
This cssociation of seismicity with gravity gradients is not only intuitively appecling, but also builds on the research and observations of Kane (1977) who hcs noted such correlations at seven major zones of seismicity elsewhere in the United States.
t.$. M7 6-3 IERA CORPORAilCN
TABLE 4-1 HISTORICAL RECORD OF MM SITE INTENSIT!ES (ADAPTED FROM WPPSS,1976)
Epicentral Maximum Date Location intensity Site Intensitv 14 December 1872 Lake Chelen Arec Vill VI
.I November 1918 46.7 N I 19.5 W V-VI IV-V 15 July 1936 46.0 N I 18.5 W VII IV 23 April 1943 47.8 N 120.6 W V
IV 29 April 1945 47.8 N 121.8 W Vil IV 14 Februcry 1946 47.5 N 122.5 W Vil IV 13 April 1949 47.3 N 122.5 W Vill IV
' August 1959 47.5 N 120.0 W VI IV 17 August 1959 44.5 N Ii1.1 W X
IV 29 April 1965 47.4 N 122.4 W Vill IV d$ bl.c1'd TERA CCRPCRAilCN
s/,
i w.-
~._
(~
(
., s t
I
{
(
-=
\\
(
\\
k L,
,,p g
i a
s w
(N.
%Nr..
.s
~:
\\~
s
~
s Tn
~
)
g,l j @,
- N *'
y o-u 6
C'7 r f
e y,
- M. >,
V
$g[M '
c-s
%~
s.
l, y
, L
\\
I,"
\\
\\
Q WQQ 4
~.
e vca%c-
'; t z
s.
u.
s.
e./
~
/
f%..L /
i
=
m
?&).
J A
J---------
. y:.
... a A%.*- 1.(
~
e.
s v{-
os 44
~5
'; % %y %3.. d,.
.~
?.[yMs;,,g."
.)*
4' e
%E
,* ' ' ' 3 O -
~
e m 4 It't
.Y M.~.
- f
'T s..g
,'34 A
p".9' 9. v.*
~.
n#' g O,,~
.e-
,o s
4
~
),
..y_
4%.
k o
e i.,.
s_\\.
i(R<g t
4 5
,!, x)
A.2
'.7 n
SITE m.
,eg" 1 2. * '. -
- r.. r.._._
3 m.
~p 6Y e
{
a 4
so..
=
3
?'
r s
g
.r e
3
(-
C
__}
(
y INTENSITY MAGMITUK f
h 211 dN7 FQ [,
~'%
FIGURE /d2 '
S2viii A2s a 2 in a2s SEISMIC SOURCE REGIONS USECf "
'^
2 mi a24 IN THE SEISMIC RISK ANALYSIS
- c i 1,.3.q
~~v 2iv
.>3 TERA CORPORATION
5 M
e',
\\
s, i
tj}
~%
2;m. -M
-[-
g
.. J d
. Q -~ -
l
,-=',..;
y 1
w;),
(
w.
e f
n p
p..
~Q_-
?.
fy f
t-
); h
'~ f j
(
hw_
nn -
, ~ ~
4 n
e l
gag ::"'.* T2 g
f~
.z
\\
1 3
8 j
./
3 i
smasse g
. c.,.
l I
1r p
%n t
i' l-p.s
, All x
A
_. Q -
g N
=W~
. s e a sse * * ' * ",,,,.... **
=
g 1
e-
)
j/
i g.~
R mi.
_7,
s.
(
m p -
1 s
c
,,.e A a'*-
i
..:,g^
z ma r.sm t,-.
ef g.
, =H
\\
/'~,
4-~
b
\\
- Q '"W~
z J
?!
t
\\
- = =
\\
m., -'
w g
)
[,-
....e'
-l s-$"
~I a
\\
/-
> ~,.
a
(.
)
s
- 4 i
- /
1 t
~.
\\
.c s
a manuaa 6
s l
I-
~
t, 1
's s
A-
~
~
. m MO um FIGURE 4-3 PHYSIOGRAPHIC PR VINCES OF THE PACIFIC NORTHWEST h
ROM WPPSS,1977) p CORPOdBON
.e e.
('
f',
s b
h
"\\
N:sw.y
' -c ' _
^ -
7
{~
,~,,
~~
e g
s
./
s l}
l m,
l
/
~~
2.* c
~
17.
Ll s
- s. p-l".
+ 5, E.
5 1
4
.. \\. '
g
.I
, j t.
6,4 N
(
[
o j *i..
q
-i O h dh
~
m:
[t*.aw.3
\\
.A '
l 4 -~
y..................
_.-,,.;g /,
.f 5
f I
l
,.s. -
- \\
f.
1 Y
L kO.
d
. g w
_+
3-m f.,..
g t
~
_>a,,.... -
/g_ _ _ _ _
7-
_1 t
I f.
m...
unu
' 3.
-3
/
/ s.):"_ i 9--
s, i
,g,
.. -' ~f 3
3~
h.
\\.,,
.t-s ame. m.. ay
- p 3- %.
" d.
l
.I i j
/
t.
a
,/
g, owc isounams A
-~
FIGURE 4-3 PHYSIOGRAPHIC PROVINCES OF THE PACIFIC NORTHWEST h
(FROM WPPSS,1977)
+
TERA CCRPORATION
e Os %
f
}
'd I
t.gft
' './ ' ' :,....... /
I't p
?
s1<
q _,j.
.g r
o.
ye
= = -
s
.d'.
g1 i
t.
}
/
l-l.
\\;-
j.,V /
2 j
f
}
n
\\
/
/,:: cast a
- q. f r,. -
I cui.ca i
i-g,a. -.g,
o
..m - -
,p o
.I G ftv1' Al.Lpi ii-fs k
\\
d Ei CPvst u k '. I
/ "*"*
GCL7 s
i i
W hi
~
l s
1
- f, eCL T (
- 1. s p
- .Y 6
s p sllR x.l 1
\\
(b-A w
'd r y -
Z.
m _
l '* 1. } //
~~' h5\\ a +? & *$ ' '.e
\\
-s 1'
L
- /' 9 k 12s
.\\
r V
~- W /'
h~
%I4 a
j j'
N.---g l#
.g
-i c.
..:: A.
~
E, h f.k,,['.['d.
-Ad/
i; ^
\\.
e" b :b
%~,\\,. M.,O I
d i
s N'+'
%b 7
- ~,
- >CL aes ve0visect i
s, l
g-j (^
e r
.a r
,, 4 a*
.i
\\\\
s..<.,
~
- s
~.,. -. s
'.v. 3 '4
~
f i
- t..
~
m
. w'S k
w g,
Se.aTANbuser
-d+ %s%
m
%.1
- g r,+3 _;c.
- y. %)
a' w< -... ~ _,. 2
- .,, e. -
.e.~.
v%
~
.,lc
~
l 1
n erw r
. v _ A (..,
- y. s,M +
w g
- &Q 3..
C-M.*:@Q.u'r g,_ w)'
, cr anM c
1 i
M-g.f,, J,p/,M"%.
- -s f f
.s-1"'
$q
, s.e.
4.
125 ms r.fl ade
,p.
/
..w r
4-
- s
/ -'
.o V.. g. l
. 4.e 1
.. g-
-,i d.
/.V o I L
4 3
43,
/i-~
we
-~
x i
c.
m i A ~-
m
- y FIGURE 4-4 L w... 31mn
-r GEOLOGIC PROVINCES OF THE PACIFIC NORTHWEST
.. 0 h
+
h TERA CORDORATION
f
\\
.4
- N;
- (L>u!
V
,D l\\
h.3
=
VD wi 0 1p s,M
'5 E
~
-l p
N u
s
'1 9.
uj
~. - -
~
( '% ftj )L S Ay
~
-Q.p%,w.2.
W
\\
n-e
~
p
,(,
.u J
%r
-- (
an i
-r e
.S i
6
's I,
t u._
,p
=
7-
,f -
~ ?j
,i.,
L
-- 1 Ims :: :
ij..
6..
E !!!
I:5' ? O S ?.3f f. 9. 3 1 ?
C-L -.
p-
.s.
- .c...".
=. = :::::::
E--
- : :n :..::--~~ --~
g!j ii i :
E5 3EE Eat 53 5 FIGURE 4-5
%w:O.w o
.s BOUGUER GRAVI FY AND EPICENTERS IN THE PACIFIC NORTHWEST (FROM WPPSS,1977)
TERA CORPORATION
In consideration of these dato, we therefore judge that a most recsoncble portioning of seismicity in the region is es presented in Figure 4-2.
EARTHGUAKE STATISTICS in this sectio a, we onclyze the seismicity in each of the source zones in Figure 4-2 to estimate the recurrence relationship for earthquakes in ecch zone. This relationship is one of the keystones to a proucbolistic seismic risk cssessment, c,w) is urvally characterized by the two parameters, a and b, in log N = c - bM where N is the cumulative annual number of ecrthquckes greater than ecrthqucke magnitude M.
Where we require conversion of an epicentrol MM intensity to mcgnitude, we employ the correlation presented by Toppozcdc (1975):
M = !.85 + 0.491 In our cnclysis of the recurrence reictionship, we direct porticulcr cttention to the sicbility of the activity rate, b, of ecrthquckes. The pcrameter b, of the two parameters, is the most significant beccuse it logarithmically determines the recurrence of large earthquakes.
Instebility in b con crise when the subinterval of time within which earthquakes are considered is either too long, in which case the data cre incompletely re-ported, or is too short, such that the mean activity rate is unstable. The overall cpprocch in this report is to calculate, for each source region, the overage annual numtur of earthquakes in ; specified magnitude interval, and to then sum these over magnitude to provide en estimate of the cumulative number of ecrthquckes greater than a given magnitude.
A linecr least squares fit of log N = c - bM results in best estimates of a and b.
Within this procedure, it is the estimates of the average annual number of events in a specified magnitude interval that can introduce significant errors in the ecrthquake parameters.
We use the method developed by Stepp (1974) to i>9 M' N minimize these errors.
k 4-5
+
TERA CORPORATION
Following Stepp, we first group ecrthquakes in each source zone into magnitudes intervals of 0.5. For each mcgnitude interval, we calculate the overage annuci number of ecrthquckes, A for several subintervals of time, T.
The standcrd deviction of a is given by A
o T
must behave like T-N if, as we have assumed, A is constant.
and therefore a Plotting the subinterval values of a ellows one to readily determine over wnct interval c begins to behave like T-D. This is, therefore, the subintervci of date that is long enough to give o good estimate of the mean cctivity rate, but short enough that it does not include intervals with incomplete data.
The results of this onelysis are presented in Table 4-2 where we indicate that c and b values that cre developed from the regression cr. clysis, cs well as the coefficient of determination for the fit.
Brecuse prelimincry analysis showed that the seismic risk at the site was domi-r:ated by the risk due to local ecrthquakes in source zone five, we use this zone to illustrate the detcils of our statisticci analysis and to further -loborate on the technique.
It is cpparent from Figure 4-2 that zone five is relatively aseismic compared to surrounding cource zones. Because of the significance of this zone end its mccro-cseismicity we rely heavily on date from the University of Wcshington microseismic array centered on the Henford Reservation.
Applying the sichility techniques described above to the historical record of earthquakes in zone five results in the data indicated in Figure 4-6.
Figure 4-7 indicates how the annuct activity rates for each mcgnitude interval cre defined, and Table 4-3 summarizes ^hese rates cnd the interval of time appropricte for each mcgnitude interval.
Because these results are supported by only 48 ecrthquakes in the magnitude range of 3.5-5.5, we turn to supplementary data from the microseismic network.
(.9 h b b 4-6 TERA CORPORAT1CN
TABLE 4-2 EARTHQUAKE PARAMETERS FOR EACH SOURCE ZONE Seismic Coefficient of Zone loc N = a-bM Determination Site ID 2
(Fic 4-2)
Descriptor c
b r
1 Puget Sound 4.30 1.05 0.99 2
S. Cascades 3.50 0.93 0.92 3
N. Cascades 3.50 0.93 0.92 4
Milton-Freewater 3.31 1.00 0.87 5
Site Source Zone 4.5 ';
!.16 0.99 bSiil.ht) 4-7 TERA CORoCRATION
xc -
Log o
- 4. M - 1.16
- g 10C 10 L
eo L
a c-
?
2 v. _
~
e e
C.1 8'
i i
2.0 3.0 4.0 5.c 6.C
. usum uminema mcm..ie u.m e usiteini uco" MGNITUDE
- es
. n' U ' ' d r u *-
FIGURE 4-6
~.,
RECURRENCE RELATION FOR SEISMIC ZONE 5 3'
TERA CORDCr[ATION
1C~
O g
- 3. C < ft ; :.!
.5_
e
.6-O A
4-3.5 < M < 4.0
- e Ae C
O N
.4
~
5 4.5 < V.
< 5.
O g
O O
O I
O
.ic-o
.s-
~.06-o 3.0 < y 5 3,5 e
3.5 < w s 4.0 A
4.0 < M s 4.5
. 04 -
0 4.5<M: 5.c
./
,. s
. 02 -
.01 4
a 6 i**
1 2
4 6 '
8' ' 1'0 2'0 40 60 80 10 T
- time in years bObIbb FIGURE 4-7 STANDARD DEVIATION OF THE ANNUAL NUMBER OF HISTORICAL EARTHQUAKES IN ZONE 5 s
e IERA CORDORATION
TABLE 4-3 RESULTS OF EARTHOUAKE STATISTICS ANALYSIS FOR ZONE 5 (SITE SOURCE ZONE)
Magnitude Time Annual Interval Interval Rate M
Dates Duration 3.5-4.0 1945-1969 25 0.508 4.0-4.5 1910-1969 60 0.204 4.5-5.0 1890-1969 100 0.092 "v3 [.). db 4-8 IERA CORPCRADON
This network has been operational since 1969 and since then has been relicbly reporting earthquakes over a large cree that includes much of zone five.
Accounting for the coverage crec, o parallel enolysis of the micro-earthquakes located in zone five results in the remaining indicated data in Figure 4-6. A best fit to the data of the form, log N = a - bM is obtained by the values a = 4.54 cnd b = 1.16 (Tcble 4-2).
Because of the significance of this particulcr seismic zone, we account for the uncertainty in this relationship through the weighting of results from a para-metric survey on this, and other, important parameters. The next section, RESULTS, describes these parameters, their voriction, and the overall results.
t:S.du0 4-9 TERA CORPORATION
5.0 CALCULATIONS AND RESULTS in the previous sections, we hcve discussed the advcntages of seismic risk analysis relative to deterministic cpprocches and presented o state-of-the-crt calculational approach to probabolistic seismic risk. We have also described the tectonic structure and seismicity of the crea cround Richicnd. In this section, we apply these concepts cnd data to a seismic risk analysis for the Exxon site.
The detailed input to the calculational model is described below, followed by a presentation of the results.
INPUT As described in Section 2.0, Seismic Risk Methodology, the input to a probcbil-istic seismic risk assessment is comprised of ecrthqucke occurrence frequency relations, ettenuation functions and a specification of local source regions.
Because risk assessment calculations are very sensitive to the particular com-position of the input, we consulted with several knowledgeable seismologists during the prepcration of input for the Exxon facility analysis. Major suggestions in this regard were mcde by Professor S. W. Smith.
Source Recions cnd Ecrthcucke Statistics The source regions, Figure 4-2, considered most oppropriate for a seismic risk analysis were earlier described in Section 4.0. Preliminary calculations indicate that earthquakes in the source region containing the site dominate the risk at the site, and thus pcrticulcr attention was directed to the validity of this region cnd its statistics. The basis for identification of this and the other source zones is physiogrophy, surface geology, Bouquer Gravity datc, cnd historical macro-and microseismicity. The other zones include the Puget Sound crea, the Northern cnd Southern Cascades, and the Milton-Freewater area.
The recurrence relations for each zone, modelled cs log N = a - bM, were also developed in Section 4.0.
In Table 5-1, we summcrize these results and also present the mcximum mcgnitude ecrthquakes for each zone. The magnitude G $ 3 D 1.
5-1 TERA CORPCRAT!ON
TABLE 5-1 RECURRENCE RELATIONS USED IN THE ANALYSIS Largest N=N e B(M-M,)
Historical Seismic Zone o
Earthoucl<e JD, Descriotor 4
B g
%gx 1
Puget Sound I.27 2.42 4.0 7.5 7.I 2
S. Ccscades 0.60 2.14 4.0 5.7 Vil 3
N. Cescodes 0.60 2.I4 4.0 7.0
\\ 111 4
Milton-Fre-water 0.20 2.30 4.0 6.0 Vll 5
Site Source Zone 0.80 2.67 4.0 5.0 V
ijfDIS 5-2 TERA CORPORATICN
indicated is our best estimate based on the historical record and our interpre-totion of the regional tectonics. Our maximum mognitude is generally one-hcif magnitude icrger then the forgest historical magnitude, which is clso listed in Tcble 5-1. We recognize the uncertainty in this parameter, and we account for the uncertainty through a parametric survey, as described below.
ATTENUATION One of tne keystones to any seismic onclysis is the specification of decoy of peck accelerotion with distance from the earthqucke. Credible attenuation relations have been difficult to develop for two reasons. First, the large sectter in the data mckes a deterministic evoluotion very difficult and second, the dato are very sparse in the near-field, thus allowing for a vcriety of interpretations.
Because of its pcrticulcr significance, a very careful re-evoluation of c!! the data was performed in order to ensure state-of-the-crt interpretation and mcximum credibility.
The overall cpprocch to development of cn attenuotion re!ction for this cnolysis is cs follows:
1)
Use oppropriate dato in the rcnge 20-100 km to estimate the far-field attenuation.
2)
Focus on the acceleration data at = l0 km to fix the trend in the necr-field.
3)
Rely on all available dato points at ranges less than 10 km to establish the very necr-field accelerations.
Peck occeleration data for development of the ottenuation relation were col-lected from the following sources:
e Brune, et al.,1977 Golcnopoulos and Drakopoulos,1974 e
Hcnks cnd Jchnson,1976 o
7, if;G OO 5-3 TERA CORDCRATION
e Boore, et al. to be published,1977 e
USGS Circulars; 672,1972 717-D,1976 713,1974 736-A, I?76 717-A,1975 736-B,1976 717-8, 1975 736-C,1977 717-C,1976 736-D, !?77 To introduce the ottenuation relationships used in this onclysis, consider Figure 5-1, which shows the data in the magnitude rcnge 6.0 to 7.0.
The emphcsis in this figt,re is on the 20-100 km ranges, but the ovci!cble very near-field daic (less than 10 km) are also plotted. In this plot, range is usually the distance to the necrest point on the fault or, the distance to the one of princioci energy release, if this con be determined. The dato cre for both rock and soil sites since there cppears to be no statisticolly valid recson for their discriminction (Boore, et al.,1978). The relations cre, therefcre, for a typical site. Superimposed on the dato are two citernative attenuation relations that represent equally picusible necr-field cecelerations. We use both of these relationships in our analysis. The scatter about these mean relationships is best characterized by a log normal one standard deviction of 0.45. This fits the data reasoncbly well cnd is consistent with the dispersion calculated by other investigators (e.g., Donovan, 1974).
These attenuation relations, while based on the most current data set ovcilable, do not represent a radical depcrture from previous recommendations.
The acceleration spanned by these two relctionships, in foet, cover all previously published attenuation relations.
This overall consistency serves to further validate our recommended relations. The functional form of these relationships and their grcphical representation are presented in Figures 5-2 and 5-3.
- 12. ) T. b'l 5-4
' ERA CCRPCRATiCN
2.0 -
O Pocoimo 1.0 -
0.8 -
C.6 -
KoW
- Lee:s 0.4 -
e C
Meogx 0
O mm
^
C.2 -
s3 O
OO g C.I 2 0.06 O
~
O
- 0.06 -
"A C-n"
'_ c.0 -
o' -
C e
A 3
h
! 0.02 O
O x<
E 0.01 -
OKS-0.006 -
0.004 -
02 21 73 Pt. M 4 A 04-09-68 Borrego Mountein 0 02 09 7i Sm Fernem 0#0 '
i i
i i i Il lt I
t i I I IIii l.0 2.0 4.0 8.0 10 20 W
60 90 100 200 RANCE (Km)
FIGURE 5-1
. < r.;
- 1,!.,
DATA BASE FOR THE MAGNITUDE 6.5 ATTENUATION RELATION k+
TERA CORPORATION
ATTENUATICh I (2.92
.58" *.04v2)M (r + 12)*i*7*
A = 0.193 e
,,,i i
i iii<
i
! ! Ii,il i
i i iiii t
i I : i i
!l i' l
1 l l!'!
l li l
i I i i' j'
j l
lli i
i
' ; j ::
i i lIlI'i i
j l 7.' 5 ' ~'l l
i i'
i 1.0 x x 7
- 6. E' N
5.5 N
'\\
E
. N N:
N i
NNN N
i
%'N N
- iik'i,;-
c.1 s s
\\ \\
\\
xs s '
t
\\ \\.\\ i i.\\
i
,\\\\ hi i N i
i: i i
I I ili I M Klll\\
l i
ll un i
i AN;\\ii 3
\\
l l
l l
l D'
l 0.01 1.0 10.0 100 10 %
,"'s
-,s r
n
' l.
9 e
FIGURE 5-2 b., s r' e. >,J '. >
ATTEN'JATION P,EL ATIONSHIPS TERA CORPORATION
AMENUATION 2
(.277 +.Ci;M)" ( r + 10 ).1. 75 A = 6.7 e T
T j
i f I $
l I
f j
r I ! !I'I i
i i
i i l i i !
I Jjl l'!i i i llI!'
i i
i i
,t i
l I
I l ll l l!i l
! i 4 l1 ii l
l ii' j
j i
I l
l jI; j
'llii t
i 7.5 1.0 x
6*x x
N' N
e o.a N
.x i N i
i i
N.I \\
'N i
i, E
4.5 NN\\\\
N N'NN i,
i g
! ' j,,,
t I !:
ili i
! i ii i
i x
s c.1 s
N ', N
\\
\\
s s
\\
N N N.
x r
, i i
I i
I ii,
'\\ :\\ ! \\
- \\
l i
i l l ii r i \\hi h l\\
t i
l l
i l
KN{ fk '
i i
\\ '\\'N j I!i l
, i
\\
'l' i
Ni i
i i
c.01 1.C 10.0 10; 1000 7,
/
af m.
P
'i FIGURE 5-3
' '*s
. 'm e ATTENUATION RELATION 5HPS e:
6 3 r ;J ~
g e
TcRA CORPORATION
RESULTS The results were obtained by computer calculations with a risk cnolysis code (McGuire,19766) that is based on the work of Cornell (1968). The basis for this coprocch was summarized in Section 2.0.
As described in Section 2.0, the computer code calculates, f ir circular sectors within each so>rce region at the si*e. the expected number of earthquckes per year cousing accelerations greate that, a specified ceceleration cnd this is done for each source region. The expected numbers cre summed for ecch region, cnd the resulting risk calculated from risk : 1.d - exp(- expected number per year).
The return period essociated with ine specified accelerotion is the reciprocci of the expected number per year. It follows from the definition of return period that accelerations with a porticuicr re+ urn period have a 63 percent probcbility of being exceeded withir the return period.
Our estimate of the seismic risk represents the weighted results from seven individual calculations. The seven cciculations represent a bcse case and our perturbations of input parameters cbout this base. The perturbations are weight-ed and combined by subjective estimates of their probcbility of occurrence end thus the combination con be considered a synthesis of sensitivity cnolyses.
The parameters that are considered uncertain and which are included in the risk cnolysis are the maximum ecrthquake in ecch source zone, the magnitude of the dato dispersion cbout ' ie mean acceleration attenuation relationship cnd the recurrence relation for the spurce region containing the site. The verictions of these parameters reprew.,t tne overall uncertainties in:
Our obility to define the strain energy limit of faults e
around the si'e, The correlation between ecrthquake mcgnitude and inten-o
- I'
@ fUj8 S-5 TERA CORPCRATICN
e The shape and level of the attenuation relation, and The seismicity of the host seismic zone.
e The bcse ecse is considered to consist of the following input:
Maximum ecrthqvcke equal to our best estimate value e
(Tcble 5-1) e Accelerction one standord deviction dispersion o'
InA =.45 a c d b values of the zone 5 recurrence relation given by e
a = 4. 34, b = 1.16.
Both attenuation relctionships, equally weighted e
Note that the value of the dispersion is based on the results of statistical anci-yses on the dato. For example, McGuire (1974) determined that the dispersion for western data ee responded to a multiplicative factor of 1.67 in the far-field while Esteva and Villaverde (1974) obtained 1.90. The values selected for the the enclysis are bcsed not only on these calculations but also on the fact that the analysis addresses a specific site in a specific tectonic province, thus reducing the variables.
We chorocterize the uncertainty in all these data by first :onsidering that it is roughly 70 percent probable that the maximum ecrthqucke is as specified cbove and that it is 15 percent probable that the mcximum ecrthquake is roughly one-half mcgnitude unit larger cnd ccrrespondingly 15 percent probcble that the maximum earthqucke is roughly one-half magnitude unit smaller. The values of these parameters are presented in Tcble 5-2.
Similarly, we consider that it is 70 percen' w 8; thct the acceleration disper-sion is cs specified cbove and that it is re ' ective..
? percent probcble that the dispersion will be:
gnA =.55 o
cnd bob 1.69 tnA =.35 e
e 5-6
.5-TERA CORDCRATION
TABLE 5-2 DISTRIBUTION OF MAXIMUM EARTHQUAKES USED IN THE ANALYSIS Seismic Zone Mcximum Earthoucke
.lD_
Descriotor lower bound best estimate upoer bound 1
Puget Sound 7.0 7.5 7.7 2
S. Ccsecdes 5.2 5.7 6.2 3
N. Cascades 6.5 7.0 7.5 4
Miiton-Freewater 5.5 6.0 6.5 5
Site Source Region 5.0 5.0
.5
~ 63.>I.70 5-7 TERA CORDCRATION
Finally, we characterize the uncertainty in the zone five recurrence relation by assigning a 70 percent subjective probability to our best estimate a and b values, and assigning 15 percent to the respective values e
c = 3.45 b = 0.94 e
a = 3.99 b = 0.93 These values represent emphasis of respectively the microseismic data and the historical record.
The results from these seven sets of pcrameters have been combined to produce our best estimate in accordance with the probability of the combination. These results cre presented in Figure 5-4, Also shown in this figure is our estimcte of the one stcndard devictions about our best estimate results. These were derived by assigning equal weights to the three mere conservative and less conservctive sets of parameters respectively.
We have also estimated the duration of strong motion essociated with these risk curves. We first note that for return periods of interest the acceleration-risk curves are domincted by the effects of small, nearby ecrthquakes. The strong motion duration from these postulated earthquakes is very difficult to estimate in that it depends on several pcrameters, including the hypocentral depth, the stress drop, the czimuth, and the rupture chorecteristics. We attempt to cover the uncertainty in o>r duration estimates by considering data from two small ecrthquckes whose properties partly soon the cbove pcrameters.
We have examined the duration function (duration cs a function percentage of peck acceleration and distence) for the following two ecrthquakes:
April 8,1961 MS.6 at Hollister, Calif.
June 27,1956 M5.6 at Pcrkfield, Calif.
,g.r....
b.; d.._..
5-8 TERA CORPORATiCN
10,000
/
/
.f
/
I i
i
/
/
i
/
/
e t
/
i l
/
I /
/'
I i /
!/
,/
/
/
l l/
/
l
't
/
/
/
\\
/.
/
/
I
/
I,000 i
/
/
J J
/
/
/
/
s i
/
/
i
/
/
n E
I
/
/
.i
/
/ :/
m r/
/
5
'/
/
a' p
p C
/
C.
/
I
/
,e 100
/
/
il
~
l //
L'
//
I//
///
ll l III i
/
/
i 10 '
i i
i i
i i
i
\\.-,
l i
I l
l l
l 1
I l
I s
0 5
10 15 20 25 30 s
PEAK HORIZONTAL ACCELERATION (%g) t>Ob17f FIGURE 5-4 EARTHQUAKE RISK AT THE EXXON FACILITY
+
b TERA CORPORAilCN
at distences ranging from less than me kilometer to over 20 kilometers, cnd from this we find thct a reasoncble estimate of duration for the Exxon f acility can be taken as follows:
% of oeck ceceleration cycles seconds 80 3
0.40 50 25 3.00 5
70 7.00 RESPONSE SPECTRUM These results define the peak horizontal ccceleration at the facility for vcrious return periods. We have also determined cn cppropriate response spectrum f or the site since some structures cnd equipment at the Exxon facility have sufficiently low fundcmental frequencies to experience spectral cmplificction of the ground motion.
The response spectrum for the site clearly cennot be developed in essociation with a specific earthqucke; our return period ceceler-otions represent en integrated effect at the site from en extraordincry variety of ecrthquckes cnd the response spectrum must reflect this. Accordingly, we judge that, the shcoe of the spectrum should be similcr to the Newmcrk-Blume statistically-based spectra from which Regulatory Guide 1.60 cvolved. Accord-ingly, it is our judgment that the mecn response spectrum for alluvium presented in WASH 1255 is the most cppropriate for analysis of the Exxon facility.
In summary we have combined the best cvoilcble input data with the most credible tools of seismic risk cnolysis to determine the return period of occeleration et the Exxon facility. The results, shown :n Figure 5-4 cecount for all the significant uncertainties in the input. Response spectral accelerations ccn be determined by secling the mean response spectrum in WASH 1255 to the desired peck acceleration.
b$uI_73 5-9 TERA CCRPCRAilCN
6.0 BIBLIOGRAPHY Algermissen, S.T.,1969. Seismic Risk Studies in the United States. 4th World Conference on Earthqucke Engineering, V.1, p.14.
Alger. nissen, S.T.,1974. Seismic Risk Studies in the United States. 5th World Conference on Earthqucke Engineering.
Algermissen, S.T. cnd Perkins, D.M.,1976. A Probcbolistic Estimate of Mexi-mum Acceleration in Rock in the Contiguous United States: USGS Open File Report 76-416.
Binghem, J.'N., Londquist, C.L. and Baltz, E.H.,1970, Geologic Investigation of Faulting in the Honford Region, Wcshington: U.S. Geol. Survey Open File Report.
Boore, D., 1977. Attenuation of Peck Accelerations, internal TERA report, August 13,1977.
Boore, D., et al.,1978. Estimation of Ground Motion Porometers, USGS Open File Report 78-509.
Brazee, R.J.,1976. An Anclysis of Ecrthquake intensities with Respect to Atte-nuotion, Magnitude and Rote of Recurrence.
Brown, D.J., l958. Subsurface Geology of Hcnford Separctions Arec: General Electric Co., Richland, Washington, HW-16780.
Brown, D.J.,1960, An Eolian Deposit Beneath 200-West Area: General Electric Co., Richland, Washington, Rept. No. HW-67549, I I p.
Brown, D.J.,1962, Geology Underlying Honford Recctor Areas: General Electric Co., Richland, Wcshington, HW-69571, 20 p.
Brown, D.J.,1963, Geology Underlying the 241-AX Tank Farm: General Electric Co., Richland, Washington, Rept. No. HW-79805.
Brown, R.E.,1968, Reported Foulting in the Pasco Basin: Battelle Northwest Laboratory, Report No. BNWL-SA-l704.
Brown, R.E.,1969, Some suggested rates of deformation of the boscits in the Pasco Basin, and their implications: Battelle Northwest Laborotory Report BNWL-SA-2443,10 p.
Brown, R.E.,
1970, Interrelationships of Geologic Formations and Processes Affecting Ecology as Exposed at Rattlesnake Springs, Hanford Project:
Battelle-Northwest Lcboratory, Arid Land Ecology Program, Reprint Series No. 3, BNWL-B-29.
Brown, D.J.,1971, Prelimincry Honford Stratigraphic Framework-Interim Re-g> J+A' -
port: U.S.A.E.C. Report, ARH-2170.
6-1 TERA CORPCRATlCN
Brown, D.J. cnd Brown, R.E.,1958, A Subsurface Aeolien Deposit at the Hanford Project, Washington: Mcnford Atomic Products Operation, Genercl Elec-tric Co., Richland, Washington, Brune, J.N., J. Prince, S. Hertzell, L. Mungic, cnd A. Reyes,1977. "Accoulco Strong Motion Record of October 6,1974,' Abstract with Programs 9, Geol. Soc. Am. 394.
Cornell, C.A.,1968. " Engineering Seismic Risk Analysis," BSSA g(5) p.1583.
Cornell, C.A., and Merz, H.A.,1974. A Seismic Risk Analysis of Boston: Jour.
Structural Div., A.S.C.E., i 10, no. ST 10, Proc. Paper 21617, p. 2027-2043.
Cornell, C.A. and Vcnmarke,1976. The Major Influences on Seismic Risk. 6th World Conference on Ecrthqucke Engineering.
Dames & Moore,1970. Geology cnd Seismology, Proposed Mixed Oxide Fuel Fcbrication Fccility, Richicnd, Washington, for the Jersey Nuclear Com-pony: 8837-001-03, Danes, Z.R., Bonno, M., Brcu, E., Gilhen, W.D., Hof fmen, T.F., Johansen, D.,
Jones, H.H., Malfeit, B., Mcsten, J.,
Teague, G.G.,
1965, Geophysicci investigation of the Southern Puget Sound crec, Wushington:
J. Geophys.
Res, Vol. 70, p. 5573-5580.
Donovan, N.C.,1974. "A Stratistical Evaluation of Strong Motion Data including the February 9,1971, San Ferncndo Earthqucke," World Conf. Earthqucke Eng., 5th, Rome 1973, Proc. Vol. I, p. 1252-1261.
ERDA,1976, Draft EIS: High Performance Fuel Lcboratory, Hanford Reserva-tion, Richland, Washington, ERDA 1550-D.
Esteva, L.,1970. " Seismic Risk cnd Seismic Design Decisions," in Hansen, R.J.,
ed., Seismic Design for Nuclear Power Plants; Cambridge, Massachusetts Inst. Technology Press, p. 142-182.
Esteva, L. and Villaverde, R.,1974. Seismic Risk, Design Spectro, cnd Structural Reliability,5th World Conf. Earthq. Eng.
Evernden, J.F.,1970. " Study of Regional Seismicity and Associated Problems,"
BSSA @(2) 0.393.
Galenopoulus, A.G. and J.C. Drakopoulos,1974.
"A T Phcse Recorded on en Accelerogram," Bull. Seism. Soc. Am., Vol. 64, pp. 717-719.
Grolier, M.J.,1965, Geology of Part Big Bend Area in the Columbia Picteau, Washington: Johns Hopkins Univ., Ph.D. (unpublished).
Gunning, H.C. and White, W.H. (eds),1966, Tectonic History cnd Mineral De-posits of.ne Western Cordillera: Concdion Inst. of Mining and Metallurgy, Spec. V-8.
b$2UtM 6-2 TERA CCRPCRATICN
Gutenberg, B. cnd Richter, C.F.,1956. Earthquake Magnitude, Intensity, Energy, and Acceleration, BSSA V. 32, p.163.
Hanks, T.C. and D.A. Johnson,1976. " Geophysical Assessment of Peak Acceler-otion," Bull. Seism. Soc. Am., Vol. 66, pp. 959-968.
Holden, E.S.,1898. Catalog of Earthquakes on the Pacific Coast, 1769-1897; The Smithsonicn Institute Collection No.1987.
Howell, B.F., and Schultz, T.R.,1975. Attenuation of Modified Mercalli Intensity with Distance from the Epicenter: Seismol. Soc. Amer. Bull. V. 65, p. 651-665.
Hunt, C.B.,1974, Natural Regions of the United States and Canada: W.H. Free-man and Company, San Francisco, 725 p.
Kane,M., 1977.
" Correlation of Eastern Earthquake Centers with Motic/UI-tramatic Scsement Messes" USGS Open File Report,77-134.
King, P.B.,1969, The Teevonics of North America - A Discussion to Accompany the Tectonic Mcp of North Americc: U.S. Geological Survey, Prof. Paper 628, 94 p.
Lui, S.C. and Fogel, L.W.,1972. Earthquake Environment for Physical De:ign: A Statistical Analysis. The Bell System Technicci Journcl, Vol. 51, No. 9, p.
1957.
McAdie, A.G.,1907. Catalog of Ecrthquakes on the Pccific Cocst, 1897-1906.
The Smithsonian Institute Collection No.1721.
McGuire, R.K.,1974. " Seismic Structural Response Risk Analysis, incorporating Peak Response Regressions on Ecrthquake Magnitude and Distence," Massa-chusetts inst. Technology, Dept. Civil Eng., Research Rept. R74-51, 371 pp.
McGuire, R.K.,1976a. "The Use of Intensity Data in Seismic Hczord Analysis."
6th World Conference on Earthquake Engineering.
McGuire, R.K., 1976b.
" FORTRAN Computer Program for Seismic Risk Analyisis," U.S.G.S. Open File Report 76-67.
McGuire, R.K.,1977c. "A Simple Model for Estimating Fourier Amplitude Spec-tra of Hor'zontal Ground Acceleration." Submitted to BSSA.
McGuire, R.K., 1977b. Effects of Uncertainty in Seismicity on Estimates of Seismic Hczcrd for the East Coast of the United States. BSSA VOL. 67 No.
3,p.827.
Newcomb, R.C.,1958, Ringold Formation of Pleistocene Age in Type Locality, the White Bluffs, Washington: Am. Jour. Sci., Vol. 256 D. 328-340.
b3UUU 6-3 TERA CORPORATICN
Newcomb, R.C.,1959, "Some Preliminary Notes on Groundwater in the Columbia River Bescit", Northwest Science, v. 33, no.1, p.1-18.
Newcomb, R.C.,1961, " Storage of Groundwater Behind Subsurface Dams in the Columbia River Bosc!t, Washington, Oregon cnd ldcho", U.S. Geol. Survey Prof. Pcper 383-A,15 p.,12 illus.
Newcomb, R.C.,1962, F'ydraulic injection of Clostic Dikes in the Touchet Beds.
Washington, Oregon and Idcho: Geol. Soc. of the Oregon Country, Vol. 28, No.10.
Newcomb, R.C.,1970, Tectonic Structure of the Main Port of the Bescit of tne Columbia River Group, Washington, Oregon, and Idaho: U.S. Geol. Survey, Misc. Geol. Invest. Mcp l-587.
Newcomb, R.C.,1972, " Quality of the Groundwater in Scsc!t of the Columbic River Group, Washington, Oregon and Idaho", U.S. Geoi. Survey Wcter Supply Pcper 1999-N, 71 pgs.,10 illus.
Newcomb, R.C., and Brown, S.C.,1961, "Evoluction of Bcnk Storage Along the Columbia River Between Richland cnd China Bar, Washington", U.S. Geol.
Survey Woier Supply Pcper 1539-1,13 pgs.,5 illus.
Newcomb, R.C., and Strand, J.R., " Geology and Groundwater Chcrocteristics of the Hanford Reservation of the Atomic Energy Commission, Washington",
U.S. Geol. Survey interdepartmental Report, Hanford Doc. U.S. Geol. Sur-vey WP-8, 265 pgs.,42 illust.,1953. (Now published c:, USGS Professional Paper 717).
Newcomb, R.C., Strand, J.R., and Frank, F.J.,1973, " Geology cnd Groundwater Charceteristics of the Henford Reservation of the Atomic Energy Commis-sion, Washington", U.S. Geol. Survey Prof., Paper 717.
Oliveric, C.,1974. " Seismic Risk Analysis," Earthquake Engineering Resecrch Center Report EERC 74-1.
Page, R.A., D.M. Boore, W.B. Joyner and H.W. Coulter,1972. " Ground Motion Values for Use in the Seismic Design of the Trans-Alcskan Pipeline System" U.S. Geol. Survey Circular 672.
Rasmussen, N.,1967. " Washington State Earthquckes 1840-1965." Bull. Seis.
Soc. Amer. V. 5, (3): 463 p.
Rcymond, J.F. and Tillson, D.D.,1968. Evoluction of a Thick Bcscit Sequence in South Central Wcshington: Battelle Northwest Lcboratory, Report BNWL-776.
Richter, C.F.,1958. Elementary Seismology. W.H. Freeman and Co., Scn Fran-cisco, Californic.
@.":[fi ~.'
TERA CORPORAiiCN
Schncbel, P.B. and Seed, H.B.,1973. Acceleratiore in Rock for Earthquakes in the Western United States, BSSA V. 63, p. 501.
Seed, H.B., cnd Idress, l.M.,1970. A Simplified Procedure - Soil Liquefaction Potential. Earthquake Engineering Research Center Report no. 70-9.
Seismic Engineering Program Reports,1974. U.S. Geol. Survey Circular 713.
Seismic Engineering Progrcm Pseports,1975. U.S. Geol. Survey Circular 7!7-A.
Seismic Engineering Program Reports,1975. U.S. Geol. Survey Circulcr 717-B.
Seismic Engineering Program Reports,1976. U.S. Geol. Survey Circuter 717-C.
Seismic Engineering Program Reports,1976. U.S. Geol. Survey Circular 717-D.
Seismic Engineering Program Reports,1976. U.S. Geol. Survey Cirevicr 736-A.
Seismic Engineering Program Reports,1976. U.S. Geol. Survey Circular 736-B.
Seismic Engineering Program Reports,1977. U.S. Geol. Survey Circular 736-C.
Seismic Engineering Program Reports,1977. U.S. Geol. Survey Circular 736-D.
Stcotz, M.H., 196C, The Republic Groben, A Major Structure in Northeastern Washington: U.S. Geol. Survey, Prof. Paper, 400-8, p. 304-306.
Stepp, J.C.,1974. " Analysis of Completeness of the Earthquake Sample in the Puget Sound Arec and its Effect on Statistical Estimates of Ecrthqucke Hazard." Proceeding of Conf. on Microzonctiom, Seattle.
Taubeneck, W.H.,1966, An Evaluation of Tectonic Rotation in the Pacific North-west: Jour. Geophy. Res., Vol. 71, No. 8, p. 2113-2120.
Tillson, D.D.,1970, Analysis of Crustal Changes in "he Columbia Plateau crea from Contemporary Triangulation and Leveling Mecsurements: Battelle Northwest, Doc. Identification No. BNWL-CC-21.'4.
Toppozada, T.R.,1975. Earthquake Magnitude os a Function of Intensity Datc.
Bull. Seis. Soc. Amer., Vol. 65, No. 5, pp.1223-1238.
Townley, S.D. anc M.W. Allen,1939. Descriptive Catalog of Earthquckes of the Pacific Coast of the United States, 1769-1928. Bull. Seis. Soc. Amer., Vol.
29,No.1.
U.S. Atomic Energy Commission,1973. A Study of Vertical cnd Horizontcl Earthqucke Spectrc, WASH-1255.
U.S. Geological Survey, 1964, Mineral and Water Resources of Idaho: Idcho Burecu of Mines and Geology, Sp. Rpt. l.
69Mbb g
6-5 TERA CORPORATION
u t
U.S. Geological Survey,1969, Mineral and Water Resources of Oregon: Oregon Dept. of Geology cnd Mineral Industries, Bull. 64.
U.S. Geological Survey, 1966, Mineral cnd Water Resources of Washington:
Washington Division of Mines and Geology, Reprint 9.
Vogliente, V.,
1973.
Forecasting the Risk inherent in Earthquake Resistant Design. Ph.D. Dissertation, Civil Eng. Dept. Stanford University.
WPPSS Nuclear Project No. I, 1973,1300 MiW Nuclecr Power Plant, orelimincry Scfety Analysis Report, vol. 2, Docket-50397.
WPPSS (Washington Public Power Suoply System),1976. Unit 1.
Preliminary Sofety Analysis Report, Amendment 9.
ig8b iO3 c.;
TERA CORDCRATION