HomeMy WebLinkAboutReport on Bradley Lake Hydroelectric Project Design Earthquake Study 1981Woodward-Clyde Consultants
REPORT ON THE
BRADLEY LAKE HYDROELECTRIC PROJECT
DESIGN EARTHQUAKE STUDY
Submitted to
Department of the Army
Alaska District, Corps of Engineers
P. 0. Box 7002
Anchorage, Alaska 99510
Contract No. DACW85-79-c~oo45
Modification P-00005
, ..
•
-
-
--
-
Woodward-Clyde ConsuHants
REPORT ON THE
BRADLEY LAKE HYDROELECTRIC PROJECT
DESIGN EARTHQUAKE STUDY
Submitted to
Department of the Army
Alaska District, Corps of Engineers
P. o. Box 7002
Anchorage, Alaska 99510
Contract No. DACW85-79-C-0045
Modification P-00005
-
-
...
-
-
-
... -
-
4791 Business Park Boulevard
Suite 1
Anchorage, Alaska 99503
907-276-2335
10 November 1981
Project No. l4844B
Department of the Army
Woodward-Clyde Consultants
Alaska District, Corp of Engineers
P.O. Box 7002
Anchorage, Alaska 99510
Attention: NPAEN-PM-PS
SUBJECT: REPORT FOR
DESIGN EARTHQUAKE STUDY
Gentlemen:
Enclosed is our report on the evaluation of the design
earthquakes and our assessment of the likelihood of fault
rupture at the Bradley Lake site. In this study, we have
reviewed the available literature in order to update
previous reports concerning the seismic hazard potential
at the site. The most significant new data are from the
microseismic network installed for the Bradley Lake project
by the u.s. Geological Survey. There is still only limited
information in.the literature regarding the geological
characteristics of local and on-site faults, particularly
with regard to their potential activity, and key parameters
that provide input into an assessment of their potential
maximum earthquakes and potential for surface fault rupture
at the site.
In order to conduct this evaluation, we have used what
data are presently · available, and have made professional
judgments and applied analogous data in order to estimate
the maximum earthquake for each potential source fault
considered, assuming that the faults are active~ These
estimates led to the Corps of Engineers selection of two
design maximum earthquakes: 1) a magnitude 8-1/2 event
occurring on the Megathrust located approximately 30 krn
beneath .the site, and 2) a 7-1/2 magnitude earthquake
occurring on the Border Ranges fault or the Eagle River
fault, both located within 3 krn of either the darn or power-
house sites. We have provided estimates of expected ground
motions at the site for these earthquakes, and have devel-
Consulting Engineers, Geologists
and Environmental Scientists
Offices in Other Principal Cities
-
111111
-
---
10 November 1981
Page 2
Woodward-Clyde Consultants
·oped corresponding response spectra. The results of this
ground motion evaluation indicate that the local faults
(i.e., the Border Ranges or the Eagle River faults) dominate
the response spectra for the design maximum earthquake.
If further field geological studies can demonstrate that
these two faults are not active, then the response spectra
for the site will be governed by the magnitude 8-1/2 event
occurring on the Megathrust beneath the site.
To assist the Corps of Engineers in their evaluation of a
potential operational basis earthquake, we have provided:
1) a response spectra developed for ground motions approx-
imately one-half those of the design maximum earthquake, and
2) a seismic exposure analysis. The results of the seismic
exposure analysis are expressed as the probability of
exceedence of given levels of ground motion at the site.
From this information, an operational basis earthquake can
be selected once the Corps has evaluated the relative risk
deemed acceptable in the project design.
The results of the seismic exposure analysis also provide
a ranking of the potential source faults in terms of their
relative contribution to the overall seismic exposure
at the site. This ranking will be useful in prioritizing
the local faults for future investigations that may resolve
questions concerning the potential activity of the faults.
If the local faults are active, as we have assumed for the
ground motion assessment, then there is a potential for
future fault rupture at the site during an earthquake
generated on either the Eagle River, Border Ranges, Bull
Moose or .Bradley River faults. By making analogies with
other faults throughout the world where secondary fault
rupture has been associated with an earthquake, and by
providing assumptions concerning the fault characteristics
at the Bradley Lake site, we have estimated the potential
slip on surface ruptures may be in the order of 20 to 300
em, and the probability of these events occurring are in the
order of 4 x lo-3 to 2 x lo-4. Ruptures of this nature
could impact the proposed components of the project such
as the dam, power tunnel, and the lake tap facilities that
are crossed by faults or 1 ineaments that are suspected of
being faults. This potential for fault rupture should be
-
---
-
10 November 1981
Page 3
Woodward-Clyde Consultants
considered in the design evaluation of those facilities or
evaluated in future geological field studies designed to
resolve whether or not the faults are active.
We appreciate the opportunity to have worked with you
on this phase of the Bradley Lake project. Should you have
any questions regarding the contents of this report, please
do not hesitate to contact us.
Rupert G. Tart, Jr.
Geotechnical Manager
JLoL-
s. Thomas Freeman
Project Geologist
-
-
-
-
...
--
-
Woodward-Clyde ConsuHants
TABLE OF CONTENTS
LETTER OF TRANSMITTAL
TABLE OF CONTENTS
1.0 INTRODUCTION .................................... 1-1
1.1
1.2
1.3
1.4
Purpose ........................... · ....... .
Scope of Work .••...•• ~ ..• · ••..•••.••.••...•
Geologic and Seismologic Data
Limitations ............................ .
Report Organization •••••.•....••...•.•.•••
1-1
1-2
1-3
l-4
2.0 EVALUATION OF DESIGN EARTHQUAKES •.••••.•.••••.• 2-1
2.1 Seismic Setting .•..•..••...••••..•...•••.. 2-l
2 . 2 Approach. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
2.3 Results................................... 2-4
2.3.1
2.3.2
2.3.3
Literature Review .••.•.••.••••••.••
Seismic Evaluation Assumptions •.•••
Seismic Source Evaluation ..•..••••.
2-4
2-10
2-12
3.0 DESIGN GROUND MOTIONS •••.•••.••.••••••.••..•.•• 3-1
3.1
3.2
Maximum Earthquake •••••...•.••••.•.••••.••
Operational Base Earthquake •••••.•••••••••
3-1
3-2
4.0 SEISMIC EXPOSURE ANALYSIS...................... 4-1
4.1
4.;2
Seismic Exposure Inputs ••....••• ·· . • • . . • • • .. 4-2
Estimate of Total Seismic Exposure
at Sites................................ 4-3
4.3 Relative Earthquake Source Contributions
to Total Seismic Exposure •••••••••••••.• 4-4
4.4 Earthquake Magnitude Contribution to
Acceleration Levels by Source ...••..•••. 4-5
5.0 LIKELIHOOD OF ON-SITE FAULT RUPTURE •••.••.•.•.. 5-l
5.1 Introduction ............................. .
5. 2 Assumptions .............................. .
5 • 3 Methode logy .............................. .
5.4 Results of On-Site Faulting •••..•••.•••.••
5-l
5-2
5-3
5-3
...
...
,.
Woodward·Ciyde Consultants
TABLE OF CONTENTS (Continued)
6.0 CONCLUSIQ_NS.................................... 6-1
6.1
6.2
6.3
6.4
6.5
Design Eartl1quakes ...•••..•.•...•.••...••.
Ground-Motion Analysis .•.•.•..•••.•.•.•••.
Seismic Exposure Analysis .•..•.•.•..•...••
Fault Slip Analysis ..••...••.•..••..•••••.
Limitation of Results ••....•••.•••.•...•..
6-1
6-2
6-3
6-3
6-4
7.0 RECOHMENDATIONS ..•.•••...•...•.•...•.•....•..•• 7-1
APPENDIX A -Methodology for the Seismic Exposure
Assessment
APPENDIX B -Methodology for Evaluations of Fault
Slip and Its Likelihood of Occurrence
APPENDIX C -Bibliography
TABLE 1 -Earthquake Source Characteristics
TABLE 2 -Summary of Possible Secondary Slip Occurrences
FIGURE 1 -Southern Alaska Regional Faults
FIGURE 2 -Lower Cook Inlet Region Section LA-LA'
Schematic Tectonic Model
FIGURE 3 -Schematic of Significant Local Faults and
Project Components
FIGURE 4 -Southern Alaska Historical Seismicity
FIGURE 5 -Flow Diagram of Approach to the Design
Earthquake Study
FIGURE 6 -Seismograph Stations Near Bradley Lake
FIGURE 7 -Microearthquake Epicenter Map
FIGURE 8 -r1icroearthquake Epicenter Map
FIGURE 9 -Seismicity Cross Section A-A'
FIGURE 10 -Seismicity Cross Section B-B'
FIGURE 11 -Seismicity Cross Section C-C' of the
Aleutian Subduction Zone
FIGURE 12 -Focal Mechanism Solutions
FIGURE 13 -Mean Response Spectra for Maximum Earthquakes
FIGURE 14 -one-Half of Response Spectra for Maximum
Earthquakes
FIGURE 15 Estimates of the Probability of Exceedence at
Dam Site
FIGURE 16 -Estimates of the Probability of Exceedence at
Powerhouse Site
FIGURE 17 -Distance to Secondary Faulting Versus
Main Earthquake (Ms)
FIGURE 18 -Slip on Secondary Fault Versus Main
Earthquake (Ms), (Strike-Slip Faulting)
FIGURE 19 -Secondary Slip as Percent of Maximum Slip
on Main Fault Versus Maximum Distance to
Main Fault
-
... ..
-
Woodward-Clyde Consultants
1.0 INTRODUCTION
This report presents the results and conclusions of
a design earthquake study completed by Woodward-Clyde
Consultants (WCC) for the Army Corps of Engineers (COE),
Alaska District. The study is part of the continuing
investigations being performed by the Corps of Engineers
for the Bradley Lake Hydroelectric Project located in the
Kenai Mountains of south-central Alaska. This report is
presented to provide guidance to the COE in its evaluation
of seismic design considerations for the various components
of the project. These components include a small dam,
power tunnel, powerhouse, and the related facilities.
This report was preceded by several reports of previous
investigations for the Bradley Lake project completed by
the COE, the U.S. Geological Survey (USGS), and WCC; these
earlier reports are listed in the accompanying bibliography
(Appendix C). This design earthquake report relies
heavily on three of those reports: two completed by wee
(Reconnaissance Geology Bradley Lake Hydroelectric Project
[1979] and Seismicity Study Bradley Lake Hydroelectric
Project [ 1980]), and a report completed by Lahr and
Stephens (1981) of the USGS entitled "Review of Earthquake
Activity and Current Status of Seismic Monitoring in the
Region of the Bradley Lake Hydroelectric Project."
1.1 Purpose
The purpose of this Design Earthquake Study was threefold:
1) to establish a design maximum earthquake based on
the current level of knowledge and to provide data
that will assist in the future selection of an
operational basis (OB) earthquake;
...
...
-
-
-
-
·-...
-
Woodward-Clyde Consultants
1-2
2) to evaluate the expected ground motions at the
site during the design maximum earthquake; and
3) to make an assessment of the likelihood of dis-
placement and amount of slip on faults that
intersect the power tunnel alignment and the dam·
site.
1.2 Scope of Work
This study was a level-of-effort investigation that relied
on the available data obtained from existing literature,
reports for the Bradley Lake Project, and consultations
with individuals who have worked on the geology and
seismicity in the area. No new geologic or seismologic
field data were generated during this study.
The "design maximum earthquake" is defined by the COE as
the most severe earthquake that is believed to be possible
at the site. In order to evaluate the factors that are
likely to influence the design maximum earthquake for the
Bradley Lake site, a review was completed of seiected
literature regarding the regional and local faults and the
tectonic regime in which the site is located. A review was
also made of the historical seismicity in the area and
ground motion attenuation relationships. On the basis
of data collected and evaluated during this review, an
estimate was made on the maximum earthquake for each known
potential earthquake source fault that appear to be of
significance to the project. These potential earthquake
maximum magnitudes provided a base from which the COE could
select the design maximum earthquake. Estimates including
response spectra, have been made of site ground motions
associated with the design maximum earthquake.
..•
-
..
-
1-3 Woodward-Clyde ConsuHants
The OB earthquake is generally less severe than the
design rna xi mum earthquake and is defined by the COE
as the earthquake most likely to occur during the life of
the project. The COE's guidelines for selecting an OB
earthquake are that it should be based on a probabilistic
evaluation developed from the understanding of regional
and local geology and seismology. The selection of an
operational basis earthquake depends largely on social and
economic considerations that are beyond the scope of this
study. To provide the COE with information that can assist
in future evaluation and selection of an OB earthquake, we
have provided response spectra that are equal to one-half
the response spectra for the design maximum earthquake. We
have also provided a seismic exposure analysis using
specific input for the Bradley Lake project. Results of
this analysis are presented as curves of probability of
exceedance versus level of peak ground acceleration. The
analysis also provides a mechanism for ranking the various
potential earthquake sources in terms of their relative
impact on the seismic exposure to the site.
To make an assessment of the likelihood of displacem~nt on
faults at the site, we have examined the literature for
empirical data on fault ruptures occurring during historic
earthquakes throughout the world. From this data base
and our current knowledge of the site, we have made some
judgments with regard to the likelihood of on-site fault
displacements using both deterministic and probabilistic
approaches.
1.3 Geolosic and Seismologic Data Limitations
Although a large body of information on the geology of the
region is available in the literature, the information is
very general and regional in nature and does not directly
...
-
-
--
, ...
l-4 Woodward-Clyde Consultants
address site-specific or project-specific seismic geology
or seismic design concerns. This places limitations on the
suitability of the data and, in turn, on the utility of the
results. The short time period over which local and
regional seismic data have been recorded also limits any
evaluation of design earthquakes and the potential for
future fault rupture. Thus, the scope of this study
addressing design earthquakes and the potential for fault
rupture at the Bradley Lake site is restricted by the
limited body of data available on the seismic geology and
the seismicity of the Kenai Peninsula.
Data limitations have been considered in developing
conservative assumptions regarding the activity of faults
in the area. Specifically and most importantly, the major
conservative assumption in the analysis is that the faults
(i.e., Eagle River, Border Ranges, Bradley River, and Bull
Mouse faults) were considered to be active for the seismic
exposure study.
1.4 Report organization
The main part of this report provides a summary of the
methodology, limitations, and results of the study. Where
more detailed discussion of methodology or procedures is
warranted, it has been included in the Appendices. The
report is organized to reflect the three primary purposes
of the study.
Section 2.0.of this report identifies, on the basis of data
from the available literature and reports, the possible
earthquake sources and associated maximum magnitude esti-
mates for those sources.
-
•
..
--
-
Woodward· Clyde Consultants
1-5
After an assignment of the maximum magnitude earthquake for
each specific source, the apparently most significant local
and distant maximum magnitude earthquakes were selected
for ground-motion analyses. Estimates of site ground-
motions are presented for these maximum earthquakes in
Section 3.0. Section 4.0 includes discussions regarding
the seismic exposure analysis. Results are presented as
curves of the cumulative probability of exceedence of peak
ground acceleration at the site, based on the identified
sources. We have also provided tabulations of the percent
contribution to acceleration levels from various sized
earthquakes postulated to occur on each of the seismic
sources.
Section 5.0 deals with the evaluation of the likelihood of
future fault rupture at the site and the amount of expected
slip. Sections 6.0 and 7.0 present our overall conclusions
and recommendations regarding the seismic hazard potential
at the project site.
...
-
Woodward-Clyde Consultants
2.0 EVALUATION OF DESIGN EARTHQUAKES
2.1 Seismic Setting
Discussions of the regional seismic setting, regional
tectonics, and local and on-site faults were presented in
earlier reports (WCC, 1979 and 1980a). As presented in
more detail in wee ( 1980a), the south-central Alaska
region has a high level of seismic activity because of the
northward underthrusting of the Pacific crustal plate
beneath the North American crustal plate along the Aleutian
subduction zone (Figures 1 and 2}. Great earthquakes
(surface wave magnitude Ms 8 or greater) and large earth-
quakes (greater thari Ms 7) have occurred historically
throughout the region (Figure 4) and can be expected to
occur in the future.
Bradley Lake is situated on the overriding crustal block
above the subduction zone and between the Castle Mountain
fault to the north and the Patton Bay-Hanning faults to
the southeast on Montague Island (part of the Offshore
Deformed Zone); all of these faults have documented
Holocene or historic surface ruptures (see Figures 1
and 2). Because of the active tectonic environment, it is
prudent to be concerned about the activity of other faults,
such as those found near or on the project site (see Figure
3), that are also located in the overriding crustal block
and between these known active faults mentioned above.
The concerns are amplified by the fact that several of
the local faults--the Border Ranges, Bradley River, and
Bull Moose faults--form striking topographic lineaments
visible on the smallest scale satellite imagery. In
addition, the Border Ranges and the Eagle River faults are
major crustal structures several hundreds to a thousand
-T--154o
' ~-' I 1520
,, / 1~ ' '' 148' ' "" J"" ,> 1 .. ' , !/,' ' -~-1 1 , ' --?' ! ..
140°
142° __ ~
...._...,....-. ' ....... ' ' / -'\! \. ,..ol' ', ',, ', ', -'\'<~. .Sc ', ', ',' I ,-,, •• ·-\ ',', '•
'"' 1-o.q '---.... ', ' v -.., ........ " ', ', ', *
·--····:;;;--'lu( ~" ',, ',
* ·· ...... No! '-.,, • * tAapped
*
J J
/ ' I 1
I I
I I
I I
I I
I
I
I
'. ~.. ' • "" --o-// .• ;.-"' r: I I , ··-_.,..:::--r--'" •
""'
~ * D ... -/ , • I'J_,,<~>--•·~ ' ''fs ~ "';~\~ ... ···· ~-----· ... ------'-
... o\.1~ .• ••••• b ~ " ·" . ' ' • -... " ' I. '.Y ', --• '•
/ .• A> ..__; ' • " -, ' ' A '
""'
_,,/ ' c "'[! J!"' ' _.. ' ·-·-· " -•• ' ,.." e~ . 1 , •' JV tP · \l ... , " r ,.. ,..,.... ~......... y· ~..:: ' ' ---', k • ' _,_..... -----~ "• --
' c ->•Y G<o .-, "' •• · ' ' •' .-". ' "·"· •• '""" (· ., "=''l'" "' .,._<:>)'<. Crackt~rf ~..,..~'--_.,.. ~~--"' ~ ,ifi..~~ ''"' ~.;:,_~ .~"'•· ~ G , ¢one,) f' .. 41! -.._;~ ' '• >'• "A / ' ,· ., ' ' . ~ ----·---'--~ 7 ·~-/ ' ," o/ ', ' -ll',' -------. ~....?/' __ / :x' ·:f" /.Y •s ·""-~" ---·--''" J)" ,• ', , ~ ,;;' ,!#' , ,..,..,'I" . '' •, '--C,
" .0\l, ;.:" o -v•L ' . )'' • R-J f ' i!' • '~-< i§ ... '• ------
• ,. ' v • • -~ • ' / '• / ' '\9 ', ' ,. " ' ;f "b ·~ .
• , ' ' <f~• ' •' I ~ -··· --
/ F ' -., ~ " ;,.; ., ~"' i• -~ > , ~ • ' J' ~ ~; i"'~omffll ~ / ~, c§J ..1? ~·! ..1? ··' 'cr ~~--...,
"'' ' . '' ' ' . ' ' ' -~._____/-' " <' ·•~•••v 'lrt.. .!'' 'fiR ' •'
'"'" ,! · ~ • / Hom" ' U.k~Si e 1,1. ·
0 ~ 1/!J ,_ ~
r-.. -·}'"" q,"' "' ~·' AUGus,-~achemak Ba ' it >< . ., •• ," "o MONTAGUE /I <' .;::;· r•ou,;"' v ' ' ' .' / -pow '
r
( • /, •-"" ''it' I / ' / I (_ '""'
0
,• ,' I Sold~•• -""'""' " , ,
0
' , , I ,-~·'" ,.-,-' .. • I ' ' -,.., ·' ·' • ----I --."'" -
' ,, q, :1 v· ~ ' •n ·-' --" '), ~' / I• -' "" <A / I ' '";;-' ~ --"'" -' (' ' / /--' '• , .. ' ~/ ( '";; " .-' / '· I / •._:,J" ' , •. --, } * h/. ,'x / ~ .. ~·• ~...,...,. ................ , •-I ' / ,_ --... ~'J. -' "" ,,, I ,. ' ·-~
~ -N-
~
o , ,-.,__ I I ' ·• .. .--' ,"'-( / .. ,,_ ----·-.,t/7··~/ ,/ .. ~j -.,.--
• * ' .' ·~ ..... I( .J. I( ~-j ' 0 * . ',, ' I ;t'l 4 ~----------* 0 • ~ I / I .~ -., •' * / ',./.4 1 / ··-. * • I I
0
0 /./ '-._ I •-. ... ,.~< ~<~ r .~ "'"<. I ---• • " -. ,. ' ' -., --/ ' o• ,,o/ • '-1 ,
1
,. ~~-stu
1
---•••
•' " . L <f ~ "' " I_.. fJ ;-.> / :._I r 1° .!;' I d $J, '·-.,
""" / .(> ' / ·" ' 'I . ~, ·----n ,r ,. " -,
.,·· J" I( ---. .... .,.. J'' .. s, '' I• ,r / ' I ' ) ' 1 '"" '•, ,.;' / I < c .. _,-~ ) "'• [''
, ' / /1 o•""•-·.Z• / ' G U L F / P<tg;·--• : ,-> ' /' \ J ~ '" ' -> "A.,;--. -•' / '\. ,f '\, ,,., 0 F ·-ftiJ·-
' /
0
•''} '\, ..,~'/ A L A 5 K A S)'•
' / 0 / ' .v / .
' ' ~
// ' \< /
-;..-/"x Y
'C;) ~ LA'
' ~ I ,. v ~_L
140°
I ~----~~----JL ______ L-----l I --I 146° 0 ...L.--------: ~ 144 1420
150° 148°
EXPLANATION
Act..,. flluolu
--r--r""" -r-· ··· · F ...,Jt. ~ wh«!t' .ooro••~leiV k'lrot»tltd,
dotl'ltd wfJoeN C"'Ondaield Ot quhhOO~
~ tt\QIUt• N'l•tt...ty d()wl"'
dtOPP*O SlOt' of t.-.lt.
-~~_._H.... Thrvtt F.tult. ~ 'llll'ht<re •r;:l()fo,urnat•ly
located. ciott.-d IIIPMrl c~*'ed 01
Quntion~. &art. incnaue-r!!lauwty
upthr!Mm ~t6t' of tauH.
----.... ,. S1rike-Siip F.-..h, ~ whltr11 IIPPf'Oxi!T\illely
loc.atltd. dOttlltd 'llll'h«r• i;Of'M»~ Of
QueUtOf'l-.... AtrOWI .ndteate rela1ive
t:hsplaa,.,.nt.
- - - - -Subtnarii'W rurl~ I;MJ/t cw scar-p tMoqnirfli
from gtOPhVitt:~l (Yta. H.achutH on
dowmNown .. de of fault.
___ ,._,. ____ .. -•-lii'W~t infefte-d fOf thit t"Cudy tot. an
•...-tt~.QUP•pr~inq nructur•.
ln&<:ti..,. Feulu
--,-_...... -.-Fault. ~d 'lfltlket"• apcwQlli:tNit•lv IOCM:ed.
donltd wMi"t-~~ 01 QUn1t'l)fYble.
H.churn ~tr ,...t~ly down
dr~ t.idll of fauiL
-_._ _._ ......_ · Thrust F&~lt. d.1llhtd ~ .,proaim.tltty
k>c.atlltd, dott•tt *""""« c:onc:»•d 01
quntt<W'l•bk'. a.rt. •ndi~ matiwly
ucnhrown li~ of tauh.
~ -Strlk?-SHp Fault. dastu:d whet• t~PC>toxinvtety-
10QteQ, dotted 'llthmt c.oneuhtd (II
quntionable. Arrow. itw:heate: rd•tiwe
tt•~tx:a-ment.
- - --Svbmarine IUI"f~ fault or SCMP tecoqniud
*
from groph""''cal <"hlt•. H...chutft on
dowothrc:nwl side.
lliQIC<Jnlc ctnt~r. went Of c:one.
1------------1 LiM Of <:TCJ;!..~~~OM.
NOTES·
Ac1iw-Fault-A subaoerial or submarti'W buh: th.-t bt~Ollk$ 01 i! i11"'ned
to ~.-.It Hoi~ or unoott10lid.a:ttd: sediments
!Notub.lv young but without ""Y age de1l'rmtnatiD1'1)
or a tubm.vi"C' t-.,tt is ~"'posol'd oo U. •• flocf".
ln.cti~ Fault A $1Jbaerial cw submatiM f<MJit th•t dQe1i not btuk
Holoune-nor uncornohdat~td wttifntf'\H tNot~.oiy
young but without ;,ny age drttrmJn.-tionl not f't
it e"'povd Oi'l !'he< •• floor
8..,r MJJp Comp)~ from World Aen;lntvtial Otwu.
2:5 o 25 50 N• .. ttcal ~hln
25 0 2:5 50 !00 Kd•••t•n
25 0 2:5 50 St•hlh -.ohln
Source:
Woodward Clyde Consultants, 1978, Offshore
Alaska Seismic Exposure Study
WOODWARD-CLYDE CONSULTANTS
SOUTHERN ALASKA REGIONAL FAULTS
Project No. 148448 Fig.
BRADLEY LAKE 1
LA
---N45W
I Kenai
Kenai Lowland Mountains Continental Slope Abyssal Plain Cook Inlet Continental Shelf
o Kachemak I ~y ~ ~2§ I I BRADLEY~/(}
.:::1 0 ~ $" ~ ~ LAKE ~ ~ ~ b A:! J........ .::; CQ ~fl.'? I ,!JJ..... ;;'~~ c:-0J..... <.."' 0 ~J..... § i;J..... R0J..... (:1-..J;;: ... ~,c:-$ ~ ._~ ~ CJ ~ ~ t.t ~ ~ QJO.f~ ~"r.; ~ DEFORMED -t ~
/1 I i I I / -~ ;-:"".·:::::::?:::::::<:~··...:.-'--' I 1 1: ~~ I / --=.~:~:s::~-·~_;,.--
N u j / --.-:-:-::-:.:-:-:-:-:-·-:~·
j /"¢;E>o;;~ \SUBDUCTION ZONE
. f' ~:::::::£=-~-~
North American
Lithospheric Plate
EXPLANATION
Tertiary and younger bedded
sedimentary sequence.
l ~ • . . ·: ~ ·j Highly folded and/or metamorphosed
·_,_,,. .. -. . .. basement complex.
... -:._.::.·:.~..::~...,'-:--::~~:·,
v:::::::::::::::::::::l Basalt or undifferentiated volcanic
·.·.·.·.·.·.·.·.·.·.·.·.· rocks. ·.·.·.·.·.·.·.·.·.·.·.·.·
I
- -----Lithologic contact, dashed where inferred
--~ --Fault, dashed where inferred. Single and
double-sided arrows represent the
relative component of displacement
inferred for this cross-sectional
orientation.
Pacific
Lithospheric Plate
25 0 25 50 100 Kilometers
25 0 25 50 Statute Miles
'""'""' ----I I
Vertical Scale Equals Horizontal Scale
NOTE:
Lithologies and structures that are not exposed at
the surface are inferred.
SOURCE:
Woodward Clyde Consultants, 1978, Offshore Alaska
Seismic Exposure Study
LA'
WOODWARD-CLYDE CONSULTANTS
~~ ~~ Granitic Rocks
[ ~J:~:~;jJ.t,~:.?:~ /Xl Magma or resorbed oceanic crust.
LOWER COOK INLET REGION
SECTION LA -LA'
SCHEMATIC TECTONfC MODEL
Project No. 148448 Fig.
BRADLEY LAKE 2
Kachemak Bay
Cl)'
I.J.J· '"'· <:.
<:(,
a::.
.~· #•
Q~.
c}·
o"'l>.
-§· . ... . ...
< • •
--~-------------------------
Project:
Project No.
BRADLEY LAKE
148448
1-
-1
:::J
<{
L!..
UJ
U)
0
0
~
I
I
I
I ,,,
I
I
I
I
I
I
I
-1---... ... ~
I '-~
0
LEGEND
__.....,..,.
~ ,,
High angle fault,
dashed where inferred,
dotted where concealed,
queried where uncertain
Thrust or reverse fault,
·dashed where inferred,
dotted where concealed,
sawteeth on upthrown side
-· -• -Lineament
... • • • • • • • • • • Lowland and Marshland Area
:·KITE: For more detail refer to Drawing Nos.
and 2 from wee (1979).
~
-N-
~
2
Bradley Lake
~
~~
\.1-~
<:Y..y .....
-"7~ ~
.................
&=;; I I
V)~
%~ Scale in Kilometres
SCHEMATIC OF SIGNIFICANT LOCAL FAULTS
AND PROJECT COMPONENTS
Fig.
3
WOODWARD· -CLYDE CONSULT ANTS
lb'l 15t'
l 1 f
.:;_~-:·=~
cj"
9 OCTOBER 1900
600
4 SEPTEMBER 1899~
/-iL-;;MVA-~"if/ tl ~~~~:te .. -~~ co .. cC)'":__:
w"l--./' r h' ' .. ' ' TJ --
'// C~ I>(" (J;;·l " w, ·--(f-(1-" & : ,, l, ·;~:~· ~-:!.-(' rc;"' ---li'J (l-) \-! ~~0"F ~~(·:~--~)ill~~\ ·--,-~~< ~-~," "'~" -
1
... -•
1
'"rHI r;" '-" ... _ .. , ,;,C':J ·--'·_f) ~ l' ), W!i -~' .. ",J' -( b-.'' ' I ~~ G(T)" .. _., ,_ \, __
,/, 01 -. ""' " m __ ,~n " ,. --
I ,.f
/;\
58~
'< ·:·.::~ , r ' --/i , .. I :""' ~ ll_) (D'" 'l'l1'br' ·-
--•. ~) _,. J' -~ --~~~-" '"'"' rn;~ ) .. , ··-o' -,
--. ( '! Jr ~-0 "' / ~-·'· :r '~!:h 1 ~ ! 'r 'I:\' \If~ ~·1; .. r !'"' / ·
. . J· ,_ ·>-~J . I~ ,i ·.@· .: ... r~.:.._~~~:< ~_),_!'· , y'f' r.) i ed;. ·v;;l-: I :·
' t\ , 0 ' . ·--~\ ~ .. ' f.-l' 1/ ' ' -v~f~S{·"i . '"'~-, ,. ";:;-;·: -, -, ,,. : ... , --
0
' _. -·' ,. ' w I I " .'-'--"
o··
o·· ' , ·r-f 11 ·,, I'~'"' i l,l,c_tl i.'
. I I r ~· \ • ~-, .,~.-, "~ , 'I, r ' ·. ~. ,, . ,''· -·~·'·i :,-lt~.E'~,.( • ' I' I I, ·' .~'1 •···' '' (' '
C)02
G U L F OF
(£).
ALASKA
~ -~~-,-·-' 1' .,.,,.: ' ' .
1
r ~1 ·, .{t''tt-?1···~ 1 I-;-~· ;• -r•.uc•· ~--• .-~ .. -~ \ . ·--""" '" II' ·-· .
,54
0 '1' I,·--··-I)',' rl ,,, r n·) _l :y-./
I I ' ' • 4 ' ·.':.L -./ ~I):-: .--· .·I ' I .-:-•
152° . ' ) I
I ,
1
>18° '
1
I 1
146° ~---~---_L-
1440 142°
156°
NOTES:
EXPllMATION
C)
R!PORHO
I I 8.Q
(11
' ' ),0
~)
C) 6.0
C) 5.0
MlGNITUO[
Magnitude symbol SllU 1ft !ltloWft
on a ton11nuous ft0nline1r Stalt
1) Minimum magnitude = 5.0
2) Events are numbered chronologically. Numbers refer to
entries in the Earthquake Data Bank catalog under The
Lo_, Cook Inlet and Gulf of Alaska areas, which is
included in WCC, 1978, Offshore Alaska Seismic
Exposure Study Report.
3) Selected earthquakes have been added and are
discussed in the text .
57°
1390
25 0 25 50 N<1.utrcet ._,r!ll
'!""------2~ 0 25 50 !<?O K1lorut1rs
25 0 25 50 ~tatuta t.fllu
-........._~ =-.!
Source:
Woodward Clyde Consultants, 1978, Offshore
Alaska Seismic Exposure Study and NOAA
Hypocenter Data File 1638-1975
WOODWARD-CLYDE CONSULTANTS
SOUTHERN ALASKA HISTORICAL SEISMICITY
Project No. 148448 Fig.
BRADLEY LAKE 4
2-2
Woodward-Clyde ConsuHants
kilometers long (see Figure 1). This observation, by
itself, raises concerns regarding their impact on the
seismic hazard potential at the site.
Two primary considerations are involved in evaluating the
seismic hazard potential of the Bradley Lake Project site:
first, the maximum expected earthquakes and the related
shaking that may be generated from movement on the local
or regional faults: and second, the potential for surface
rupture along the faults in and adjacent to the project
site. This section of the report deals with the first
concern7 surface faulting is addressed in Section 5.0.
2.2 Approach
The approach used to develop a basis for evaluating the
maximum design earthquake and to develop a basis from which
the eoE can select an OB earthquake is summarized in
Figure 5. As discussed in Section 1.2, the study is based
entirely on available data from the open literature,
project reports, or from consultations with individuals who
have or are completing research in the Kenai Peninsula
area. wee's (l980a) original report on the seismicity of
the Bradley Lake area provided the initial springboard for
this Design Earthquake Study. The wee ( l980a) report was
primarily based on wee's work for the Offshore Alaska
Seismic Exposure Study {OASES) ( 1978), which, at the
time, was the most comprehensive seismic evaluation of
Alaska's coastal areas. That 1978 report provided a basis
from which more site-specific investigations for the
Bradley Lake project could be directed; however, it was
not a site-specific document and was intended only as a
regional study. Therefore, a part of the effort for this
study was to utilize the existing data to address more
site-specific seismic concerns to the degree that the data
base would allow.
Project:
TASK I TASK II
Recent literat•Jre
review and consultation
with researchers
M icroearthquake
data analysis
TASK VI
Fault displacement
assessment
TASK Ill
Update Existing
Seismicity Study (WCC 1980)
Refine estimate of:
1. potential source fau Its
2. source fault parameters
3. recurrence estimates
Input
TASK IV
Conduct seismic exposure analysis
to estimate probability of exceedence
of different levels of peak acceleration
TASK VIII
Draft and
F ina! Report
pr,eparation
Final Report
TASK VII
Design ground
motion analysis
Project No.
BRADLEY LAKE
148448
FLOW DIAGRAM OF APPROACH
TO THE DESIGN EARTHQUAKE STUDY
Fig.
5
WOODWARO-CL YDE CONSULTANTS
2-3
Woodward-Clyde Consultants
The initial steps in the process leading to the maximum
design earthquake selection involved a rev1ew of the
available literature, primarily the most recent literature,
to provide data, for updating the earlier wee reports (1978
and 1980a). A key part of this process involved a review
and analysis of the earthquake data collected for the eoE
in the Bradley Lake area by the USGS ( Lahr and Stephens,
1981).
The results of the review of the available literature led
to updating and modifying some of the earthquake source
parameters listed in the original Bradley Lake seismicity
report {wee, 1980a). These revised earthquake source
parameters are presented in Table 1 of this report.
Some of the earthquake sources listed in the wee ( 1980a)
report were not included for further consideration in the
present study because their potential effects were judged
to be negligible (that is, very low ground accelerations
were expected at the site as a result of earthquakes on
these sources). A more detailed review of those earthquake
sources retained for consideration in this study is
presented in Section 2.3.3.
The apparently most significant earthquake sources and
associated earthquakes were selected in collaboration with
the eoE for development of seismic response spectra and
ground motion parameters (discussed in Section 3. 0). The
earthquake source characterization data presented in Table
1 were used as input to a seismic exposure analysis that
uti 1 iz ed a computer program developed by \'lee for the
National Oceanographic and Atmospheric Administration. The
results of the analysis provide data on the likelihood of
various ground acceleration levels occurring at the site.
The seismic exposure analysis also allows a ranking of the
TA[)LE l
EARTHQUAKE SOURCE CHARACTERISTICS
Minimum
Minimum(a) Distance To Earthquake Type
Name of Earthquake Distance to Powerhouse Source of
Estimated
Maximum Rupture Estimated(e)
Historic Plane/geometry Area of
Rupture(f) Rupture(g)
Area/ Length/
Maximum Maximum
Source Dam Site Site Len;1th Source Earthquake Length (dip) Rupture Magnitude Magnitude
(km) (km) (km) (Ms) (km) (km2) (Ms) (Ms)
Subduction Z0ne
Aleutian (b) 30 30 Underlies Plate
Megathrust (depth) (c) (depth) (c) entire boundary
8-l/2 22,400 8-l/2 N/A ( i)
region megathrust
fault
Benioff Zone (b) 30 30 Underlies Deformation
(depth) (c) (depth) (c) entire within
7-l/2 N/A ( j) N/ A ( j) 7-l/2 N/A (j)
region Pacific
Plate
Local Sources
Border Ranges fault 7-10 1-l/2 1,000+ Oblique --100 (d) 2 '100 7-l/2 7-l/2
Reverse? (70" dip)
Eagle River fault l-1/2-2 4-1/2 750+ Oblique --? ? N/A 7-1/2
Reverse?
Bradley River fault 1 4 19 Oblique --19 190 6-l/ 2 6-l/2
right
lateral
Bull Moose fault 3-1/2 2 ll Oblique --11 110 6-l/4 6
right
lateral
Random Source (k) 5 5 Underlies Nonassociated 5-3/4 N/A N/A N/A N/A(L)
entire earthquakes
region
(a) Measured from Beikman, 1980
(b) The subduction zone is divided into two segments, each with its own seismic characteristics. The first segment is termed the Aleutian Megathurst,
which separates the overriding North American Plate from the subducting Pacific Plate. The second segment termed the Benioff Zone is a zone of
earthquakes corresponding to stress relief in the upper 20 km of the subducting Pacific Plate (Lahr and Stephens, 1981).
(c) Estimated from data presented by Lahr and Stephens (1981). Approximately+ 10% of distance is judged to be the value for uncertainty in the estimate.
(d) Estimated from data by WCC, 1978; WCC 1980a and 1980bi (U.S. NRC, 1981) and comparison to known length of Quaternary activity along the Castle
Mountain fault.
(e) Area of rupture based on length of rupture and 20 km depth for major faults and 10 km for minor faults.
(f) Magnitude estimated from Wyss, 1979; rupture area, and expressed as Ms (surface wave magnitude).
(g) Magnitude estimated from Slemmons, 1977 for strike-slip faults, reverse-oblique, and reverse slip, and expressed as Ms (20 second surface wave
magnitude).
(h) The Eagle River fault is divided into two segments to accommodate the computer seismic exposure analysis.
(i) Maximum magnitude assigned on basic of historic earthquakes and area of rupture plane.
(j) Maximum magnitude based on historic earthquake seismicity on this s?urce and is not associated with a rupture on any well defined fault plane.
(k) The random source was not included in analysis. See .Section 2.3.3 ~n text for a discussion.
(1) Maximum magnitude based on historic earthquake seismicity.
--No available data or record of occurrence.
N/A Not applicable.
2-4 Woodward-Clyde Consultants
various earthquake sources relative to their individual
contribution to the total seismic exposure of the Bradley
Lake project area.
2.3 Results
2.3.1 Literature Review
This section summarizes the results of our review of the
most recent data identified during the investigation that
is of use in evaluating the level of activity of local
faults and estimating the design maximum earthquake for the
Bradley Lake project. The majority of new data are from a
microearthquake network installed and operated for the COE
by the USGS (Lahr and Stephens, 1981).
The microearthquake data are derived from recordings made
by the USGS at permanent, high gain seismograph stations
(Figure 6) operating over two time periods. The first time
period extended from October 1971 through November 1980,
during which time a regional network of stations on and
adjacent to the Kenai Peninsula was in operation. The
station locations for the regional network are shown as
diamonds annotated with the station name and year of
installation in Figure 6. This regional network detected
earthquakes in the approximate magnitude range of Richter
.local magnitude (ML) 2.0 and greater.
The second period of seismic network coverage discussed
in this report extended from December 1980 through July
1981 and included data from five additional stations that
were added to the original regional network by the USGS for
the COE and the Bradley Lake Project (see Figure 6) .
The augmented station array provided both an increased
earthquake location accuracy (to a few kilometers within
about 50 km of Bradley Lake) and a reduction of the
60
Project:
Project No.
MSP ....
..J
(73l :::> _ ... ~
~z u..
zw cc w:E w
~4. > w cc z 'J. cc, . w u·
4.
..J a..
<) Regional Seismograph Stations
0 Bradley Lake Seismograph Stations
NOTE:
1. Fault locations modified from Beikman 1980; seismicity and base map
modified from Lahr and Stephens (1981).
BRADLEY LAKE
14844B
SEISMOGRAPH STATIONS NEAR
BRADLEY LAKE
Fig.
6
WOODWARD-CLYDE CONSULTANTS
2-5
Woodward· Clyde Consultants
earthquake detection level (to the ML l .0 to 1.5 range).
Data collection and processing for the initial operation of
the Bradley Lake stations was discussed in detail by Lahr
and Stephens (1981). Data for the period January through
July 1981 were provided to wee by Lahr and Stephens in
advance of publication (Lahr and Stephens, open-file report
in preparation, 1981).
The available earthquake data were evaluated by wee in
cooperation with the USGS to assess the nature of currently
active deformation within the shallow crust of the southern
Kenai Peninsula and to evaluate the seismic potential of
the subduction zone underlying the peninsula. The results
of our review regarding the seismic potential of the
subduction zone are presented in Section 2.3.3 and serve to
update the earlier wee (1980a) report.
Microearthquake activity can generally provide usefu 1
information relating to the level of activity and charac-
teristics of rupture on fault planes at depth. Functional
associations between the seismicity and known faults
may be inferred by examining spacial patterns of earthquake
hypocenters in relationship to the fault plane. Analysis
of the first motion characteristics of the events asso-
ciated with the fault (focal mechanism solutions) can
lead to judgments regarding the style of faulting, for
example, strike-slip, thrust, or normal fault displacement.
There are limitations in the use of microearthquake data.
First, the number of events occurring in any spacial
pattern around a fault must be sufficient to rule out
random or fortuitous occurrences of events. Second,
the geometry of the fault plane needs to be reasonably
well known in order to relate it to focal mechanism
Woodward-Clyde Consultants
2-6
information derived from the microearthquake data. Third,
available microearthquake data are generally not conclusive
with regard to demonstrating that a particular fault is not
active over long-time periods. Long periods of relatively
quiescent seismicity are known to occur on major active
faults {for example portions of the southern segment of the
San Andreas fault in California). Therefore, caution is
required in using microearthquake data in resolving the
activity of a fault.
In the Kenai Peninsula, the earthquake activity shallower
than about 20 km is considered to be distinct from the
deeper, subduction zone activity and it is apparently
associated with present-day deformation of the crustal
rocks. In order to assess possible associations of this
seismicity with the known faults in the vicinity of Bradley
Lake, hypocentral plots and cross-sections were examined
for spatial patterns that may be indicative of functional
associations.
Figure 7 shows the recorded data set for October 1971
through November 1980 from the regional networks. No clear
spatial association between the earthquake epicenters and
the Border Ranges, Sterling, or Eagle River faults is
apparent. A few events lie generally along the trends of
the three faults, but do not indicate a linear trend
suggestive of active faulting. The greatest concentration
of activity shown in Figure 7 is in the vicinity of the
Kenai Lineament and the Placer River fault, as discussed by
Lahr and Stephens (1981). In this group, the largest
event, assigned a magnitude of 5 .1, occurred 5 February
1976 at a depth of 21 km. There is a broad zone of
activity trending northwest near the center of Figure 7:
this zone is not associated with any known geologic
Project:
~i
z
OCT 71-NOV 80
DEPTH:$ 20 km
_v
0 ~
7 ~IJ/(o
.!./ v
PLACER RIVER FAULT
A./ lA'
~" ~"?" c
A.'?"
~ CJ
~~---------L--~----------------L---------~ uw rdw t:lw f:tlw 149'w
NOTE:
1. Fault locations modified from Beikman 1980; seismicity and base map
modified from Lahr and Stephens (1981 ).
0 25 km
Project No.
BRADLEY LAKE
148448 MICROEARTHOUAKE EPICENTER MAP
~
-N -
~
Fig.
7
WOODWARD-CLYDE CONSULTANTS
2-7 Woodward·Ciyde Consultants
features and is probably a coincidental alignment. Most of
the located earthquake activity lies in the topographically
high pgrtion of the Kenai Peninsula, with little activity
in the lowlands or in the adjacent portion of Cook Inlet.
Figure 8 shows the later data set, recorded between
December 1980 and July 1981. The features of the seismi-
city in this period are generally similar to those of the
earlier period, but the lower detection level and increased
station concentration in the southern Kenai Peninsula have
probably enhanced the numbers of earthquakes identified
south of 60°N latitude. There is a tendency for the
located earthquakes to scatter about the Border Ranges and
Eagle River faults. This pattern is not coherent enough to
define the presence of active faulting. There is a general
pausity of earthquakes in other areas, including the Kenai
Lowlands and around the Sterling fault.
Three cross sections were plotted for the study area, as
shown in Figure 8. Only the earthquake locations with
epicentral errors less than 5 km were used in constructing
these cross sections. In the more northerly section
(Section A-A', Figure 9), through latitude 60°N, the
crustal earthquake activity is seen to occur primarily
above a depth of 20 km. The concentration of activity
noted in the Placer River fault area does not clearly
define a planar trend. No other suggestions of fault
planes or localized deformation are obvious.
In the southern cross section (Section B-B' 1 Figure 10),
the crustal earthquake activity between the Border Ranges
and Eagle River faults occurs primarily above a depth of 15
km. The clusters of events seen southwest of Bradley
Project:
600N
DEC 80-JUL 81
DEPTH :$· 20 km
0
SINGLE EVENT
FOCAL MECHANISM ~
0
PLACER RIVER
FAULT
KENAI
LINEAMENT
0
8
EVENT@l0440
08.24.75
.... / J11'/fl-I I I \. 'vf ( I / ,B' i I \: I I 7
0
0
EVENTS INCLUDED IN
COMPOSITE
C'
590N~------------~--~~----------------------------------~ 153ow 152ow 151ow 150°W 1490W
NOTE:
1. Fault locations modified from Beikman 1980; seismicity and base map
modified from Lahr and Stephens (1981 ).
2. See Figure 10 for focal mechanism solutions.
0 25km
Project No.
BRADLEY LAKE
148448 MICROEARTHQUAKE EPICENTER MAP
~
N
~
Fig.
8
WOODWARD-CLYDE CONSULTANTS
OCT 71-JUL 81
E-W SECTION
40 KM WIDE AT LATITUDE 60°
60 EVENTS
PLACER
A KENAI RIVER A' LINEAMENT FAULT
0.0 2~.0 59.0 ~ ~ 1y.O DISTANCE IN KM q ---,
0 -o 0 ... 0
~ 0 0 0 0 0 6l 0 0 0 0 :E 0
0 Do 0
0
0 ::.::: 0 0 0 oo 0
z oo 0
80
0 0 -Q 0 0 00 0 8 :c 0 0 0
1-0
0 0. 0 w
Q
0
0 0
0
0 0 0 .
1.0-
C?
NOTE:
1. For location of cross section refer to Figure 7.
Project: BRADLEY LAKE Fig.
SEISMICITY CROSS SECTION A-A' 9 Project No. 148448
WOOOWARD-CL YDE CONSULTANTS
OCT 71-JUL 81
E-W SECTION
40 KM WIDE AT LATITUDE 59°30'
92 EVENTS
/BORDER rAGLE B ~RLING RANGES B' FAULT FAULT RIVER
FAULT
0.0 25.0 50.0 75.0 100.0 DISTANCE IN KM q I I i _l j _l J
0
0 :I 0 • 0 " 0 t9 llj ~ o 0 o 0
0 o 0 0 O z 0 % 0 0 0 -0
% c ....
0 0 0 0 Q. w
Q 0
0
q 0
0 0 IlL 0 0 C'?
NOTE:
1. For location of cross section refer to Figure 7.
Project: BRADLEY LAKE SEISMICITY CROSS SECTION 8--B' Fig.
Project No. 148448 10
, ... ,....,...,.....,.,.," nn ,.., vnc: f"'('\~IC:1tl TAfi.ITC:
Woodward-Clyde Consultants
2-8
Lake, and noted in Figure 8, appear in Figure 10 as a
near-vertical lineation just west of Bradley Lake and as a
diffuse zone east of and beneath the Border Ranges fault.
In the first case, the vertical lineation seen in Figure 10
does not strongly suggest the presence of an active fault
because the epicentral locations of the events, seen in the
map view (Figure 8), are so closely spaced; in the case of
an active fault one would expect them to be distributed
more along the map trace. The vertical lineation is
likely the result of a combination of scatter, due to
uncertainties in depth control for the individual event
locations, combined with some true variation in depths.
Such clusters are common in many tectonic environments and
are not necessarily indicative of throughgoing faulting.
In the second case, the events plotted near the mapped
trace of the Border Ranges fault have hypocentral locations
east of the postulated westward dipping plane for that
fault. In addition, no consistent dip orientations can be
inferred from the hypocentral locations seen in Figure 10.
In cross-section C-C' shown in Figure 11, the shallow
seismic activity is seen to occur primarily beneath the
topographically high portions of the peninsula, with
fewer events beneath the Kenai Lowlands or Cook Inlet.
Several south-east-dipping alignments of hypocenters are
seen, but they are defined by such a limited number of
events that they appear to be coincidental. The deeper
seismic zone is evident in this cross-section and is
discussed in Section 2.3.3.
Focal mechanisms were analyzed to assist in an assessment
of possible associations of the microearthquake activity
with the mapped regional and local faults. These focal
Project:
Project No.
::;
~
~
J:
I-
C. w c
OCT 71-JUL 81
80 KM WIDE
1031 EVENTS
NW SE
BORDER RANGES BRADLEY LAKE
FAULT l rr: 200 km
STERLING -:1 r:EAGLE RIVER TO TRENCH
FAULT FAULT ,
Co 50 100 150 2oo 25oC
Ql ~ ~ "I I I e 1 ~ :~g._oo ~ o 0 00 o
n .~ ~ 0 0 00
o 0 o w ~o o io ,..t:J o o o o 0 ~o 1964 RUPTURExxxx 0 xxxxxxxxxxxxxx 0
-x.D -j} 0 ALEUTIAN MEGATHRUST
00 ZONE 00 0 .......,_
0
0 .....
50 .O 6.5 cm/yr
100
150
0
0 0 / /
0 ,t>
//
///
~ / 0 // ..,""
/1"
BRADLEY LAKE
14844B
,'
/''
/'
0
0
___ --a -cr------------------------------
NOTE:
1. Figure modified from Lahr and Stephens (1981).
2. For location of cross section refer to Figure 7.
SEISMICITY CROSS SECTION C-C' OF THE ALEUTIAN SUBDUCTION ZONE I Fi;~
WOODWARO-CL YDE CONSULTANTS
Woodward-Clyde Consultants
2-9
mechanisms are used to infer possible fault planes for the
earthquakes considered as well as likely causative stress
axes. To develop the focal mechanisms 1 first motions at
the various seismograph stations for better-recorded events
were plotted for individual events and for composite groups
of nearby events.
The second period of recordings (1980-81) provides the most
useful data for assessing focal mechanisms because of the
increased earthquake location accuracy that resulted
from the increased station density. The composite focal
mechanism solution shown in Figure l2A for the group of
events nearest the Eagle River Fault (see Figure 8}. Other
solutions are generally consistent with this one and
indicate predominantly normal faulting, with the principal
least compressive stress axis oriented approximately
horizontal on a trend of N70°E 1 as shown in Figure l2A.
This mechanism of faulting for the recent earthquake
activity is not consistent with the geological history of
the faults (thrust or left lateral oblique slip, see
d cussion in Section 2.3.3) nor is it consistent with
nOrmal slip on fault of the mapped orientation (northeast}.
Focal mechanism solutions were developed for a few of the
larger (ML 3.5 to ML 5.1), shallow events recorded during
the time period between October 1971 and November 1980.
One event 1 which occurred on 24 August 197 5 at 0440,
yield~d sufficient data to allow for a focal mechanism
solution. This solution is presented in Figure 12B and is
essentially identical to the solutions obtained previously.
The occurrence of normal-faulting focal plane solutions
rather than thrust or lateral-slip solutions and the lack
w
N
s
A)
E
LEGEND
A) COMPOSITE SOLUTION OF EVENTS OCCURRING NEAR THE
EAGLE RIVER FAULT, DECEMBER 1980-JULY 1981. SEE
FIGURE 7. NORMAL FAULTING, PRINCIPAL AXIS OF
TENSION ORIENTED N60°E
T Axis of minimum ~ horizontal compression
stress
0 Axis of maximum
P vertical stress
C Compression first motion
D Dilatation first motion
+ Low confidence
compression
-Low confidence
dilatation
N
s
B)
B) SINGLE EVENT FIRST MOTION PLOT.
EVENT OF 24 AUGUST 1975, h = 6.5 KM, ML = 3.8.
NORMAL FAULTING, PRINCIPAL AXIS OF
TENSION ORIENTED N70°E
Project: BRADLEY LAKE FOCAL MECHANISM SOLUTIONS Project No. 14844B
Fig.
12
WOODWAAD-Cl YDE CONSULTANTS
2-10 Woodward-Clyde Consultants
of well-defined spatial association of the microearthquake
activity with the mapped faults in the Kenai Peninsula
suggest that the present mode of crustal deformation is
different than that associated with primary movement
of the regional faults. Due to the recent occurrence of
rupture on the Megathrust in 1964, the current crustal
microearthquake activity may reflect low-level adjustments
to the 1964 stress release.
The microearthquake data do not show a compelling associa-
tion with the mapped faults in the area of the Bradley Lake
Project. However, the period of relatively high-resolution
monitoring was short and the limited existing data base
with its inherent uncertainties of hypocenter locations,
is not sufficient to rule out the association of micro-
earthquakes with known faults. Additional monitoring time,
with an increased number of recorded events, and improved
knowledge. regarding the characteristics of faults in
the area, may provide a more definitive data base. In
addition, these microearthquake data are not conclusive
with regard to the long-term seismic activity of faults. ,
As seen in other regions of the world, known active faults
can have long periods of no appreciable microseismic
activity, and networks monitored over short periods of time
can give incomplete or apparently contradictory results.
Thus the assessment of fault activity must combine an
evaluation of the geologic data along the faults with the
seismic data.
2.3.2 Seismic Evaluation Assumptions
It is useful to review the key assumptions made during the
course of this and past evaluations of seismic hazards at
the Bradley Lake site before proceeding with the results of
our reevaluation of the seismic sources that may affect
Woodward-Clyde Consultants
2-ll
the site. These assumptions were required because the
available literature does not contain sufficient data
regarding seismic geology particularly the characteristics
of the faults in the area.
For the purpose of this study, we have conservatively
assumed (as was done in the wee [l980a] report) that
several faults are potential earthquake sources even though
the geologic evidence for or against their activity
under the current tectonic environment is not presently
available. This applies in particular to the near-site and
on-site faults that have mapped lengths of hundreds of
kilometers or that have distinctive topographic lineaments.
We have also made assumptions regarding the nature of fault
slip and the length of potential fault rupture on several
of the near and on-site faults that are being considered as
seismic sources. These parameters were used to estimate
the maximum magnitude earthquake for a particular seismic
source.
To develop a reasonable basis for estimates of the charac-
.teristics of some of the local and on-site seismic sources,
a geologic model of the region was evaluated which incor-
porated a fault with known Holocene surface rupture, the
Castle Mountain fault. Using this model, estimates were
made of the fault parameters for some of the faults near
the site through analogies with the Castle Mountain fault.
Two methods were used during this study to relate the size
of the potential seismic source with an estimate of the
maximum earthquake magnitude that can be generated by that
source. Wyss (1979) provided relationships correlating the
area of a fault rupture plane with earthquake magnitude.
2-12
Woodward· Clyde Consultants
Slenunons ( 1977) provided a correlation between length of
surface fault rupture and maximum earthquake magnitude.
Both of these relationships are based upon empirical data,
and both depend on judgments of what fraction of the total
fault length will rupture during the maximum event. The
Wyss (1979) relationships also depend on estimates of the
down-dip depth of rupture during the maximum event. Aside
from the Megathrust and Benioff zone, where the geometry of
the rupture plane is estimated with reasonable certainty
and where the maximum historic earthquakes govern our
estimates of the magnitude of the maximum event, the
estimates for maximum magnitude on the remaining seismic
sources are based upon the surface rupture length criteria
of Slenunons ( 1977) . The Wyss ( 1979) relationships were
applied to these sources for comparison purposes. In using
the Wyss (1979) relationships, the maximum depth to which a
rupture plane on the local faults could extend was assumed
to be 20 km for the long sources and 10 km for the shorter
on-site sources.
The specific assumptions made for each fault will be
discussed in the following section.
2.3.3 Seismic Source Evaluation
In the following section .we review the. changes and modifi-
cations to the data, which were originally presented in
Table 3, "Earthquake Source Characterization," of the
WCC (1980a) report. These changes and modifications were
incorporated into Table 1 of this report.
The wee (1980a) report represented a preliminary regional
evaluation of potential seismic sources that might affect
the Bradley Lake Project~ its primary goal was to establish
which sources were of priority concern to later, more
Woodward· Clyde Consultants
2-13
detailed studies. These preliminary results largely were
taken from the regional OASES report (WCC, 1978).
We have reviewed the wee ( l980a) report, and it 1.s our
judgment that several of the regional potential seismic
sources presented in that report can be eliminated from
further consideration. That judgment was based on the
estimated maximum site ground acceleration as presented in
the wee (l980a) report and a cutoff value equal to or less
than 0.05 g. This procedure eliminated the Volcanic Chain,
Bruin Bay fault, Castle Mountain fault, Contact fault,
Johnstone Bay fault, Placer River fault, Kenai Lineament,
and the Offshore Deformed Zone from further consideration.
Five potential earthquake source faults remain after
applying the screening criterion: l) the Border Ranges
fault, located about 1.5 km west of the powerhouse site at
its point of closest approach; 2) the Eagle River fault,
which dips beneath the dam site with an estimated closest
approach of l-l/2 to 2 km; 3) and 4) the Bradley River and
Bull Moose faults, located between the powerhouse and the
dam site; and 5) the subduction zone which underlies the
entire region (see Table 1).
The random source seismic event, or floating earthquake is
that earthquake that has the chance of occurring anywhere
in the upper crust including beneath the Bradley Lake
site. In evaluating the seismic hazard potential of a
given site the random source event is generally the largest
historic earthquake within the region that cannot be
justifiably assigned to a potential earthquake source
fault. Given the historic seismicity in the Cook Inlet and
Kenai Peninsula region of southern Alaska a Ms 5. 5 event
was assigned as a maximum random source by wee (1978
Woodward-Clyde Consultants
2-14
and 1980a). At the present time larger events would
be expected to be associated with mapped faults in the
region. This maximum magnitude for a random source event
is presented in Table 1. However, for the purpose of this
investigation, the random source was not included in the
analysis. Considering the nearby location of the Border
Ranges, Eagle River, Bradley River, and Bull Moose faults,
with respect to the site, these faults would probably
over-shadow any effect from a random source event should
these faults prove to be active earthquake generators. If
these local faults prove to be inactive with additional
field investigations, then the random source should be
considered in the evaluation of the seismic design for the
project.
Table 1 summarizes the estimated characteristics of the
five potential earthquake sources considered in the
analysis. For cases in which the data were adequate, the
characteristics were based on geological or seismological
data; otherwise, they were based on our judgments, taking
into consideration as many variables as possible. These
source characteristics form the basis for the selection of
the maximum design earthquakes and are key input data into
the probabilistic assessment of the seismic exposure of the
site. Detailed descriptions of these potential sources
are presented in wee {1979 and 1980a). The following
subsections address those characteristics of individual
sources for which new information has been obtained since
the wee {1980a) report was written or for which additional
clarification is necessary for the purposes of this study.
The Subduction Zone
As noted in earlier reports {WCC, 1980a; Lahr and Stephens,
1981), the subduction zone lying beneath the Bradley
Woodward· Clyde Consultants
2-15
Lake site is a major and frequently active source of
earthquakes. The data obtained from the Bradley Lake
Microseismic Network (Lahr and Stephens, 1981) have
resulted in refinements to our estimates of the geometry of
the. subduction zone and its location with respect to the
Bradley Lake site. The earthquake data shown in Figure 10
clearly define the subducting slab. Several features of
the zone are noted in Figure 10. Using the criteria of
Davies and House ( 1979), we have divided the subduction
zone into two portions--the main thrust zone (Aleutian
Megathrust) that is the source of periodic very large
earthquakes, and the Benioff zone that is the source of
smaller and more continuous earthquake activity.
The Aleutian Megathrust is the source area of the magnitude
Ms 8-1/2 1964 earthquake, which ruptured along the inclined
plate boundary from the eastern Gulf of Alaska to the
vicinity of Kodiak Island. The 1964 rupture zone is
indicated in Figure 10 by the "x"ed hachures. The western-
most extent of the zone is not clearly known but was
estimated on the basis of aftershocks (Plafker, 1971). The
pattern of major earthquake activity reviewed by Davies and
others ( 1981) suggests that the Mega thrust Zone in this
area produces earthquakes of a size similar to the 1964
earthquake approximately every 160 years. Th~ maximum
magnitude of the Aleutian Megathrust is estimated to be
~1s 8-1/2 for engineering purposes.
The Benioff portion of the subduction zone is restricted to
the upper 20 km of the descending Pacific Plate. Within
the Benioff zone in the depth range of 40 to 70 km, no
earthquakes larger than about Ms 7. 5 are known to occur.
Therefore, a maximum magnitude of 7-1/2 is estimated for
the Benioff zone. Slightly larger earthquakes may be
Woodward· Clyde Consultants
2-16
possible at depths greater than 70 km due to increased
shear modulus in that portion of the Pacific Plate.
The boundary between the two zones is a poorly defined
transition region. On the basis of the hypocentral
pattern, the closest point of approach of the Benioff
zone is considered to be the locus of increased down-dip
seismicity, at a distance of about 30 km from beneath
the Bradley Lake site. For all practical purposes, the
boundary 1s located directly beneath the site; thus the
closest approach of the megathrust and the Benioff zone to
the site is essentially the same.
The Border Ranges Fault
The northwest front of the Kenai Mountains forms an abrupt
topographic lineament that extends from northeast of
Anchorage nearly the length of the Kenai Peninsula.
MacKevett and Plafker (1974) mapped this lineament as part
of the Border Ranges fault, which is a northwest dipping
high-angle reverse fault that juxtaposes Paleozoic con-
tinental sediments on the northw~st over Mesozoic deep
water marine deposits on the southeast. MacKevett and
Plafker (1974) reported that this fault zone is probably an
ancient Hesozoic subduction zone. Tectonic activity has
gradually diminished on the ancient subduction zone as
crust has been accreted to the continent and the active
subduction zone has migrated southeastward to its present
location in the Aleutian trench.
Magoon and others, 1976, have postulated that the Border
Ranges fault, as exposed on the southwest end of the Kenai
Peninsula, ties in with the Sterling fault located in the
Kenai lowlands (see Figure 1 and 5). They base their
interpretation on oil well and geophysical data taken from
Woodward· Clyde Consultants
2-17
the Tertiary deposits that underlie the Kenai lowland.
However, the location of the old Mesozoic suture line
through the Kenai area is still not certain, and its
association with the Sterling fault remains speculative.
As yet, no offsets within Quaternary sediments along either
postulated trace of the Border Ranges fault are reported in
the literature. However, no detailed systematic investiga-
tion geared toward resolving seismic hazard problems has
been applied to the Border Ranges fault. John Kelley of
the USGS, Anchorage office (oral communication, 1981) has
suggested that the topographic lineament associated with
the northwestern Kenai Mountain front may be, in fact, a
younger high-angle oblique-strike slip fault that truncated
and is on or near the ancient Mesozoic subduction zone.
'
If this hypothesized younger Border Ranges fault is
the cause of the sharp topographic lineament along the
Kenai Mountain front, it would be consistent with other
structural features in the Cook Inlet region. The Kenai
Mountains' northwestern front appears to be the south-
eastern boundary of the subsiding tectonic basin forming
the Cook Inlet area. This basin has accumulated several
thousand feet of Tertiary sediments and appears to be
bounded on the northwest by the Bruin Bay and Castle
Mountain faults. Thus, the Border Ranges and Bruin Bay-
Castle Mountain faults could be complimentary structures on
either side of a geologically young basin.
The oblique right-lateral Castle Mountain fault is the only
one of the three major faults along which Quaternary
displacements have been reported and on which we have
reasonable seismic geology data. By examining the charac-
teristics of the Castle Mountain fault reasonable estimates
can be made about the characteristics of the Border Ranges
Woodward-Clyde Consultants
2-18
fault. Holocene evidence for displacement on the Castle
Nountain fault is located in the Susitna Lowlands, north
of Cook Inlet (see Figure 1), where treriching across a
topographic lineament revealed evidence suggesting that 2.5
meters of apparent dip-slip separation has occurred within
the past 255 to 1700 years (Detterman and others, 1974). A
magnitude Ms 7 earthquake in 1933 has been associated
with the Castle Mountain fault (WCC, 1980b). In estimating
a maximum magnitude event on the Castle Mountain, we have
estimated a rupture length of approximately 20 percent of
the total fault length of 500 km, or about 100 km, based on
fault-rupture-length versus total-fault-length relationship
development by Slernmons (U.S. Nuclear Regulatory Commis-
sion, 1981) . This is in reasonable agreement with the 80
km length of Holocene rupture reported by Evans and others
( 1972) for the Castle Mountain fault across the Susitna
Lowland. A rupture length of 100 km leads to an estimate
of Ms 7-1/2 maximum earthquake for the Castle Mountain
fault (Slemmons, 1977), which appears reasonable when
compared to the historical Ms 7 .0 recorded on the fault.
Estimating the rupture length and style of rupture on the
Border Ranges fault remains speculative because of the lack
of adequate geologic information on the fault. If one
hypothesizes that the Border Ranges fault is part of the
same tectonic system as the Castle Mountain fault and that
its characteristics are similar to that of the Castle
Mountain fault, then, lacking any conflicting data, a
rupture length of 100 km seems reasonable for the Border
Ranges fault. Using this rupture length and empirical
relationships of fault rupture length related to earthquake
magnitudes (Slernmons, 1977), a maximum earthquake magnitude
of Ms 7. 5 was estimated for the Border Ranges fault.
Woodward-Clyde Consultants
2-19
Eagle River Fault
The Eagle River fault is also an ancient Mesozoic, north-
dipping thrust fault. No Quaternary activity has been
reported in the literature for the fault. Unlike the
Border Ranges fault, there is not a striking topographic
lineament associated with the Eagle River fault. However,
because of its mapped length (over 750 km) and its tectonic
setting, we were conservative in assuming that it could be
active until sufficient data are accumulated to warrant
judgment about its state of activity.
Although little is known about the present seismicity of
the Eagle River fault, if it is active, its location,
topographic expression, and mode of displacement suggest it
is not a part of the same tectonic system as the Castle
Mountain and Border Ranges fault. Thus analogies between
the Eagle River fault and the Castle Mountain fault do not
seem appropriate. However, it is our judgment that the
maximum magnitude earthquake on the Eagle River is probably
no higher than the maximum magnitude earthquake on either
the Border Ranges or Castle Mountain fault. Therefore a
Ms 7-1/2 was assigned to the Eagle River fault.
On-Site Faults
The information regarding the characteristics of the Bull
Moose and Bradley River faults is essentially unchanged
from its original characterization in the Woodward-Clyde
(1980a) report.
Woodward·Ciyde Consultants
3.0 DESIGN GROUND MOTIONS
3.1 Maximum Earthquake
At the 8 October 1981 meeting in Anchorage, the COE
reviewed the maximum magnitudes estimated for the various
potential earthquake sources presented in Table 1 and
discussed in Section 2.3.3. The following earthquakes were
adopted as design maximum earthquakes by the COE for
the purposes of this study: a magnitude 8-1/2 Ms earth-
quake occurring on the Megathrust zone beneath the site at
a closest distance of about 30 km; and a magnitude 7-1/2
Hs earthquake occurring on a shallow crustal fault within
a distance of 3 km from the site. Data do not currently
exist that would allow for a detailed estimate of the
maximum earthquake magnitudes for the Border Ranges and
Eagle River faults. In fact, the data base does not allow
for an appropriately informed judgment as to whether or not
these faults should be considered active. In view of the
above, we believe that the 7-1/2 Ms estimated maximum
earthquake magnitude assigned to these shallow faults
is a conservative estimate. Considering the apparent
conservatism in the selection of the design maximum
earthquakes, the COE decided that the design ground motions
should be based on best-estimates (i.e., mean or average
values) of the ground motions for each earthquake.
The following are our estimates of mean values of peak
horizontal ground acceleration, velocity, and displacement,
and the duration of strong ground shaking (significant
duration) on rock for the design maximum earthquakes:
Woodward-Clyde Consultants
3-2
Design Peak Peak Peak Significant
Maximum Acceleration Velocity Displacement Duration*
Earthquake (g) (em/sec.) (ern) (sec.)
(Ms)
Magnitude 8-1/2 0.55 55 40 45
on Megathrust
Magnitude 7-1/2 0.75 70 50 25
on shallow
crustal fault
* Significant duration is defined as the time required to
build-up from 5 percent to 95 percent of the energy of
and accelerogram (See Dobry and others, 1978).
Estimated mean horizontal response spectra (damping ratio
of 0.05) of ground motions on rock for the design maximum
earthquakes are presented in Figure 13. Based on the
results of this analysis, if the nearby shallow crustal
faults such as the Border Ranges or Eagle River faults are
active, as we have assumed they are, then these potential
earthquake sources appear to dominate the response spectra.
Should the local faults prove to be inactive through
additional field investigations, then the respons.e. spectra
for the site would shift to the level defined by the 8-1/2
Ms earthquake occurring on the Megathrust.
3.2 Operational Base Earthquake
At the 8 October 1981 meeting, the COE indicated that it
was not an appropriate time to select an OB earthquake. To
assist the COE in making initial assessments of possible
ground motions for an OB earthquake, response spectra, are
provided corresponding to one-half of those for the design
maximum earthquakes. These response spectra are shown in
'·~ ,,
':]
l
~}
Project:
Project No.
-o, -co
CJ)
c:
0
·~ e
QJ
Q)
(,J
(,J
<(
2.5
2.0
1.5
"§ 1.0 .....
~
0.
CJ)
0.5
0
0.01
ap = 0.75g
ap = 0.55g
0.03
BRADLEY LAKE
148448
0.1
Megathrust
0.3
Period (sec)
Damping Ratio= 0.05
Nearby Shallow
Crustal Fault
MEAN RESPONSE SPECTRA FOR MAXIMUM EARTHQUAKES
Fig.
13
WOODWARD-CLYDE CONSULTANTS
Woodward-Clyde Consultants
3-3
Figure 14. This approach of using one-half of the maximum
earthquake ground motion for an OB earthquake has been used
for some critical facilities (e.g., for nuclear power
plants). However, the COE may not wish to base the OB
earthquake on the same methods used for nuclear power
plants. Therefore, a seismic exposure analysis was
completed and the COE can utilize the results of this
analysis in selecting an appropriate OB earthquake once an
acceptable level of risk established for the Bradley
Lake project.
I 1.25 I ---l
I I I I I I I I I 1 I ,. I I I I I I I I I I I I I 1 1 1 1.:
Project:
Project No.
2
co
Cf)
1.00
c· 0.75
.2 .....
E
~
Q)
(.)
(.)
<(
E 0.50 .....
rrl a.
Cf)
0.25
0
0.01
a~= 0.375g
ap = 0.275g
0.03
BRADLEY LAKE
148448
0.1
Megathrust /
0.3
Period (sec)
Damping Ratio= 0.05
Nearby Shallow
Crustal Fault
ONE-HALF OF RESPONSE SPECTRA FOR MAXIMUM EARTHQUAKES
Fig.
14
WOODWARD-CLYDE CONSULTANTS
Woodward-Clyde Consultants
4.0 SEISMIC EXPOSURE ANALYSIS
The objectives . of the seismic exposure analysis are: ( 1)
to provide estimates of the likelihood of exceeding various
maximum ground acceleration levels at the Bradley Lake dam
and powerhouse sites during a design period of 100 years;
and (2) to provide estimates of the relative contributions
of various local and regional earthquake sources to
the seismic exposure at the site. The results of this
analysis will be useful to the COE as an aid in selecting
the operational basis earthquake. We emphasize that the
results of this analysis should be viewed as best-estimate
values in terms of useful comparisons of the impact of
various sources on the seismic exposure of the site.
As discussed in Section 3.0, the response spectra for
maximum earthquakes are controlled by the local faults.
These local faults were conservatively assumed to be
active for the purposes of this study because data in
the available literature are not sufficient to resolve
questions of the potential fault activity. In addition to
aiding in the selection of an OB earthquake, the results of
the seismic exposure analysis also provide a means of
ranking the earthquake sources with respect to their
contribution to seismic exposure of the sites. This
ranking of earthquake sources provides a basis for priori-
tizing the local faults for future field investigations.
Through such field investigations it may be possible to
resolve the questions regarding the activity of the local
faults and may allow the COE to eliminate some of those
local faults from further consideration.
The seismic exposure analysis was largely based upon
results generated by the computer program SEISMIC-EXPOSURE,
4-2 Woodward· Clyde Consultants
which uses seismic source geometry information and regional
historical seismicity data as inputs. Results are ex-
pressed in terms of the likelihood of exceeding various
maximum ground acceleration levels at the Bradley Lake dam
and powerhouse sites for design time periods of 40 and 100
years. The analysis was completed for the 100 year design
period selected by the COE for the Bradley Lake project.
For comparison purposes, the analysis was also completed
for a 40 year design period which is typical of many other
types of engineering projects.
A summary of the methodology for applying the seismic
exposure analysis is presented in Appendix A.
4.1 Seismic Exposure Inputs
It is useful to briefly review some of the key input
parameters to the analysis and also review the uncertain-
ties and limitations associated with these parameters.
Developing inputs for the analysis required the identifi-
cation of the potential seismic sources and the estimation
of fault plane geometry for each source.
The input parameters for ·the five potential seismic sources
identified in Table l are discussed in Appendix A and
illustrated in Figure A-2. Each source was modeled in
3-dimensions as a series of planar segments with orienta-
tions based upon available geologic data. Faults of short
length, such as the Bradley River fault, were modeled with_
only one segment, while long faults that have a number of
curves and bends in their mapped traces were modeled with
several segments. Each segment was assigned a number for
reference purposes (e.g. Eagle River fault, Segment 2) .
4-3 Woodward-Clyde Consultants
Earthquake recurrence and maximum magnitude earthquake
assessments for each source were required. It was also
necessary to specify fault rupture length versus earthquake
magnitude information and to describe ground motion
attenuation. The details on the definition and implemen-
tation of these parameters are presented in Appendix A.
4.2 Estimate of Total Seismic Exposure at Sites
Figures 15 and 16 summarize the seismic exposure results at
the dam and powerhouse sites in terms of estimates of the
probability of exceedance of various ground acceleration
levels over 40-and 100-year design time periods.
The dashed set of curves in Figures 15 and 16 represents
the probability of exceeding a given acceleration level
from all identified seismic sources. The solid set of
curves represents the probability of exceedance of all
sources except the closest segment of the Megathrust
to the site (Segment 2). The greatest contributions
to the seismic exposure of the site is from Segment 2 of
the Megathrust (see Table A-1 and A-2). These tables are a
breakdown of the percentage contributions by source to the
accelerations at various probabilities shown in Figures 15
and 16.
The historical seismicity data used for this study (from
the wee I [1978 and 1981]) of the Bradley Lake project is
dominated by the 1964 Ms 8.5 earthquake and its large
number of aftershocks, which occurred on the Megathrust.
To examine the effect of the Megathrust on the analysis
results at the sites, a sensitivity study was made in which
Segment 2 of the Megathrust was excluded (see Figures 15
and 16). However, segments 1 and 3 of the Megathrust
Project:
Project No.
C1! u c: .,.
"0
C1!
C1! u
X w
0
;::-
ii
"' .0 e
0..
C1! > ·;::;
~
E
:::J u
0.5
0.1
0.05
0.01
" ' ' ' \ \
\ \
\ \
\ \
\ \
\ \
I \
\ \
\ \
\ \
I \
\ \
I \
I I
I \
I I
\ I
I \
I I
I \
\ I
\ I
\ \
\ \
\ \
\ \
\ \
\ \
\ \
\ \
\ \
\ I
L~
\ \
\ \
All sources
All sources except the
megathrust (Segment 2)
\ "'-----"'"? 100 year Period of Interest \
\
\
\
I
\
I
I \
I
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
' \
' ' '
0.0 0.1 0.2 0.4 0.6 0.8 1.0
Maximum Horizontal Acceleration (g)
BRADLEY LAKE
14844B
ESTIMATES OF THE PROBABILITY
OF EXCEEDANCE AT DAM SITE
Fig.
15
W00DWARD-CL YDE CONSULTANTS
Project:
Project No.
"' <J c:
"' '0
"' "' <J
)( w
0
~
ii
"' .c
0
~
"' ·~
"' E
::J u
0.5
0.1
0.05
'
40 Year
All sources
"-.. -... -.. )~IL_5P.U!C:!!_5.!!)(C:ePLt!'Hl ..
megathrust (Segment 2)
Period of ln~erest · ~----'H-"'k--.....1
I
0.01
0.0 0.1 0.2
BRADLEY LAKE
148448
0.4 0.6
I
I
\
\
I
I
I
I
\
\
I
I
I
\
\
\
\
\
\
\
\
\
\
\
\
\
\
' \
\
\
\
\
\
\
\
\
\
\
\
\
\
\
' \
0.8 1.0
Maximum Horizontal Acceleration (g)
ESTIMATES OF THE PROBABILITY
OF EXCEEDANCE AT POWERHOUSE SITE
Fig.
16
WOODWAR D-·CL YDE CONSULT ANTS
4-4 Woodward-Clyde Consultants
were included as they contribute little to the exposure
analysis. Results show that given their magnitudes and
distances from the site, earthquakes occurring on Segments
1 and 3 of the Megathrust consistantly contributed little
to the seismic exposure at the site for both time periods.
The results presented in Figures 15 and 16, are summarized
below for maximum horizontal accelerations for all sources
combined and at probability of exceedance levels of 50
percent, 30 percent, and 10 percent for the design time
period of 100 years:
Maximum Horizontal Acceleration
Probability of Powerhouse Site Dam Site
Exceedance 100 xears 100 years
50% 0.37 g 0.37 g
30% 0.43 g 0.44 g
,'"'"~~,
10% c::. 58\ g 0.58 g
These results illustrate that the total seismic exposure
levels are similar for both the dam and powerhouse sites
for a 100-year time period. If the Mega thrust were
excluded, the values of the acceleration levels for both
sites would be reduced by about 0.10 to 0.16 g.
4.3 Relative Earthquake Source Contributions to Total
Seismic Exposure
Tables A-1 and A-2 summarize percent contribution of
various sources to the total seismic exposure levels
at the site. These data are summarized below as a ranking
of the closest segment of these sources from each of
the sites. The ranking is ordered from the most dominant
source contribution to the least dominant source cont.ri-
bution:
4-5
Powerhouse Site/Dam Site
Mega thrust
Benioff Zone
Eagle River Fault
Border Ranges Fault
Woodward-Clyde Consultants
Refer to Figures l and 3 and Figure A-2 for identification
of these faults. These results show that the closest
segments of regional faults impact the sites base<;] upon
their respective proximity to the sites, and the estimated
levels of seismicity mode led on the faults. The Bull
Moose and Bradley River faults are not indicated as being
significant to the total seismic exposure at the sites. In
terms of possible future field geologic studies on local
faults that affect the design maximum earthquake (discussed
in Section 3 .0), the relative ranking of the sources
indicate that the Eagle River fault has more relative
impact on the seismic exposure of the site than the Border
Ranges fault. This suggests that future investigations
should address the Eagle River fault first before concen-
trating on the Border Ranges fault.
4.4 Earthquake Magnitude Contribution to Acceleration
Levels by Source
From the earthquake source ranking, insights can be gained
as to what earthquake magnitude ranges contribute most to
the seismic exposure at the sites due to the earthquake
sources. These tabulations have been made and are included
in Tables A-3 and A-4. A summary is presented below of
these percent contribution data for acceleration levels of
engineering significance ( 0. 5 g and greater). Since the
nearby shallow crustal faults govern the response spectra
presented in Section 3.0, only these sources are considered
here.
Source*
Eagle River Fault
Segment 2B
Border Ranges Fault
Segment 2
Eagle River Fault
Segment 3A
Woodward-Clyde Consultants
4-6
Dominant Earthquake
Magnitude Ranges (Ms)
Powerhous Dam Site
6-1/2 -7-1/2
6-1/2 -7-1/2
5-1/2 -6-1/2
*see Figure A-2 in Appendix A.
This tabulation indicates which of the earthquake magnitude
ranges associated with the local crustal faults contributes
most to producing higher accelerations (in therange 0.5 g
and higher) at the sites. As can be seen in Tables A-5 and
A-6, at lower acceleration levels, the smaller magnitude
earthquakes begin to have an important contribution to the
seismic exposure because of their more frequent occurrence.
Woodward-Clyde Consultants
5.0 LIKELIHOOD OF ON-SITE FAULT RUPTURE
5.1 Introduction
The purpose of this section is to present the basis used
for assessing the possibility of fault rupture at the
facilities of the Bradley Lake project using data available
in the literature and project specific reports prepared
for the COE. The faults that might intersect the power
tunnel and the darn site and that are considered in this
assessment are shown on the reconnaissance geologic map of
the Bradley Lake area (Drawing Number 1, wee, 1979) and are
summarized in Figure 3. These faults are collectively
evaluated in this study to provide an estimate of the
amount of possible slip and the probability of occurrence
of slip during the 100-year expected life of the project.
The largest of the faults that intersect critical planned
facilities are the Bradley River and Bull Moose faults.
Both of these faults appear prominently in the topography
and on aerial photographs, but data that might be used to
evaluate their potential for future fault rupture do not
exist. The other faults and lineaments on the geologic map
are probably smaller faults than the Bradley River and Bull
Moose faults and are likewise without data regarding future
fault rupture potential.
In order to provide an assessment of the potential surface
rupture hazard, given the lack of data on specific faults,
an empirical and probabilistic approach has been developed.
The approach utilizes existing data on the seismicity,
structure, and tectonics of the Kenai Peninsula in order to
provide guidance for our attempt to compile compatible
empirical data on historical faulting from throughout the
world that is consistent with geologic. conditions at
Woodward-Clyde Consultants
5-2
Bradley Lake. These empirical data are
estimates of possible slip on the faults
used to make
in the Bradley
Lake area should they be active. The evaluation then
incorporates the probability of occurrence of various
magnitude earthquakes based on regional seismicity combined
with subjective judgments to yield a probability of fault
rupture.
5.2 Assumptions
In evaluating the fault rupture potential, assumptions
about the behavior of the faults at Bradley Lake had to be
made in order to accommodate uncertainties inherent in the
data base. Of particular importance is whether or not
the faults are active~ for example, it is assumed that
the Border Ranges and the Eagle River faults have some
possibility of being active and that the Bradley River and
Bull Moose faults could respond as secondary ruptures to a
large earthquake on either the Border Ranges or Eagle River
faults. It is also assumed that the Bradley River and Bull
Moose faults could generate their own earthquakes, and thus
sur face fault rupture. Earthquakes of magnitude Ms 5. 5
or greater, up to the maximum magnitude assigned to each
fault, are considered capable of surface rupture in this
analysis (Table 1). The probability of occurrence of the
various magnitude earthquakes on each of the faults is
based on the same data base that was used in the seismic
exposure analyses (Section 2.3.4). Finally, it is assumed
that the smaller unnamed faults around the Bradley Lake
site could be subject to slip as secondary faulting
resulting from earthquakes on either the Bradley River or
Bull Moose faults~ these smaller faults are assumed not to
be capable of generating earthquakes independently.
Woodward-Clyde Consultants
5-3
Inherent in this evaluation of the likelihood of fault
rupture is the assumption that the characteristics of
secondary or branch fault rupture seen during events ~n
other parts of the world are similar to the characteristics
that will occur at Bradley Lake in the event of a future
earthquake on the seismic sources of concern.
5.3 Methodology
The faulting evaluation is presented in two parts. The
first part is a deterministic evaluation of whether or not
faulting can reasonably be expected on faults at the
site and the evaluation of the amount of slip possible.
Empirical data collected for historical earthquakes
throughout the world were analyzed for trends in the data
that would help make estimates of faulting at Bradley
Lake. The second part is an estimation of the probability
of occurrence of faulting given certain geologic and
seismologic conditions. The methodology for each of
these evaluations is presented in Appendix B.
5.4 Results of On-Site Faulting
The analysis of on-site faulting relies on a series of
conditional steps summarized in Appendix B• Both the
amount of displacement and the likelihood of occurrence
were evaluated.
The evaluation of amount of offset was based on several
empirical relations developed from historical earthquakes
throughout the world. The empirical data and the suggested
bounding conditions are presented ~n Figures 17, 18,
and 19.
Figure 17 is a plot of earthquake magnitude on a main fault
versus the maximum observed distance of secondary faulting
-E
~
(!) z
~
.J
::::>
<(
u.
>-a:
<(
0 z
0
(.)
w
tfJ
0 ....
w
(.) z
<( ....
tfJ
0
~
::::>
~ -X
<(
~
Project:
Project No.
100 ,.---,.,-,........., ,,r--.--'1.---..,-........... ,,-.,.---, ,,r-r,-r--r,,-.. ,1-.,....-, ,,r-r,-r--r,,-r-r ,,-.,.--, ,,r-r,-r-r,,-r-r ,,-.,.--, ,r-r-,r-r,l
Associated with Reverse and ~ 6
........... Oblique Reverse Faulting
....... .? • Associated with Strike-
6 • ~ Slip and Oblique Strike-
6 6 .......... Slip Faulting • ..........
10 --6 • .......... A
6 6 • .......... • • . .......... 6 • • • 6 • • .......... • ~q 6. • 6 ~(JI) • • • 61 • • • • ~y • • ~'Ve 6 61 • • 6 I • .......... 1.0 • • ~-~ • • 6 • • • • • ·• • • •
• A • •
0.1 --
I I I I I I I I I Ill 1l111 J 1 I I 0 I I I I
8.0 1.0 6.0 5.0
EARTHQUAKE MAGNITUDE (M 5 )
NOTE: Data points with question marks are discussed in Appendix B.
BRADLEY LAKE
148448
DISTANCE TO SECONDARY FAULTING
VERSUS MAIN EARTHQUAKE (M 5 )
Fig.
17
WOODWARD-CLYDE CONSULTANTS
140 ,. I I I I I I I I I I I I I I I I I I I I I I I I I I I I
' 120 ,... • ' -
" -" E u " -100 f--(.!)
" z
1-' ....1
:::::>
<t }. u.. 80 f-->-• " a:
<( • "0<$)'-0 z ~ 0 • u 60 f-0~ -w ~ t/) Q)')..
z • ·~ 0 • ~,.,
t-(9
0... . " ....1 40 --
t/)
• " .
• • • " •
20 -• • '\. -
• •• • '\. • • •
_j_Jt!l I I l I I I I
~
I I
8.0 7.0 6.0 5.0
EARTHQUAKE MAGNITUDE (M 5 )
I
Project: BRADLEY LAKE SLIP ON SECONDARY FAULT VERSUS MAIN Fig.
Project No. 14844B EARTHQUAKE (Ms),(STRIKE-SLIP FAULTING) 18
WOODWARD~CL YOE CONSULTANTS
I 80
'
I I I I I I
\ c. Associated with Reverse and
1-~\ Oblique Reverse Faulting ....J 70 ::J -
<( \ • Associated with Strike-Slip
LL \ and Oblique Strike-Slip Faulting z \ \
<( \ ~ 60 ~ \ \ -
z \ 0 \ \ a.. \ ....J .\ (/) \
~ 50 c. -
\ \ ::J c. \~ ~ \~ -\L X \~ <( :P
~ 40 r--\~ \~ -
LL \U, \ c.?
0 \~
?f?. Vo ,~
(/) • • \ <(
30 f-\"j1~ VU'& -a.. ~ ,o
....J •\'00 '(}
(/) / "j1 c. c. '"j1.)-' ~ >-• /o ',o1--a: "1--', . <( 20 ~· • c. 0 • '-'-...._ SData Boundary -
z
0 .........._ ~ ''...o, Curves
(.) • • w ....._ ...__ ..........
(/) ~· • ---• •• ----10 ------·-• c. .. c. • c. •• j:> • c. ••
0 I I I I I I
0 5 10 15 20 25 30 35
MAXIMUM DISTANCE TO MAIN FAULT (km)
NOTE: Data points with question marks are discussed in Appendix B.
Project: BRADLEY LAKE SECONDARY SLIP AS PERCENT OF MAXIMUM Fig.
Project No. 148448 SLIP ON MAIN FAULT VERSUS 19 MAXIMUM DISTANCE TO MAIN FAULT
WOODWARD-CLYDE CONSULTANTS
Woodward· Clyde Consultants
5-4
from the main fault trace. This graph was used to screen
out the smaller earthquakes on the main fault that, because
of distance, would have a low likelihood of resulting in
secondary faulting at the Bradley Lake site.
The other two graphs present two sets of empirical data
that aid in evaluating amount of slip possible on secondary
faults. Figure 18 is a plot of main-fault earthquake
magnitude on strike-s lip faults versus measured slip on
associated secondary faults. Reverse faults occasionally
have associated secondary slip as much as three times
greater than for strike-s lip faults. Figure 19 presents
the offset on a secondary fault as a percentage of the
max1mum slip on the main fault plotted against distance
from the main fault. This is the same format as a graph
presented by Bonilla (1970) except that the data set has
been limited to the styles of faulting applicable at
Bradley Lake, and the data have been brought up to date.
In order to apply this relationship to the problem of
faulting at Bradley Lake, the maximum slip on the main
fault is estimated for various magnitudes using data from
Slemmons (1977) and a regression of slip as a function of
magnitude. The data that have been included for slip
determinations relate to strike slip, reverse slip,
and reverse-oblique slip faults.
expressed in the following formula:
logS= a+ b (Ms),
This relationship is
where "S" is the slip in centimeters and Ms is a selected
magnitude. Based on the existing data base, "a" equals
-3.106, and "b" equals 0.481. Applying the resultant slip
on the main fault to the percent derived from Figure 19
yields the subsidiary slip estimate.
5-5 Woodward-Clyde Consultants
The amounts of secondary slip corresponding to various
earthquakes on different faults at Bradley Lake are
presented in Table 2. The corresponding probabilities of
occurrence of each slip event are also listed in Table 2.
These probabilities are conditional: they depend on
estimated earthquake recurrence and on several conditions
that were assigned subjective values of probability. The
probability analysis is based on a logic tree format that
is described in Appendix B.
The evaluation of potential future fault rupture at Bradley
Lake has been made for the Bradley River, Bull Moose, and
minor faults on the basis of various assumptions about
fault activity and earthquake recurrence. Because there are
several possible causes of rupture on each of these faults,
each potential cause was evaluated for the amount of
resultant slip and probability of occurrence. These data
are presented in Table 2. The sum of the probabilities
in each of the three columns provides estimates of the
probability of surface faulting on each of the local
faults. By adding the number of events for each possible
rupture case, the summed probability of rupture at the
power tunnel occurring in 100 years on the Bradley River
fault is approximately 4xlo-3: the summed probability
on the Bull Moose fault is also approximately 4xlo-3.
If the power tunnel, lake tap, slide gate, or dam are
located astride a mapped minor fault, the summed proba-
bility of rupture at these facilities is estimated to be
approximately 2xlo-4 during a 100-year period.
TABLE 2
SUMMARY OF POSSIBLE SECONDARY SLIP OCCURRENCES
MAIN FAULT ( 1) SECDNDARY FAULT RUPTURE SECONDARY FAULT RUPTURE SECONDARY FAULT RUPI'!JRE
EARTHQUAKE (Ms BRADLEY IUVER BUlL MOJSE MINOR FAULT
Sli;e (an) (2)
No. of Events ·No. of Events No. of Events
;eer 100 Years Slip (em) ( 2) per 100 Years Slip (em) (3) per 100 Years
BORDER RANGES
7.5 98 120 &do-4 98 174 8xlo-4 NA 66-96 4xlo-s
7.0 78 64 3xlo-4 78 100 3xlo-4 NA 35-55 7xlo-6
6.5 58 36 2xlo-4 58 57 4xlo-4 NA 20-31 8xlo-6
6.0 NA J::.1A NA 38 28 lxw-4 NA NA NA
EAGLE RIVER
7.5 300 143 lxlo-3 300 117 lxlo-3 NA 79-64 7xlo-5
7.0 200 82 7xlo-4 200 67 6xlo-4 NA 45-37 4xlo-5
6.5 130 47 6xlo-4 130 38 4xlo-4 NA 26-21 2x1o-s
BRADLEY RIVER
6.5 105 105 7xlo-6 58 58 3xlo-7 58 58 3xlo-7
6.0 60 60 lxlo-5 38 33 4xlo-7 38 33 4xlo-7
5.5 34 34 2xlo-5 18 19 3xlo-7 18 19 3xlo-7
BULL M:X)SE
(1)
(2)
(3)
(4)
6.0 38 33 2xlo-7 60 60 7xlo-6 38 33 2x1o-7
5.5 18 19 2xlo-7 34 34 Bxlo-6 18 19 2xlo-7
Earthquake magnitudes that have a likelihood of causing secondary faulting are screened by comparing
distance to main fault to earthquake IllCignitude in Figure 18. ·
The tv.u estimates of slip are based on the oonibination of earthquake magnitude on rrain fault, distance
to rrain fault, and Figures 19 and 20, respectively.
The range of slip estimates is base::l on the am:unt of secondary slip on either the Bull Moose or
Bradley River faults and on Figure 20.
N.A. refers to being not applicable either because the slip cannot be estimated with the given
geologic parameters or the secondary faults are too distant fran the main fault to be affected.
Woodward-Clyde Consultants
6.0 CONCLUSIONS
6.1 Design Earthquakes
To evaluate the characteristics of the maximum design
earthquake for the Bradley Lake site, a review was cor.~
pleted of select literature regarding the regional and
local faults and the tectonic regime in which the site is
located. The literature review indicated that although
pertinent information 1s available, it is general and
regional in nature. Further, this information does not
address seismic design issues of a site-specific or
project-specific nature. The primary data missing from
the available literature 1s adequate information on the
characteristic of the Border Ranges and Eagle River faults
and information from which judgments can be developed
regarding the activity of these faults and the on-site
faults.
Review of the microearthquake data indicated that no
associations could be made between the recorded seismicity
and the mapped faults in the Bradley Lake Project area.
First-motion plots indicate normal-faulting mechanisms on
north to northwest trending failure plans for the low level
of earthquake activity recorded in the Kenai Peninsula.
These results are apparently contrary to the crustal
deformation associated with primary movement of the
regional faults. Shallow crustal seismic activity of less
than 20 km focal depth appears to be distinct from the
subduction zone, which is actively deforming in response to
present day northwest-southeasterly trending compressive
tectonic forces.
The historical earthquake recordings by seismographic
instruments for the Kenai Peninsula have been kept for less
Woodward-Clyde Consultants
6-2
than 75 years. The microearthquake data cover a time
period of 10 years for the regional network and of only
a few months for the more accurate local Bradley Lake
network. Seismicity data collected over these short time
periods are inconclusive with regard to the potential
long-term seismic activity of faults in the project area.
Under these circumstances, an accurate assessment must
ultimately come from an evaluation of the geologic data
along the faults. However, such geologic data do not now
exist. Because of the lack of data in the available
literature on the level of activity of the local and
on-site faults, a conservative position wa!3 taken in
assuming that these faults are active.
Table 1 lists the maximum magnitudes assigned to the
regional, local, and on-site faults. The design maximum
earthquakes for the purpose of this study, were selected
during a meeting with the COE. The design maximum earth-
quakes include a 8-1/2 Ms on the Mega thrust, located 30
km beneath the site, and a 7-1/2 Ms on either the Border
Ranges fault or the Eagle River fault, both located within
3 km of either the powerhouse or the dam sites.
6.2 Ground-Motion Analysis
Estimates of earthquake ground motions on rock at the
site for the design maximum earthquakes are presented in
Section 3.0.
As shown in Figures 13 and 14, the response spectra of
ground motions at the site are expected to be dominated by
events occurring on either the Bradley River or Eagle River
faults, for which the closest approach of the fault plane
to the project facilities is less than 3 km. If this level
of response proves to be unacceptable in the economic and
Woodward-Clyde Consultants
6-3
design evaluation of the project, additional geologic
studies on these nearby faults may produce data that could
decrease or even eliminate the impact of these faults on
the design ground motions. However, it is unlikely
that any additional studies would have any impact on
reducing ground motions below those associated with the
maximum design earthquake of 8-1/2 Ms on the underlying
Megathrust zone.
6.3 Seismic Exposure Analysis
To assist the COE in its evaluation of the design OB
earthquake, we have conducted a seismic exposure analysis.
The results of the exposure analysis are expressed as
probabilities of exceedence of different acceleration
levels at the site. The exposure analysis indicates
that the earthquake sources that dominate contributions to
the probability of exceedence of a given level of ground
acceleration at the site, over a 100-year period, are in
order of their dominance the Megathrust, the Benioff zone,
the F,:agle River fau 1 t I and the Border Ranges fault 1
respectively.·
6.4 Fault Slip Analysis
To address the likelihood of future surface rupture
along the on-site faults that may intersect key facilities
of the Bradley Lake Project, we have reviewed the available
literature on southern Alaska and other areas of the
world in order to develop an empirical data base for
making deterministic and probabilistic assessments of the
problem. Assuming the faults in the project area are
active as has been done in other sections of this study,
the results of the analysis indicate that there is a
possibility of subsidiary fault rupture on the Bradley
River, Bull Moose, and other smaller local faults in
6-4
Woodward-Clyde Consultants
response to displacement on the Border Ranges or Eagle
River faults. In addition, fault rupture appears possible
on the Bradley River and Bull Moose fault acting as
independent structures and possibly inducing displacements
on the minor faults located in the project area. If fault
ruptures were to occur on these faults, it could directly
affect structures such as the dam, the lake tap facilties,
and the power tunnel. The impact of this potential for
fault rupture problem should be addressed in the design of
these structures.
According to the results of our analysis, amounts of slip
could range from as low as 20 em to as high as 300 em on
the Bradley River and the Bull Moose faults, and from 20 to
100 em on the minor faults in the area. All of these
estimates are derived by encompassing the upper end of the
range in emperical data and are thought to be conservative.
The probabilities of these displacements occurring in the
next 100 years are estimated to be in the range of 4xlo-3
to 2xlo-4. These probabilities are largely controlled
by the recurrence data based on historic seismicity.
Because of the uncertainty associated with available
recurrence data, it is our opinion that these bounding
probabilities could be in error by as much as a factor
of 10.
6.5 Limitation of Results
The limitations imposed by the lack of adequate data in the
available literature should be recognized in using the
results of this study. This particularly pertains to
the uncertainties in the level of activity, recurrence
relationships, and characteristics of the local and on-site
faults. Additional field geological studies on the
Woodward-Clyde Consultants
6-5
Border Ranges, Eagle River, and the on-site faults may
provide additional data that could alter the results
and conclusions regarding the design maximum earthquake,
the probability assessment of seismic exposure, and the
potential and amount of possible on-site fault rupture.
Woodward-Clyde Consultants
7.0 RECOMMENDATIONS
Critical data regarding the level of activity, recurrence
relationships, and displacement characteristics of the
Border Ranges, Eagle River, Bradley River, and Bull Moose
faults are either lacking, of poor quality, or of only
a general character. It was not possible to develop
new fault-specific field data within the scope of work
addressed in this report. In order to proceed with the
present exposure study, we made assumptions about fault
parameters based on the regional historic seismicity and
the current state of knowledge regarding the tectonics of
the southcentral Alaskan region. Because of the character
of the existing data base, the results presented in this
report have large associated uncertainties; we have not
attempted to quantify these uncertainties.
As they stand, the results of this study should allow the
COE to assess the general seismic exposure and identify the
sources of greatest risk for their design evaluations. We
make the following recommendations to expand the basic data
base and to provide data to improve and perhaps modify the
present conclusions stated in Section 6.0.
0 If the COE analyses suggest that the design alter-
natives cannot economically accommodate the ground
motion levels estimated in this report, or if the COE
would like to reduce the uncertainties in the data, we
recommend that geological field studies be performed
to address the seismic geologic issues regarding the
Eagle River fault, Border Ranges fault, and the
several on-site faults. These field studies may
produce data sufficient to reduce the size of the
0
7-2
Woodward-Clyde Consultants
local seismic source design maximum earthquake or
possibly even eliminate from further consideration,
such as the Eagle River or Border Ranges faults that
control the local source design maximum earthquake.
If the COE elects to conduct additional seismic
geology field studies on these faults, we recommend
that the approach to the studies involve a series of
tasks to be completed in sequencia! order. These
·tasks include: l) the analysis of existing remote
sensing imagery; 2) a brief aerial reconnaissance;
3) low-sun-angle aerial photography aquisition, and
analysis: 4) helicopter supported ground reconnais-
sance mapping: and 5) trenching across selected fault
traces. The goal of such a program is to collect as
much data as possible on the characteristics of
the fault ( s) utilizing the less expensive regional
investigation methods first. This allows one to be
more effective and efficient in the more expensive,
site specific investigations such as trenching across
fault traces. This approach is considered to be the
current state-of-the-practice and it is generally well
accepted and often required by regulatory agencies.
We understand that at least a portion of task.s l and 2
have been completed during previous investigations by
the COE. We recommend following through with the
remaining tasks of the fault study sequence.
Strong-motion instrumentation should be installed at
the earliest opportunity on the abutments of the dam
site. At least two instruments should be installed to
maximize the likelihood of collecting data in the
event of an earthquake sufficiently strong to trigger
Woodward-Clyde Consultants
7-3
the recorders. Operation prior to construction could
provide strong-motion data useful in comparison with
the selected design motions.
In addition to these recommendations, it would be useful to
continue monitoring the microearthquake network established
for the project at least until construction of the project
begins. These data are useful for interpreting local
stress fields and geological data on the activity of faults
in the site vicinity. These data become more significant
and more representative of the area the longer the survey
continues. In conjunction with continuation of monitoring,
the velocity models for the Kenai Mountains and Kenai
Lowlands should be reviewed and calibrated in order to
improve interpretation of earthquake locations and the
possible associations with faults.
Woodward-Clyde Consultants
APPENDIX A
METHODOLOGY FOR THE
SEISMIC EXPOSURE ASSESSMENT
INTRODUCTION
The objective of the task described in this appendix was to
assess the probability of exceeding various acceleration
levels at the Bradley Lake dam and powerhouse sites
during the design life of the project.
METHODOLOGY
Estimates of the probability of exceeding various levels
of maximum ground acceleration at the sites were made
using the approach illustrated in Figure A-1. As indicated
in that figure, the probability analysis requires the
characterization of certain input parameters. Speci-
fically, these include:
0
0
0
0
0
identification and geometry of seismicity sources:
seismic activity (recurrence and maximum magnitude)
of each source;
relationship between rupture length and earthquake
magnitude:
ground acceleration attenuation relationship; and
selection of anticipated design time period of
interest.
INPUTS
Source Seismicity
• Location and Source Geometry
• Recurrence
• Maximum Magnitude
Attenuation
• Site Conditions
• Transmission Path Conditions
• Magnitude and Distance
Exposure Evaluation Criteria
• Period of Interest
{40 and 1 00 years)
Project:
Project No.
BRADLEY LAKE
148448
-
ANALYSIS
Exposure Analysis
Obtain Cumulative
Distribution Function
based on Contributions
from all Sources
Repeat for Each Site
RESULTS
(For the Dam and
Power House Sites)
Peak Acceleration
Levels
• average return period
in years
• probability of at least
one occurrence in
period of interest
Percent Contribution
Tables for given
Sources
SCHEMATIC DIAGRAM OF THE ELEMENTS
OF THE SEISMIC EXPOSURE ANALYSIS
Fig.
A-1
WOODWAAD-CL YDE CONSULTANTS
Woodward-Clyde Consultants
A-2
Inputs we.re defined in a manner consistent with the present
level of understanding of the tectonic environment of
the sites, the seismology, and the attenuation of ground-
motions. The exposure analysis was conducted with these
inputs in order to calculate the mean number of occurrences
for which a given level of ground motion would be exceeded
at least once at each of the sites during the time period
of interest~ this analysis was performed by combining the
contributions of different magnitude earthquakes occurring
on the various sources at different distances from the
sites. The analyses were made for two possible design time
periods--40 and 100 years. The calculations were made
using the computer program SEISMIC-EXPOSURE that has
been developed by WCC (1981) for the National Oceanic and
Atmospheric Administration. The program provides for a
general earthquake hazard analysis methodology suitable to
subduction zone and/or other tectonic environments. The
theoretical basis for the analysis methodology can be found
in Mortgat and Shah ( 1979) and Patwardhan and others
(1980).
The resulting mean number of occurrences by which a given
level of ground motion at a site would be exceeded within
the time period of interest may then be used to evaluate
the average return period of that ground motion level and
the probability of that level being exceeded at least once
during that time period of interest. The analysis also
provides an indication of the relative importance of an
individual source based on its contribution to the total
exposure at each site.
ASSESSMENT OF INPUTS FOR ANALYSIS
Identification of potential sources of earthquakes and
their geometry with respect to the dam sites were provided
A-3 Woodward-Clyde Consultants
as a result of the seismic geology field and office work as
reported in WCC {1979 and 1980). Conclusions from the
reevaluation of available microearthquake data were helpful
in refining the geometry of the regional and local faulting
to be modeled as earthquake sources. Discussions of the
reevaluation of the sources of earthquakes are included in
Section 2.0 of the text.
The sources modeled in the seismic exposure analysis
were the Eagle River fault, the Border Ranges fault,
Bradley River fault, Bull Moose fault, and the underlying
Megathrust and Benioff Zone. Since it was necessary to
model faults with dips that apparently vary along the
length of the faults, such faults were modeled as segments
consistent with dip and maximum earthquake assignments.
The sources with long mapped traces were modeled using
depths from 0 km to 20 km. The Bradley River and Bull
Moose faults were modeled as vertical planes with a depth
of from 0 km to 10 km. The Megathrust dips underneath
the sites from a 6 km depth at the Aleutian Trench axis
{see Figure 2 in text) to a 40 km depth about 30 km
northwest of the sites. The Benioff zone dips from a depth
of 30 km near the site to a depth of 125 km in the vicinity
of the Iliamna Volcano {see Figure 1 in the text}. A
schematic of the sources modeled are shown in Figure A-2
along with the project location.
Maximum earthquakes assigned to the sources are given
in Table 1 in the text. These maximum earthquake magni-
tudes were used along with historical seismicity data to
assess the recurrence character is tics of each of the
sources used in the analysis.
//
A ~~~
/ \
//\· _//'
I
'IV'
I
/ ~,·' ~.K _ .. /
/A~'
u' ... '(;
'\ v
v.;
~"-'~\
~~
/~e~~
/ se~ \
\
Border Ranges Fault
Eagle River Fault
\
\
\
\
\ / \
// \
\ \
Bull Moose Fault\
Approximate location
Bradley Lake \
Project Site \
~ ~fl's:-....Y se~
/
/
\
\
\
\ / y
\
\
\
Bradley River Fault
\ /
\ \ \ /
\ -<,Y..'<-0"'"' «'~ /
<::>t'-\ se<:$
\
~'<(;: Y NOTE: Segment labels shown for each fault segment;
/ . dips of fault planes vary, see Table 1. //
\, /
\ /
/ .~ ~
Woodward-Clyde Consultants (I
SCHEMATIC OF EARTHQUAKE SOURCES
MODELED IN SEISMIC EXPOSURE ANALYSIS
Project No. 14844B Fig.
BRADLEY LAKE A-2
Woodward· Clyde Consultants
A-4
Examination of historical earthquake data based upon
instrumental locations and felt reports for the time period
1904 through 1978 (National Oceanic and Atmospheric Admin-
istration, 1980) has shown that no clear or compelling
correlations can be made between historical seismicity and
the faults as defined by geologic and seismicity studies to
date. However, the rna jor i ty of the reported historical
seismicity is associated with the Megathrust and the
Benioff zone which underlie the region and the site. This
interpretation is consistent with the more accurately
located microearthquake data which were reevaluated for
this study (see Section 2.3.1 of the text).
To estimate the recurrence of earthquakes of different
magnitudes for each source, the activity level of each of
the sources was defined using procedures developed for
OASES (WCC, 1978) and the NOAA studies (WCC, 1981), which
involved normalizing regional historic seismicity to a
given fault in terms of events per unit area per unit of
time. Earthquake distributions for each fault modeled in
the anlaysis is shown in Figure A-3. Recurrence of larger-
magnitude earthquakes (Ms ~8.0) on the Megathrust was
defined by NOAA project estimates of numbers of earthquakes
and their likelihood of occurrence (WCC, 1981). These
estimates were made on the basis of historical data
for large earthquakes in the Gulf of Alaska region and
on the basis of the waiting time since the last large
earthquake (Ms 8-1/2 in 1964).
In the analysis, it was assumed that on a given fault,
earthquakes of a certain magnitude could occur with equal
likelihood at any location on the fault. The release of
energy during any such event was assumed to be along a
surface rupture. The distance to the sites is an important
1000~---------------------------------------------------------,
...
1:
E
::J z
"' > ·;;
"' ::J
E
::J
(.)
Project:
0.0001
Project No.
'
l
E~gle Rjver Fault
I
r
I
I
!
' ! ; ! l i i-
' ' I /~radlev!·:. River Fault!.
I . I
i i ! j
' I ' I I ~ ~
I !
I l I Bu~l Moose Faul~
4.0
BRADLEY LAKE
148448
I
5.0 6.0 7.0
Ms
EARTHQUAKE RECURRENCE RELATIONSHIPS Fig.
USED IN SEISMIC EXPOSURE ANALYSIS A-3
WOODWARD-CLYDE CONSULTANTS
Woodward-Clyde Consultants
A-5
parameter in the attenuation of ground motion. Therefore,
it was necessary to characterize the extent that the fault
would rupture for given magnitude events. The relationship
between rupture length and earthquake magnitude selected
for the analysis is illustrated in Figure A-4. The
approximate
account for
relationship was
the observation
used
that
in the NOAA work to
along the Aleutian
Megathrust, rupture widths seem to be limited to about 200
km. This relationship is empirical and is based on the
Wyss (1979) relationship for rupture-area versus magnitude.
ATTENUATION OF EARTHQUAKE GROUND MOTION
Available literature and the results of on-going wee
studies were used to select attenuation relationships to
describe the variation of peak ground acceleration on rock
at the sites with earthquake magnitude and distance of
earthquakes from the site. The published work of Schnabel
and Seed {1973), Seed and others {1976), Woodward-Clyde
Consultants ( 1978) , Idr iss ( 1978) , Patwardhan and others
(1978), and Crouse and Turner (1980) were reviewed, and two
attenuation relationships were developed:
0
0
A relationship for earthquakes occurring on the Mega-
thrust and Benioff zones beneath the sites was based
primarily on analysis of recordings from South America
and Japan for subduction zone earthquakes.
A relationship for earthquakes occurring on the regional
and local surface faults around the sites was based
primarily on recordings from locations in California and
other parts of the western United States.
u, ... u. A L" w* ---
5.0 7.3 2.7 2.7
6.5 22.4 4.7 4.7
6.0 70.8. 8.4 8.4
6.5 224 15 15
7.0 708 27 27
7.5 2240 47 47
8.0 7080 90 80
8.5 22400 180 175
9.0 70800 400 t75
9.5 224000 1100 200
300
e 200~ ~
.r:: ;
3:
-~
0
i 1QO-0
0
0 I I I I I I I
0 100 200 300 400 500 600 700 800
Rupture length lkml
*Fault rupture dimensions from Wyss (1979) and NOAA (1981).
Project: BRADLEY LAKE FAULT RUPTURE LENGTH -Fig.
Project No. 148448 MAGNITUDE RELATIONSHIP A-4
WOOOWARO~CL YOE CONSULTANTS
Woodward-Clyde Consultants
A-6
Several recordings from Alaskan earthquakes were also
examined in developing these relationships. The mean
(average) attenuation relationships used in this study
are illustrated 1.n Figure A-5 for shallow focus crustal
earthquakes and in Figure A-6 for Hegathrust and Benioff
Zone earthquakes.
For a probabilistic evaluation, it is also important to
include the uncertainty of the predicted acceler ion
values for any given earthquake magnitude and distance. A
random error term was used in the analysis to represent
that uncertainty as a statistical distribution about the
median values. A lognormal distribution was assumed and
the standard error term taken to be s 0. 40 for the
shallow focus relationship and s == 0. 60 for the Benioff
zone relationship. The methodology also provides for
constraining the probability distribution of peak acceler-
ation so that unrealistically high values of peak acceler-
ation are not included in calculating probabilities of
exceedance. Using bounds suggested by empirical data,
an upper bound on peak acceleration was specified to be
three standard deviations.
SEISMIC EXPOSURE RESULTS
Figures 15 and 16 in the text summarize the seismic
exposure results 1.n terms of probability of exceedande of
various acceleration levels over the 100-year design
period of interest. Tables A-1 through A-6 provide a
detailed breakdown as to the contribution of various
sources to the total seismic exposure of the sites.
These tables also provide the relative contribution to
acceleration levels for specific earthquake rnagni tude
ranges by source. These results provide a means of
assessing the relative importance of earthquake sources and
earthquake magnitudes to acceleration levels at the sites.
Project:
Project No.
::§
1.0 ~~~~~~~~~-------.-----.--,---,-,-~,-IT--------~~--~
0.8
Note:
Curves are applicable only within the distance
range shown; at distances less than 3 km.
peak accelerations are constant and equal to
the values at 3 km.
c.
I'll
c:
0 ·;;
I'll
~
QJ
Qi
(.)
(.)
<(
~ .... c:
0
N ·;:
0 ::c
~
I'll
QJ
0..
0.6
0.4
0.2
0
3
BRADLEY LAKE
148448
10 30 100
Distance from Rupture (km)
MEAN ATTENUATION RELATIONSHIPS FOR
SHALLOW FOCUS EARTHQUAKES
300
Fig.
A-5
wnnnWARO-f:l YOF f:ONSULTANTS
C'l -a.
~
c::
.Q
+-'
~ ....
Q)
Qj
u u <t:
~
+-' c::
0
N ·;:
0
I
.:,t
~
Q) a..
Project:
Project No.
0.8
0.6
0.4
0.2
0
10
BRADLEY LAKE
14844B
30 100 300
Distance from Rupture (km)
MEAN ATTENUATION RELATIONSHIPS FOR Fig.
DEEP FOCUS (BENIOFF ZONE) EARTHQUAKES A-6
WOODWARD-CLYDE CONSULTANTS
TABLE A-1
PERCENT CXNI'RIBUTIOO BY s::JURCE
'Kl 'IHE 'IDTAL PIDBl\BILITY OF EXCEEDENCE
OF EACH LEVEL OF ACCELERATIOO
FOR A 100-YEAR DESIGN TIME PERIOD
BRADLE'£ LAI<E -I:W-1 SITE
PFAK GROUND ACCELERATIOO LEVELS ( 9)
0.22 0.30 0.38 0.46 0.54 0.62 0.70 0.78 0.86 0.98
Eagle River Fault
t 1
Eagle River Fault <1 <1
Segment2A
Eagle River Fault <1 <1 <1 1.0 1.8 2.7 4.2 5.6 8.1 12.7
t 2B
Eagle River Fault <1 1.2 1.5 2.1 2.8 3.1 3.3 3.5 3.6 3.9
t 3A
Eagle River Fault <1 <1 <1 <1 <1
t 3B
Eagle River Fault <1
t 4
Eagle River Fault
t 5
Border Ranges Fault
t 1
Border Ranges Fault <1 <1 <1 <1 <1 <1 <1 1.0 1.2 1.4
t 2
Border Ranges Fault <1 <1 <1 <1 <1 <1 <1 <1 <1
t 3
Border Ranges Fault <1 <1
t 4
Border Ranges Fault
t 5
Bull Mcx:lse Fault <1 <1 <1 <1 <1 <1 <1 <1 <1 <1
Bradley River Fault <1 <1 <1 <1 <1 <1 <1 <1 <1 <1
Mega thrust 1.9 1.1 <1 <1 <1 <1 <1 <1 <1 <1
t 1
Mega thrust 69.4 72.7 74.5 76.0 76.6 76.$ 76.8 76.6 76.0 74.2
t 2
Megathrust 1.2 <1 <1 <1 <1 <1 <1 <1
t 3
Benioff Zcne 1.2 <1 <1 <1 <1 <1 <1
t 1
Benioff Z01e 24.9 23.1 21.6 19.6 17.7 16.3 14.5 13.1 10.8 7.4
t 2
Benioff Z01e <1 <1 <1 <1
t 3
TABLE A-2
PERCENT <XNI'RIBUTION BY SJURCE
'ID 'lliE 'IUI'AL POOBABILITY OF EXCEEDENCE
OF EACH LEVEL OF ACCELERATION
FOR A 100-YFAR DESIGN TIME PERIOD
BRADLEY U\KE -~OOSE SITE
PEAK GROUND ACCELERATIOO LEVELS (~)
0.22 0.30 0.38 0.46 0.54 0.62 0.70 0.78 0.86 0.98
Eagle River Fault
t 1
Eagle River Fault <1 <1
Segment2A
Eagle River Fault <1 <1 <1 <1 1.2 1.8 2.6 3.4 4.6 6.7
Segment 2B
Eagle River Fault <1 <1 <1 1.1 1.3 1.3 1.2 1.2 1.2 1.0
t 3A
Eagle River Fault <1 <1 <1 <1 <1
Segment 3B
Eagle River Fault <1 <1
t 4
Eagle River Fault
t 5
Border Ranges Fault
t 1
Border Ranges Fault <1 <1 <1 <1 1.3 1.7 2.2 2.7 3.4 4.7
t 2
Border Ranges Fault <1 <1 <1 <1 <1 <1 <1 <1 <1 <1
t 3
Border Ranges Fault <1 <1
t 4
Border Ranges Fault
t 5
Bull tob::se Fault <1 <1 <1 <1 <1 <1 <1 <1 <1 <1
Bradley River Fault <1 <1 <1 <1 <1 <1 <1 <1 <1 <1
Mega thrust 1.9 1.1 <1 <1 <1 <1 <1 <1 <1 <1
t 1
Mega thrust 65.9 68.8 70.5 72.0 72.9 73.4 73.9 74.2 74.5 73.9
t 2
Mega thrust 1.2 <1 <1 <1 <1 <1 <1 <1
t 3
Benioff Z<ne 1.4 <1 <1 <1 <1 <1 <1
t 1
Benioff Zcne 28.4 27.2 26.0 24.3 22.6 21.3 19.7 18.2 16.0 13.2
t 2
Benioff Z<ne <1 <1 <1 <1
t 3
2 (Eagle River 2A)
3 (Eagle River 2B)
4 (Eagle River 3A)
5 (Eagle River 3B)
9 (Border-Ranges 2)
10 (Border Ranges 3)
13 (Bull Moose)
14 (Bradley River)
15 (Megathrust 1)
16 (Megathrust 2)
17 (Megathrust 3)
18 (Deep Benioff 1}
19 (Deep Benioff 2)
20 (Deep Benioff 3}
TABLE A-3
PERCENT CONTRIBUTION OF EACH EARTHQUAKE MAGNITUDE
RANGE FOR EACH SOURCE TO THE TOTAL SEISMIC EXPOSURE
FOR A 100-YEAR DESIGN THlE PERIOD
BRADLEY LAKE -DAM SITE
TABLE A-4
PERCENT CONTRIBUTION OF EACH EARTHQUAKE MAGNITUDE
RANGE FOR EACH SOURCE TO THE TOTAL SEISMIC EXPOSURE
FOR A 100-YEAR DESIGN TIME PERIOD
BRADLEY LAKE -POWERHOUSE SITE
PEAK GROUND ACCELERATION LEVELS (g)
Source 0.22 0.30 0.38 0.46 0.54 0.62 0.70 0.78 0.86 0.98
River 2A)
<l <l
3 (Eagle River 2B)
5 -5-l/2 <l <l <l <l <l <l <l <l <l <l
5-172 -6-172 <l <l <l <l <l <l <l <l <l <l
6-172 7 172 <l <l <l <l <l 1.4 2.2 2.8 4.0 6.0
4 (Eagle River 3A)
5 -5-l/2 <l <l <l <l <l <l <l <l <l <l
5-172 -6-172 <l <l <l <l l.l 1.2 l.l l.l l.l l.l
6 l 2 -7 l 2
5 (Eagle River 3B)
5
<l <l <l
<l <l <l <l <l
9 (Border Ranges 2)
5 -5-l/2 <l <l <l <l <l <l <l <l <l <l
5-172 -6 172 <l <1 <l <l <l <l <l 1.0 1.3 1.7
6 172 7 17 2 <l <l <l <l <l <l 1.0 1.3 1.9 2.8
10 (Border Ranges 3)
5 -5-l/2 <l <l
5-172 -6-172 <l <l <l <l <l <l <l
6-172 -7-172 <1 <1 <l <1 <l <l <l <l <1 <l
13 (Bull Moose)
5 -5-l/2 <l <l <l <l <l <l <l <l <l <l
5-172 -6-172 <l <l <l <l <l <1 <l <l <l <1
6-l 2 7 1 2
River)
<1 <1 <l <1 <1 <l <1 <l <1 <1
<1 <1 <l <1 <1 <1 <1 <l <1 <1
15 (Mega thrust l)
5 -5-1/2 <1
5-172 -6-172 <1 <1 <1 <1
6-172 -7-172 1.0 <1 <l <1 <1 <1 <1 <1
16 (Mega thrust 2) \.'
5 -5-l/2 29.1-23.4 19.3 14.9 10.8 7.5 4.0 2.5
5-172 -6-172 23.3-24.6 25.4 25.6 25.2 24.6 23.6 22.5 20.5 15.8
6 172 7-172 14.1. 20.0 24.7 35.8 34.4 38.1 42.0 44.2 47.7 50.2
3)
5 <1
5 <1 <1 <1
6 <1 <1 <1 <1 <1 <l
l)
5 <1
5 <1 <1 <1
6 1.0 <1 <1 <1 <1 <1 <1
19 (Deep Benioff 2)
5 -5-1/2 3.3 2.5 1.6 <1 <1 <1
5-172 -6 172 9.2 8.0 7.1 6.0 4.9 4.2 3.3 2.6 1.7 1.0
6-172 -7 172 14.9-16.6 17.3 17.6 17.3 17.0 16.4 15.6 14.3 12.2
3)
5
<1
<1 <1 <1 <1
Source
Eagle River 3A
5 -5-1/2
5-172 -6-172
6-172 -7-172
Eagle River 2B
5 -5-1/2
5-172 -6-172
6-172 -7-172
Border Ranges 2
5 -5-1/2
5-172 -6-172
6-172 -7-172
Border Ranges 3
5 -5-1/2
5-172 -6-172
6-172 -7-172
Bull MCX>Se
5 -5-1/2
5-172 -6-172
6 172 -7-172
Bradley River
5 -5-1/2
5-172 -6-172
6-172 7 172
TABLE A-5
EARI'HJUAKE Ml\.GNI'IUDE RANGE PERCENT CJJNTRIBUTION
'ID EACl-l ACCELERATION LENEL FOR LOCAL S:URCES
100-YEAR DESIGN TIME PERIOD
BRADLEY lAKE-DAM SITE
PEAK GROUND ACCELERATICN LEVELS (g)
0.22 0.30 0.38 0.46 0.54 0.62 0.70
51.1 44.6 39.7 35.2 31.5 27.2 22.9
49.9 55.4 60.3 64.8 68.5 72.8 77.1
0.0 0.0 0.0 0.0 0.0 0.0 0.0
33.9 28.8 24.6 21.3 18.7 15.9 12.9
31.0 29.9 28.7 27.4 26.5 25.4 24.3
35.1 41.3 46.7 51.3 54.8 58.7 62.8
31.8 19.4 10.7 5.5 3.4 1.8 0.8
42.6 45.1 43.2 39.2 35.1 30.4 25.6
25.6 35.5 46.1 55.3 61.5 67.8 73.6
31.8
42.6 32.7 21.1 12.7 10.6 6.8
25.6 67.3 78.9 87.3 89.4 93.2 100.0
78.0 74.7 69.6 62.7 56.6 49.8 43.1
22.0 25.3 30.4 37.3 43.4 50.2 56.9
65.8 60.3 54.3 48.4 43.3 37.0 30.9
34.2 39.7 45.7 51.6 56.7 63.0 69.1
0.78 0.86 0.98
20.2 16.3 12.0
79.8 83.7 88.0
0.0 0.0 0.0
11.1 8.8 6.6
23.7 22.8 21.6
65.2 68.4 71.8
0.6
22.9 19.3 14.9
76.5 80.7 85.1
100.0 100.0
39.8 36.4 35.0
60.2 63.6 65.0
27.2 22.9 18.9
72.8 77.1 81.9
TABLE A-6
E'.ARI'H;2UAKE MA.GNI'IUDE RANGE PERCENT ffiNTRIBUTION
'ID FAal Acx:ELERATICN LEVEL FDR LOCAL SCURCES
10(}-YEAR DESIGN TIME PERIOD
BRADLEY LAKE -Pa\IERHOUSE SITE
PEAK GIDUND ACCELERATION LEVELS (g)
Source 0.22 0.30 0.38 0.46 0.54 0.62 0.70 0.78 0.86 0.98
Eagle River 3A
5 -5-1/2 44.4 34.6 25.9 19.2 15.0 11.4 7.3 6.1 7.3 o.o
5-172 -6-172 55.6 65.4 74.1 80.8 85.0 88.6 92.7 93.9 92.7 100.0
6-172 -7-172 0.0 0 0 0 0 0 0 0 0 0.0 0.0 0.0 0.0 0.0
Eagle River 2B
5 -5-1/2 28.3 18.4 10.7 6.0 3.7 2.1 1.1 0.6 0.4 0.3
5-172 -6-172 31.4 31.8 30.4 27.5 24.4 20.8 17.2 15.0 12.4 10.1
6-172 -7-172 40.3 49.8 58.9 66.5 71.9 77.1 81.7 84.4 87.2 89.6
Border Ranges 2
5 -5-1/2 46.3 38.8 32.1 26.3 22.0 17.5 13.5 11.2 8.2 6.0
5-172 -6-172 35.5 38.2 39.6 40.0 39.9 39.7 38.9 38.0 36.9 34.7
6-172 -7-172 18.2 23.0 28.3 33.7 38.1 42.8 47.7 50.8 54.9 59.3
Border Ranges 3
5 -5-1/2 3.1 0.7
5-172 -6-172 42.8 29.0 18.7 12.9 9.5 7.3 5.4
6-172 -7-172 54.1 70.3 81.3 87.1 90.5 92.7 94.6 100.0 100.0 100.0
Bull Mcx:>se
5 -5-1/2 79.1 76.9 73.3 68.3 63.3 57.4 51.0 47.2 42.7 39.4
5-172 -6-172 20.9 23.1 26.7 31.7 36.7 42.6 49.0 52.8 57.3 60.6
6-172 -7-172
Bradley River
5 -5-1/2 60.6 49.6 38.3 29.0 22.1 17.3 14.4 11.2 7.3 7.5
5-172 -6-172 39.4 50.4 61.7 71.0 77.9 82.7 85.6 88.8 92.7 92.5
6-172 -7-172
Woodward-Clyde ConsuHants
APPENDIX B
METHODOLOGY FOR EVALUATIONS OF FAULT SLIP
AND ITS LIKELIHOOD OF OCCURRENCE
EVALUATION OF FAULT SLIP
This study emphasizes the evaluation of slip on secondary
faults as the result of an earthquake on an adjacent
larger fault. This emphasis stems from our knowledge of
the Bradley Lake project site. Most of the faults or
lineaments that are suspected of being faults in the siting
area are relatively minor features adjacent to two major
faults, the Border Ranges and Eagle River faults. These
minor faults or lineaments are of concern because several
cross near or through the location of several key project
components such as the power tunnel, dam, and lake tap
facilities. Thus, the potential for surface rupture on
these features are of concern in the design of the project.
Secondary faulting can be defined in different ways
according to varying casual mechanisms. These mechanisms
include: 1) passive adjustments (slip) on local faults
that are coseismic (may include some post seismic creep) to
the main earthquake on the main fault: 2) triggering of
earthquakes on adjacent faults after the main earthquake,
due to changes in the stress field in the earth • s crust
resulting from the main event; and 3) sympathetic, aseismic
slip on adjacent faults. This study considers the passive
slip on faults as the most likely nature of secondary
faulting at the Bradley Lake site and it is most commonly
observed in association with historic events on a world-
wide basis. Triggered earthquakes may be possible on
faults adjacent to the Border Ranges and Eagle River
B-2
Woodward· Clyde Consultants
faults; however, the possible structural relations of the
Bradley River and Bull Moose faults to the larger regional
faults does not seem comparable to structural relations of
worldwide examples of faults that had triggered earth-
quakes. In consideration of the third mechanism, sympa-
thetic slip, too few instances of such occurrences have
been recorded to be of use in this evaluation, and most
often the slip is quite small {approximately l-2 em or
less}. Thus, the possibilities of triggered earthquakes
and sympathetic slip as secondary faulting are not treated
in this evaluation.
The evaluation of slip on the Bradley River, Bull Moose,
and other minor faults is based on a compilation of
data on historic secondary faulting observed throughout
the world. The data have been selected so that they
are consistent with the local structural and tectonic
setting of the Bradley Lake area. The secondary faulting
could relate to the Border Ranges' style of faulting,
suspected of being an oblique reverse fault, or to the
Eagle River style of reverse or thrust faulting. In
addition, secondary faulting could result from primary
earthquakes generated on the Bradley River or Bull Moose:
the primary earthquake on these fault should not be
confused with a triggered earthquake as discussed above.
All of these possibilities were considered in data collec-
tion.
Three classifications of secondary faulting exist in
the data: 1} branch faulting or splays from the main
fault that have the same sense of slip, 2} conjugate faults
that branch away from the main fault and have the opposite
sense of slip but are responding to the same local stress
regime, and 3} subsidiary slip that occurs on adjacent
B-3
Woodward-Clyde Consultants
faults regardless of style or known structural affinity to
the main fault. These three classifications are shown
graphically in Figure B-1 as they might occur in the two
possible local structural settings being considered.
Three graphs have been plotted from the worldwide data set
collected from published literature (see Figures 17
through 19 in the text, Section 5 .0). The raw data are
presented in Table B-1 for strike-slip and oblique strike-
slip earthquakes, and in Table B-2 for reverse and oblique-
reverse-s lip earthquakes. Figures l 7 through 19 in
the text were constructed to aid in judgments of whether or
not slip is possible, and if it is possible, of what the
maximum slip might be.
All three figures present a bounding limit to the data:
however, in two instances data appear to exceed the
bounding limit. These data are marked with question marks
on Figure 17 and 19. This may be explained by inaccuracies
in the data set, by special geological conditions not
reported in the literature and thus not considered in
screening the data, or by the possibility that there are
rare cases that will exceed the majority of the data set.
For example, the point plotted above the bounding line on
Figure 17 results from a zone of secondary faults that
apparently ruptured coseismically with the 1976 Motagua,
Guatemala earthquake (Ms 7.6). However, this zone of
faults also experienced a large triggered earthquake (Mb
5.8) that cause surface faulting. Maps of the secondary
faulting do not differentiate between the coseismic
ruptures and the triggered earthquake rupture. Thus, some
uncertainty exists of the actual distance of coseismic
rupture. The point plotted on Figure 19 that is above the
line is from a 1957 earthquake (M 8.3) in Mongolia. Again
Project:
THRUST FAULT
(CROSS SECTIONAL VIEW)
Project No.
BRADLEY LAKE
148448
OBLIQUE STRIKE-SLIP AND
OBLIQUE REVERSE FAULT
(PLAN VIEW)
Subsidiary Fault
/Branch Fault
,.......-Main Fault --.,;;;__
LEGEND
__.:::::' Arrow Indicating Relative Slip
U Up
D Down
(±) Block Moving Toward Viewer
8 Block Moving Away From Viewer
SECONDARY FAULTING,
POSSIBLE ASSOCIATIONS FOR
TWO STRUCTURAL SETTINGS
Fig.
B-1
WOODWARD-CL YOE CONSULTANTS
TABLE B-1
HISTORIC SECONDARY-FAULTING DATA
Strike-Slip and Oblique Strike-Slip Faults
Secondary
Maximum Slip as %
Distance To Slip on of Maximum
Earthquake, ~lain Fault Secondary
Ms Fau~~l Reference Sources ··---··---------
l Hayward, 1868 6.7 2.2 45.7 50 Bonilla, 1970
California
2 San Francisco; 8.3 2.4 61 10 Bonilla, 1970
1906, Califor-nia 2.1 15 3
. 5 122 20
. 3 30 5
1.0 122 20
1.0 76 13
3 Kagi, 1906 7.1 1.0 81 23 Bonilla, 1975
Taiwan Ambraseys and
Tchalenko, 1968
4 Tagima, 1925 6.8 2.0 Research Group for
Japan Active Faults, 1980
.6 Ambraseys and
Tchalenko, 1968
5 Tango, 1927 7.5 1.3 Reseatcch Group for
Japan Active Faults, 1980
5.0 70 20 Ambraseys and
Tchalenko, 1968
6 lzu, 1930 7.0 4.0 30 8.5 Research Group for
Japan Active Fault, 1980
4.5 Ambraseys and
4.0 Tchalenko, 1968
7 Taiwan, 1935 7.0 6.0 Bonilla, 1979
3.0
2.0 50 15
3.0
8 Ericincan; 1939 8.0 13.0 Ambraseys and
Turkey Tcha1enko, 1968
27.0 9 Ambraseys, 1970
20.0
9 Ladik, 1943 7.6 3.0 Ambraseys and
Turkey Tchalenko, 1968
5.0 Ambraseys, 1970
6.0
10 Gered, 1944 7.6 10.0 Ambraseys and
Turkey Tchalenko, 1968
Ambraseys, 1970
ll Gonen-Yenice, 7.4 2.5 Ambraseys and
1953, Turkey Tchalenko, 1968
Ambraseys, 1970
12 Tur-key, 1957 7.1 1.8 Ambraseys and
Tcha1enko, 1968
Ambraseys, 1970
13 Buy in Zara, 7.2 1.5 Ambraseys and
1962, Iran 1.5 Tchalenko, 1968
1.9 Ambraseys, 1963
4.8 13.4 15 Berberian" 1976
1.9 Ambraseys, 1965
.8
.8
3.2
14 Parkfield, 1966 5.6 1.3 2.4 32 Brown and Veddar,
California 1967
TABLE B-1 (Continued)
Secondary
Maximum Slip as %
Distance To Slip on of Maximum
Earthquake, Main Fault Secondary Slip on
Year, Location Ms Trace (km) Fault (em) Main Fault Reference Sources
15 Murundu, 1967 7.2 . 5 Ambraseys and
Turkey 1.0 Tchalenko, 1968
1.0
5.0
16 Dasht-e, Bayaz, 7.2 2.0 Tchalenko and
1968, Iran 1.5 28 6 Ambraseys, 1970
1.6 20 4 Tchalenko and
7.5 Berberian, 1975
.6 32 7
.4
1.5
17 Borrego, 1968 6 .. 4 1.1 25 7 Clark, 1972
California 1 .8 40 11
• 5 50 13
.5 40 10
1.5 10 3
2.0 10 3
1.2 20 5
18 Motagua, 1976 7.6 28 15 4 Langer and
Guatemala Bollinger, 1979
19 Homestead, 1979 5.2 .8 3.2 33 Hill and others,
California (ML) . 7 1980
20 Imperial Valley, 6.8 5.3 10 l3 Lei vas and others, 1980
1979' California 4.5 15 19 Sieh, 1980
1.1
21 Livermore, 1980 5.8 .25 2 80 Bonilla and
California (ML) . 3 . 5 20 others, 1980
.7
TABLE B-2
HISTORIC SECONDARY-FAULTING DATA
Reverse and Oblique Reverse Faults
Secondary
Maximum Slip as %
Distance To Slip on of Maximum
Earthquake, Main Fault Secondary Slip on
Year, Location Ms Trace (km) Fault (em) Main Fault Reference Sources --------
22 Arvin-Tehachapi, 7.7 2.8 30 25 Bonilla, 1970
1952, California 1.4 30 25 Kupfer and others,
8.0 9 8 1955
23 Bagdu, 1957 , 8.3 20.0 500 39 Ambraseys and
Mongolia 20.0 200 16 Tchalenko, 1968
7.0 250 20
15.0 200 16
5.0 650 50
24 Meckering, 1968 7.0 5.0 20 7 Bonilla, 1970
Australia 1.25 ----Everingham and
2.0 15 5 others, 1969
25 San Fernando, 6.6 2.7 15 6 Kamb and others,
1971, California . 3 5.8 2.4 1971
.9 113 47
26 Tabas-E-Golshan, 7.7 5.0 ----Berberian, 1979
1978, Iran 10.0
7.5
14.0
27 El Asnam, 1979, 7.3 7.0 100 20 Burford and
Algeria 1. 5 ----others, 1981
2.0
B-4 Woodward-Clyde Consultants
the data for this event show all surface ruptures in the
vicinity of the earthquake without any differentiation of
cause of rupture or possible effects from triggered
events. For purposes of this analysis, the bounding limit
line is considered to represent the worst case for the
majority of data.
For given earthquake magnitudes and geologic conditions,
amounts of fault slip can be read directly from the graphs
constructed from the empirical data.
PROBABILITY OF FAULT RUPTURE
In general, the surface-rupture capability of the small
faults near the Bradley Lake project depends upon: 1)
the capability of the Border Ranges and Eagle River faults
to produce earthquakes; 2) structural relations to the
major faults and regional stress regime; and 3) recurrence
intervals of earthquakes in the region. However, large
uncertainties exist in the literature concerning the
activity of the faults, the structure, and recurrence
intervals of earthquakes in the Bradley Lake region of the
Kenai Peninsula. Therefore, in preparing an analysis of
the probability of occurrence of slip on faults at the
site, the basic approach has been expanded to allow for
uncertainties 1n fault and earthquake parameters. These
parameters include surface faulting capability, maximum
earthquake magnitude, tectonic associations, and recurrence
intervals of different magnitude earthquakes on the various
faults.
For this analysis, recurrence data of various size earth-
quakes are derived from the data set used in the seismic
exposure assessment where they are expressed as number
B-5 Woodward-Clyde Consultants
of earthquakes in the 7 5-year period of observation.
The upper end of the magnitude range over which recurrence
was considered is limited by the maximum magnitude assigned
on each source. The maximum earthquake is defined 1n
section 2.0 of the text for each source. The lower end of
the magnitude range considered in this analysis is governed
by the minimum magnitude earthquake that is suspected of
having associated surface fault rupture for a particular
style of faulting. It is generally accepted in the
practice that magnitudes in the range of 5. 5 and 6.0 are
the lower limits where surface rupture can be expected.
For this study, we considered recurrence interval in terms
of the number of earthquakes that might occur close enough
to the site to affect secondary faults in a 100 year time
period for each magnitude increment (i.e. the increment
from 6. 755 to 7.25 is shown as Ms 7.0). This recurrence
value also incorporates the effect of distance from the
site to the main fault trace rupture shown in Figure 17
(Section 5.0). For example, a magnitude 7.0 earthquake
could rupture on either side of the site on the Border
Ranges fault and approach to within the distance shown in
Figure 17; the recurrence estimate included the number of
events in that combined area of rupture that could result
in secondary faulting at the site.
It is assumed that probability of subsidiary fault rupture
depends upon the magnitude of earthquake on the main fault,
the recurrence interval, and the distance from the main
fault. The various
geologic conditions
fault parameter values and range of
upon which the on-site offsets depend
have been combined and presented in a logic tree format.
An exar~le of a logic tree is presented in Figure B-2. The
logic tree is constructed to provide a means of formally
Secondary Fault
Secondary Fault
Project:
Project No.
No. Events That
Earthquake Magnitude, Rupture on May Result in
I Capability I Tectonic Association I No. Events in Specific Secondary
100 Yr. Period Secondary Fault Rupture in
100 yr. Period
Yes
7.5 . ....,.---
No
p(Nk I Tj• Ci) p(R 11Nk, Tj• Ci)
CASEiikl Yes
7.0
I
p(TIC-) J I
Border Ranges Y 6.5
Yes
~
No
Yes ~ 0
.,;; 5.5 No
·~ Capable ~ 0
7.0 Yes
Ea~ ~
Yes ~ 0
I \ \ .,;; 6.0 No
Independent <:; : Yes
Yes
Non-Capable
Probability of CASEijkl = p(C1) • p(TjiCi) • p(NkiTj, C1) • p(R 11Nk, Ti• Ci)
BRADLEY LAKE
148448
EXAMPLE OF LOGIC TREE FOR
EVALUATION OF SECONDARY FAULT RUPTURE
Fig.
B-2
WOOOWARD-CL YDE CONSULTANTS
Woodward-Clyde Consultants
B-6
accounting for the uncertainty in input parameters of the
probability assessment. A logic tree is composed of nodes
and branches. Each node represents a point at which a
choice is possible between alternative states or values of
an input parameter. The branches represent discrete
alternatives such as yes or no for the capability of a
fault to sustain rupture. The alternatives may also
represent a continuous distribution, such as recurrence
rate for various magnitude earthquakes.
At each node conditional probabilities are assigned to each
branch that represent the likelihood of that branch being
the best value of the input parameter. As the branches
at each node are all possible alternatives, the sum
of the conditional probabilities is equal to one. The
probabilities at each node depend on the condition that the
events leading to the node have occurred.
For this study, several data gaps exist in the literature
that affect the evaluation of potential fault rupture.
Specifically, these data gaps are: 1) capability of the
main fault; 2) tectonic association; and 3) the probability
of the secondary fault rupturing if the main fault ruptures
during an earthquake. To these unknowns, a conditional
probability value has been assigned by several geologists
who based their judgments on subjective evaluations of
available data for the Cook Inlet-Kenai Peninsula area.
By combining the recurrence probabilities with the sub-
jective probabilities, a total probability of occurrence
is obtained. In so doing, the logic tree utilizes all
available data, as well as experience and professional
judgment, in arriving at probability estimates. As such,
Woodward-Clyde Consultants
B-7
the probabilities reflect relative degrees of confidence in
the parameter values on each branch.
These probability data were evaluated in a logic tree
format for the Bradley River and Bull Moose faults.
Another logic tree was prepared to estimate the likelihood
of slip on any one of the minor faults in the vicinity.
Each option in the logic process is identified and account-
ed for, thus probabilities of individual earthquake events
and associated faulting can be identified separately or
collectively, on the basis of a wide range of assumptions.
A summary of the results of our assessment of the likeli-
hood of fault slip are presented in Section 5.0 and 6.0 of
the text.
Woodward· Clyde Consultants
APPENDIX C
BIBLIOGRAPHY
Agnew, J. D., 1980, Seismicity of the Central Alaska
Range, Alaska, 1904-197 8: University of Alaska,
Fairbanks, Thesis, 95 p.
Ambraseys, N. N., 1963, The Buyin-Zara (Iran) earthquake
of September, 1962, a field report: Bulletin of
the Seismological Society of America, v. 53, no. 4,
p. 705-740.
1965, An earthquake engineering study of the Buyin-
Zahra earthquake of September 1, 1962, in Iran: Third
World Conference on Earthquake Engineering, New
Zealand, 1965, Proceedings, v. III.
1970, Some character is tic features of the Anatolian
fault zone: Tectonophysics, v. 9, p. 143-165.
Ambraseys 1 N., and Tchalenko, J. 1 1968, Documentation of
faulting associated with earthquakes (Part 1):
London, Department of Civil Engineering, Imperial
College of Science unpublished report, 37 p.
Beikman, H., 1979, Preliminary geologic map of Alaska:
U.S. Geological Survey, scale 1:2,500,000.
1980, Geologic Map of Alaska: u.s. Geological Survey,
Scale 1:2,500,000.
Woodward-Clyde Consultants
C-2
Berberian, M., 1976, The 1962 earthquake and earlier
deformations along the Ipak earthquake fault:
Geological Survey of Iran, Ministry of Industry and
Mines, Contribution to the Seismotectonics of Iran
(Part II), Report No. 39, p. 419-427.
1979, Earthquake . faulting and bedding thrust asso-
ciated with the Tabas-E-Golshan (Iran) earthquake
of September 16, 1978: Bulletin of the Seismological
Society of America, v. 69, no. 6, p. 1861-1887.
Bonilla, M. G., 1970, Surface faulting and related effects,
Chapter 3 in Earthquake engineering: Englewood
Cliffs, Prentice-Hall, p. 47-74.
1975, A review of recently active faults in Taiwan:
u.s. Geological Survey Open-File Report 75-41, Menlo
Park, California, 58 p.
1979, Historic surface faulting--Map patterns,
relation to subsurface faulting, and relation to
preexisting faults: u.s. Geological Survey, Open-File
Report 79-1239, Menlo Park, California, 21 p.
Bonilla, M. G., Lienkaemper, J. J., and Tinsley, J. c.,
1980, Surface faulting near Livermore, California,
associated with the January 1980 earthquakes: U.S.
Geological Survey Open-File Report 80-523, Menlo Park,
California, 27 p.
Brown, R. D., Jr., and Vedder, J. G., 1967, Surface
tectonic features along the San Andreas fault,
California, in Brown, R. D"' Jr., and others, The
Parkfield--Cholame, California, earthquakes of
Woodward-Clyde Consultants
C-3
June-August 1966--Surface geologic effects, water
resources aspects, and preliminary seismic data: u.s.
Geological Survey Professional Paper 579, p. 2-23.
Burford, R. 0., Harsh, P. w., and Espinosa, A. F., 1981,
Last October, 7. 3 quake in Algeria reviewed: Geo-
times, v. 26, no. 5, p. 26-28.
Buwalda, J. P., and St. Armand, P., 1955, Geological
effects of the Arvin-Tehachapi· earthquake in Kern
County, California, 1952: California Division of
Mines and Geology, no. 171, 283 p.
Clark, M. M., 1972, Surface rupture along the Coyote
Creek fault in The Borrego Mountain earthquake of
April 9, 1968: u. S. Geological Survey Professional
Paper 787, p. 55-86.
Crouse, c. B., and Turner, B. E., 1980, Processing and
analysis of Japanese accelerograms and comparisons
with u.s. strong motion data: Seventh World Con-
ference of Earthquake Engineering, Istanbul, Turkey.
Davies, J. N. , House, L. , 1979, Aleutian subduction zone
seismicity, volcano-trench separation, and their
relation to great thrust-type earthquakes: Journal
of Geophysical Research, v. 84, no. B9, p. 4583-4591.
Davies, J., Sykes, L., House, L., and Jacob, K., 1981,
Shumagin seismic gap, Alaska Peninsula: History
of great earthquakes, tectonic setting, and evidence
for high seismic potential: Journal of Geophysical
Research, v. 86, no. B5, p. 3821-3855.
Woodward· Clyde Consultants
C-4
Detterman, R. L., Plafker, G., Hudson, T., Tysdal, R. G.,
and Pavoni, N., 1974, Surface geology and Holocene
breaks along the Susitna segment of the Castle Mountain
fault, Alaska: u. s. Geological Survey Miscellaneous
Field Studies Map MF-618, scale 1:24,000, l sheet.
Dobry, R., Idriss, I. M., and Ng, E., 1978, Duration
characteristics of horizontal components of strong
motion earthquake records: Bulletin of the Seismo-
logical Society of America, v. 68, no. 4, p. 1487-
1520.
Evans, c. D., Buck, E., Buffler, R., Fisk, G., Forbes, R.,
and Parker, W., 1972, The Cook Inlet environment--a
background study of available knowledge: Department
of the Army, Alaska District, Corps of Engineers, in
accordance with Contract No. DACW 85-72-C-0052.
Everingham, I. B., Gregson, P. J., and Doyle, H. A., 1969,
Thrust fault scarp in the western Australian shield:
Nature (August 16, 1969), v. 223, p. 701-703.
Gedney, L., and Estes, S., 1981, A recent earthquake on
the Denali fault in the southeast Alaska range:
Fairbanks, Geophysical Institute University of Alaska
unpublished report, 8 p.
Hill, R. L., Pechmann, J. c., Treiman, J. A., Mcr.tillan, J.
R., Given, J. W., and Ebel, J. E., 1980, Geologic
study of the Homestead Valley earthquake swarm of
March 15, 1979: California Geology, v. 33, no. 3, p.
60-67.
Woodward· Clyde ConsuHants
C-5
Idriss, I. M., 1978, Characteristics of earthquake ground
motions, State-of-the-art paper: Specialty Conference
on Earthquake Engineering and Soil Dynamics, Geotech-
nical ~ngineering Division of the American Society of
Civil Engineers, Pasadena, California, June 19-21,
Proceedings, v. III, p. 1151-1263.
Idriss, I. M., and Power, M. s., 1978, Peak horizontal
accelerations, velocities and displacements of rock
and stiff soil sites for moderately strong earth-
quakes: submit ted to Bulletin of the Seismological
Society of America for possible publication, April:
contained in Idriss (1978).
Kamb, B., Silver, L., Abrams, M.,
T., and Minster, J., 1971,
and nature of fault movement
Carter, B., Jordan,
Pattern of faulting
in the San Fernando
earthquake in The San Fernando, California, earthquake
of February 9, 1971: U.S. Geological Survey Profes-
sional Paper 733, p. 41-54.
Kupfer, D. H., Muessig, S., Smith, G. I., and White, G. N.;
1955, Arvin-Tehachapi earthquake damage along tie
Southern Pacific Railroad near Bealville, California,
in Oakshott, ed., earthquakes in Kern County, Cali-
fornia, 1952: California Division of Mines and
Geology, no. 171, 283 p.
Lahr, J. c., and Stephens, c. D., 1981, Review of earth-
quake activity and current status of seismic monitor-
ing in the region of the Bradley Lake Hydroelectric
Project, U.S. Geological Survey, Menlo Park, Califor-
nia, preliminary report, 21 p.
Woodward-Clyde Consultants
C-6
Langer, c. J., and Bollinger, G. A., 1979, Secondary
faulting near the terminus of a seismogenic strike-
slip fault: Aftershocks of the 1976 Guatemala earth-
quake: Bulletin of the Seismological Society of
America, v. 69, no. 2, p. 427-444.
Leivas, E., Bart, E. w., McJunkin, R. D., and Real, c. R.,
1980, Geological setting, historical seismicity and
surface effects of the Imperial Valley earthquake,
October 15, 1979, Imperial County, California, in
Leds, D. J., ed., Reconnaissance Report, Imperial
County, California Earthquake, October 15, 1979:
Berkeley, California, Earthquake Engineering Research
Institute, p. 5-19.
MacKevett, E. M., and Plafker, G., 1974, The Border Ranges
fault in south-central Alaska: U.S. Geological Survey
Journal of Research, v. 2, p. 323-329.
Magoon, L. B., Adkison, w. L., and Egbert, R. M., 1976, Map
showing geology wildcat wells, Tertiary plant fossil
locations, K-Ar age dates, and petroleum operations,
Cook Inlet area, Alaska: u.s. Geological Survey Map
I-1019, scale 1:250,000.
Mortgat, C. P., and Shah, H. C., 1979, A Bayesian model for
seismic hazard mapping: Seismological Society of
America Bulletin, v. 69, no. 4, p. 1237-1251.
National Oceanic and Atmospheric Administration, 1980,
Hypocenter Data File for Period of Coverage 1638
to December 1978: Environmental Data Service,
Boulder, Colorado.
C-7
Woodward· Clyde Consultants
Patwardhan, A. s., Kulkarni, R. B., and Tocher, D., 1980, A
semi-Markov model for characterizing recurrence of
great earthquakes: Seismological Society of America
Bulletin, v. 70, no. 1, p. 323-347.
Patwardhan, A., Sadigh, K., Idriss, I. M., and Youngs,
R., 1978, Attenuation of strong ground motion--effect
of site conditions, transmission path characteristics,
and focal depths: unpublished report to be submitted
to the Bulletin of the Seismological Society of
America for possible publication.
Patwardhan, A. s., Tocher, D., and Savage, E. E., 1975;
Relationship between earthquake magnitude and length
rupture surface based on aftershock zones: 70th
Annual Meeting of the Seismological Society of
America.
Plafker, G., 1971, Tectonics in The great Alaska earthquake
of 1964, Geology: National Academy of Sciences,
Committee on the Alaska Earthquake, Division of
Earth Sciences, National Research Council, Part A,
p. 47-122.
Research Group for Active Faults, 1980,
Japan; Sheet maps and inventories:
of Tokyo, 380 p.
Active faults in
Japan, University
Schnabel,
rock
P . B. , and Seed, H • B. ,
for earthquakes in the
1973, Accelerations in
vlestern United States:
Bulletin of Seismological Society of America, v. 63,
no. 2, p. 501-516.
Woodward-Clyde Consultants
C-8
Seed, H. B., Hurarka, R., Lysmer, J., and Idriss, I. M.,
1976, Relationships of maximum accelerations, maximum
velocity, distance from source and local site condi-
tions for moderately strong earthquakes: Bulletin of
Seismological Society of America, v. 66, no. 4,
p. 1323-1342.
Sieh, K. E., 1980, Imperial Valley faulting, 1979, in
Earthquake Engineering Research Institute, Reconnais-
sance Report, Imperial County, California, Earthquake,
October 15, 1979, p. 32-34.
Slemmons, D. B., 1977, State-of-the-art for assessing
earthquake hazards in the United States; Part 6~
Faults and earthquake magnitude with appendix on
geomorphic features of active fault zones: U.S. Army
Engineer Waterways Experiment Station, Vicksburg,
Contract No. DACW 39-76-C-0009, 120 p., plus 37 p.
Appendix.
Soward, K. s., 1962, Geology of waterpower sites on
the Bradley River, Kenai Peninsula, Alaska: u.s.
Geological Survey Bulletin 1031-C, 70 p., 1 plate.
Tchalenko, J., and Ambraseys, N. N., 1970, Dasht-E-Bayaz,
Earthquake fractures: Geological Society of America
Bulletin, v. 81, no. 1.
Tchalenko,
fault,
J. s., and Berberian, M., 1975, Dasht-E-Bayaz
Iran, earthquake and earlier related structures
in bedrock: Geological Society of America Bulletin,
v. 86, p. 703-709.
Woodward-Clyde Consultants
C-9
u.s. Army corps of Engineers, 1978, Reanalysis of the
Bradley Lake Hydroelectric Project, Appendix lB
Technical Report, Appendix 2, Pertinent Correspon-
dence.
u.s. Geological
culprit fault
Survey, 1976,
identified:
Guatemalan earthquake,
u.s. Geological Survey,
Earthquake Information Bulletin, v. 8, no. 3.
U.s. Nuclear Regulatory Commission, 1981, Safety evalua-
tion report related to the operation of San Onofre
Generating Station, Units 2 and 3, Docket Nos. 50-361
and 50-362, Southern California Edison Company:
NUREG-0712, Supplement No. 1 Appendix E, 28 p.
Wallace, R. E., and Roth, E. F., 1967, Rates and patterns
of progressive deformation in Brown, R. D., and
others, eds., The Parkfield--Cholame, California,
earthquakes of June-August 1966, Surface geologic
effects, water resources aspects, and preliminary
seismic data: u.s. Geological Survey Professional
Paper 579.
Woodward-Clyde Consultants, 1978, Offshore Alaska Seismic
Exposure Study: Unputlished report prepared for
Alaska Subarctic Operators' Committee (ASOC), March,
1978, v. 1 through 5.
1979, Reconnaissance Geology, Bradley Lake Hydro-
electric Project: Unpublished reort prepared for the
Department of the Army, Alaska District, corps of
Engineers, Contract No. DACW 85-79-C-0045, 65 p.
C-10
Woodward· Clyde Consultants
1980a, Seismicity Study Bradley Lake Hydroelectric
Project: Unpublished report prepared for the Depart-
ment of the Army, Alaska District, Corps of Engineers,
Contract No. DACW 85-79-C-0045, 35 P•
1980b, Interim Report on Seismic Studies for Susitna
Hydroelectric Project: Unpublished report prepared
for Acres American Incorporated, Alaska Power Auth-
ority, Susitna Hydroelectric Project, Subtask 4.01
through 4. 08, by Woodward-Clyde Consultants, Orange,
California.
1980c, Development and Initial Application of Software
for Seismic Exposure Evaluation, v. I, October.
1981, Seismic exposure software: Development and
initial application of software for seismic exposure
evaluation: two volumes prepared for the National
Oceanic and Atmospheric Administration, in press,
Three Embarcadero Center, San Francisco, California.
Wyss, M., 1979, Estimating maximum expectable magnitude
of earthquakes from fau 1 t dimensions: Geology,
v. 7, p. 336-340.