Download Seismic Hazard Characterization of the BNL- HFBR Site ... and more Lecture notes Geographic Data and Information in PDF only on Docsity!
UCRL-ID--
DEgl 004671
Seismic Hazard Characterization of
the BNL- HFBR Site
(Upton, New York)
Jean B. Savy
October, 1990
ii:i:i:i:i:i:i:i:!:i:i:i:i.......:i..._,.....
This is an informal rei_ort intended p r imarily for internal or limited external distribution. The opinions arid conclusions stated are those of the author and may or may not be those of the Laboratory, Work performed under the auspices of the U.S. Department of Energy by the Lawrence L i vermore Nafionai Labo r ato ry under Cont r act W-7 4 05-Eng-48.
IIISCL+AIMLI{ i+ Thisdo c um e nt ,+ 'a ps r ep ar e dasan acc o L 1nt o iwork sp on s or e d U v sn ag e n cy ,o 't the U ni te d5 ta l e s( + , o v e rn me nt. Neithe r th e Unit e d _ tat es (,ov e rn m en t north eU niversityotCalifornia n orany o ith e ir e mploy ee s ,m ak es an ,,' warranty,ex p res s ors m p[ied,orassumesany legalli a biliorrespon t y si bilitytorth ea c cur a cycompleten , e ss or, us e f uln e sso t any mtor m ation ,p paratus, p rnduct , or processdis c los e d,or r e pr e sentsthatitsu s e would not intringprivately e owned riRhtsR ;e f e renceh e reinto a ny s p e c l tico c mme rcialproducts,proc e ss,ors e rv l ceby t rade name , trademark , m anutactu L e r , or o t herwis e, doesno t nec e s s ar i l y c on sti tu t e or i m p ly its end o rsement. recommend a tion,or f a vor in g by th e United t ate s (.,overn m ent or t h e Un i vers _ .tvo f Ca l i f ornia. I he view s and o p i mo ns o i authors ex p ressedhere i n do no t necessa r i l y stateor re fl ect thoseot the United S t **ates (,** o vern m ent or the Univers i ty ot Ca l iforni a._ and sha ll no t be used for adver t isin _ or p r o duc t endor s em e n t purpo s es.
T h i s r e p o rl h a s b ee n re p rodu c e d dir e clI¥ from th e be s t avai l ab l e co py : k V, ll lab l e to I) ()E + and t)(}[ co n tcactors l tom t he ( )i t i c e o! ci entt ti c and [e t hn i cal I n | orma l ton 1 '{) Ho lt( 2, Oak Rid g e , I N 1 7 81! l' ti ces aVal ilble f rom (bl S l 5 '7 6 - + 401. I - l5 h. _ 6" _ 40 1
N.,lion+_'_s ai lla bi [-e ¢ c^ hnI o lca^ l Ihe^ ln pub lorm.+li c^ ttI+ rom .vn^ hUl+Vl+.tthe ' I _ Dep._r t men l o l ( ommef c e . _ Pot! Royal I (d.. **,pr** i ngtie l d , VA 22 1 bl
Price Page Code Range A0 1 Microfiche ,'
Papercopy Prices i, A02 1- l A03 1 I- 50 A04 51- 75 A05 7 6- 100 A06 1 01- A07 126- A08 151-1 75 A09 1 76- Al0 201- All 226- A12 251- 2 75 A 13 276- A14 301- A 15 326- A16 351- A 1 7 37 6- A18 40]- AI9 426 -4 50 A20 451-4 75 A21 476- A22 501-5 2 5 A23 526- A24 551- A25 576- A99 o0] & UP
Table 1. 1 Site Soil Category: Rock Site ............................................................. 4
Table 1.2 Site Soil Category: Deep Soil Site ...................................................... 5
,t
. List of Figures
Figure lR Arithmetic average hazard curves (Minimum contributing magnitude = 5.0) of the PGA for the BNL-HFBR rock site characteristics and with five ground motion experts' input ................ 7
Figure 2R 5, 15, 50, 8 5 and 9 5 percent constant percentile hazard curves (Minimum contributing magnitude = 5.0) of the PGA for the BNL-HFBR site, under the assumption of rock site characteristics and with five ground moti_m experts' input .......................................... 8
Figure 3R Arithmetic average Uniform Hazard Spectra for three return periods (Minimum contributing magnitude = 5.0) for the BNL- HFBR site, under the assumption of rock site characteristics, and with five ground motion experts' input .......................................... 9
Figure 4R 5, 15, 50, 85 and 9 5 percent constant percentile 1,000 year return
. period Uniform Hazard Spectra (Minimum contributing magnitude = 5.0) for the HFBR site, under the assumption of rock site characteristics, and with five ground motion experts'
- input ............................................................................................................... I
Figure 5 R 5 , 15 , 50, 8 5 and 95 percent constant percentile 10,000 year return period Uniform Hazard Spectra (Minimum contributing magnitude = 5.0) for the BNL-HFBR site, under the assumption of rock site characteristics, and with five ground motion experts' input ............................................................................................................... 11
Figure 6R 5, 15, 50, 85 and 95 percent constant percentile 100,000 year return period Uniform Hazard Spectra (Minimum contributing magnitude = 5.0) for the BNL-HFBR site, under the assumption of rock site characteristics and with five ground motion expert_' input ............................................................................................................... 12
Figure 1S Arithmetic average hazard curves (Minimum contributing magnitude = 5.0) of the PGA for the BNL-HFBR site, under the assumption of deep soil site characteristics and with five ground motion experts' input .................................................................. 13
- Figure 10 R 5 , 15 , 50 , 85 and 9 5 percent constant percentile 1,000 year return period Uniform Hazard Spectra (Minimum contributing magnitude = 5.0) for the BNL-HFBR site, under the assumption of rock site characteristics , and with four ground motion experts' input ............ 22
Figure 1 1R 5, 1 5, 50, 85 and 95 percent constant percentile 10,000 year return period Uniform Hazard Spectra (Minimum contributing magnitude = 5.0) for the BNL-HFBR site, under the assumption of rock site characteristics, and with four ground motion experts' input ............ 23
Figure 12R 5, 15, 50, 85 and 95 percent constant percentile 100,000 year return period Uniform Hazard Spectra (Minimum contributing magnitude = 5.0) for the BNL-HFBR site, under the assumption of rock site characteristics, and with four ground motion experts' input ............ 24
Figure 7 S Arithmetic average hazard curves (Minimum contributing magnitude = 5.0) of the PGA for the BNL-HFBR site, under the assumption of deep soil site characteristics and with four ground motion experts' input ................................................................................. 25
- Figure 8S 5 , 15 , 50 , 85 and 95 percent constant percentile hazard curves (Minimum contributing magnitude = 5.0) of the PGA for the BNL- HFBR site, under the assumption of deep soil site characteristics
- and with four ground motion experts' input ........................................ 26
Figure 9S Arithmetic average Uniform Hazard Spectra for three return periods (Minimum contributing magnitude = 5.0) for the BNL- HFBR site , under the assumption of deep soil site characteristics, and with four ground motion experts' input ........................................ 27
Figure 10S 5, 15 , 50 , 85 and 95 percent constant percentile 1 , 000 year return period Uniform Ha z ard Spectra (Minimum contributing magnitude = 5.0) for the BNL-HFBR site , unde r the assumption of dee r soil site characteristics , and with four ground motion experts' input ............ 28
Figure 11S 5 , 15 , 50 , 85 and 95 percent constant percentile 10,000 year return period Uniform Hazard Spectra (Minimum contributing magnitude = 5.0) for the BNL-HFBR site, under the assumption of deep soil site characteristics, and with four ground motion experts' input ............ 29
Figure 12S 5 , 15 , 50 , 85 and 95 percent constant percentile 100,000 year return period Uniform Hazard Spectra (Minimum contributing magnitude = 5.0) for the BNL-HFBR site, under the assumption of deep soil site characteristics and with four ground motion experts' input ............. 30
• ACKNOWLEDGEMENTS
This work was performed by the Lawrence Livermore National Laboratory (LLNL) for the Department of Energy / Brookhaven National laboratory (BNL) under purchase order 45947 0. Mr. Charles Hofmayer was the Project Manager, Mr. Jean B. Savy was the LLNL Project Leader.
viii
. 2 .0 METHODOLOGY
The hazard model used here is based on the now well accepted model developed by C.A. Cornell (1968).
The important aspects of the NRC / LLNL methodology consist of:
- Using many experts' opinions to include the knowledge uncertainty, as well as the physical uncertainty which is already included in the hazard model itself.
- Pro p a ga ti ng a l l un ce rtain t i e s , i n th e s e is m i c ity m od e l ing a nd g round m otion m od e lin g, throu g h t h e us e o f a Mont e Carlo si m ul a tionpro ce ss.
Th e d e t a ilso f the methodolo gy, includin g th e h aza rd mod e l a nd th e proc e sso f eli c it a tiono f the experts'opinion , a r e g iv e ni n B e rnreuter e t a l.(1 985) , Vol.I a nd 2 a nd B e r n reuteret a l. , Vol. I , 6 , a nd 7 ( g iv e nin App e ndix C ). In p a rti c ul a r Volu me 7 d e t a ilsallt he qu e stionn a iresus e d in th e e xp e rts'opinioneli c it a tions. L t
- 0 INPUT DATA
The input data used for this HFBR study , is exhaustively described in Bernreuter et al. (198 8 ) , Vol. 1. The minimum magnitude of the earthquakes contributions to the hazard in the base case , is magnitude 5.0 , and ali five ground motion e x perts inputs are used.
4.0 SEISMIC HAZARD ESTIMATES AT THE HFBR SITE
4.1 Site Description
The HFBR site is part of the Brookhaven National Laboratory in Upton , New York. The coordinates of the site provided to LLNL in a transmitted fax nn 4 / 1 7/ 90 (in a memo from P. Tichler to J. Tichler and H. Bernstein , dated 7 / 28 / 88) and used in this analysis are:
- Latitude: 40. 87 01 ° N (decimal)
: • Longitude: 72. 87510 W (decimal)
Formally, the site characteristics were described in the statement of work as follows:
- "The HFBR Site is situated at BLN on a 200 foot thick layer of dense glacial deposits of gravels, sands and clays. Below
this morainal deposit exists additional series of dens e
sediments about 1,200 foot thick. Bedrock is at a depth of
about 1,400 feet. The site can be classified as a deep soil site
(category 8)." [Referring to LLNL site categories described
in Bernreuter et al., 1985.]
Following a later request by BNL staff, the site characteristics was assumed
to be either a deep soil, as defined in Bernreuter et al., Vol. 1, 1988, or a
rock site.
4.2 Domina_at Zonal Contributions
The dominant zonal contribution tables (Tables 1.1 and 1.2) are an attempt
at identifying which of the seismic zones given by the seismicity experts
(S-experts) contribute the most to creating the seismic hazard at the I / FBR
site. The percentages given in these tables are the ratios of the hazard
provided by a zone, to the total hazard at the site, given ortly for the four
highest contributing zones, (See Bernreuter et al. 1988 , Vol. 1 ,
Appendix C).
- However, one needs to be careful in interpreting the results presented in
Tables 1.1 and 1.2, since the calculations are made _only with the best
estimate hazard curves (BEHC) and not with the constant percentile
hazard curves ( C PHC).
The BEHC for a given S-expert is obtained by setting each of the uncertain
parameters equal to what each expert has defined as the most likely value
of the parameter (zonation, seismicity parameters) and the most likely
models of the ground motion experts (G-experts). Thus, the table of zonal
contributions is only indi , 'ative of the relative contributions but it does
not always represent L_CC.aratelythe relative contributions when all
pf. , ssible alternatives, and ali uncertainties are included.
In these tables, the percentage values have been rounded-off to the nearest
integer value and only the five zones contributing the most have been
reported. Thus, the sum of ali the percentages for a given S-expert, either
at low PGA levels (0 125g) or high levels (0.56g) do not necessarily add up
to unity. The zo,_e indexes refer to the indexes for each of the zones in the
maps given in the Appendix B.
- 4.3 Hazard Estimates with all Seismic and Ground Motion Experts
4.3.1 Hazard Estimates Assuming the !_ F BR to be a Rock Site
The results are first presented for a site at the location of the HFBR facility assuming that the characteristics are those of a rock site. The results are surface free field. Figure lR represents an arithmetic mean hazard curve of the Peak Ground Acceleration (PGA); Fig. 2R gives the 5 th, 15 th, 2Sth, 35 th, 45rh , 50th, 65rh, 7 5rh, 85th and 95th percentile hazard curves, including ali types of uncertainties and expert variabilities;Fig. 3R gives the arithmetic average 1,00 0 , 10, and 100,000 year 5 percent damping Uniform Hazard Velocity response spectra; Fig. 4R gives the 5 percentiles (5, 15, 50, 85 and 95), 1,000 year, 5 percent Damped Velocity response spectra, Fig. 5R gives the same information as Fig. 4R for 10,000-year return period and Fig. 6R the same as Fig. 4R for 10 0 ,0 0 0-year return period.
4.3.2 Hazard Estimates Assuming the HFBR to be a Deep Soil Site
Figures 1S to 6S give the same results as Figures lR to 6R for deep soil site.
4.4 Sensitivity Analysis
The results presented in this section correspond to exactly the same cases treated in section 4.3• the only difference is that only the ground motion models from four of the five G-experts were used. In this case, the Trifunac-Anderson ground motion model was not considered. Figures 7 R to 12R show the results of using four G-experts' input and considering the site to be a rock site. Figures 7 S to 125 show the same results for the deep soil case.
SEISMIC HAZARDAl DOE FRC[LIT]ES CASE Ml (I,1>5.g), LliTH 5 Gli-EXPERT
Figure 2R 5, 1 5 , 5 0, 85 and 9 5 percent constant percentile ltazard curves (Minimum contributing magnitude = 5.0) of the PGA for the BNL-
HFBR site, under the assumption of rock site characteristics and
with five ground motion experts' input.
8
SEISI'4]C HAZflROlit DOE FRC[LITIE CASE Hl IH : ,5.01, WITH 5 GM-EXPERT
1000,, ]B0 0 0., 100000. YERR8 REfURN PEP,[ ARITHMETIC RVERRGEBPECTRRFOR ALL EXPER'I E 03. 1 I 1 I - 1 1 I ¢! '"' 1 t 1! ' | t 1 1 I ' I 1 l
_ 4 - 3 -
E-OI " , i i i i i i i 1 i i _ i t 1 l i i I i I
' (^) I,.¢.1' PERIOD (SEC) (^) _" I.,I,,.I'"
BROOKHAVEN
.i
Figure 3R Arithmetic average Uniform Hazard Spectra for three return periods (Minimum contributing magnitude = 5_._ 0) for the BNL- HFBR site, under the assumption of rock site characteristics, and
. with five ground motion experts' input.
J
9
SEISMIC HAZARD AT DOE FACILITIES
- CASE Hl (M>5.0i, WITH 5 GM-EXPEBT
1000O.- ' fEAB BETUBN PEB[O0 CONSTANTPEBCENf]LE SPECTBAFOB : PEBCENT[LES , 5., 1 5., 50., 85., AND 9'!i.
Figure 5R 5, 15, 50 , 85 and 95 percent constant percentile 10,000 year return period Uniform Hazard Spectra (Minimum contributing magnitude = 5.0) for the BNL- H FBR site, under the assumption of rock site
- characteristics, and with five ground motion experts' input. i
11
SEISMIC H_ZARO AT DOE FRCILITIES CASE Hl IM>5.0). WITH 5 GH-EXPEBT iOBB_B,-YEAR RETURN PERIOD CONSTANT PERCENTILE SPECTRA FOR : PERCENTILES ' S., 15., 50., 85., AND 95,
E @3+ I I I I ' I I I I ' I l l I I I I , I I I I i
E -O I
' ' PERIOD (SEe) + +
BROOKHAVEN
Figure 6R 5, 15 , 50, 85 and 95 percent constaz,.t percentile 100 ,0 00 year return period Uniform Hazard Spectra (Minimum contributing magnitude = 5.0) for the BNL-HFBR site, under the assumption of rock site
- characteristics and with five ground motion experts ° input.
f
12
