ML16054A009

FOIA/PA-2015-0294 - Resp 2 - Final, Agency Records Subject to the Request Are Enclosed. Part 8 of 8

ML16054A009
Person / Time
Site:	Palo Verde, Columbia, Diablo Canyon
Issue date:	02/12/2016
From:	NRC/OCIO
To:
References
FOIA/PA-2015-0294
Download: ML16054A009 (281)

Text

Columbia - Hazard Curves by Seismic Source Figure 10.44 SSHAC Report

Seismic Sources Characterized in the SSC Model -

SSHAC Figure 8.1

• Zone B

• Zone C

• Zone D

• YFTB Background

• YFTB Faults 12ZW 120W E[M]

1.85 to 2 0 5 to 6 -

• Fault Sources 0

0 2 to 3 3 to 4 0 6 to 7 c:J Source Zones 0 4 to 5 0 >= 7 2

Seismic Sources Characterized in the SSC Model -

SSHAC Figure 8.1

• Zone B

• Zone C

• Zone D

• YFTB Background

• YFTB Faults 3

Ta ble 8.5. Fault soW'Ce~ including fault M!gments.

Fault So1irce Abbreviation 20 Fault Sources Ahtamun Ridtte Arlmgtou Cleman Mountain Columbia HiUs AR AF CM CH Columbia Hills-Centrnl-East CH-C-E Columbia Hills-East CH*E

" ~l")(r'Q'\¥ QI' ' tr.)QO'W Colm11b111 Hills-West CH-\\"

FH*E SFZ Columbia Hills-Ce11m1I CH-C Columbia Hills-Ce111ral-W~1 CH-C-W

~--

+~

Freuchmilll Hills FH Hom Rapids Fault HR

>>f""

Table 8.5. (contd)

~.

Fault Source Abbreviarion Horse Hea\en Hills HHH Horse Heaven Hills-Central HHH*C Horse H~\'eu Hill>-Cenn*aJ-East HHH-C*E CCf>' 0 100 200 Kilometers Horse Heaven Hills-Central-West HHH-C-W Horse Heaven Hills-\Vei;t HHH-W Laurel 1.F Lw13 Bune LB

VlanaMa>h Ridge MR Mauastasb Ridge-Cenrral MR*C Mauastasb Rid{te-East MR-E Manastash Ridge-West MR-W Mau pm MF Ranle$ of tbe Ranlesnake-Wallula Ahgmueru RA\\"

Ranlesn:ike Rills RH Rau lesnake !\fountain RM Snd<Ue Mountains S~I Saddle Mow1tn1M-E.'lst SM-E Saddle Mo1unains-Wes1 SM-W Sean I~ Fault SFZ Selah Bune SB Topptubh Ridge TR TOJlpemsh Ridge-Ehl TR-E Toppeni>IJ Ridge-West TR-W Umtauum Ridge UR tJnu:mmn Ri~e..Southeast Anticline UR*SA Umtauuru Ri~e-Oable Mountain UR-GM Umtamun Ridge.('eorrnJ UR-C Umtamuu Ridire-East UR-E Um1a111u11 Ridge-W~r UR-W Figure 8.43. Fault sources and fau lt segments. Teeth are shown on the hanging wall of the faults and Wallula Fault WF Yakima Ridge YR squares define the segment boundaries. Acronyms for fault sources are given in Table 8.5.

Yakwia Ridge-Ea>t YR*E The insert shows the location of the Seattle fault relative to the Hanford Site and to other Yakima Ridge-West 4.'R-W fa ult sources. Yakinia Ridge-Southeast YR-SE

Subduction Zone Sources - CSZ and JDF (used sources from BC Hydro Report - modified?)

CSZ - Cascadia Interface JDF - Cascadia lntraslab Source Downdip seismic limit ,. *.

Near c oa tal cities closest approach Plate Interface in Black red labeled ""-"----=z~one Seismogenic zone = ~~!!:...~----~ Arrows.=

Intra Slab Cascadia Interface Source (Licensee's (Licensee's CSZ Source)

JDF Source)

Figur~ 8.8. Diagrammatic depiction of lhe seismic sources and other elements related to the CSZ (Hyndman 2013). The plate interface source is shown m red and labeled "'sei.smogemc zone." The mtraslab source is shown by the do\\-n-gomg blatl arrows withm the Juan de Fuca plate. Episodic tremor md slip (ETS) earthquakes are labeled ~-ETS zone." The landward extent of the plate mterface. which is a ke)' aspect of the SSC model. is labeled "clowndip seisnnc limit" (Hyndman 2013) The location of Hanford Site lies east (to the right) of the Cas.cade volcanic arc.

5

SiteC mean TOT

--* -*- CSZ

..... ..lf.."*o* JDF

--0-*- ZONES

....-.9 ... . - ZON EC

--*&- YFTB

- ..-& ..... ZONED AF ARH CH CM FH HHH HR LB LF MF MR RAW RM SB SFZ SM TR UR YR WF 1o*8 L-'----1.__._._..............J_..1-..1..._.__._................_,,L..........L-..L...A.....L..u..i..L:--'"-.1.-._........_i~~~!.J&.w..~.......___.__._~ o 10"6 10*5 10-4 10*3 10*2 10*1 10 6 T1 Osec Spectral Acceleration [g] f = 1/period = 1/10 sec = 0.1 Hz

% Source Contribution - 0.1 Hz

% Contribution 0.1 Hz at lE-04

% Contribution 0.1. Hz at lE-05 UR RM UR RAW 1% RM SM HHH HHH HR No Data No Data 1%

11% 11%

CM or YR 1 5%

3%

3%

Note: Sources that contribute less than 1% and w ith no data are not labeled 7 Hand digitized, uncertainty= +/-10%

SiteC

- - meanTOT

• --* -- CSZ

• -***-*.. JDF

- &-- ZONEB

.....- &-..- ZONEC

- ..**0 - YFTB

~-a--.- ZONED

- - AF

- - ARH CH

- - CM

- - FH

= = HHH

- - HR

- - LB

- - LF MF

- - MR

- - RAW RM SB

- - SFZ

- - SM

- - TR UR

--YR

- - WF 1o*s L.....__.___.__._.a....LL.&l_...__._..................i...u.L___.__.__...............~~......._,,...............uu;up.~~~ o ....Jt.al-..__._.~ 1 10-5 10-4 10*3 10*2 10*1 10 10 8

T1 .Osec Spectral Acceleration [g] f = 1/ period =1/ 1 sec = 1 Hz

% Source Contribution - 1 Hz

% Contribution 1 Hz at lE-04  % Contribution 1 Hz at lE-05 No Data\

SM 3% \

1%

No Data UR SM 8% 6%

UR RAW 2% LF. RM.

6%

5%

HHH RAW 2%

HR 4%

CM or YR 1 6%

ARH 2%

Note ~ Sources that contribu te. less than 1% and with. no data. are not labeled. 9 Hand digitized,. uncertainty= +/-10%

SiteC

- - meanTOT

- -+**-- CSZ

................. J OF 10-1

-*-&*- .. ZONES

....- 9 ........ ZONEC Cl) -*-&-..... YFTB 0

c "

0

-a . .... ZONED cu - - AF "t:J - - ARH Cl)

Cl) - - CH 0 - - CM

>< 10-3 w - - FH

....0 HHH

- - HR

~ 10 4

- - LB c - - LF Cl)

J - - MF C"' - - MR Cl> 1o-5 - - RAW

~

LL RM cu SB

J - - SFZ c 10-6 c - - SM

<( - - TR UR 10-7 --YR

- - WF 10~.____.__..___._.. . . . . . . ...___._..._._. . . . . . . ......____.__.__._.....................____...~.....__..........................._...._.......................__............_............_.............,

4 3 2 1 1 2 10 10- 10- 10- 10° 10 10 T0.1 sec Spectral Acceleration [g] 10 f = 1/period = 1/0.1sec = 10 Hz

% Source Contribution - 10 Hz

% Contribution 10 Hz at lE-04  % Contribution 10 Hz at lE-05 No Data -4 RAW UR No Data I RM 3%\ 3%

1%. I 4%

~

CM or YR 1 4%

HHH 3 Note : Sources that contribute less than 1% and with no data are not labeled 11 Hand digitized, uncertainty= +/-10%

SiteC

- - meanTOT

• - ---*- CSZ

--***-- JDF

- &-- ZONEB

.......e-*- ZONEC

- &-- YFTB

• • -*0-- ZONED AF

- - ARH CH

- - CM

- - FH

- - HHH

- - HR

- - LB

- - LF

- - MF

- - MR

- - RAW RM SB

- - SFZ

- - SM

- - TR UR

--YR

-~ WF 10*B .__......___._._._..............._....__.__._._................_....__._........................._~_.........._.......U-1.1.i.~........................~...__.__.__._~

10-4 10-3 10*2 10* 1 10° Peak Ground Acceleration (g] 12 f = 100 Hz

% Source Contribution - PGA

% Contribution PGA at lE-04  % Contribution PGA at lE-05 No Data No Data 12%

17%

SM UR 5%

UR 5%

SM 2% J RAW 1%

RM 7%

RAW CM.or. YR.1 4%

2%

CM or YR 1 ARH 3% 1%

Note: Sources that contribute less than 1% and with no data are not labeled 13 Hand digitized, uncertainty= +/-10%

Results, Questions, and Notes

• At 0.1 Hz CSZ, but JDF subduction zone source doesn't contribute to hazard at lE-04 and lE-05 hazard levels

• At 1, 10, and 100 Hz the YFTB background source dominates the hazard, followed by fault sources in the YFTB (CSZ contributes a 1 and 10 Hz).

• Top 5 sources at lE-04 and lE-05 for 0.1, 1, 10, and 100 Hz are:

- RM - Rattlesnake Mountain

- CM or YR 1- Cleman Mountain or Yakima Ridge

- UR-. Umtanum. Ridge

- HHH - Horse Heaven Hills

- HR - Horn Rapids Fault

- ARH - Ahtanum Ridge (AR)+. Rattlesnake Hills (RH)? 14

Results, Questions, and Notes (continued)

• Ask Licensee?

- Provide data not available for curves (outside plot area) - those sources less than lowest percent?

- Can't read color difference between CM and YR (black). Hazard for black curve (either CM or YR) matters at 10 Hz where it is 15% at lE-04 and 20% at lE-05 and at 0.1 Hz and lE-05 (second highest contributing source = 5%)

- Is ARH for the hazard curves AR + RH on the map? How do they decide when to split and combine faults?

- Double counting deformation YFTB? - fault sources+ YFTB background seismicity- do they interpret them to be connected?

• For Site B - there is an odd kink in TR curve for 10 Hz (O.lsec) between 0.1 and 1 Hz (Hazard Curves at 10 Hz add up to >100% for sources with data)?

Over estimated hazard? Why does. this. kink exist?~ not seen for site C

• Note x and y axis extents on hazard curve plots changes between plots

• Need to check - Do these results match with deaggregation plots?

15

Figure 8.43 . Fault sources and fault segments. Teeth are shown on the hanging wall of the faults and squares define the segment boundaries. Acronyms for fault sources are given in Table 8.5.

The inse1t shows the location of the Seattle fault relative to the Hanford Site and to other 16 fault sources.

5/19/2015 Columbia R2.1 Review - Questions for Public Meeting from Lisa SSHAC Report Section 8.4.3.4 - Structural Relief to Net Slip Conversions In SSHAC Report Section 8.4.3.4, the licensee indicated that "the dip angle of thrust or reverse faults beneath the YFB anticlines was derived using a simple model to relate the fault geometry to the plan dimensions of the folds." The licensee indicated that a range of alternative dips and seismogenic depths were. used to model. the deformation of surface topography from faulting .. The licensee then used the parameters from the range of models that best fit the observed topography above each fault and the measured surface relief to predict the net fault slip.

1. Limited information on structural relief to net slip conversion provided -

Staff acknowledges that variation and fault dip angle and depth of faulting will create variations in the surface topography above the fault and that with limited information the exact fault geometry could be unknown . . Please provide additional information on the. following

i. Provide the depth of faulting and dip value parameters used. to derive the slip rates with their models and the range of values used in alternative models to estimate the uncertainty ii. Show all observed topographic profiles in cross-section and their location in map view and how they compare to modeled topography predicted by model used for fault slip estimat ion iii. Provide a table with structural relief, displacement estimate, and slip rate iv. Show profiles with varying depth of faulting and dip angles and how the associated modeled topography compared to the observed topography

2. Listric or complex geometries not fully considered or discussed-

1. Observations show that some of the faults have a back thrust, why isn't this incorporated into the three geometries indicated by the licensee? In one of the models, the. licensee uses a steep fault that the. licensee indicates. gives a slight concavity to the resulting topographic profile .. Field. and geologic mapping in the region indicates faulting is more complex (i.e. presence of back thrusts and listric geometries). Are these geometries described in the geologic mapping sections of the SSHAC report? They are cited in published literature for the YFTB. These geometries don't appear to be incorporated into the modeling used to estimate the net slip.

2.. Please provide a more in depth discussion of why the. thin-skinned model. was not selected, do not have access to zacharanin report.

3. Previous research by Watters et al. of similar styles of faulting in basalt-like material indicated that using listric geometry over a simple fault could produce net slip rates 1.5

- 2 or more t imes greater. If they have evidence for listric geometry or backthrust for the faults in the YFTB, why aren't they used in their models, especially if it wou ld produce a higher net slip rate? I still need to read more in the geology sections to see what the licensee describes about the faulting from field and geophysical observations.

4. I need to look up the reference, but I know that there were geophysical survey's done in the YFTB that show complex fault geometries beneath the anticlines in the YFTB.

5/19/2015 Ideally, those geometries should be used in the net fault slip estimation rather than a simple flat dipping fault.

5. Average not maximum relief used to derive slip rates -

a. The licensee stated that the average, not maximum, structural relief along faults used to estimate fault slip (See Appendix E page 5.10). Is this valid? With uncertainty, why not use maximum measured relief? Is the maximum relief considered in the logic tree as a possibility for rupturing?

b. Rattlesnake Mountain unfaulted quaternary deposits= not maximum estimated slip rate? Or timing of slip stopped before quaternary?

Columbia R2.1 Review - Questions for Public Meeting from Lisa Seismic Source Characterization Structural Relief to Net Slip Conversions Please show additional details of modeling used to. convert structural. relief to net slip, including

1. Please show comparisons of the actual observed topographic profiles with topographic profiles predicted from elastic dislocation modeling used for slip calculations for each fault source in the YFTB

2. Please provide a table with measured structural relief from topographic profile, displacement estimate from. modeling, and slip rate for each fault source in the. YFTB

3. What range of depth of faulting and dip value parameters were used in the modeling? How were. these alternative models used to predict uncertainty?

4. Was a model considering a listric geometries or complex geometries (i.e. presence of a backthrust) considered?

5. Please explain why the average, rather than the maximum, structural relief along faults was used to estimate fault slip (See Appendix E page 5.10).

Thin- vs. Thick-skinned Tectonic Environments.

6. Please provide additional background information on the findings in Zachariasen et al. (2006) and supporting evidence. that led your selection of 100% Thick-skinned over Thin-skinned tectonics. If the models.were weighted equally in previously studies, what is the new evidence.

that establishes that weight should be given solely to the thick-skinned model?

7. If the regional tectonics is solely thick-skinned, can that type of tectonic environment readily explain t he nearly regu lar spacing of faults in the YFTB that could be explainable by faulting on a decollement via the thin-skinned model? (i.e. Schultz and Watters, 1995)

8. If you used the thin-skinned tectonics model in your structural relief to net slip calculations, how much would it change the predicted structural relief, thus net slip and resulting hazard curves at lE-04 and lE-05 at 1 and 10 Hz for the GMRS?

1.

Columbia - Source% Contribution - 0.1 Hz.

SlteC

- meanTOT

- + - CSZ JOF ZONEB ZONEC

- YFTB

-E ZONED

- AF

- - ARH CH

- - CM

- FH HHH

- - HR

- LB

-LF MF

- - MR

- - RAW RM

- - se

- - SF2

- - SM

- - TR UR

- - YR

- - WF 10*.___..__...._...___..___.,......____.___..............__~_,__...__,_,,~...............-....................

10* 10~ 10~ 10-3 10 2 10*1 10° T10sec Spectral Acceleration [g]

RJ6 Contribution 0.1 Hz at lE-05 UR 2%

HHH c~

ARH 1%

YFTB 3%

1

Columbia - Source % Contribution - 1 Hz SiteC

- meanTOT csz

- JDF ZONEB ZONEC YFTB ZONED

- AF

- - ARH CH

- - CM

- FH HHH

- - HR

- LB

--LF

- MF

- - MR

- RAW RM SB

- SFZ

- - SM

- - TR UR

- - YR

- WF 10~--~............-----...... --~--.___...__._._~...._.,.._.........................._..._....

10"5 10"' 10*3 10 2 10*1 10° 10 1

T1 .Osec Spectral Acceleration [g)

No % Contribution 1 Hz at lE-04 UR  % Contribution 1 Hz at lE-05 Data~ 6 8%

UR

%~ata

% \

SM 6% RAW 2% 5%

RAW 2%

HR 9%

6% 2%

2

Columbia - Source % Contribution - 10 Hz SiteC 1a6 ----~----~~----~--~~~~---~~---~---.-. -------------~ - meanTOT 10*

1Ir------- csz JOF ZONEB

--&- ZONEC YFTB

-< ZONED

- - Af

- ARH CH

- - CM

- FH HHH

--HR

- - LB

--Lf MF

- -MR

- - RAW RM SS

- - SFZ

--SM

- -TR UR

- - YR

- - WF 10~ .__--~........._ ____~......._____~......._............._............ua......._._..~------_...,

10.. 10 l 10~ 10** 10° 101 101 T0.1sec Spectral Acceleration [g)

% Contribution 10 Hz at lE-04  % Contribution 10 Hz at lE-05 No Data RAW UR No Data

-4 RM 3% 3%

4% I 4%

3

Columbia - Source% Contribution 100 Hz {PGA)

SiteC 10° ., ., T I I l

- meanTOT l csz

__...,__ JOF 1

Cl>

10*

.r ZONEB ZONEC

-& VFTB 0 ZONED c - AF ca "O - - ARH Cl>

Cl> CH u - - CM

)(

w - - FH 0

HHH

- - HR

~ 10"4 - - LB c --LF CD j r - MF CT - - MR

£ 10.st - - RAW RM ca j

J

- sa 10""1 - SFZ c - - SM c

ct - TR UR 10*7 - - YR l - - WF f

io""

10.. 10-$ lO~ 10 I 100 10 1

10 1

Peak Ground Acceleration [g)

% Contribution PGA at lE-04  % Contribution PGA at lE-05 UR UR 5% SM %5 SM """'\. RM 2% \ 6%-!!:__1 RAW J HR l% \

4%

HHH 2%

CM or HHH 4% YR 1 ARH 4% 1%

3% 1%

4

Seismic Sources Characterized in the SSC Model -

SSHAC Figure 8.1

• Zone B

• Zone C

• Zone D

• YFTB Background

• YFTB Faults 4

5

20 Fault Sources ......_... a.Mie<

~

l1CllUllM-11u1

~ .. ma.

Alt U'

CM (1J C<lhllllli* llllk<ftllllli.~ C1H'-.t

('.......... Htll>-Ealc ru-r

.:*"WWW ......... ('o-*Hdi..WN Ol-W f1oi-E SFZ ('olulltiit Rollt-C""""I Ol*<

('o,....*lloll..C-.1.\\ ..1 rn..c.w

~- rn

+~

FJtD..ii:1.,. lhlh Hom llapab r ..11 Hit 0

w

~. 100 2 0 0 -....

11..... ~.... 11.n.

t i - ffn\'9 lltlM'-

• I H.cltoe Ko\ t'i6 lu..a.-t ftllft.IS,..hlll 11......-llillo.c_..1-.......

1toc.. lb,.. HJ.11o.\\n1 llHH HID!.(

HlW.('.f HHll-<:'-1\

llHlf.\\

Ulftl Lf

~-llullt ~

\1.W1iu.i.lt,o(h;< \Qt

~i.w....-.o>l>~t><°<<UlL>I \IR-('

M-Rldf:t*E.obl MR*£

~l->J,tatb lbd;*-\\'"" \lfl.\\'

M.w...  !\IT R.wtle.ofo11t~i.....i........u.,.~ RA'f.

IW.i.-l:t HlU* IUI

&Mllc>A.l*t Moc.1:1111 RM s.JJlc M-*la S1'j Wiit~....&" SM*t

"-"l.lk\i.-.....w... '~M\ "

'W"Nl l~f<lc~ '<Fl

!><lab 8'11!0 ~a T"l'1-,b~ rR fqll-!Wp-C.'4 fk-l 1.,,.....r1 !W~w.., 111-\I t_!W,,. lTR

""'"""" Rllbe-'io111....... Antl<fu>o IJJl. A

'-'--~1.,1 ......... L'l\-G.\f L*,.~- IW*e-<'n111ll U"R.C 1*,......, RJ4li ...r_... Vll-1'.

l\i.

V*ro*u R.Mltt"-\\ '~1 l'll-1\

Figure 8.43. Faull source and fault segments. Teeth are sllowu on me haugiug wall of the faults and \\anu1araodl WT squares define the segment bow1druies. Acronyms for fault sources are given in Table 8.5. \ 'ol..... \R

\ "ol.11. . R*.ljle-i:... \R.(

Tite insen shows the locariou of 1he Seanle fanh relative to the Hanford Sue and to 01her \" llJ.o111R~\\ ... 5,R*V.

fault ources. Yal:rono Rl4tt-~- nt.!>J:

6

Subduction Zone Sources - CSZ and JDF (used sources from BC Hydro Report- modified?)

CSZ - Cascadia Interface JDF - Cascadia lntraslab Source Near eoa tal cities Down dip closest approaeh Vole .

seismic limit 1* * ~

Y*ncouver Seattle Pord8nd Plate Interface in era ck red labeled S.!!isnioge.nic zone= ~~--H-_....--- Arrows=

Intra Slab lntraslab Source Source (Licensee's (Licensee' s CSZ Source)

JDF Source)

Fi:ure 8.8 Diagrmuna:tic dep1ctiou of the SfiSDDC sowces and other elements related to the CSZ (Hyndman 2013)_ The plate interface source 1$ shown m Rd and labded ..~lSmogemc zone .. The mtraslab source is shown by the down-gomg black arrows wllhm the Juan ~

fuea plate Ep1$od.te tremor and slip (ETS) ewquakes are l~beled ..ETS zone... The landward extent of~ plate mterface. which is a key aspect oflhe SSC model. 1S labeled

downdip seismic li.mlt.. (Hyndman 2013) ~location of Hanford Site lies east (to the nght) of the Cascade volcanic arc..

2 7

~--

+ ~ +

....... 0

~ 100 200 Kilome~

Figul'e 8.43. Fault sources and faulr segmeuts. Teeth are shown ou the hanging wall of the faults and quares define the egmenc botmdarie . Acronyms for fault sources are given in Table 8.5.

The insert shows the location of the Seattle faulc relative 10 the Hanford Site and to other 11

,.. *~

8

115*w 0 0 00 O<fj O 0 Q) 0

'IC"t>

0 0 0 0 o 0

0 0 0 z

o'f' ' I~ ~ z I

- - USGS Quaternary Fault Database Earthquakes C!: Magnitude 1.85 IU 0 Licensee Cataog

• Crustal

• 11124/1866 to 413012013 z 0 0 11> 0 Licensee Catalog - Subduction M1

• 1112411866 to 4130/2013

" 0 Licensee Catalog* Subduction M2 - 11/24/1866 to 4/30l2013 o\ o i

'~ Coordinate System: World G-wdetic Sysl'l!m 1984 ~

0 NRC Confirmatory Calalog. ANSS 5/112013 to 6/1/2015 (post-licensee catalog)

~I \'~ Projec\ion: Transverse Mercator, UTM Zone 11N 0 NRC Confirmatory Catalog* ANSS 1/1/1898 to 4130/2013 Esri, ~b::>~rne, GEBCO, NOAA NGOC, and other contributors 12s*w 11s*w

Digitized from Columbia SSHAC Report Figure 10.44 (electronic files not available)

Opt1Hz_10sec Color. of Line SA value at 1 OOE-04 for all is 0 0 % Contrib SA value at 1 OOE-05 for all is 0 037 % Contrib mean TOT 1.00E-04 mean TOT 1.00E-05 csz 0.000075 75 csz 0.0000075 75 subduction zone sources JDF 0.00000002 0.02 JDF NO DATA NO DATA ZONEB 0.00000008 0.08 ZONEB 0.000000015 0.15 four seismic source zones in the SSC mode ZONEC 0.00000003 0.03 ZONEC NO DATA NO DATA red dashe YFTB 0.0000026 2.6 YFTB 0.0000003 3 ZONED NO DATA NO DATA ZONED NO DATA NO DATA AF NO DATA NO DATA AF NO DATA NO DATA ARH 0.0000007 0.7 ARH 0.00000005 0.5 CH 0.00000002 0.02 CH NO DATA NO DATA black (can CM or YR 0.0000028 2.8 CM or YR 0.0000005 5 FH 0.000000015 0.015 FH NO DATA NO DATA HHH 0.0000016 1.6 HHH 0.000000075 0.75 HR 0.0000007 0.7 HR 0.000000045 0.45 LB NO DATA NO DATA LB NO DATA NO DATA LF NO DATA NO DATA LF NO DATA NO DATA MF NO DATA NO DATA MF NO DATA NO DATA MR 0.000000012 0.012 MR NO DATA NO DATA RAW 0.00000065 0.65 RAW 0.000000045 0.45 RM 0.0000028 2.8 RM 0.0000002 2 SB NO DATA NO DATA SB NO DATA NO DATA SFZ 0.00000003 0.03 SFZ NO DATA NO DATA SM 0.00000055 0.55 SM 0.00000003 0.3 TR 0.00000009 0.09 TR NO DATA NO DATA UR 0.0000016 1.6 UR 0.00000016 1.6 black (can YR or CM. NO DATA NO DATA YR or.CM NO DATA NO DATA WF 0.0000001 0.1 WF NO DATA NO DATA NO DATA HAZARD CURVE OFF CHART AND CANT EXTRACT VALUE

SSC model Source AFE PGA 1E-4 %

csz NO DATA NO DATA JDF NO DATA NO DATA MR AF NO DATA NO DATA TR LB NO DATA NO DATA 0.2%

LF NO DATA NO DATA SB CH MF NO DATA NO DATA 0.04%

SFZ NO DATA NO DATA No Data (CSZ, JDF, AF, LB, LF, YR or CM NO DATA NO DATA MF, SFZ, YR or 1 YFTB 0.000055 CM2) 2 RM 0.000007 16.9%

3 UR 0.0000048 4 HHH 0.000004 5 CM or YR 0.0000032 3.2 6 HR 0.0000024 2.4 7 SM 0.000002 2 8 RAW 0.0000016 1.6 9 ARH 0.0000012 1.2 10 ZONEC 0.00000045 0.45 11 WF 0.00000038 0.38 12 ZONES 0.00000035 0.35 13 CH 0.00000018 0.18 Columbia-13 FH 0.00000018 0.18  % Contribution 14 TR 0.00000016 0.16 by Seismic Source 15 ZONED 0.00000012 0.12 (PGA, lE-4) 16 MR 0.00000008 0.08 17 SB 0.000000035 0.035 No Data (CSZ, JDF, AF, LB , I 16.865

Source AFE (PGA%

ZONED NO DATA #VALUE!

TR csz NO DATA NO DATA ZONEC ZONES WF JDF NO DATA NO DATA AF NO DATA NO DATA CH NO DATA NO DATA FH NO DATA NO DATA RAW LB NO DATA NO DATA 1.3%

LF NO DATA NO DATA MF NO DATA NO DATA MR NO DATA NO DATA SB NO DATA NO DATA SFZ NO DATA NO DATA YR or CM NO DATA NO DATA 1 YFTB 6.SE-06 65 2 RM 5.5E-07 5.5 3 UR 4.5E-07 4.5 4 CM or YR 3.9E-07 3.9 5 HR 3.5E-07 3.5 6 HHH 1.8E-07 1.8 7 RAW 1.3E-07 1.3 Columbia -

8 SM 9.5E-08 0.95  %. Contribution 9 ARH SE-08 0 .8 by Seismic Source 10 ZONEC 1.1E-08 0.11 (PGA, lE-5) 11 ZONEB 1E-08 0 .1 12 TR 9.5E-09 0 .095 13 WF 9E-09 0.09 No Data (Zone D, CS 12.355 meanTOT 1.00E-05

Source AFE 110Hz. Double fault sources onlv  %  % double !Source AFE  % I csz NO DATA NO DATA NO DATA NO DATA ZONES 1.30E-08 0.13 JDF NO DATA NO DATA NO DATA NO DATA ZONEC 1.25E-08 0.125 AF NO DA TA NO DATA NO DATA NO DATA YFTB 0.000008 80 LB NQDATA NO DATA NO DATA NO DATA ZONED NO DATA NO DATA LF NO DA TA NO DATA NO DATA NO DATA AF NO DATA NO DATA HR, 3.8 MF NQDATA NO DATA NO DATA NO DATA ARH 4.SE-08 0.48 SB NO DATA NO DATA NO DATA NO DATA CH NO DATA NO DATA SFZ NQDATA NO DATA NO DATA NO DATA CM or YR 4E.Q7 4 YR or CM NODA TA NO DATA NO DATA NO DATA FH NO DATA NO DATA YFTB 0.000075 0.000075 75 75 HHH 8E*08 0.8 ZONED 2. 1E--08 0.00000002 1 0.021 0.021 HR 3.4E-07 3.4 RM 0.0000075 0.000015 7.5 15 LB NO DATA NO DATA UR 0.000005 0.00001 5 10 LF NO DATA NO DATA CM or YR 0.0000042 0.0000084 L 4.2 8.4 MF NO DATA NO DATA HR 0.0000038 0.0000076 ( 3.8 7.6 MR NO DATA NO DATA HHH 0.000003 0.000006 3 6 RAW SE--08 0.8 RAW 0.0000016 0.0000032 I 1.6 3.2 RM 3.SE-07 3.5 SM 0.0000015 0.000003 1.5 3 RM, ?.S SB NO DATA NO DATA TR 0.0000012 0.0000024 1.2 2.4 SFZ NO DATA NO DATA ARH o.ooooo~J 0.000002 1 2 SM NO DATA NO DATA ZONEC 3.SE-07: 0.00000035 0.35 0.35 z~!ll.taia TR 4.SE--08 0.48 WF 9E--08 0.00000018 0.09 0.18  % Contiibution UR 3.4E*07 3.4 ZONES 1.3E-07 0.00000013 0.13 0.13 by Seismic Source YR or CM NO DATA NO DATA FH 4.2E-08 0.000000084 0.042 0.084 (10Hz Hz, 1E*4) WF NO DATA NO DATA CH 3.9£-08 0.000000078 0.039 O.D78 CSZ NO DATA NO DATA MR 1.2E-08 0 000000024 0.012 0.024 JDF NO DATA NO DATA No Data -4.484 0 104.484 UR, 10.0 Columbia -

% Contribution b y Selsmlc Source (10Hz Hz, lE-4)

• fault sources doubled

Source AFE 110H; % !source AFE (10H; %

csz 7E-07 7 ZONED NO DATA NO DATA JDF NO DATA NO DATA AF NO DATA NO DATA ZONEB 2.8E-08 0.28 FH NO DATA NO DATA ZONEC 1.7E-08 0.17 CH NO DATA NO DATA YFTB 0.000006 60 LB NO DATA NO DATA ZONED NO DATA NO DATA LF NO DATA NO DATA AF NO DATA NO DATA MF NO DATA NO DATA ARH 1E-07 1 MR NO DATA NO DATA CH NO DATA NO DATA SB NO DATA NO DATA CM or YR 9E-07 9 SFZ NO DATA NO DATA FH NO DATA NO DATA SM NO DATA NO DATA HHH 1.8E-08 0.18 YR or CM NO DATA NO DATA HR 5E-07 5 WF NO DATA NO DATA LB NO DATA NO DATA 1 CSZ NO DATA NO DATA LF NO DATA NO DATA 2 JDF NO DATA NO DATA MF NO DATA NO DATA 3 YFTB 0.000008 80 MR NO DATA NO DATA 4 CM or YR 4E*07 4 RAW 1.5E-07 1.5 5 RM 3.SE-07 3.5 RM 5E-07 5 5 UR 3.4E-07 3.4 SB NO DATA NO DATA 6 HR 3.4E-07 3.4 SFZ NO DATA NO DATA 7 RAW 8E-08 0.8 SM 7.5E-08 0.75 8 HHH 8E-08 0.8 TR 1.4E-08 0.1 4 9 ARH 4.8E-08 0.48 UR 6E-07 6 10 TR 4 .8E-08 0.48 Columbia.-

YR or CM NO DATA NO DATA 11 ZONES 1.30E-08 0.13 % Contribution WF NO DATA NO DATA 12 ZONEC 1.25E-08 0 .125 by Seismic Source No Data 2.885 (lOHz Hz, lE-4)

-...~~~l!!L"D!."W"~- *ll C!lll!Lo

...O l f l t - U . h W...* - -...... ~*-.........._.. _ _ ....._elWI-

.*~

1-QbJ~- . i *"*

g -;._--

. . :. \ *. *:

5. ii-~-il~ ~ '9 I* . ~

• ]**

=

1111 SSE RG 1.60 Frequency SA [g] Frequency SA [g]

0.4 0.12 0.1 0.019 2.05 0.6 0.25 0.118 6.1 0.6 2.5 0.783 18.9 0.25 9 0.653 100 0.25 33 0.25

freq 5% Damped.Spectral Acceleration [g]

0.500 0.283 0.980 0.608 1.937 0.881 3.214 1.007 4.770 1.034 6 .861 0.963 9 .720 0.810 19.985 0.522 29.709 0.396 49.765 0.395

Columbia Generating Station Site Response (Part 2)

Vs Measurements Shear Wave Velocity (ft/s) 0 2000 4000 6000 8000 10000 12000 0

I I I I

"'"'IL_

200 II I

~ - WNP-2 I

400 I - WNP-1 Crosshole

- WNP-1 Downhole I

E' .

I

- 600 I - WNP-4 Crosshole

.s=

0 Q.

Q) 800 I

I I - WNP-4 Downhole

---

* I I

- WTP Downhole

.......... I 1000 I I I

I ---- WTP P-S suspension I

I I

1200 .... .:

a 1400

Licensee Vs Profile Shear Wave Velocity (ft/s) 0 2000 4000 6000 8000 10000 12000 0 ...... I I I I I Zone 1 200 -

Zone 2 400 -

Zone 3

.I E'

.c:

+J

~

600 -

I I

I

~ 800 -

---

. I I

1000 - I* ----

I Zone 4 I

I t

1200 -

• --*I 1400

- -vs P1 (Wt= 0.67) ----Vs P2 0Nt = 0.33)

Vs Profiles Shear Wave Velocity (ft/s) 0 2000 4000 6000 8000 10000 12000 0 ---,_,=---~~*~~~~-*~~~~-*~~~~~~~~~~~~~~

• ~ -

200

,_I _. . .I I I I r.mm _ _,

- Licensee Pl (Wt =0.67}

I 400 I I - - - - Licensee P2 (Wt= 0.33)

I I I I I * - - NRC BC

--.... 600 ~~~~~~~~~~~~~~---

I

• ~--

I I I

' .I

! ---* NRC LBC/UBC

~

.c c:::::::::::::::::::::.:::::;;;:::=~-~. -.

I cr1n = 0.15

~ I I I Q) I I I 0 I I I 800

..-_~...................*. ......~-----L - -r --*l I I I I I I 1000 t, ~------------------llllllil----11111111-----

11 : ---,--*

I I I I I I I t I I I I 1

~~~~~~~~~~~~~~~~*~~------~*~

1 1200 r,_

I

-..m11111 ii

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . __ ~~--~

I

_~--~~~

Vs Profile Questions

• Is it. adequate to only lJse aleatory uncertainty in upper 525 ft?

• Does use of the two Vs profiles and associated weights for the SMB stack adequately capture the epistemic uncertainty at CGS?

• Should random profiles consider limited.

lateral extent of interb eds (i.e. zero thickness 1

interbed)?

Damping

• Suprabasalt Sediments

- Strain. dependent dam1:>ing

• SMB stack

- Q model used to determine small strain damping

- Constant strain in. basalt layers.

- Strain dependent dam1:>ing in interbeds

Development of Q Model Used for SMB Stack Damping 0.08 y =1/0.0345*x (d}

Q = Yl1s 0.07 R2 =0.5033 u

Q) 0 .06 0 .05

• Kd amping 'H*

= L,. t. v;Si ~*t

.e> 0 .04

~

IV

~ 0.03 0.02 0 .01 0

0 0 .0005 0.001 0.0015 0.002 SUM(Hl/Vsll\2) (sec2 /m)

• Data points Fit Data points represent Kdamping for stations HAVvA, E07 A, E08A, F07 A, D08A, E09A and associated Vs profiles.

Damping in SMB Stack Table 7.2-'. Damping propert ies in the S:MB stack .

Profile 1 Profile 2 Uuit V s (km/sec) s V s (km/sec) ~

Ice Harbor (Martindale flow rap 1) 1.4 1 1.03% 1.77 0.82%

Ice Harbor (Ma1tindale flow rop 2) 1.67 0.87% 2.11 0.69°0 Ice Harbor (Manindale flow rap 3) 1.93 0.75% 2.44 0.59%

Ice Harbor (Manindale flow) 2.3 1 0.63% 2.9 1 0.50%

Levy Inrerbed 0.85 1.71% 0.85 1.71%

Elephant Mountain (flow top 1) 1.41 1.03% 1.77 0.82%

Elephant Motmtain (flow top 2) 1.67 0.87% 2. 11 0.69%

Elephant Mountain (flow top 3) 1.93 0.75% 2.44 0.59%

Elephant Mow1tain 2.31 0.63% 2.9 1 0.50%

Rattlesnake ridge Interbed 0.8.i 1.73% 0.83 1.74°0 Pomona (flow top 1) 1.43 1.01% 1.77 0.82%

Pomona (flow top 2) 1.70 0.85% 2.11 0.69%

Pomona (flow top 3) 1.97 0.74% 2.44 0.59%

Pomona 2.52 0.58% 3. 12 0.46%

Selah Interbed 0.88 1.65% 0.97 l.49°'o Esquatzel (flow rop 1) 1.49 0.97% 1.74 0.83%

Esquatzel (flow top 2) 1.80 0.81% 2.11 0.69%

Esquatzel (flow top 3) 2.00 0.72% 2.35 0.62%

Esquatzel 2.52 0.58% 2.95 0.49%

Cold Creek Interbed 0.82 1.76% 0.76 1.9 1°'o

Baserock K Kbasero ck = Ksit e-Kstack Uncertainty in SMB stack damping absorbed into Kbaserock Note: At this point, Kstack and Ksit e only considers damping portion Table 7.25. Target Kbasernck at the five hazard calculation sites at Hanford.

Profile 1 Profile . . .

Site Case All Sed. Half of Sed. No Sed. All Seel. Half of Sed. o Sed.

Lower Km\* 0.0297 0 .0189 0.0058 0.0378 0.0233 0.0058 Central 1'.:inv 0.0441 0.0285 0.0096 0.0556 0.03-l8 0.0095 Upper Kin\* 0.0491 0.0319 0.0110 0.0618 0.0388 0.0109 Site C Lo\ver KAH 0.0164 0.0100 0.0022 0.0240 0.0144 0.0028 Central 1'.:AH 0.0286 0.0182 0.0055 0.0370 0.0228 0.0056 Upper K ..\H 0.0485 0 .0315 0.0108 0.0563 0.0352 0.0097

Shear Modulus and Damping Curves SSHAC Example Licensee

• Sands (H2)

• Pasco Gravel (Sand with

- EPRI and Menq Scattered Gravel)

• Gravels - EPRI and Peninsular

- Rollins et al. and Menq

• Ringold

• Ringold - EPRI and Peninsular

- Peninsular

• lnterbeds

• lnterbeds - Darendeli and Stakoe

- Darendeli and Stakoe

G/Gmax for Pasco Gravel {Sands) 0.8

- MenqCu=5

- Menq Cu= 15

>< 0.6 cu - MenqCu=25 E

(!)

- - EPRI 20 - 50 ft

(!) 0.4

- - Peninsular 0 - 50 ft

- -

• Darendeli and Stokoe 35 ft 0 -+-___._-'-~~+--~~~--1~~~~--+-~~~~~

0.0001 0.001 0.0 1 0.1 1 Shear Strain (o/o)

Damping for Pasco Gravel (Sands) 10 - MenqCu=5 C)

- MenqCu= 15 c

Q.

- MenqCu=25 E

C'CS c - - EPRI 20 - 50 ft

- - Peninsular 0 - 50 ft

- -

• Darendeli and Stokoe 35 ft 0 -+-~----'--~~r-----~~~---+~~~~~~~~~~

0.0001 0.001 0.01 0. 1 1 Shear Strain ( 0/o)

G/Gmax for Ringold 0 +--~~~~~~~~~~~~~~~~~~~~-----!

---c:--, ,,,, ,

0.0001 0.001 0.01 0.1 0.8 ' '

Shear Strain(%) ' '

\

)( 0 .6

\

\\

\

' \

\

ca - - EPRI 120

• 250 ft E \ \

<!> 0.4

\

\

\

\

\

\

\

- - Peninsular 50

• 500 ft

- Darendefi and Stokoe 185 ft

\

0.2

0 ~~~~~~~~~--,~~~~~-.-~~~~4 0.0001 0.001 0.0 1 0.1 Shear Strain (%)

Damping for Ringold I

I I

I I

I 10 I I

I I - - EPRI 50

• 120 ft I

I - - Peninsular 50

• 500 ft I

,,I - Darendeli and Stokoe 85 ft 5

-- <Il !  ;

0 +-~~~~~~~~~~~~~~~~~~~~----<

0.0001 0.001 0.01 0.1 Shear Strain(%)

10 O>

c - - EPRI 120

• 250 ft

• a.

E - - Peninsular SO

• 500 ft ns c - Oarendeli and Stokoe 185 ft 5

0 -t-~~~~--.-~~~~--,.~~~~~..--~~~~~

0.0001 0.001 0.01 0.1 Shear Strain(%)

Strain Levels in lnterbeds Strain Profile (Target PGA=O.Sg)

Maximum Shear Strain (%)

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0

200 400

~

600

~

800 a.

~ 1000 1200 1400 1600 _.____ _____.___ ___..___ _ _ _____.__ _ _ _ _ _______,

- ~ Nonlinear Rock Behavior o Linear Rock behavior

Cold Creek lnterbed Shear Modulus Degradation

-NRC 0 .9 --* Licensee Max. Strain Range Target PGA = 0.5 g 0 .7 0 .6 -l---+---_,_..__-~_,__..__--.-1 _ _ _ _._..

>< 0 .5 _,__ _ _

ca ~ _ ____,______.___ __..

E a o.4 C) 0 .3 -!---------'-----+---~

0 .2 0.1 --i---~-----+-----.-+----i--1 0 +------.----------+----~---____,

0.0001 0.00 1 0.01 0. 1 1 Shear Strain ( /o) 0

Cold Creek lnterbed Damping Curve 20 I

---Licensee Max. Strain Range

-NRC Target PGA = 0.5 g 1---.. .........

I 15 -- I

~

I I

C>

D = b ( GG max y- 1 DMassing + Dmin I

'/

1 c

I . ,*

I

c. 10 I E ,

I I

cu c

5 ~

~

v --- I ,

I I

0 0.0001 0.001 0.01 0.1 1 Shear Strain (0/o)

Damping Equations G )0.1 D = b(G D Massing + Dmin max where:

- y- y, 1n(Y+YrJ -

0

_ 100 Yr _1 (Vo) - fI 4 D Afasing.a=l.0 y -' -

y +yr c1 = -1.1 143a 2 + 1.86 18a + 0.2523 I vr =(¢1 +¢2

• PI

• OCR¢3)* a 0 h 2

c 2 = 0.0805a - 0.0710a - 0.0095 a = ¢s J

c 3 = - o.ooosa - + o.0002a + 0.0003 Dmin = (¢6 + ¢7

• PI

• OCR ~)* a 0

'~ *[l + ¢10

• h1(fi~q)]

b = ¢11 + ¢12

• ln(1V)

Questions on G/Gmax and Damping

• Does. the use. of EPRI and Peninsular adequately characterize the epistemic uncertainty for the Pasco Gravel?

• Does. the use. of a single sand curve (Darendeli and Stakoe) adequately capture the epistemic uncertainty. in the san<:Jstone. and claystone interbeds?

- Should curves with less degradation and near constant damping be considered?

Randomization

• Layer Thickness

- Uniform distribution

• Velocity

- Suprabasalts: Lognormal, USGS B Correlation Model

- SMB Stack: Lognormal, LJncorellated

• G/Gmax and Damping

- Lognormal distribution at y = 3.16E-2

- (Tin( G ) = 0.3, CT1n(D) = 0.15 Gm ax

- cr1n(D) = 0.4 for strain inclependent damping ratios

Questions on Randomization

• Why did you randomize the damping for the basalt layers in the SMB stack when uncertainty for small strain damping in the SMB stack was incorpc>rated into the base rock kappa?

Conditional Mean Spectra 2.5 , . . . - - - - . . . - - - - - . . . . - - - - - r - - - - - - - . . - - - - - - . - - - - - - .

- - Predicted median spectrum, M-7.03, R- 12.2 km

,. - - *Predicted median+ 2o spectrum

--en c

2 I \'

- - Individual record sp ctrum

- - Conditional Mean Spectrum 0

'+:; 1.5 \

'° QJ I

CV u

u

<(

'°'-

tJ CV Q.

V'l 0.5 o.__~~---~~---i.~~~--~~__._~~~..._~~~

0 0.5 1 1.5 2 2.5 3 Period (s)

Application of CMS at CGS 101 CMS 1.0 Hz. AEF: l.Oe-04 Site C (CGS): 10-4 MAFE: T=1.0 sec SA "4 C<>ntnbutlon 25%

20'lfo

\S'tl o;

vv u io* 1 1()q(,

~u

~

Vl ta.

E 10-l IO 0

~

11"1 UHRS OISllln<:e (~m) \f : 5.6, R: 7 km , Wt: 0.15

\I* : 6.5, R: 11 km, Wt 0 39

\f : 7.2, R: 20 km. Wt: 0.25 II : 9 0. R: 336 km , Wt : 0 21 Frequency [ Hz]

CGS Site Response Logic Tree CMS2 (Wt= 0.39)

Profile 1 (EPRI Curves)

(Wt= 0.33)

CMS3 (Wt= 0.25)

Profile 2 (EPRI Curves)

(Wt= 0.167)

Profile 3 (Peninsular Curves)

(Wt= 0.33)

Profile 4 (Peninsular Curves)

(Wt= 0.167

• Note: CMS weights change with MAFE (Jln,T = Ii Wi [ (µln,i - µln,T )

2

+ ((Jln,i) 2

]

Minimum Amplification b .--__,-...-.....-....-_.........-...-...-...----.__,_....__.......__-...-...-...__,_,..__..............__.....,__,__,__,...,_.,.___,..,.,

b w,

u. 0

'\

' \

\

\

\

\

\

\

\

\

\

'9

~ I--__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,....,

Bo$erock '\

Soil M tt "'I Min mum Amr; O1 '

sou n% arn1 9:ilh't . !'Annum Arnp: o I '

Soi Meari. \4inlmurn Amp 0 5 \

son StM~ an:f 95th%. Mnlrnum Amp c 0.5 I

'b --~__,__,__,__,__,__,__,__,__,__,__,__,~~~---~~~~~--........~---~~~--..._.....

\ .-1....

-0 (10 0.01 O. t 1 1()

PGA (g)

Minimum Amplification

• SSHAC

- "The effect of imposing th*~ EPRl(2013) minimum amplification is to change the shape of the soil hazard curve such that it parallels the baserock curve at large PGA values. Evaluation of an a1p propriate minimum level of site amplification is an im1lortant assessment."

• Licensee

- "The 0.5 limit is not imposied. here. in the calculation. of the surface hazard because thE~ intended. purpose. of this report is to obtain. the best estimate of the mean and.

fractile levels of the seismic response for plant risk assessment with no added conservatism."

Questions on Minimum Amplification

• The example site respc)nse calculations from the SSHAC report shovv the impact of reducing the minimum site am~~lification from 0.5 to 0.1 and states that evaduation of an appropriate minimum level of site amplification is an imp~ortant assessment.

Please describe the assessment you made on the effect of not implementing a minimum amplification function *?

Columbia Generating Station Site Response 1

CGS Vs Model Shear Wave Velocity (ft/s) 0 2000 4000 6000 8000 10000 12000 0 ...... I Zone 1 200 -

Zone 2 L...-

400 -

Zone 3

.I

~ 600 -

.r:.

I ---..........

---

* - I

~ I

a. I

~ 800 - ........... I

____, I 1000 -

,. ---. Zone4 I

I 1200 - __ ..I I

1 1400

- vs P1 (Wt = 0.67) ---- Vs P2 (Wt = 0.33) 2

Shear Modulus and Damping Curves

• Sand Gravels, 0 - 525 ft

- EPRI and Peninsular

• Basalts, 525 - 1290 ft

- Linear

• lnterbeds, 525 - 1290 ft

- Darendeli

• Note: Effective strain is magnitude dependent

- ELSRAP uses a constant factor of 0.65 3

Kappa

• Kappa at top of SMB stack provided by SSHAC.

study

• Kappa above SMB stac:k based on low strain damping from damping curves

- Computed using site re~sponse

- Was damping from SMB stack double counted?

4

Vs Randomization

• USGS B. correlation model above SMB stack Thickness S-Wave Velocity Stratum Unit Variatlon Uncertainty, 01nvs P-1 Pasco Gravel +/-22% 0.25 P-2 Pasco Gravel +/-22°/o 0.25 P-3 Pasco Gravel +/-22°/o 0.25 P-4 Pasco Gravel +/-22°/o 0.25 MR-1 Middle Ringold +/-10°/o 0.25 MR-2 Middle RinQold +/-10°/o 0.15 MR-3 Middle RinQold +/-10°/o 0.15 LR-1 Lower Ringold +/-15°/o 0.15 LR-2 Lower Ringold +/-15% 0.30 LR-3 Lower Ringold +/-15°/o 0.20

• SMB stack profiles from SSHAC report 5

Input Motion

• Based on conditional rnean spectra provided in SSHAC report

- 4 events for each of th~e 20 frequencies and 27 mean annual frequenc y of exceedances (2160 1

spectra)

• Duration

- Estimated using magnitude and distances from CMS text files along with WUS duration model.

6

Vs-Kappa Factors Vs-Kappa Correction Cmnp..tte h<l!>t G} IPE FCSJ><.ll16'! ~CU'J. Ill abon dt.lil..~.s unJ on ~IT 9:lil e>r rook t \~pU1c r.!!1(1\l!'I.,~ spect1.t<C:Omj1'3uhl~ FAS us*~ IR \.T

+

D1\ 1J.-1.h~ FAS bylh.l ho.l.-r~10tt ..1tt amphlirotJM fxtob

.i E~im~ hOlft o. IJl~ fr~ I.he fti¥h fr<1qucnc) ~lopt vfU-.: FAS u11J 11vc1~

hOt!l a; <W a 111 cc:>!l5Jdcrcd '<X'ruinos I

l***~(i>-

+ l{O(ll1.":'

I I I Yes No

', '~

Ditfinr! .I lll<W h.....t FAS >UCh thJl ti~ lu~- '\J'lllr t. aailmg tiy muhlplyir:g dic ~ f \S by ttt<jlll'tlC)' ~Of'~ follows K scalin:g exp ( -rrf("car11.i - "11o!J1))

,'lf'pl)*;.. .calln;l bv mu!upJyu1s the lust FAS by erp (-11f(~uu11 tt - K110si))

.1 Jt'.1c-i1Calcd FAS tru:.:al F/\S Id hJSh frC<lllffiCl~I.. force th.? t.-scaltd FAS Co ti.? e.:iwJ to the ltUlJal o:MJ'E FAS Jt th.?!>c: tttqu.:nc1cs y C'<ltW<lrt ... ~FA~ le> N,,.~ lf1'.'Ct1a usms R\'T t+-

Cumput.l ' a:\lhni f~* ~ Ji~ 1J1.. lh<.. A. *"'11lk!J r~*n.t ~Ciro.I

\*lllu.!~ by th.! OMPE rup.ir.se "J'l'ctraJ v:l!Ub Figur~ 9.21. Steps for dennng kappa sca1.tng factors nsing the CRVT approach.

Host Kappa ompute host Glv1PE response spectra at short distances and on stiff soil or rock Compu te r sponse sp ctra-compatible FAS using IRVT Di ide the FAS by the host-region site amplification factors Estimate host!<. values fro1n the high-fn~quency slope of the FAS and average host K over all considered scenarios

Host Kappa 3.5 ------------~~~-------

• Calculate Host.

GM PE/FAS 9 magnitude/distance pairs

• Divide Host FAS by 0.1 1 10 100 Host. Amplification. - Profile Al aoi o -

Freq uency (Hz)

Profile 81 aoi o - Profile Cl aoi o

- Profile 01 - aol O - Profile El - aol O - - -

• Profile A2 . aoi O

- - -

• Profile 82 -aol o - --* Profile 0- aoi o ---*Profile 02 - aoi o

---

ProfileE2-.ioiO - Hm lWUS760 Figure 9.23. QWL linear site amplification fac1ors for the host WUS Vs profile with V530 of 760 m/sec compared to target site amplification factors at the Hanford hazard calculation sites.

Host Kappa (a) 10*1 ~~-~---~~

- - Cale FAS

- - 25perFAS - k =0.039

• Frequency Range for Kappa Calculation Method 1

( /'J (b) 10*1 . . - - - . - - - . . - - - - - - - - . - - - . .

- - Cale FAS

• fl at 25% below peak - - 25perFAShigh - k =0.041

• f2 at 1.5 x PGA but no greater than 20 Hz \"/

1o*1 .-------.------r-----r-----.-----.

Method 2 - - Cale FAS

- - 25perFASlow - k =0.036

• fl at 25% below peak

• f2 at fl plus 10 Hz Method 3

• f2 at 1.5 x PGA but no greater than 20 Hz 10-4

• fl at f2 minus 10 Hz 0 10 20 30 40 50 Freauencv (Hz)

Host Kappa Calculation 0.050 I I I I t I I I I I I I I I I I I I I I

---

r---.,----.,----,---- ----,----,----7----T----

1 I I f

t I

I I I I

I I I

I I

---

T*---}----~----~----

I I I

I I I I t I

• I I I '

I I

I I

I  !* I I I t

---

~----t-----1----i- -- ---1----i----i----t---*

' t t I I I I t-1

-f --~----t---- ----~----+-

I J  :  :  :

---1-- -

I

---

L--- . ----'----.J----

I I I I I I I f

---

~----:-----:----~----

' I f I

• ---1----1----;----t---- ----1---t*---~----r--- 1 I I I

I I

I I' I II II I I I I

• I I

-*-* I I I t 0.045

---

~---.L.---

1 I

I I I

f

... ---~---

I f

I I ...... __......:

I I

I

---

~----~----~----~----

I I

I

' I I

I I

I I

I I

I

• --~----L**--~----

I I I

l I I

---..4---~----~----

I a

I I

I I

I I

I I

I I

I iI I

I I

I I

~-- T' I

I I t I I  : I

___ .,... ___ I I I I I f I I I I I

---

~---~~----~----*---- ---~---~---~----~----

---

~---.,._ ~---

cu

' I I t

I

• ' *I ' I I I

.. ---~---- '...

I I I a.

I

---

~--_....,.

I f

___ ...,.I ___ I I I

~----

I I

• ---~----4----*----*----

I I I - I I

_.. _..'I ____II..,.

---

I *

---

:---~---1----1----*

a. I I I I I

---

~---...-----.----1----

I I

I I

. *-1---....

I I

~

I I I cu

• I I

I I

I t

I I

I I

I

!  ! O Lower

~ 0.040 I I o

I I I I

I I I I I t I t t

---

l-~-...,..--~~--4----

I I I I I I I I 0 ----1-----:-----:----4---- ---1---1 -~---t---~---

• Central a=  : I  !

--:----~

I

---

1-- -.l I

---

~

-~-

t' -~--

I 1I

• f' 4-I z I-I ---:--

t I f I t1 I j

-- 1 & Upper

---

~----r---i----~----

I I I I I

---

r----t-----1----i----* - ~--r----*----~----t--~

.J.I I I I i - Equality

-+---- ----t-I I I  : I  :

---

L----'----..1...*

I I J I .L. --~----~ ----:------:----~-- t t I

I I

I I I  : I  : 1 I

I I

I

• ~

t 0.035

.. I I I

_' J' __..._

I I I I I

I I

4 I

L I

I I

-'---..1-I 1

I

--~---

____ .,'____ ., ...... I' I I I I  : l I I I

' ----~- --~ -~ ---+---.. . .---~----~----

I I I ___ .,. ____ ..___. _

I I I I I

• I I I I

I I

1 I I I

"'4---

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

I

I I

I I

- -i----.!-- ~----J - -

I lt I

--..-----r---~---- ----1----1----,----~---- ----r----~----~----~---- ---~---~---~----~----

1 I I 1 I I ' I I I I I I' II II I I I  :  !

I I I I I 0.030 0.03 0.035 0.04 0.045 0.05 Licensee Kappa

Kappa Scaling Target K < Host K?

Yes No

~ ' ,,

Define a new host FAS such that the high- Apply K scaling by mu ltip lying the host FAS by fre quency slope follows K scaling exp ( - nf( Ktarge l - Khost) )

H*

Apply K scaling by mulliplying the host FAS by exp ( -nf ( Ktarger: - Khosr:) ) ]

~

If K-scaled FAS < initial F .AS al high :frequen cies, force the K-scaled FAS to be equal to the initial G:MPE FAS at these foequencies

,.... Convert K-scaled FAS to response spectra using RVT f.E-Compute K scaling factors by dividing the K-scaled response spectral values by the GMPE response spectral alues

Logic Tree for Vs-Kappa Correction Hose K II Target K I

Target Profile Approach to Estimate y Within-Approach Profile Depth I U ncettainty I l\'.uiv

• 1. 1 (0. 1)

H igh Profile 1 Inversions Kev All Sub-Basal! Sed (0.3) (0.67) (0.5)

(0.67)

(0.33)

Knv / 1.4 Central (0.4)

(0.-l)

(0.34) lCAJi*exp{l.6cr)

(0.2)

Profile 2 Anderson & H ough No Sub-Basalt Sed Low (0.33) KAH

( 1984)

(0.33)

(0.3) (0 ~ ~) ( 0.6)

KAH /exp( l.6a")

(0.2)

Target Kappa Table 7.25. Target Kooserock at the fiye hazard calculation ites at Hanford.

Profile 1 Profile 2 Site Case All Sed~ Half of Sed. ~o Sed. All Sed. Half of Sed. No Seel.

LO\Yer Kmv 0.0297 0.0189 0.0058 0.03 - 8 0.0233 0.0058 C entt*al Kmv 0.0-Wl 0.0285 0.0096 0.0556 0.03-J8 0.0095

. O.O-J91 0.0319 0.0110 0.0618 0.0388 0.0109 pper Kim*

Site C Lo\Yer KAH 0.0164 0.0100 0.0021 0.0-40 0.01-W 0.0028 Central K.oIB 0.0286 0.0182 0.0055 0.03 .,0 0.0228 0.0056 Upper KAH ~

O.O-l85 0.0315 0.0108 0.0563 0.0351 0.0097

Vs-Kappa Factors NRC Calculation

- - *SSHAC Branch 1

- - -SSHAC Branch 2

- - -sSHAC Branch 3

- - -ssHAc Branch 4

- - *SSHAC Branch 5 SSHAC Branch 6

- - -SSHAC Branch 7 10* 1 ._________.____...__.__...__..__.._._..........._______.._____...___.__.__.__.._._............________.____.___ _.__.__.___._......_.~

10* 1 Frequency (Hz)

Reviewing a SSHAC

" ... effective implementation of a good elicitation process cannc~t NUREG-1563 guarantee acceptance of the BTP on the Use of technical conclusions; howeve~r, Expert Elicitation use of a flawed process or (1996) improper implementation of a good process cannot help but cast serious doubt on the quality of the conclusions."

Simply put, NRG needs to C(Jnclude that the SSHAC process developed a PSHA that appropriately represents the center, body, and range of the informed technical community.

Reviewing a SSHAC NUREG-2117, Sect. 4.10.5: PPRP review of documents Process Considerations

• Have all of the essential ste~ps of a SSHAC process been followed and documented?

• Is the data evaluation proce~ss sufficiently justified?

• Is there clear documented E3vidence that the views of the larger technical community have been considered?

• Has the integration process been sufficiently documented such that the c~enter, body, and range of technically defensible inter~>retations are well justified?

Reviewing a SSHAC NUREG-2117, Sect. 4.10.5: PPRP review of documents Technical Considerations

• Have all data used in the assessment been identified and documented?

• Have all elements of the mc)del been defined in sufficient detail?

• Have the model elements alnd expressions of uncertainty (e.g., logic-tree branches and their weights) been technically justified?

• Are there any technical issues that have not been sufficiently addressed?

Reviewing a SSHAC Practical Considerations based on staff experience

• Were contrarian views given appropriate (e.g., unbiased) presentation at workshops?

• Was some potentially relevc:int information excluded from presentation at workshops?

• Does documentation providle a clear and traceable basis to support selection of mod,els and data used in PSHA?

• How engaged was PPRP in SSHAC process?

• How were PPRP review co1mments resolved?

WUS SSHAC Review Guidance Goal of Review: Establish confidence that the center, body, and range of the informed technical community have been considered appropriately in the PSHA.

Three Perspectives for SSHAC Reviewers

- Hanks et al. (2009): "It is simply not possible to verity that the center, body, and range of the full ITC have been successfully captured without repeating the entire process of expert interaction with a different group of experts and perhaps different Tiff Fis as well. As a matter of practical reality, this has not occurred."

- NUREG-1563 (1996): "effective implementation of a good elicitation process cannot guarantee acceptance of the technical conclusions ; however, use of a flawed process or improper implementation of a good process cannot help but cast serious doubt on the quality of the conclusions."

- The acceptability of a SSHAC ultimately depends on the transparency and traceability of its documentation.

Review Approach: To determine if the SSHAC process was acceptable,. reviewers should address the following 7 key questions that focus on the important attributes of an acceptable SSHAC. Not every sub-topic needs to. be addressed, or answered affirmatively, to conclude that the SSHAC process is acceptable. Nevertheless, the information provided in the SSHAC documentation should provide a traceable basis to answer these 7 key questions, and many of the. sub-topics as well. Staff will. use information from these review. questions to develop assessments on the acceptability of the SSHAC process in the hazard reevaluations.

SSHAC Review Considerations Organized Around Considering 7 Key Questions

1) Did the SSHAC process reasonably follow guidance in NUREG/CR-6372 and NUREG-2117?

- Was the potential for cognitive bias discussed early in workshops?

- Did Tl lead identify or address potential cognitive bias issues in workshops?

- Did the scope of the workshops support logical development of a PSHA?

2) How effective was the Participatory Peer Review Panel?

- Was the PPRP engaged throughout the SSHAC?

- Were early comments/concerns by PPRP addressed in later workshops/meetings?

- How were PPRP review comments resolved?.

- Are there unresolved PPRP review comments that might affect results significantly?

3) Has applicable data been considered?

- Was a common database developed?

- Was the database easily accessible to participants (including PPRP)?

- Were updates to database made, and were all participants aware of updates?

- Were local experts engaged in identifying potentially relevant data?

- Is there indication that some potentially relevant data was not considered?

- Are the data summarized and presented sufficiently to support use in the PSHA?

4) Were data uncertainties identified and considered?

- Were uncertainties in original measurements maintained in compilations?

- If data were collected in different studies, were issues of calibration/bias addressed?

- Were data quality issues considered by the T l team and PPRP?

- How have data uncertainties been propagated in resulting models/calculations?

5) Was an appropriate range of potentially applicable models considered?

- How was the range of potential models developed/documented?

- Did an appropriate range of model proponents participate in workshops?

- How well did the T l team engage with the proponents?

- Were contrarian views given appropriate representation?

- Are there indications that potentially relevant models were not considered?

6) How were models selected and weighted in the analyses?

- Is the basis for inclusion or exclusion of models well supported?

- Is there appropriate documentation for how model weights were developed?

- Were models tested or supported by independent information?

- Were there interactions to openly discuss inclusion or exclusion of models?

- If new models were developed, are limitations in existing models documented?

7) How were models and data assembled into the PSHA?

- How were models abstracted into logic trees and weighted?

- How were sensitivity studies used to refine. logic trees or weights?

- How was uncertainty partitioned into aleatory and epistemic components?

- How were these uncertainties propagated into ensemble results?

- Is. it clear how. ensemble results were generated?.

Kishida et al. (2014) report provides estimates of kappa values for Arizona seismic stations based on the data recorded by TA array. As stated in the above mentioned report, these. data have high-frequency range limited by the sampling rate. The earthquake recordings used for kappa estimates are mostly from earthquakes with magnitudes not exceeding M 3.4.

Considering the relatively high corner frequencies of these events and limited recording instrumentation frequency range please comment on the. reliability of kappa estimates.

Kishida et al. (201 4) report concludes that the comparison showed that overall the recorded 5%

damped response spectral ordinates were over predicted by the NGA-West2 models developed for California. It also provides estimates of kappa in average similar or lower than the average kappas in the NGA-West2 GMPEs. The estimates of 0-values in Arizona (Phillips et al. , 2014) are generally higher than in California for which GMPEs were developed Considering the estimated values of Q and kappa please discuss seismological effect contributing for potentially same or faster attenuation of seismic radiation with distance in Arizona relative to California.

The approach to ground-motion models for PVNGS from sources located in central and southern California and Mexico is based on as-published NGA-West2 GMPEs with path-specific adjustment factor to take advantage of the available ground-motion data in Arizona from these sources.

Please provide assessment of the influence of this path term correction effect on hazard calculations from California and Mexico sources.

Please clarify if any path corrections were performed to the 6 models used for Greater Arizona sources.

Section 5.5 of the SWUS GMM report provides recommendation of using Bindi et al. (2014) and Akkar et al (2014) attenuation models for greater Arizona sources in addition to the models developed for California.

The rationale for adding Bindi model is that this GMPE is based on the ground motion database that contains significant number of normal fault events typical for Arizona. This model also satisfies other requirements/criteria specified for model selection including publication in the peer-review journal. Bindi model is characterized by a specific magnitude scaling differing from other models that may be considered as alternative to other models representing the range of opinions.

Please clarify the reasoning for the limitation on magnitude in applying this model to PVNGS.

Please discuss potential effect of this limitation on hazard calculations.

Four NGA-West2 GMPEs are. applied to both distant California & Mexico. and greater Arizona sources. These NGA-West2 GMPEs include basin effect dependency as a depth to the 1.0 or 2.5 shear-wave velocities.

Please clarify if the. basin effect was accounted in using GMPEs. for distant and/or Arizona sources.

Diablo Canyon ESEP Discussions, 6/9/15 3.0 r.===========-------------,

IDiablo Canyon , ..

2.5 - 2015PSHA I

- DE I I ' \

\

- DOE

- HE

' \

\

\

- LTSP \

- - LTSP margin \

0.0 .....__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __,

0.1 1.0 10.0 100.0 Frequency (Hz)

- PG&E (3/11/15) states:

"PG&E has determined that it is not necessary to perform an expedited seismic evaluation process as PG&E's interim evaluation provides reasonable assurance that it is safe to operate DCPP while the updated/upgraded seismic PRA is developed." and "However, as discussed in Section 5.0, the DCPP GMRS is bounded by other previous seismic evaluations, including the design/licensing basis 1977 HE evaluations and the 1988 L TSP evaluation. Therefore, there are no additional benefits in performing this activity in parallel with the more robust risk evaluation associated with updating/upgrading the SPRA PG&E will devote the critical skilled resources to expediting the update/upgrade of the SPRA in order to gain additional risk insights in a timely manner." (p. 51).

- Nevertheless, EPRI (2013, Evaluation Guidance section 3 .2) states:

"In responding to EA 12-049, each plant will have defined an essentially indefinite coping capability for scenarios involving an extended loss of alternating current (AC) power condition .... Plant-specific evaluations for FLEX will determine the specific equipment and strategies to be employed in these three phases. The scope of the ESEL is limited to installed plant equipment and FLEX equipment connections."

- In responding to Order EA-12-049, what plant-specific equipment did PG&E identify for ESEP?

- Is. that plant-specific equipment evaluated in the. LTSP analyses?

Palo Verde Site Response My understanding ...

• Scale input ot appropriate input amplitude (11)

• Apply FAS Correction (9)

• Calculate site responsE~ using corrected FAS

• Amplification function is surface response/original bedrock response

- iRVT does not recover IHF

Site characteristics

• Geology is relatively uniform

- Exception if Fanglomerate

• Extensive Geophysical s1urveys

- Initial siting

- Downhole (R2.1)

- SASW across site (R2.1)

• Three profiles are easily within the randomization of the base case profile

- Does their model of epistemic really represent what is meant by. epistemic unce~rtainty?

Confirmatory proposal

• Develop site. profile based on updated information

- Single profile

- Three profiles represer1ting geologic differences under reactors

• Damping model is oka,y and consistent with SPID approach

• Reduce Aleatory uncertainty to be in line with the epistemic proposed by the SHSR (Table 3)

Amplification 4

3.5 3

- Licensee 1 Hz 2.5 - Licensee 10 Hz c:

0

• - Licensee PGA

"'u

E 2 - ShallowlHz a.

E - Shallow 10 Hz

<(

1.5 - Shallow 100 Hz

- DeeplHz 1 - Deep 10 Hz Deep 100 Hz 0.5 0

0.001 0.01 0.1 1 10 Spectral Acceleration (g)

GMRS

- Licensee GMRS (g) 1.4

- Confirmatory GMRS (Shallow) 1.2 1

~

c 0

IV (ii 0.8 Qi 8

ct IV

~ 0.6 Q,J a.

Ill 0.4 0.2 0

0 1 10 100 Frequency (Hz)

Palo Verde SSHAC Level 3 s~;c 09/01/2015 PVNGS 1

120° w 117° w 114° W 111° w 1os 0 w 36° N 33° N 30° N

>e1smotecron1c areas a1ternat1ve

120'W 117'W n**w 1tl'W 1oa*w USGS Quaternary fault and fold database z lasi updated 11-3-2010 m

~

Shp Rate (mm/yr) 36' N Less lhan O 2 0 2 to 1 0 1Oto5.0

..' Greater than 5.0 "I

Unknown or not reported

.* Earthquake Locations from PfO]ecl Catalog (MW 2.7

• 7.3) 33' N SBR

\

JO"N 0 80 1887 rupture trace

- mt m::i- km 0 80 Figure 9-7. Certain Seismotectonic sources are defined based on trends in the density and orientation of Quaternary fau lts. The CP and TZ sources are differentiated by changes in crustal thickness (Figure 9-6), as well as the lack of Quaternary faults in the interior of the physiographic Colorado Plateau. The 1'1H and SBR sources are differentiated based ou interpreted cmstal twckness (Figure 9-7), as well as the lleiglJtened deusity of Quatemary faults in the southeastern portion of the southern Basin and Range.

ElMI

. ~ )'

Figure 9-23. ~tap mowing the PVNGS catalog (color coded by magnitude bin) and Sei~t:ectoruc sources

120° w 117° W 114° w 111° w 10s* w

-- UT~

z m

~

' ~*

m 35* N

x-()

' \ 10

\

l

\

' I

• __r~ 33* N

)

\ I

(

\ J' It

///1:\~)

- ,, { 30° N 0 80 mi km 0 80 Fault sources (168) in PVNGS SSC

Slip rate mm/yr

>3.0

=: 1and<3 0

?. O 11nd < 1

!001and<0 1

<001 30* 111 U169 0

From Figure 10-1: High slip rate plate boundary faults (red) modeled with UCERF 3 and layered alternative.

Basin & Range and low slip rate CA modeled with characteristic fault alternative (blue and green)

Sand Tank and Ballenas have their own special detailed logic tree (greater uncertainty)

120°w 111°w 114° w 111° w 108"W 0.., 1'~

( /~ "':-1()

- ... - UTAH

.... ..... z

.... rn Pisgah- o-? ~ ~

Bullion Mtn- 1'/ / ' '

' ' l~

Mesquite lk ..,/ 36° N

/

San Andreas Calico-Hidalgo ----

\

I

/

/

/

B_ig Chin.o- ~

Little Chmo ,_ Wlliamson Valley

\

\

()

0

\

\

grabens I \

Plomosa East l Horseshoe \

1

' Carefree

• )

Sand Tank 33° N

\

Laguna Salada ________ ,

I

\ I I

/

/

' /

/

' /

30° N 0 80

- c:=--* mi

• km 0 80 Final fault sources (18) in PVNGS SSC

Hazard Curves in general.

Total Hazard is driven by the Areal sources

@ 10 and 1 Hz, Areal sources ( 2 zone and seismotectonic (6) zones)> all fault sources Closer look @ Areal sources for 10 Hz: seismotectonic zones> 2 zone> CA faults (mostly)> AZ faults for 1 Hz: seismotectonic (mostly) >CA faults> 2 zone> AZ faults Separating the areal sources:

for 10 Hz: SBR > East> TZ >everything else for 1 Hz: SBR mostly> East mostly> TZ mostly> everything else By fault:

for 10 Hz: SAF >Cerro Prieto >San Jaciinto >everything else, exception SD TK For 1 Hz: SAF >San Jacinto> Cerro Prieto > everything else, exception SD TK

10 Hz Site Specific Rock Hazard at Palo Verde, by Source Type 1E*3 QI v

c QI "O

QI 1E*4 QI v

)(

QI 0 - Mean v> lE*S c:

QI

~

- Area er QI IV c

~

1E*6

- Fault c:

1E*7 +---------....._._ __,________._____--'-+_ _ _ _ ____..___ __

0.01 0.1 1 10 10 H l cno,.tr::>I ,,.,.,..,1or::>nn n l a\

1 Hz Site Specific Rock Hazard at Palo Verde, by Source Type 1E*3 QI v

c QI "O

QI lE-4 QI v

)(

QI 0 - Mean

~ lE-5 c

QI

~

- Area

-r:::r QI IV

~

c 1E*6

- Fault c

lE-7 -<--------~-----------------------~

0.01 0.1 1 10 1 Hz spectral acceleration (g)

-f.a--llA.....,. ____

c=i Olt1 & .,..._ _ RC'l_ dlt &Qr~ ~ wl'l1u"140'*...,*'"' c::::J Cif*-*~--

llll)OiL' Cl- ~\NT c::::J r*v-Figure 37. Geologic cross-section showing the shallow and deep stratigraphy at the PVNGS site; modified from UFSAR Figure 2.4-27. The map in plan view on the right shows the cross-section line, as noted by A-A'; map is modified from PSAR Figure 2.5-13. Note that Units 1-3 are west of the crosssection line, between borings PV-21 and PV-24. Source: Figure 8 from LCI (LCI, 2015d).

PPRP identified issues

• SSC site region is mostly. in S B&R and as such is devoid of Q faults. Is this valid? are we missing some Q faultsi?

• Geodetic rates are gre,a ter than fault slip rates in central and souther1n AZ

• There are no modern cjetailed geologic maps of the PVNGS site area1 and vicinity.

• There are 3 faults near the site that have uncertainty about characteristics.

'E

' )<

' ti 0

' \

.. ' \

r \

~

\

\

\

I

'I M' I

\

i:*u 0* - "1!1

...... 4(1 Ftgun 7-1. Qu,atenwy fanlts m Arizona and neighboring sta1es from d:te USGS Quaternary Fault and Fold Databa!.e (USGS. '.!006. last updated 2010). Arrow points 10 the Sand Tanlc fault ~ast oftbe PVNGS sae (black stat) .

l *O' W

' \

' \

'~

' \

\

I

'I a4' "

'I l:' 11

)O' H fiprt -1. Quatemmy fanlts in Arizona am neighboring '1ates from me USGS Quatemiuy Fault and Fold Database (USGS. 2006. last updated 1010). Arrow points to the Sand Tank fault ~asl oftbe PVNGS sue {black star).

\SW 11""\V lll' W 11~*w l WW 110' YI 1CJ6" W IS\

ov

'I

"+&~ *

,ii,.

1) ~*ti

it"' I

.SI I fl/

"""' 0

I I

I

~A1i1GE

~~

12' N

}, I ~) i

..* (

.. (, ' ' 'l

\

l' \

\

31" 4 0

-==-*40km 40 ni

..._~___......__...._____~~~~~~--.~-'-l-1__~~~~--~-~----~~~ ' ......._..___.____~-'--~--------~--~----'10" U Figure 7-7. Green lint ~ nre po1em~11 Quatemary ac ti~e faulL' in northern $(mora i<k'Tltirled b)' the! n Team b~ a1'Pl) tn~ Appro~h I. Purple li11-e'

' ho" faulh t1lnt appear"" "iet'\ icio C"ioologico Me\ic.ano (SCiM) geolosic mar, of fl( 11hem Soll(lrn 1ha1 are \\ithin or bounJ Quatemar; unit . Rod hrn: an: Qu~uemary foulh f1vrn I he USC. ' Qu~1enrnry ra ull nud fold D111nbu..'e (U ' C..S. 2006. I~"' ' up<l41ted 2010).

PHOF.N X NORTH Figurr -11. Location m.ip show~ extents of ~tailed geologic: mapping oft.be SIU! Acea (8-km radius) aod recOllDiili.sanc~level mapping of the Site Viciwty (40-bn radim). .Pwple lines sho\V extents oftbe Pboemx North and Pboeni.""t South 30- x 60-minute geolopc map sheets. Green shadi.ag show:. area of ma.pplng m'ised by Phil P~ (AZGS) for this pcoject. Orange shading s.bo\\'s ~a mapped by Jeri Youog (AZGS) Btne 'lhading shows areas mapped by LCL The Saud Tani.: fault is represented by solid aod dashed black lines northwest of the SU TOUlkMountains. Modified aftecLO (2014).

IU' W Sand Tan k fau lt Elf!Yation

~IOMl\rO Feui~CIYP n 065 Tonal lmeamenl 0

0

--=z=-

-==-- * - 8 Ftgure-l-LExteut ofSand Tank fwlt scarp (sohd bladd1oe) and toe.al lineament {dashed blackhne) modified after I>cimey and Pearthrtt (1990). Sc-aip hes outsi~ of the 40-bn P\lNGS S1te Vicinity.

113.is* w 113* \V n?1s*w t%5* VY J:7S k Piedmont Alluvium Yat1rgtr Uni

~

ft OdcJUnts

'tr

~

l

~

tl

~

(\#

>>S'N

..,?

~....

%l'..,

Ri;1el Gi\O 33 25' ~

0

-c=:i- -mi 0- - 8 km 8

Figure 7-1 J. Simplified QU'1tt:ntaf') geotugic map vf lhl! Site: Vicuttt) ( 10-"m mdiu!>) Yellow .,,how di 1ribu1ion of )ot111gc.:r Jt'pO"\it ~ 12 ka

{Qy), orange ~ow' olderQuillcrn!l1') umt' >SO kn <Qa3. Oil (.)i i. and (.)o). gra~ -.llO\H. area.c; of Pliocene and older hedmck. Modified atler I.

(2014).

UU7S"W Explanation

• Bonng UFSAR June 2001 Revision 11 Appendix 2F H Profile PSAR July 1975 Ch 2 5 Part 2 Figure 39 Piedmont Alluvium Oy2 Late HoJocene aauv1um Oy1 Holocene alluvium Qy Holocene a~uvlaJ depoS1t5, und1fferenuated 013 Late Pleistocene alluvium Oi2 Middle lo late P1e1stocene alluvlum Qtc Quatemary htllslope iaws and colluvlum Bedrock Units Tbu Upper basalt Tbl Lower basalt Basatt undllferenbated 0 2,000 ft m

33 37$' N 0 400 Figure 7-15. Portion of the 1:24,000-scale geologic map of the Wintersburg quadrangle (Pearthree et al. , 2006). The unnamed fault of Pearthree et al. (2006) is shown as a black dashed and queried line. Blue line shows location of line of geologic cross-section shown in Figure 7- 16, whereas b lue circles indicate key borings that constrain the subsurface geology in the cross-section. Star indicates the location of operating nuclear Unit 2 at the PVNGS.

l' ig ure 7-24. Lran ect <lfi?eodciically "" ed ea-.t-\\e ~ e'ICl'biori rates tin mm*~ rt acm..s Arvona and \\>C"temmo'I New M~x1co tmm Kreemer

(\\ or~tll'p I p~~mat_ion. t:\! A~ndix B }.

F igure 9-t7. C-0mpanson of calculated deformation rates across the PVNGS region using the Seismotectonic model and Case 1 magnitude ~s (red-nlues) to GPS-measured extension rates by Kreemer (\11hles in black). ~!odified from Kreemer (Woikshop 1 presentmon.. see Appendi."t B).

Magnitude-Range Source Type Weighting Zone Model Sources Fault Model West N/A Two-zone 0.2 East Case 1 So. Caflfornia and Baja 0.4 SCABA NA Areal sources Case2 0.4 Gulf of California GULF Case3 So. Basin and Range (SBR) 0.2 Seismotectonic 0.8 Transition Zone TZ)

UCERF 3 I

Colorado Plateau (CP)

Fau lt Model: Mexican Hi hlands MH Ru ture Sets

• UCERF 3 0.8 San Andreas fault

• Layered Layered

• Characteristic fault 0.2 Ru ture Sets San Jacinto fault 0.8 La ered 0.2 18 faults in final SSC model Ru ture Sets Elsinore fault 0.8 La ered 111 0.2 .i::.

Algodones Fault sources N/A N/A b.O I

Big Chino-Little Chino Cerro Prieto fault La ered Carefree South of the .

Ballenas transform fault Layered }

Horseshoe Border Plomosa East 13 fau lts - - . . All other faults Characteristic Sand Tank . :===========-~ 7inAZ Wi lliams Valley grabens Figure 8-1. Master logic tree for the PVNGS SSC model.

Slip Rate, Rupture~

Model Objective Fault Model Resetting MRE? EPR Distribution Fault~

tMRE for EPR See Table 104 36,139' point sources from ~an liernard1no (0.25)

South to lmperb l ApprmcimatP I 1r~ ru:~ Rupture event set 25 mm/yrJ UCERF3. sections rupture behavior from UCERF3 294yrs No See Table 1o-4 (Appendix G) NA (0.8) (O.S)

See Table l D-4 (0.25)

Allow for fault se~menta tlon and variations in slip rate L.ayered fault model See Figure 10-Sb (0.2)

Figure I 0-Sa. San Andreas fault source logic tree.

Seismogenlc Slip rate, Rupture Slip rate Style of Dip/ Resetting EPR Rupture Mchar Recurrence Thickness layer tMRE for length (mm/yr) Faulting Direction MRE? Distribution Area (km2 ) Basis Model (km) EPR (km)

See Table 10-4 Full rupture Hi h I.a er 1 (0.25) area (HBOS) Char EQ (0.2) (0.5)

Slip rate, See See 15 90 d Layer 2 tMRE Table 10-4 Table 10-4 RL RA (O.S)

Low La er3 See Char EQ (0.2) Table 10-4 I.a er4 (0.25)

• Hal f rupture area is only applied to Layer 2. All other layers use full rupture area with a weight of 1.0.

Figure 10-Sb. Layered fault model branch of the Sao Andreas fault sourre logic tree. Layer-specific sUp rates and magnitudes are provided in Table 10-

3. Parameters fer EPR are provided in Table 10-4.

Selsmogenlc Magnitude Slip Rate EQ Recurrence Fault Style Dip/ Direction Geometry Magnitude M ethod M char Thickness (km) Regression(s) (mm/yr) M odel Sand Tank Fault

-0.2 0.001 WC94max.

displ. (a ll} Char. EQ (0.8) (1.0)

N/A Per. even (0.2)

WC94 avg.

~ -0.2 (0.2) displ. (all) 7.2 (0.2) (0.6}

+0.2 (0.2}

Avg. ofWC94 RA (all)i Stirling

- 0.2.

Ru pt. et al. 2002 Cl; Wesnousky (0.2)

Long dimensions (32 (lineaments) km length) 15km 2008 SRL (all) 6.9 Char. EQ 0.2) (1.0) (0.6} (1.0)

+0.2 (0.2)

Scarp height more reliable indicator of mag than length Distribution broader Uncertainty and distance than other normal faults from range front Figure 10-30. Logic tree for the Sand Tank fault. Mchar is calculated from scarp height (upper branch) and fault dimensions (lower branch), owing t<

uncertainty related to the full length of this fault source.

Tabk 7-2. Preseniatlon and degradation of f.ault scarps in the desert landscape.

modrlied after dePolo and Anderson (2000).

RI Degrade RI Degradt Decradt time Tbnt Tina RI (kyr) timt Timt Slip nm {)QT) (kp) tim t

'TrO Tr 0 4m WO (mm yr) lm (k.yr)

-c:;u']>

m {JcyT) ic:;up ~ c:arp

(\.:r)

K AJ'p

,cup lm sc:u *p ?m 4m 0001 l,000 25 gs* . 2.000 100 95*. -l,000 400 90-.

0.005  :!00  ::!5 ss*. 400 100 75*. 800 400 50-t 0.01 100 25 15*.  :!00 100 5~* 400 400 ..

o**

0.02 50 25 5~9 100 100 o**

• • 200 400 o*.

0.05 20 25 O!o 0 100 o~. so 400 o~ .

0.1 5 25 ()! . 10 100 o,. 20 ..$00 ()! *

\ot~:

RI: Average recuaence interval. in thousa.ods ofyean (kyr).

Degrade Tune: scarp degradation tune, m thousands of years (l-yr).

PVNGS SSC Areal (Background)

Sources SSC SSHAC Report Section 8.2 and Chapter 9

CHAPTE R 8: SEIS:\llC SOL'RCE CHARACTERIZATIO~ OVER\"IE\Y.................................... 8-1 PVNGS SSC SSHAC Report: .

8.1 Types of Seismic Sources Identified and Characterized in the SSC Model ........................................ 8-1 8 ..2 Areal Seismic Sources ......................................................................................................................... 8-.2 8.2.l Maximum Magnitude (Mmax) Assessment for Areal Sotu*ccs ...................................................... 8-2 8.2 2 Zone Boundaries for Areal Sotu*cc-s ............................................................................................... 8-3 8.2.3 Ma~n1de Freguency :M odel for Areal Sources ............................................................................ 8-3 8.2.4 Earthquake Recurrence Assessment for Areal Sources ................................................................. 8-3 8.2.4.1 Smoothing to Represent Spatial Stationarity ........................................................................... 8-4 8.2.4.2 Formulation of Penalized-Likelihood Model for Recurrence Parameters ............................... 8-5 8.2.4.3 Modeling the Joint Distribution of Recurrence Parameters and Development of Altemative Recun*ence ~laps ................................................................................................................... 8-13 8.3 Fault Sow*ces ...................................................................................................................................... 8- 14 8.3. 1 Model Approaches for Fault Sources ........................................................................................... 8-14 8.3.2 Slip Rates for fault Sotu"Ces ......................................................................................................... 8-15 8.3.3 Magnitude Scaling for Fault Sow*ces ........................................................................................... 8-15 8.3.4 Magnitude Frequency Model {Characteristic Earthquake) for Fault Studies ............................... 8-17 8.3 .5 Earthqua ke Recurrence Models (Poisson and Time Dependent) for Fa ult Sources .................... 8-19 8 .3. 5.1 Time-Independent Ea1thquake Recurrence ............................................................................ 8-19 8.3.5.2 Time-Dependent Eanhquake Recurrence ............................................................................. 8-20 8.3.5.3 Equivalent Poisson Ratios ...................................................................................................... 8-.23 8.3.5.4 Ocher Considerations.............................................................................................................. 8-25 8.3.5.5 Sununa ................................................................................................................................ 8-25

PVNGS SSC SSHAC Re ort: .

CHAPTER 9: SSC MODEL! .!IU:..U SOl"R.CI S 9 1 Critma forDetinm.g Areal Soun:~ and Data Used.

9.U Future ~uab ClwtdmStics __ ****-**

9 .l.1 Sel.SDlOgemc Tbic~s **--**************--******* ....................*-***-*****-**-**--**********-*-*-*****-****9-2

.. 9-~

9-3

... 9-3 9.3.4.l Basis for Defining Soun:e ______ ........._ .................... _ .. ______ ....... _ ...... _......... ..9-9 91.~ Futun~~~~~~~~~~~~~~~=======~*c!:'. 9-~ .5 9.3 4.2 Basis fur Source Mmax ......... ..........------*-*--***-....--........- ......-....... 9-9 9.2..5 &sis for Mmax m the East Somc::e.= . :;;**:***

• • • = *:::.: ***:*::=*= ==========:::.:*::.:.

• • • .:. **9:....-::,;

.5 9.3 4J future ~alee Characteristics ....... _ .....................- ....... _ ....................... _. ~..... ..9-10 9 2.6 futuR Eartbqulle Chanictensbcs m~the =:...Eut =!...:S:.::~~ce::...= ** ::::*========= 9-6 93 .5 Southem Cahfonua and Baja Cahforaia ....--------- *--*-... - ... *- - * *- 9-10 93 Se:ismot~mc Altematl\"e ***********- **--*-****-- - -*--*****- *-*****-*..- *- -...-.------- 6 9.3.5.1 Basts for Defimng Sowu. ____... ____ ,______________ .. _____ .. _______________ 9.10 9.3.1 Soulhem B.tsul and Range ..........- ...................- ....--..**-*-**-..... _ ..........................._._.......... 9-6 9.3.5.2 Daw for~ Source Mmax . ......- -.. ----..- -..*---*--**-**-*-**-* .9-10 9.31 1 Basis for Ddining Source ... ***--*--..-**-*--..- -........ --....*****--***--..---............ -* 9-7 93.5J future Earthquake Charactmsttcs ..... -----...- ..- ............ _. _______ ,, ____., ______ ,,,.,._...... 9-10 9.3J .2 Basu for Source Mma:t.....-*-**................._______ ..........- .... - ..............- ............_,, __,.. 9- 7 93.6GulfofCahfonm ... - - * - - * -..- - - - - - - - - - - * - * * - - - - * * - 9-11 9.3 U Future Earthqwke Cbaracterutics .......,_ ---.. - - * - * * - - - - * * - - - ____ ,,_ 9-7 9J.6.l Basis for Ddimng Source .....- .............. _ ..___ ,.................. ________ ,,_______..........._ .51-11 9.3.2 Colorado Plateau_ .....- ...*.. -**---*------*..***-..-*-*---....------*----*----.. *-*-*-*----9-7 93.6..2 Bms for Source Mmax... . ** _ -*****-**..**----*--**--*********-....._..____ .... ..9-11 9.3 2 1 Basts for0e1inmg Source .... _ .......... _, __ ,.____ ....................,.._________ ,,....,_, ____ ... 9-7 93 63 fulUR Earthquake Clw-.c:1rrutia **-*-*-------*-**----*-**----*-----....*----9-11 9.3.2.2 Bws for Source Mm.ax .........- ...........------*-*--*---*-*-*-*-------**..*****----***--*-**---*9-8 9.3.7 Seislnogemc Thickness ............ - .....- .........................................................- ........................... 9-11 9.323 Future Earthquake Cba.rac::tensti~ __ ,..... _.. _, __ ..___ ,________,,,_ ........... _ .... _ .. ____ ,, ___ ,9-8 93.3 Trmsrtion Zone .......................- ............. __ ,, ____ ....... - -............................. - .........- .... 9-8 9.3 3.1 Buis for.Defining Source ......................- .......- ...-*-****-*..**-**...................- ................ 9-8 9.4.1 l R.emo\-al of On.fault E\-ents from the P\J"NGS E~e Catalog .....- ............................ 9-13 9.332 Ba.us for Source Mmu ____ ,,,,.......- ....*------*-**-*--..**-**-***-**--*****---**-**-*9-8 9.-U.1 Cakulanons.._____ *-- -* ---**----****--**---****-*-**---*-*-**---*-*9-13 9.3 3.3 F11ture Eanhquake Characteristics.. . _ .... __ *---.... - *** - -.. *-*-..- .. 9-9 9.3.4 t.lexican Hig)ilands ........ _ ............_ ............. - .........- ....................- ..................... _. _____ ,,_..... 9-9 9 4.1 4 Coonderanon of Constant I>-\'al~ Kemtl l\J'l1f:o.aches ..........- -..* - . *-*

9.-t U C~on io USGS ~aps for the Basm and~ ~gi,~o::;; n ;.;;;

-*;* ====.;;.;.;.;.; "*..-.

9-"""1"""

8 9 4 l 6 Stram-Rate !izq>licanons of RKmrmce Puuneten .. 9-I &

PVNGS PPRP-TI Team Correspondence: .

l.rUh Co1nul1..11nh lnttn1a1ion11l. la(.

1.t.1 1 luurlt<.'j Kl"' u *r *~11

\.du..:Ja. CA QI 355 IMlt ~ll~-'/90't; l.n 16611 ~ '1-W<JU Apnl 17, 2015 Mr. Ronald Gaydos Pro1ect Manager Engineered Equipment & Major Pr()fects Westinghouse Electric Company 1000 Westinghouse Dnve CWH03-410M Cranbeny Townshtp, PA 16066

Subject:

Additional Documentation for the PVNGS SSC Report Mr. Gaydos, Letbs Consultants International, Inc. (LCI) 1s pleased to submit lhtS additional documentation associated with the Palo Verde Nuclear Generabng station (PVNGS) Seismic Source Characterization (SSC) Report (Revision 0, dated February 2015). This additional documentabon satisfies dehverable requirements for Task 1, as descrtbed in Project Impact Notice (PIN) No. 8 for Scope Changes lo the AriZona Public Service (APS) 2.1 Seismic Hazards Evaluation (SHE) Project The U.S. Nuclear Regulatory Commtssion (NRC) requested that APS provide additional information detailing the interactions between the Participatory Peer Review Panel (PPRP) and the Technical Integrator (Tl) Team. The requested information 1s provided m the three attachments that accompany this letter:

• Attachment 1: PPRP Comments on the Protect Plan and TI Team Responses.

• Attachment 2: Formal Correspondence Between the Tl Team and the PPRP.

• Attachment 3: PPRP Comments on the Draft SSC Report and Tl Team Responses Please do not hesrtate to contact us with any questions.

Sincerely, Lettis Consuttants International. Inc.

"?-, '\). \-4~-'-~

Ross Hartleb Project Manager

PVNGS SSC Areal (Background) Sources Areal sources. are Future earthquakes in the characterized with a areal sources are modeled defined: with rupture characteristics

• geometry such as:

• seismogenic thickness

• geometry

• rate of earthquake

• rate occurrence

• slip sense

• Mmax

• magnitude-frequency distribution function

SSHAC Section 8.2: .

PVNGS Areal (Background) Sources

• Two alternative depictions of PVNGRS areal sources

1. Two-Zone (0.2)

2. Seismotectonic (0.8) 1n*w lll'W 117'W 114'W 111"W 108'W

,,,z

l1

~)( 3&'N 0

0 33' N PVNGS SSC SSHAC Figures ES-2 & ES-3

10 Hz Site Specific Rock Hazard at Palo Verde, by Source Type lE-3 ~

~ - ~

-~

Cll v

c Cll

"'O lE-4 f'.. -

~ =

Cll - -- ~ "'!:::

Cll v \f :'.E f: .+

-x Cll 0

v>

c Cll lE-5

- --r--

- ~~

~ -X=+/-

-+-

r"N K ~ ~

,~

- Mean

s - Area

• ' -. -~Iii.

- CT

-- t 1t4- - ....

Cll

~

- Fault 1: .'"" \

IO

s c

lE-6

= = ~ T .r_ = p  := lt lt t T

~-

~ = 1: = I=

ct c

-~ -- - ....

lE-7

~ ~

0.01 0.1 1 10 10 Hz spectral acceleration (g) 1 Hz Site Specific Rock Hazard at Palo Verde, by Source Type lE-3 Cll v f ~~

c

• ~ i Cll

"'O Cll lE-4

,,, t;. + -= -=t= ~

I I

-c:::=

! fl FtiK Cll v

- x Cll 0

~

c lE-5

,- , ~ ~

...._.... _,... ~

...,..... _ - Mean Cll 1: - Area

s .

er I QI

~

IV lE-6 J l --\ ., .t' "1 - Fault

s c

c *-

..._~

ct -\ ~

I  :~

lE-7

. I

\. '\..

0.01 0.1 1 10 1 Hz spectral acceleration (g)

10 Hz Site Specific Rock Hazard at Palo Verde, by Source Sub-category lE-3

~~

~

Cl.I u +~

c:

Cl.I "O

l E-4 '- I~ ~ I

- Mean Cl.I

~ ._,__ '\ ~

Cl.I u

)(

....0 Cl.I

- ,._ ~ ""

i\

... '.- --~

""' ~~~ ~

- seismotectonic area

~

c:

lE-5

- Two-Zone area

-~~

Cl.I

,.._~

J CT ......

Cl.I '-" I l J Greater AZ faults

'\. - ~--1 .... -

"'c:

J lE-6

.. ~- .... , ..-

- California-Mexico c:

ct '"-'-" ' '~~ faults lE-7 0.01 Ql ~"" 1 10 10 Hz spectral acceleration (g) 1 Hz Site Specific Rock Hazard at Palo Verde, by Source Sub-category lE-3

... -- -~,:- -

~") "

Cl.I i - - _,_ ,__

u c: - ,_

Cl.I - Mean "O

Cl.I l E-4 -... - --

"~

~

Cl.I u

)(

....0 Cl.I

-'\~ *~ ~ -~* - Seismotectonic area

> - Two-Zone area

• 111 u l E-5

- '\. ~

c:

Cl.I

J

...... ~~

CT '

1n 111rn Cl.I Greater AZ faults

"' \,,

"' lE-6 - - ...

~ ~

J c:

- - Ca lifornia-Mexico

' ~ ~~

c: I faults ct ........

I l E-7 * '*

' \. l" J 0.01 Ql 1 10 1 Hz spectral acceleration (g)

10 Hz Site Specific Rock Hazard at Palo Verde, by Area Source lE-3 QI Total u - Mean c

QI

'tJ - 5BR QI lE-4 QI u

)(

East QI TZ 0

u lE-5

- GULF c

QI - MH

~

C" QI - west lE-6 - SCABA "c'

~

- CP c

<(

lE-7 +----...:....._____;:......_...:....._...:.........;:......:.._..,---~~-~...:....._...:....._......,.__,__..,-- _ __,,,,.___,......_-'--...:.........;:......:..---l 0.01 0.1 1 10 10 Hz spectral acceleration (g) 1 Hz Site Specific Rock Hazard at Palo Verde, by Area Source lE-3

~ *

~ :*~"

~

QI Total u - Mean c

QI

'tJ QI lE-4

- - SBR

- ~~~~

QI u - East

)(

QI

-~

• TZ 0 -

~ lE-5 - GULF

~~ ~ ~

c QI

~

-- - MH C"

QI -

west SCABA lE-6

"'c - '~~ ~~~ ~-

~

-f~ *,~ ~ ~

~

c

--* - cP I

<(

lE-7 0.01 0.1 1 10 1 Hz spectra l acceleration (g)

PVNGS Areal Source!;: Mmax Assessment Distribution of Mmax values and their weights for the areal sources are based on th1e expert judgment of the Tl Team and informed by

• regional geologic, geophysical, and seismic information,

• previous PSHA studies performed for the PVNGS (REI, 1993; LCI, 2012),

• regional PSHA studies performed by the USGS for the NSHMP (Frankel et al., 199 6, 2002; Petersen et al.,

1 2002,2008,2014),and

• PSHA studies performed fc>r the Yucca Mountain site in southern Nevada (e.g., Wo ng and Stepp, 1998).

1

PVNGS Areal Source!;: Mmax Assessment

• The CEUS-SSC project (EPRI et al., 2012) utilized two alternative approaches

. . for estimating the Mmax distributions for areal se1sm1c sources,

- the Bayesian approach

• Requires a robust database of earthquakes in analogous ("tectonically comparable") crust. The Tl Tearn is not aware of such a database. for actively extending crust and plate. margins that would be applicable ta. the PVNGS. model region, thus the. Bayesian approach was not used in the PVNGS. SSC to estimate Mmax distributions for areal seismic sources.

- the Kijko (2004) approach

• Relies on observed seismicity w ithin a region to provide a direct (or posterior) assessment of Mmax . This approach, however, does not provide stable results when the number of observed earthquakes is low (Kijko, 2004). The Tl Team investigated the applicability of the Kijko approach, but determined that there have been an insufficient number of earthquakes throughout much of the model region to apply the Kijko approach to estimate Mmax distributions for areal seismic sources.

PVNGS Areal Source!;: Mmax Assessment Key uncertainties with the assessment of Mmax for areal seismic sources is the possibility,. and perhaps likelihood,. that Mmax varies. spatially throughout each areal source. In the PVNGS SSC, the broad distributions of Mmax for each areal source. are assumed to be entirely epistemic and applicable throughout the source, such that the diistribution applies. to all locations within that source.

The Tl Team considered whether it is realistic to assume that a single Mmax distribution developed for an. areal source. applies to all locations within. that source, especially given. the. large extents of these sources and the broad range of magnitudes in the Mmax distributions. The Tl Team utilized the available data to the extent possible to identify areas of varying expected Mmax. Spatial variation in Mmax is one of the primary criteria used by the Tl Team to differentiate areal seismic sources, such that if there is a basis for identifying spatial variation in Mmax, that information was used to identify a separate seismic source.

PVNGS Areal Source!;: Mmax Assessment

• Mmax distributions are based on the judgment of the Tl Team, informed by the evaluation of published data, rather than quantitative estimates such as the Bayesian method or the Kijko (2004) approach.

• Mmax distribution for all areal sources ranges from M6.8 to M7.9, except CP ranges from M6.5 to M7.9.

• The uppermost values in the PVNGS Mrnax distribution (M7.5 and M7.9) are representative of the largest known on-fault earthquakes within the western U.S.:

the M7.5 1887 Sonoran earthquake (Suter, 2008a), which is also the largest known normal fault earthquake worldwide (based on rupture length), and the M7.9 1857 Fort Tejon earthquake (e.g., Townley, 1939; Sieh, 1978; Agnew and Sieh, 1978).

Fo r the 2-Zone Areal Sou rces :

Rupture Top of Seismogenk l\ la gnitu de Source Boundary Ruptw*e Ruptw*e ~Imax Rate Dip Rupture Thickness Recurrence

~fame Type ~lechanism 01ientation (-\Iw) Cases (degrees) (km) (km) ~lodel Strike-slip 70° (20%)

(80%) 80° (20%)

90° (60%)

6.8 (0.1)

N35°W (20%) 30° (20%) 12 (0.2) 7.0 (0.25)

Reverse West Leaky (lO~o)

N45°W (60%) 45° (60"0) 0 15 (0.6) 7.2 (0.4)

N55°\V (20%) 60° (20° ~) 18 (0.2) 7.5 (0.2) 7.9 (0.05) 35° (20%) 1 (OA)

Normal (lO~o) 50° (60%~ 2 ~0.4) G-R ( LO) 65°(20% 3 02) 35° (20%)

Normal N20°E (100/o) 50° (60%) 6.8 (0. 15)

(80°0)

N-S (100~) 65° (20%) 12 (0.2) 7.0 (025)

East Leaky N20°W (400/o) 0 15 (0.6) 7.2 (0.35)

N40°W (200~) 70°(20%) 22 (0.2) 7.5 (0.2)

Strike-slip Random (200/o) 7..9 (0.05) 80° (20%)

(20°~)

90°(60%)

PVNGS SSC SSHAC Table 9-1

Top of Seismogenk  :\Iagnitude Source Boundary Rupture Rupture Rupture  :\hnax Rate Rupture Thickness R ecmTence For the l'\ame Typ e )lechanism Orientation Dip (km) (km)

()Ill") C ases llodel Seismotecton ic Strilre-shp 700 (200/o) 800 (20%) 6.8 (0.15)

(90%) N35°W (20"~) 12 (0.2) 7.0 (O 25)

Areal Sources: SCABA Leaky N45oW (60°'.)

90° (60"/o) 0 15 (0 6) 12 (0.4)

N55°W(20'lo) 300 (20" o) 18 (0.2) 7.5 (0.15)

Reverse 45° (60%) 7.9 (0.05)

(10%)

600 (20"/o)

Strike-slip 700 (20"'.)

(W~.) 800 (20%) 6.8 (0.05)

N35°W (20'lo) 900 (60" '.) 12(0.3) 7.0 (03)

GULF ual")' N45°W (60"*) 0 14 (0.6) 7.2 (03)

Normal N55°W {20'lo) 35° (20"'.) 16 (0.1) 7.5 (03)

(30%) 500 (60"/o) 7.9 (0.05) 65° (20"/o) l {OA) 2 (0.4) G-R ( l.O) 35° (20"/o) 3 {0.2)

~ormal 500 (60"/o) 6.8 (0.1)

(80%) ~S(Ne) 65° (20'l'O) 12 (0.2) 7.0 (0.25)

N2()0W (400/o)

SBR Leaky 0 15 (0.6) 7.2 (0.4)

N400W (200'.) 70° (20"/o)

Strike-slip 18 (0.2) 7.5 (02)

Random (2001.)

800 (20%) 7 9 (005)

(20'lo) 900 (60"'.)

35° (20"/o)

Normal 50° (60"'o) 6.8 (0.05)

(80%) N-S('.?OO*)

65°(20%) 12 (O.l ) 7 0 (0.25)

N20"W (400.4)

MR Leaky 0 15 (0.6) 1.2(03 5)

N4<>°W (2001.) 700 (20010)

Strikr-slip 18 (0.3) 7 5 (03)

Random <2°'*) 800 (20"'0)

(20%) 7 9 (0.05) 900 (60%)

35° (20'"1o)

Normal 6.8{0.2) 50° (600fo)

(700,'.) N20°E (25~o) 14 {0.2) 7.0 (025) 65° ('.?00/o)

TZ Leaky N20°W (25°~) 0 17 {0.6) 7.2 {03) 700 (2001.)

Strike-slip Random (50"/o) 20 (0.2) 7.5 (0.2) 800 (200'.)

(300*) 7.9 (0.05) 900 (60"!.)

35° (2~.)

~ormal 6.5 (0.2) 500 ( 6Qo,.)

(80"'o) 15 (0-2) 7.0(03) 65° (2~*)

CP uaky Random (1OO'lo) 0 20 (0.6) 12(0.25)

PVNGS SSC SSHAC Strike-slip 700 (200/o) 25 (0.2) 7.5 {0.2) 800 (200 o) 7.9 (0.05)

(200/o)

Table 9-2 900 (600/o)

PVNGS Areal Source~s: Zone Boundaries.

Primary criteria for defining and differentiating areal seismic sources include changes in Mmax potential, seismogenic tt1ickness, and/or future rupture characteristics (fault orientation, fault style, etc.) between volumes of crust.

All areal source are mod eled as leaky," modeled 1

earthquakes that initiate~ in one areal source are allowed to rupture beyo1 nd the boundary into the adjacent areal sourc~e or sources.

PVNGS Are~al Sources:

Frequency Model & Earthquake Recurrence

• Recurrence of future earthquakes for all areal sources in the PVNGS SSC is treatecJ as a truncated exponential distribution (Gutenberg-Richter) with spatially variable parameters (a- and b-valuE~s).

• As discussed in the earthquake catalog development, each individual earthquakE~ in the catalog is. expressed in. the form of the expecte(J magnitude, E[M], and an equivalent count, N*.

• Using. these. two quantities in the recurrence calculations are used to pnoduce earthquake rates (a) and b-values.

PVNGS Areal Sources: Smoothing Three conceptual models for the spatial variation of v recurrence rate per unit area (v) within an areal source.

~ uniform sei:smicity; not used by PVNGS.

v

~variable, but continuous and relatively smooth seismicity; used by PVNGS. Became standard model of current practice (e.g., CEUS SSC). two general approaches have been used :

1. penalized maximum likelihood approach (EPRI ,1988 and CEUS SSC) where epistemic uncerta inty can be readily incorporated into the PSHA; used by PVNGS v 2. Gaussian "kernel" function to calculate the rate at any grid point as 1 *

• a distance-weighted sum of the earthquake counts within the areal

• t source (Frankel 1995)

~ assumes earthquakes can occur only at some discrete geographic locations within the large areal source; not used by PVNGS.

PVNGS SSC SSHAC Figure 8-2

Section 8.2.4.2 - Fornnulation of Penalized-Likelihood Model for R~ecurrence Parameters

1. Divide source zone into 0.25 degree cells

2. Formulate Poisson likelih(Jod function in each cell (depends on vi, bi, and earthquakes within cell)

3. Introduce penalty function that discourages large changes in v or b betwee11 adjacent cells

4. Introduce prior distributic)n of b (discourages solutions where b value dliffers from regional b)

5. Generate many realizations of joint penalized likelihood function ~ joir1t distribution of vi, bi, and smoothing parameters for all cell in zone

6. Generate representative 1maps from penalized likelihood functions

Section 8.2.4.2 - Fornnulation of Penalized-Likelihood Model for R~ecurrence Parameters

• Inputs to Recurrence c:alculations

- Weights to magnitude bins

- Prior distribution for b

- Priors for smoothing parameters

- Number of iterations

PPRP WS3 Comment #3: Pre~paration of a succinct narrative and more complete documentation of the analytical tool called "smoothing" during the meeting should be considered so that reviews of the analyses can be carried out in an informed and efficient manner.

Tl Response to PPRP WS3 Cc>mment #3: The Tl team acknowledges the hazard si~~nificance of the. decision. to calculate earthquake recurrE~nce parameters for areal source zones using a smoott1ed seismicity approach.

The Tl team also understancls that the soothing approach used in the project is a complex procedure that may not be well understood by all readers of the SSC report. For these reasor1s, a complete and thorough documentation o1f the smoothing process and assumptions will be pr<>vided in the SSC report.

PPRP WS3 Comment #4: Some. uncertainty in the nature of the M>4.65 seismic events mapped on the west side of the Southern Basin and Range province just east of the Gulf of California was noted in the rneeting. Conducting a review of the earthquakes comprising these events sho1uld be considered to determine if they are located on land or are associated with faulting within the Gulf. If they are pre-instrumental (or otherwise poorly located), efforts could be made to reposition the events.

Tl Team Reponse to PPRP WS3 Comment #4: The Tl Team. reviewed the portion. of the project earthquake. catalog. directly east of the Gulf of California, where. an approximately triangular wedge of seismicity appears to taper off into the Southern Basin. and Range. In order to assess the lik1elihood that: (1) the. project catalog correctly reflects a region of elevated seisrnicity rate along the western border of the Southern Basin and Range; and (2) the catalog correctly locates Mw > 4.65.

earthquakes in this region, the Tl Team reviewed the age, location uncertainty, and.

magnitude type of these earthquakes.

"'62 earthquakes. in the. project catalog are~ identified in this area, 8 occurred prior.

to 1950. The majority of these 8 earthquakes are based on instrumental data, such that only 2 earthquakes are reported with intensity-based (MMI) magnitudes. The Tl team assumes that the more recent 1963, 1969, and 1981 earthquakes are relatively well located, however, and should not be repositioned . Given this assumption, it is difficult for the Tl team to justify repositioning the 1935, 1952, and 1958 earthquakes. Therefore, the Tl team does not plan to reposition any of the earthquakes in this area. From these observations, the Tl Team assumes that the project catalog correctly reflects a region of elevated seismicity along the western border of the Southern Basin and Range.

PPRP WS3 Comment #6:.

(1) Usefulness should be reconsidered of the two-zone model (2) Reexamine. the positions of zone boundaries in the seven-zone model

• Include a possible extension of the transition zone to the west to include faults. and seismicity that are concentrated between the 320- and 400-km radii.

• Absorb the ETR zone. into the SCAB~~ zone

• Result would be 6-zone model that 'Nould be either the only model. or. the dominantly weighted model Tl Team Response to PPRP WS3 Comment #6:. The Tl team generally agrees that the geologic data suggest the two-zone model may be unrealistic, but this alternative is included to capture the range of technically defensible interpretations. Going forward, the Tl team will further evaluate the need for the two-zone alternative.

The Tl team agrees that the Eastern Transverse Ranges (ETR) zone is unnecessary, based on discussions with the PPRP and on hazard sensitivity results presented at Workshop #3. The Tl team likely will combine the ETR zone into the adjacent Southern California and Baja (SCABA) source zone.

The mathematics of the P'VNGS approach are the same as those used in CElJS SSC and are contained in PVNGS SSC SSHAC Section 8.2.4 and CUES SSC Sections 5.3.2, 6.4, and 7.,5.

Details on the application of this approach to the PVNGS are_contained in P~VNGS SSC SSHAC Section 9.4, where they discuss results of the recurrence calculations and the specific recurrence parameters for the areal seismic sourc:es. in the PVNGS SSC.

9.4.1.1: Removal <)f On-Fault Events from the PVNGS Earthquake Catalog

• Areal sources model the occurrence of future earthquakes not associated with identified active fault sources, so the recurrence in each areal source relies on a modified version of the PVNGS catalog in which on-fault earthquakes have been removed. Avoids double counting.

Earthquakes of M>5.5 were judged to be on-fault events if they produced surface rupture, or if published studies associated the. earthquake with a fault based on other data (e.g., seismological data, historical accounts).

Exception AZ earthquakes: Proponent Expert, Philip Pearthree, at WS2 asserted that no earthquakes within AZ can be definitively associated with known faults, with the possible exception of the 1992 ML 5.8 St. George earthquake (Christenson, 1995). The 1992 St. George earthquake occurred in southernmost Utah near the active Hurricane and Washington faults and may have been generated by slip on one of those faults (Pearthree and Wallace, 1992; Black et al., 1995). The St. George earthquake did not rupture the surface, however, so the causative structure is not known with certainty (Black et al., 1995).

Identified a total of 21 earthquakes that could be associated with identified active fault sources. Only the West, SCABA, and GULF areal sources are affected by this modification to the PVNGS catalog.

11e*w ns* w 9.4.1.1: Removal of On-Fault Events from the PVNGS Earthquake Catalog 20'014'4

7. 19 **

19~11?13 1 1 ~34112131 71 6.4& 191511112~ . "' * " ..

s 6 tq101 t 2n t 961> 811

~ 79 6.;3 f igutt 9-10. Eaitbquakrs remc>\-ed from~ P\~GS catalog for I.be~ of calculating re.curmice U1 areal soutces (Section 9.4 1) Earthquake epicenters are represented by red dots. associated surfucc ruptures (fromFigtn +14) att tngbliglued in red along !he causative &ult. The ma~ ofthr 193.i 11131 Cerro Pneto earthquake is taken from.Andet'son and BodU1 (1987).

9.4.1.2 Calculations - Recurrence Maps Tablr 9-4. !vlaguin1de-dependent "'*eights for east0FB (SCABA. GULF. and West) source zones.

western

\12.67- M 3.33-  :\1 4.0-  :\14.67- M 5.33-  :\1 > 6.0 Casr 3.33 4.0 4.67 5.33 6.0 1 (0.4) 0 0.8 1 1 l 1 2 {0.4) 0 0.3 1 1 1 1 3 (0.2) 0 0.2 0.5 1 1 1 Tablr 9-5. Magnitude-dependent weights for western (SBR. TZ. CP.1IH. and East) source zones.

eastern M2.67-  :\13.33- M4.0- :M 4.67- M 5.33-  :\I > 6.0 Ca st>

3.33 4.0 4.67 5.33 6.0 1 (0.4) 0 0 0 1 1 1 2 (0.4) 0 0 0 0.5 1 1 3 (0.2) 0 0 0 0.3 0.5 1

• 3 cases that were selected for the weights to the magnitude bins

• Weights are different for Eastern and Western seismic sources, b/c the catalog is very different in magnitude coverage and completeness between these two regions.

• Reasons for 0 weights:

• Stepp completeness analysis showed large spike in activity during passing of TA array suggesting low magnitude bin not complete before then

• magnitude-recurrence behavior of the data may deviate from exponential

• magnitude-conversion models or completeness models may be less reliable for lower magnitudes

9.4.1.2 Calculations - Recurrence Maps

• The mean value for the prior distribution of~ was set to 1.84 for all sources; corresponding to the regional bprior =0.85 (based on USGS)

• Strength of the prior distributiion of~ is specified by the standard deviation, where a large value indicates a weak prior.

ab:

- Eastern sources

• Case 1 crb=0.6

• Case 2 & 3 crb=0.7

- Western sources, crb=0.4

• Smoothing parameters were determined such that for each source zone, and for each of the three cases, the. smoothing parameters o 11v and a 11 ~ were allowed to vary, allowing the catalog data to determine the optimal range of values (described by the mean and standard deviation).

9.4.1.2 Objectively determined smoothing (roughness penalty) parameters+ one-sigma range Eastern Sources Case 1 (0.4) Case 2 (0.4) Case 3 (0.2) f f

I .,- -!- -

I i I f i

"i!

E

~ 0.01 b

aflv 0.001 ._____.__ _ . __ _ _- - I 1 o - - - - - - - - - - 4 .___.__ ____.__-.----------

r:

' ~ ~  ! ~

Source Zone Source Zone Sourcrt Zone 0.1 . , . - - - - - - - - . .

f

~ t t f i t II I

001 i 1~

f

...,s0 1~

1-1 0.001

~

0.0001 1 - - . . . . . . - - - - - . - - - . . - - - - -

l.L 0

.L

l 1

Ill Source Zone Sourco Zono Source Zone smaller ~ larger PVNGS SSC Figures 9-11 to 9-13

9.4.1.2 Objectively determined smoothing (roughness penalty) parameters+ one-sigma range Western Sources Case 1 (0.4) Case 2 (0.4) Case 3 (0.2)

"C" 1"' 0.1 "l

Iii E

~

,. 0.01

~ '

a~v 0.001 ii

~ ~

~ '.J I

J :l 8

0 Source Zone "' Source Zone Source Zone 0.1

.s::.

t 8E I\

0.01 t i Cl)

~

.!?.

a 0.001

~

O~p 0.0001 a u

~ ~

~

~

rn

~ (5  ::>

Cl I Source Zone Source Zone "' Source Zone smaller ~ larger PVNGS SSC Figures 9-14 to 9-16

' ' ~* 1::111 9.4.1.3 Resulting Race(M>5.0)/deg1/yr b value

_...__..__.___,_.....____.__.___. Recurrence Parameters Ca e 1 ( .4) and Maps Maps of mean recurrence rate (M>S) and b-value for Seismotectonic sources.

...... The right panel also displays the M>2.67 earthquakes in the PVNGS catalog.

Ca e 2 ( .4)

JI~ "

.2)

PVNGS SSC Figures 9-17 to 9- 19, 9-23

W *A I .If:

R:ite{M>5.0)/deg1/yr

' .. I to

• 9.4.1.3 Resulting

,.......__._----~~----1 Recurrence Parameters Ca e 1 ( .4) and Maps 1"'°' ...

C' Jt.:'

Maps of mean recurrence rate (M>S) and b-value for Two-zone sources.

t1'1 '"'

The right panel also displays the M>2.67 l~S ,, ' ~_.____.____.___.__.._____.._ . earthquakes in the PVNGS catalog.

Ca e 2 ( .4)

'=-' 1: ~

" ;llT1'o Ca e 3 0.2)

PVNGS SSC Figures 9-20 to 9-23

... 11..... """ """ u " llG'" ........

'- ~ " I *~ lJ :tu 1 9.4.1.3 Resulting

,__..__....____._...____.___._._----I ,... .., ,_....____.____.__.._____.____......__.___ Rec urre nce Pa ramet ers Cas 1 ( .4) and Maps Map of coefficient of variation (CoV) in earthquake recurrence rate (M>2.67) and sigma b-value for Seismotectonic sources (statistical uncertainties high Cas 2 ( .4) where there are few earthquakes) ur*w tu*w

.. ~'

Cas 3 ( .2)

• ~ '6~

PVNGS SSC Figures 9-23 to 9-26

Ir U1 9.4.1.3 Resulting CoVf R:m:(M>2.67)1 Sigm:i(h v.llue)

Recurrence Parameters Ca e 1 .4) and Maps Map of coefficient of variation (CoV) in earthquake recurrence rate (M>2.67) and sigma b-value for Two-zone sources (statistical uncertainties high where 1.\"'-oj '8'-S there are few earthquakes)

Ca e 2 .4)

Ca e 3 0.2)

PVNGS SSC Figures 9-23, 9-27 to 9-29

. .. ... 1\1'11 '" I ~ t 111 IJO 9.4.1.3 Resulting Rate(M>5.0)/<lcg2/yr b value Recurrence Parameters

.., I '

and Maps Two realizations of rate (M>S) and b-value for Seismotectonic sources. Magnitude weights according to Case 1.

PVNGS SSC Figures 9-23, 9-30, 9-31

l* I , ,. I 1, I u >it ! n,n tJilll II b value

• • 110 9.4.1.3 Resulting Rate(M>5.0)/dcg:/yr Recurrence Parameters and Maps t11it

~* lh'11 llA ll II' ' II ll Two realizations of rate (M>S) and b-value for Two-zone sources. Magnitude weights according to Case 1.

PVNGS SSC Figures 9-23, 9-32, 9-33

Comparison of model-predicted earthquake 1000 counts for SBR source zone (host zone) r------------~~=====:i ~--------------~

• C;Uog Case 1 (0.4) - Reaoi:nbon I Case 2 (0.4) - -1

-2

-3

- Roelt:auon 4

-~'

- Rooli:ntal s - ReolizabJR 5 i' - -s - ~e

~

~ 10 -.t--------~~-----i - ~7

- -7

~

ro

- -e RllGllclllon8 cr

..r:

t!

ro w

0 0

z 0.1 - - - - - - . - - - - - - - - - . . . . . . - - - - - - - - . - - - - - . - - - - - - - - - - . - - - - - - . - - - - . . . . . . - - - - - - - . - - - - - - . - - - - - - - - !

2.9 3.6 4.3 5 5.7 6.4 7.12.9 3.6 4.3 5 5.7 6.4 7.1 Magnitude Magnitude 1000 ~--------------~

• CilalOO Case 3 (0.2) - Reolcatm 1 The error bars are the 16%-84%

100 uncertainty associated with the data.

- Reablon4

~

- Reaizallcn 5 Predicted counts are obtained by summing

- Rleabltlon 6

~

~

~ 10 - ~7 the rates by bin for all cells within the Ill

- Rlmbtlcn9 er source taking into account the ro w

z 0

0 completeness times for each cell.

0.1 + - - - - - - - - . - - - - - - . - - - - . . . . . - - - - - - - - - . - - - - - - - . . - - - -

2.9 3.6 4.3 5 5.7 6.4 7.1 Magnitude

Comparison of model-predicted earthquake counts for East source zone (host zone)

Case 1 (0.4) 1000 , , . . . - - - - - - - - - - - - - - - . . , - - - - - - - - - - ,

• CainlOQ ,

. , , - - - - - - - - - - - - - - --=i Case 2 (0.4

- ~1 C>lalcg 100 -3 - Reallzatlon 3

- R-4 - Reammn 4

E

- -s - RIWzlllion 5

- Reali:alm 6

~

10 - -7 - Rmlclllcn 7 ca

J r:r

- -8

.c t::

ca w

0 0

z 0.1 ------.---------.-----.--------.-----4 --- - - - - - - - - - - - . - - - - . . . . - - -

2.9 3.6 4.3 5 5.7 6.4 7.1 2.9 3.6 4.3 5 5.7 6.4 7.1 Magnitude Magnitude 1000 ~--------------~

• Czlloo

- -1 The error bars are the 16%-84%

ReallZalOn 2 100 uncertainty associated with the data.

i' Predicted counts are obtained by summing

~

10 - Reaiz:lboo 7 the rates by bin for all cells within the

.x ca

J r:r

.c source taking into account the t:

ca completeness times for each cell.

w 0

0 1

z 0.1 ------.----------~-~--~------!

2.9 3.6 4.3 5 5.7 6.4 7.1 Magnitude

Rate {M>5)/deg2/yr 0

0 0

0 0 U)

~ ~

~

I I I

• I I I I USGS IKerael 1 r+

~

~

U1 I I II c(./') m n

• I usus Rates WitJiJlf loor G)

(./')

OJ 0

• I I I I I 11 QJ (J) 3 USGS Rates ~ith Fl'oo112 (,S-CA~IZ filoor /2) -* -0

l QJ QJ

l (J) c.. 0 I I I I I

Sei shi\otecton i c SBR c*ase 1, mean 1 1 l  :::0 QJ

l r+

0

- I I

SeismotectoniGSBR case 2 *me,an 1

I II :l C1Q m

c V>

I I I I  :::0 G)

• :s.e*i~ot1et"toni~ SBR Cas~ 3 nte,an1

-0 zG) m V>

C1Q s:

T *- ---*

0-*

(./')

• I I I III (./')

(./')

n QJ

~Z1ooe 1 E~sl Case 1 mean :l -0

• - ~-

II (J)

• I I I Z..Z!o oe E:ast !Case 2 mea~

~

I I II

,0

• I I 2*Z1ooe East 1Case 3 rmeani

PVNG Implementation of SSC SSHAC -

2 simplifications to area seismic sources (1) collapsing the rupture orientation branch to the central value, and (2) modeling fault dips as vertical

• APS stated

- For non-host sources, these simplifications were used, b/c non-host sources are minor contributors to the total 10-4 hazard at 1 Hz SA and the insensitivity of hazard to rupture orientation and fault dip was confirmed by performing a rupture orientation sensitivity using the SBR source {LCI, 2015a).

- Host sources are. sensitive. to. dip. and crustal thickness,. so. the difference.

in ground motion between the SSC fault dips and seismogenic thicknesses, and a vertical dip and single crustal thickness, was taken into account in the host sources by adjusting the ground motion for a vertical fault to the ground. motion for a non-vertical fault with multiple.

down-dip widths. The adjustment was calibrated to achieve accurate hazards at MAFEs of 10-4 and 10-6 {LCI, 2015a).

Simplification: (1) collapsing the rupture orientation branch to the central value 10 Hz Reference Rock SBR Hazard Sensitivity to Fault Orientation lE-3 al u

- - ~

c ""-........._

al

"'t:J t

u lE-4 '-

x al

~

- N-S 0

> r- ~ -....

~

al

J D"

al lE-5

_,, - N20W LL.

- N40W ni lE-6

\.

s *~

c 1 c 'Ill

<(

lE-7

\

0.01 0.1 1 10 10 Hz spectral acceleration (g)

Simplification: (2) modeling fault dips as vertical 10 Hz Reference Rock SBR Sensitivity to Fault Dip/Thickness

- 35 degrees 12 km Cl> - - 35 degrees 15 km u

c

~ lE-4 - t - - - r - - - . - - - - . - - - . - - , - ; - - . - r - - t - - - - ; - --rollk---r-7-i--r-:--t----r---r----7---i--:-r--.--rl - - -* 35 degrees 18 km Cl>

u

)(

Cl> - so degrees 12 km 0

u lE-5 -t---T--t----;--t---t-i--+-1---t---T--i---.--i--:--+~

c Cl>

J C- ---* 50 degrees 18 km C>

"' - 65degrees12 km E 1e-6 -j--~=====t==t=:i=i::::t=;=t -----:t==:t==::;=:t='f~=tt:-~~-~=Fr:R9=1:=T.I c -

• 65 degrees 15 km ct

- --* 65degrees18 km lE-7

• 0.01 0.1 l 10 - 90 degrees 15 km 10 Hz spectral acceleration (g)

Corrected for in the host sources by adjusting the 90° normal rupture ground motion to a dipping rupture ground motion, using: A2 = A1

• exp(C1+ C2

• ln(A1 )) (Eq'n 11-2) in which, A1 is the predicted median acceleration, A2 is the adjusted median acceleration, and C1 and C2 are adjustment coefficients.

Simplification: (2) modeling fault dips as vertical 10 Hz Reference Rock SBR Hazard, Modeled vs SSC Model 5

I:

1!-A Ii~

- SBRSSC

~ Mod'et i-o Wetstited lE*S

~

c cu

J CT - SBR: Modeled

-GI I;;

I c

1£*6 c

1<

~l 1 10 10 H-z spectral acceleration (g)

Corrected for in the host sources by adjusting the go0 normal rupture ground motion to a dipping rupture ground motion, using: A2 = A1

• exp(C 1+ C2

• ln(A1 }) (Eq'n 11-2) in which, A1 is the predicted median acceleration, A2 is the adjusted median acceleration, and C1 and C2 are adjustment coefficients. The coefficients were calibrated by performing sensitivities for the expected difference in hazard at MAFEs of 10-4 and 10-6 between the SBR source mean hazard calcu lated using dips described in the HID (Appendix F) and the corresponding hazard calculated using a goo dip. These adjustment coefficient s were used in the final hazard calculations.

PVNGS - Recurrence Rates &

Smoothing

• recurrence of future earthquakes in each area source is treated as a truncated exponential distribution (Gutenberg-Richter) with spatially variable parameters based on the smoothing of observed seisn1icity

• smoothing approach used is the penalized maximum likelihood. approach. that was implemented by the CEUS-SSC Project (EPRI et al., 2012).

• Activity rates and b-values calculated for area sources using assumptions on spatial smoc~thing of parameters and on interpretations of historical earthquakes. This process resulted in activity rates (for M>S) and b-values for each 0 .25 degree cell, for each area source used in the hazard calculations

PPRP Commen1ts and Tl Team Resolution wrt WSO (Kick-off Meeting)

PPRP. Comment #3: We concur with the benefits of having a recent SSHAC Le\lel 2 seismic source characterization for the P\/NGS site. However, care will need to be taken to a\loid the occurrence of anchoring (e.g., cognitive lbias). The Project Plan (p.

2) provides a procedure intended to address this subject using self-evaluatiions, but does not clearly include independent persJJectives that could identify a condition of bias. We would appreciate your informing us of how 'VOU plan to obtain independent views of any possible bias on the Team's part.

PPRP Commen1ts and Tl Team Resolution wrt WSO (Kick-off Meeting)

Tl Team Response to PPRP Comment#~!:

• The SSHAC Level 3 Tl team includes members who did not participate in development of the SSHAC Level 2 SSC. Specifically, PTI William Lettis and Tl team member Ross Hartleb did not participate in development of the SSHAC Level 2 SSC and thus bring fn~sh perspectives to the ongoing SSHAC Level 3 SSC effort. Gabriel Toro acted as a hazard analyst for the SSHAC Level 2 PSHA, but he did not participate in the development of the SSHAC Level 2 SSC. As such, Gabriel Toro also brings his fresh perspective to the SHAC Level 3 SSC Tl team.

• Discussions of cognitive bias will be included at the start of each workshop and working meeting by the PTI or Tl Lead. Moreover, if apparent cognitive bias arises at any point during a workshop or working meeting, the Tl Lead or other Tl team members or staff will be responsible for alerting the Tl team.

• Continual review of SSC development will be performed by the PPRP for the duration of the project. The Tl t~~am expects that the PPRP will alert the Tl team of any perceived cognitive bias at any point during the project.

PPRP Commen1ts and Tl Team Resolution wrt WSO (Kick-off Meeting)

PPRP. Comment #4: It was helpful for the PPRP to have participated via conf~~rence call with the

[GMC] team members during the SSC meeting.

There was the appearance~ of a gap in communications regardin~~ GMC-SSC interface items that came out in Norm Abrahamson's discussion. It is advantageous to have Tlhomas Rockwell of the SSC-PPRP also serving on the GMC-PPRP for the Project to help assure good coordination. Even so, we would appreciate your informing us of how you intend to maintain an effective interface between the GMC and SSC aspects of the PSHA.

PPRP Commen1ts and Tl Team Resolution wrt WSO (Kick-off Meeting)

Tl Team Response to PPRP Comment #4:

An effective interface between the SSC and GMC efforts will be maintained by the following:

• In addition to his role as PPRP member for the PVNGS SSC project, Thomas Rockwell also serves as a member of the PPRP for the SWUS GMC project. As such, he will b1e able to provide information and coordination between the SSC and GMC projects.

• The Project Plan defines the role of t he PTI as a technical expert responsible for ensuring coordination and compatibility between the SSC and GMC Projects. William Lettis is the PTI for the PVNGS SSC project and, therefore, is responsible for maintaining effective communication between the SSC and GMC projects.

• Members of the PVNGS SSC project also serve as members of the SWUS GMC project. Specifically, PVNGS SSC hazard analyst Robin McGuire is the Pa lo Verde PTI for the SWUS GMC project. Thus, he wi ll attend all PVNGS SSC and SWUS GMC workshops, be informed of SWUS GMC Tl team deliberations, and provide an important interface between the PVNGS SSC and SWUS GMC projects. Likewise, PVNGS SSC hazard analyst Melanie Walling serves as the Palo Verde hazard analyst for the SWUS GMC project. In this role, she attends all SWUS GMC workshops and working meetings. Thus, Melanie Walling wi ll be able to provide to the PVNGS SSC Tl team her first-hand knowledge of discussions and activities of the SWUS GIVIC project , and vice versa.

• PTI William. Lettis and Tl Team Lead Scott Lindvall. attended and presented at SWUS GMC Workshop. #1, which was held on March 19-21, 2013. This workshop also was attended by PVNGS SSC hazard analysts Robin M cGuire and Melanie Walling.

• Hazard analyst Melanie Walling and Tl team member and Project Manager Ross Hartleb participate in weekly status conference calls for the Palo Verde Seismic Hazard Evaluation Project. These conference calls also include participants from [APS] and Westinghouse Electric Company. The purpose of these calls is to discuss project progress, schedule, and SSC-GMC interfaces.

Backup slides PVNGS Area Source - Two-Zone Alternative 120" w 117'W ,, ... w 1n* w 1ot*w

PVNGS Area Source - Seismotectonic Alternative.

120* w 117°W 114* w 111° W 1oa* w 36° N 33* N 30° N

Section 8.2.4.2 - Formulation of Penalized-Likelihood Model for Recurrence Parameters n(m ~ M < m+dm ) =ATvj3e-f3(m-moJ [Aki, 1965}

n =.#of earthquakes b/t magnitude m & m+dm A= area of source T = duration of complete catalog (years) v = rate/unit time/unit area for earthquakes with m~m 0 (PVNGS m 0 = 3.0)

~=slope of the exponential magnitude-recurrence law (i.e., the b-value times ln[lO])

Section 8.2.4.2 - Formulation of Penalized-Likelihood Model for Recurrence Parameters Likelihood function takes the form:

f(v l 1l

/3) = IJ(v ATpe-/J(mi-mo))exp - vA T/Je-P(m-m,i)dm J i 1 v =rate/unit time/unit area for earthquakes with m>m 0 (PVNGS m 0 = 3.0)

B=slope of the exponential magnitude-recurrence law (i.e., the b-value times ln[10])

A= area of source T = duration. of complete catalog (years)

N = number of such earthquakes Likelihood function indicates the degree of consistency between the parameters that one. wants to estimate (in this case,. v and B) and the available data (in. this.

case, the number of earthquakes and their magnitudes during a time period T)

Section 8.2.4.2 - Formulation of Penalized-Likelihood Model for Recurrence Parameters N

£(v,f3) == (vAT) Ne-vAT x IT (f3e- /J(m;-mo) )

i ==l Apply simplification, because T depends on magnitude, and separate the likelihood function into a product of factors that depend on the rate v and factors that depend on the exponential slope~

Section 8.2.4.2 - Formulation of Penalized-Likelihood Model for Recurrence Parameters N

VML =

AT The separability implies that the maximum-likelihood estimates of v and ~are decoupled. In particular, take the logarithm of the previous expression, differentiating with respect to each parameter, making the result equal to 0, and solving for the parameter, the results are well-known eq'ns for maximum-likelihood estimates of v and~ [Aki (1965) and Utsu (1966)].

9.4.1.1: Removal of On-Fault Events from the PVNGS Earthquake Catalog YE'ar .Month Day FauJts RupturE'd Ea11bquake :Name Mw 1890 2 9 San Jacinto or Elsinore fault 6.8 1892 2 14 Laguna Salada fau lt Laguna Salada 7.3 1899 12 25 San Jacinto fault Christmas day 6.7 1906 4 19 Brawley fault 6.2 1915 11 21 Laguna Salada fault 6.6 1918 4 21 San Jacinto fault San Jacinto 6.8 1934 12 31 Cerro Prieto fauh 7.1

• 1940 5 19 Imperial fauh El Centro 6.9 1947 4 10 Manix fault Mani,-x. 6.5 1954 3 19 San Jacinto fault Arroyo Salada 6.4 1956 2 9 San Miguel fault San *Miguel 6.8 1966 8 7 CeITo Prieto fault 6.3 1968 4 9 San Jacinto fault BoITego Mountain 6.6 1976 12 7 Cerro Prieto fault 5.8 1979 3 15 Homestead Valley and Homestead Valley 5.5 Kickapoo faults 19 9 10 15 Imperial Brawley. and Rico Imperial Valley 6.51 faults 1987 11 24 Elmore Ranch fault Elmore Ranch 6.5 1992 6 28 Johnson Valley. LandeJs. Landers 7.28 Homestead Valley. Emerson.

and Camp Rock faults 1999 10 16 La,ic Lake and Bullion faults Hector Mine 7.12 2010 4 -t Sierra Cucapah fault system El Mayor-Cucapah 7.3 2010 6 15 Elsinore or Laguna Salada 5.8 fault

PVNGS SSC SSHAC Report: . c nAPTER 6: c ATAL oG oF L'I>EPE1'-i:DE.'i-r EARrnQc AKEs FOR PsHA ............................ 6-1 6.1 Catalog Region and Paranieter Lwllts ................................................................................................. 6-1 62 Earthqu ake Data Sourc:es ..................................................................................................................... 6-2 vi PVNGS SSC, Rev 0 6.2. 1 AISN Catalog ............................. ******-* ............................. **-**-- ---**********-*** ...................................... 6-2 6.2.2 SCSN Catalog ................................................................................................................................ 6-3 6.2.3 RESNOM Catalog.****--*-*********-********-**-*******-*******************-********************-*******-*********-*-*** .................. 6-4 6.2.4 Unified Earthquake Catalog of California (Unified CA) ............................................................... 6-5 6.2.5 ANSS Catalog ................................................................................................................................ 6-6 6.2.6 UCERF3 Catalog ........................................................................................................................... 6-7 6.3 Regional and National Catalog Selection and Standardization............................................................ 6-7 6.4 Remo\*al ofDuplicates ......................................................................................................................... 6-7 6.5 Standardize Earthquake Size Measure ................................................................................................. 6-8 6.6 Catalog Declustering ............................................................................................................................ 6-9 6.7 Magnitude Uncertainty ...................................................................................................................... 6-11 6.8 Catalog Co1Dpleteness ........................................................................................................................ 6-11

SSHAC Chapter 6: PVNGS SSC Independent Earthquake Catalog I::>" W 11r w t H' W 111' W 11111' W PVNGS Catalog

~ ,. .,., UT"H

~\ ~/,. z m E(MJ

~,,,' *""i>"o"f ~

... \ ~ c  ! 2 70 and < 3 00

~\ ~ r:I

)< 36' N

'Q. \

\

n 0 *  ! 3 00 *nd < 4 00

? ' 00 and< 500

\

~ 5 00 and < 8 00

\

\

! 6 00 and < 7 00

• 33' N e ?700 note ont-11ndepenclef\I events shown PVNGS SSC SSHAC Figure 6-9

• Eastern portion of the model region, earthquakes included in catalog if M > 2.7.

• Southern CA, N Baja CA, and Mexico earthquakes included in catalog if M > 4.7.

• Black, dashed "East-West divide" line shows line over which the criteria for inclusion in the catalog changes ..

SSHAC Chapter 6:

PVNGS SSC Earthquake Catalog

• PVNGS catalog is in SSC SSHAC report Appx E

• Tl Team used

- the catalog to calculate recurrence parameters for areal sources (SSC SSHA(: report Chp 8 & 9)

- individual inputs to the catalog that retained foreshocks and aftershocks to evaluate the distribution of seismicit~' in the model region to look for the presence of seisnnicity lineaments that might suggest activity along an unmapped fault in the southern B&R province (SSC SSHAC report Section 4.2.1)

PPRP Comment 83 from PPRP Comments on the Draft SSC Reoort and Tl Team Resoonses Location of Date Location ln N.o. PPRP Comment Comment Summ:iry of Revisions to Report Received Report' Revision runctlon of time. network operator at AZGS maintains tnat M4.0 are oomote.te since 2007 and MS.Osince 1970.

83 211712015 Chapter 6 The tlUe and subsequent terminology should Throughout As dascussed In the Chapter 4 comments and be clearer. The objecUve is to have a catalog or Chapter 6 responses. there IS no version of the catalOg that events that are statis\Jcally independent of 11dudes earthquakes below M2.7 In that case, each other. thereby not Including roreshocks, our *complete* catalOg pnor to dec1uster1ng anershoc.ks, or swarm-like occurrences. Since cannot meanangfuly be used to search for there also needs to be a complete se1sm1clty setsmicrty lrieaments The pre-dedustering catalog that Includes the events excluded from catalOg alSo has no utility in terms of PSHA. II is the statistically independent catalog, it needs a therefore not presented meaningfUI name, such as "historical catalog" In chapter 4, ANSS and AZGS catalogs aown to or *composite catalog*. There are several MO are used to search tor spatial trends in add1Uona1 names ror the "tndependenr catalog m1croselsm1c1ty.

In Chapter 6, such as "project catalOg," -rmal The catalog is now oonSistent.ly referred to as catalog", *oata catalog*, "final PVNGS catalog; "the PVNGS catalog*

• comprehensive composite catalog", etc.-

please pick two names ror the two catalogs used In the SSC study, and donl use any others.

It would be helpful to the reader to explain the characteristics of the two catalogs in this lntroductlon. pointing out that the Independent catalog Is used for eva1ua11ng recurrence of PVNGS SSC. SSHAC Report future earthquakes for earthquakes magnitude 2.7 and larger. While the

• more complete*

catalog (name not established yet) contains all CHAPTER6 reported earthquakes in the region (needs to be specified) Including earthquakes smaner than M2.7. In particular It is necessary to have a *complete* seismtclty map within the 320-l<m-radius region around the PV srte to support a discussion or the association of seismicity With CATALOG OF INDEPENDENT EARTHQUAKES FOR PSHA oeotoo1c structure and faults.

This chapter describes the development of the catalog of independent earrl1quakes for PSHA for the P\'NGS SSC (hereafter refe1red to as the PVNGS catalog). Developin,g a comprehensive catalog of independenc events is crucial to the tmdersranding of futw*e earthquake hazard. panicularly in regions where rbe causarive mechanisms for earthquakes are not well known and se1smicity rates are low. such as in central Arizona. The PVNGS catalog is provided in Appendix E.

Iu constrncting the catalog of PVNGS catalog. foreshocks aud aftershocks were removed. Tbe TI Team used the P\'NGS catalog to calculate recurrence parameters for areal sources (Chapters 8 and 9). The TI Team also used mdi\idual mp11rs to the catalog (Section 6.2) that retained dependent and smaller evems 10 eviiJuate the dtsmburiou of se1suuc11y w the model 1eg1011. ')pecillcally lookmi? fo1 lhe reseuce of se1sm1c1ty lineaments 1bat might suggest acti\ ity along au munapped faull (Figwe .t-8)

PVNGS model region extends to 400 km to include major faults in SoCA_and NW Mexico 1ZO-W u,.~ ioe*w Seismic Netwol't(

"' Authoritahve Regions Antona Int gratetf

"' se1siii1e ~etwort< (AJSN) l6* N La Red Slsl'Tllca def Noroe&te de MflUCO (RESNOMl

- NevadaS mic

\ NolWOfk tNN)

\

\ . . Southern Ca11f0tl'Ma Seis c Net~*o"k (SCSN)

I mJ Utah S.1$mograph Netlf.'Otk (lJU)

• I D'N Ul\lfi d CA J

I '

30'N 0 100

- mi PVNGS SSC SSHAC Figure 6-1 -==-*

0 100 km

The procedure used to cre<3te the PVNGS catalog is :

1) Identify and obtain regi ional and. national.

seismicity. catalogs, and. then standardize the catalog entry formats.

2) Merge the catalogs and remove duplicate events.

3) Use magnitude conversion equations to estimate a uniform magnitude measure (Mw) for each earthquake.

4) Identify independent e\,ents through declustering analysis (time-space wir1dow declustering approach [USGS, 2007]).

5) Account for magnitude uncertainty.

6) Assess the overall catalc)g completeness.

1) Identify and obtain regional and national seismicity catalogs, and then standardize the catalog entry formats.

Catalog Source AISN Arizona Geological Sun'ey (AZGS)

SCSN Southen1 California Earthquake Data Center (SCEDC)

RE NOM Center for cienrific Research and Higher Education at Ensenada (CICESE)

Unified CA University of California, Los Angeles (UCLA)

ANSS Advanced National Seismic System (a USGS Consortiu111)

UCERF3 \J\Torking Group on Califon1ia Earthquake Probabilities (\VGCEP) 1~*w nrw 11 ~ *w 111'W 1orw PVNGS Catalog Sources ACIV..nceG NllllOl\dl Set$1111C System (ANSS) 36" N e Anzuna Integrated Setsmtc Network (Al Sff) e La Red Sfsrn1ca de!

Noroeste de. M6iuco (R.ESNOM)

Souahern Celtfo11ua Setsmtc Netwarl< (SCSN)

Cal.tomla Undied Catalog (Urned CA) ll' N 0 100

- ITll 0- 100km

AISN Catalog

• Network: Arizona Integrated Seismic Network {comprises two regional.

networks operated by the AZGS and N. AZ University {NAU).

• Dates of operation : AZGS adopted eight legacy Transportable Array stations in 2008, 7 of which are currently working. Earthquake monitoring in northern AZ can be traced back to 1961 when a station was installed at Flagstaff. In 1986, formation of the Arizona Earthquake Information. Center {AEIC) and the northern Arizona Seismic network, which originally consisted of 3 stations at Flagstaff, Williams and the Grand Canyon. Currently, this network includes 8 stations (5 running) stretching from the Utah boarder to the southern edge of the Colorado Plateau.

• Mission: Monitor AZ earthquakes at a lower threshold and with more uniform coverage than USGS, and generate a n1eaningful earthquake catalog that can be used to determine seismicity rates and elucidate tectonically active seismic areas.

• Catalog

Description:

3,489 events in the AISN catalog dating back to 1852. "'60 to 100 events are added/year. Estimates of completeness as a function of time :

M4.0 since 2007; M5.0 since 1970.

PPRP Comment 1.1 on Seismicity data (PPRP Letter #3: PVNGS SSC WSl)

PPRP Comment 1.1:

It was clear from the presentations and discussions that earthquake monitoring within Arizona has been given a low funding priority historically and is operating in a fragile manner through the dedication of a few individuals. This severe restriction of resources may have led to operational practices that could have impacted the quality of the data being relied on for catalog developnnent. We suggest that the Tl Team consider performing a friendly "quality assurance" review of the earthquake. monitoring. data analysis procedures. used. for the ABN and NASN. The national standard for seismic network operations is established by the US Geological Survey's Advanced National Seismic System (ANSS); the seismic networks to the west (California) and north (Nevada and Utah) are. members of ANSS.

Tl Response to PPRP Comment on Seismicity (7/15/13 Letter "Response to observations and comments from the PPRP on Workshop #1")

Tl Response to PPRP Comment 1.1: To the extent possible, the Tl Team will perform the recommended "friendly quality assurance review" of the data analysis procedures of the Arizona Broadband Network (ABN) and.the. Northern Arizona Seismic Network (NASN), which collectively comprise the Arizona Integrated Seismic Network (AISN). To our knowledge, operational procedures for these networks are not published and are not readily available. The Tl Team will contact Jeri Young (Arizona Geological Survey) and David Brumbaugh (Northern Arizona University) to see if such procedures are available for the ABN and NASN, respectively, for Tl Team1 review. The Tl Team will use this information to evaluate quality and uncE~rtainty of the data to inform our judgment when weighting alternatives in the SSC, but will not reevaluate earthquakes in those earthquake catalog~.

NRC Review of Tl Response to PPRP Co1nment 1.1: There's no mention of "quality assurance" review in the PVNGS 50.54f submittal or the SSHAC report, but it may have been incorporated, as the Tl team stated in its comment response, into evaluation of quality and uncertainty of the data to inform catalog priority and weighting of alternatives in the SSC.

SCSN Catalog

• Network: Southern California Seismic Network (SCSN)

• Dates of operation: Been running in some form since 1921. Starting in 1932 it had 6 stations. Today the SCSN records data from 4:~4 stations.

• Catalog

Description:

As of 1/1/2015, the SCSN catalog has 608,908 local events (events for which SCSN is authoritative, excluding q1Jarry blasts and other non-tectonic events). On avera1ge the catalog is complete for Mc~ 3.2 since 1932, and Mc~ 1.8 since 1981, excluding early hours or days of large aftershock sequences and regions near the edge of the network.

RESNOM Catalog

• Network: Red. Sfsmica del Noroeste. de Mexico.

• Dates of operation: RESNiOM has been operating for more than 30 years. 9 shc>rt period instruments started operating in 1978, this number expanded. to 13 in 2013. The first broa~dband stations were installed in 2001. 14 new stations were installed during 2011 and 2012. At present, RESNOM operates 19 broadband stations.

• Catalog

Description:

7,0510 earthquakes of magnitude 2.7 or greater in the RESNOM catalog through December 2012 ,with 100s of events added per year. The minimum rr1agnitude of completeness for current reporting is 2.4.

Unified Earthquake Catalog of California (Unified CA)

• Mission: Construct and test certain hypotheses of earthquake occurrence, Wang et al. (2009) compiled a new catalog covering the vvhole of California, which lists all known earthquakes at magnitude 4. 7 and above.

• Catalog

Description:

Unifie~ d CA lists all known regional earthquakes at magnitude 4.7 and above, from 1800-2006. 23 existing catalogs 'Nere examined for this effort, enabling more accurate magnitude and location estimation; different magn1itude types were also converted to moment mag~nitude. Catalog. ends. on.

December 31, 2006.

ANSS Catalog

• Compiler: Northern. Cadifornia Earthquake.

Data Center, Universit~f of California at Berkeley Seismologica l Laboratory

• Catalog

Description:

Cjomposite national earthquake catalog th<3t is created by merging the master earthquake~ catalogs from.

contributing ANSS institutions and then removing duplicate soll utions for the same event.

UCERF3 Catalog

• Compiler: USGS

• Catalog.

Description:

Tl1e UCERF3. earthquake.

catalog is an update of the catalog compiled for UCERF2. Only eight, post 2006 CA earthquakes(> M4.7) ,were. selected from this catalog. It was necessc1ry to add these events because the primary cla talog for the Unified CA ended on DecembE~r 31, 2006.

PPRP Comment 1.2 on Seismicity data (PPRP Letter #3: PVNGS SSC WSl)

PPRP Comment 1.2:

The PVNGS site is within a part of the Southern [B&R] Province that is characterized by very low seismicity. For example, within 50 miles of the site, there is only one M3+ event in the current catalog and only one event of M<2 recorded during the three years of the TA operation. We think that it would be useful to further quantify the seismicity rate by searching for recordings of earthquakes that are large enough to be detected but too small to have enough stations to locate. One could start by looking at the three-year

[transportable array (TA)] database using the nine TA stations roughly centered. on the station closest to the PVNGS site. These data could. be helpful in providing a more refined subdivision of the seismicity patterns associated with the Southern [B&R] Province, and. thereby used for refining the areal sources used in smoothing seismicity for areal source rates. Some of the initial results for computing rates for areal sources. using the base case model, as.

discussed by Melanie Walling, were startlingly high given the observed low seismicity within 50 miles. of the. PNVGS site.

Tl Response to PPRP Comment 1.2 (7/15/13 Letter " Response to observations and comments from the PPRP on Workshop #1")

Tl Response to PPRP Comment 1.2: Mlany of the earthquakes recorded in the PVNGS study region during the three-year window of the [TA] have magnitudes that are below the lower rnagnitude cutoff (Mw 2.7) for inclusion in the project earthquake catalog. The Tl Team agrees, however, that the TA earthquakes may be usefu l, in particular for evaluating seismicity rates and patterns within the study region and possibly to provide additional information or insights on seismicity rates. As such, the Tl Team will continue to investigate the TA earthquake data to assess alternate ways to capture uncertainty in the SSC. These investigations likely will include sensitivity analyses intended to assess the impacts of various modeling decisions on seismic hazard at the PVl\JGS site.

NRC Review of Tl Response to PPRP Comment 1.2: Discussed in SSHAC.

report Section 4.2.1 regarding lineaments or patterns that would reflect activity of unmapped faults in Southern B&R.

The procedure used to. cre<3te the PVNGS catalog is :

1) Identify and obtain regi ional and national seismicity catalogs, and then standardize the catalog entry formats.

2) Merge the catalogs andl remove duplicate events.

3) Use magnitude conversion equations to estimate a uniform magnitude measure (Mw) for each earthquake.

4) Identify independent e\,ents through declustering analysis (time-space wir1dow declustering approach [USGS, 2007]).

5) Account for magnitude uncertainty.

6) Assess the overall catalc)g completeness.

2) Merge the catalogs and remove duplicate events.

• 7 authoritative zones were drawn and prioritized (figure on next slide)

• Duplicate records were identified and removed from the composite catalog when they occurred within a minute of each other and were spatially located within 0.1 degree in latitude and longitude (about 9 to 11 km) . If only the time criterion was met, a visual inspection of the records in question determined if removal was required . When duplicates were identified, the record from the highest priority catalog was retained .

Table 6-1. Catalog priority within authoritative zones. Catalogs are described in Section 6.2.

Authoritative Zones are hown in Figure 6-2. (Priority ranking: 1 is high. 5 is lo\v and 0 does not apply).

Zone Location l:nified CA scsx ANSS AIS .. '" RE S~Ol\I 1 W. California l 0 0 0 0 2 E. California l 2 .) 4 5 3 Arizona l 4 3 2 5 4 r . Sonora 1 4 3 5 2

. Baja 5 California 1 4 3 5 2 6 NW. Sonora I 2 ~

.) 5 4 7 S. Nevada 1 3 2 4 5

2) Merge the catalogs and remove duplicate events.

120* w 114*w t \I" w 108' W

- z m

~

-x 0

Ci 36 N

1) Unified CA \

2) ANSS

3) SCSN 1) Un1f1ed CA

4) AISN 2) AISN

5) RESNOM 3 3) ANSS

4) SCSN

5) RESNOM

1) Unified CA

, f 33' N

2) SCSN I

3) ANSS I 4 ) RESNON I

5) AISN I 4

1) Unified CA

2) RESNOM

3) ANSS 1) Unified CA 4)SCSN 2) RESNOM

5) AISN 3) ANSS

4) SCSN

5) AISN 0 PVNGS SSC SSHAC Fi ure 6 3

The procedure used to. cre<3te the PVNGS catalog is :

1) Identify and obtain regi ional and national seismicity catalogs, and then standardize the catalog entry formats.

2) Merge the catalogs and remove duplicate events.

3) Use magnitude convers;ion equations to estimate a uniform magnitude rr1easure (Mw) for each earthquake.

4) Identify independent e\,ents through declustering analysis (time-space wir1dow declustering approach [USGS, 2007]).

5) Account for magnitude uncertainty.

6) Assess the overall catalc)g completeness.

3) Use magnitude conversion equations to estimate a uniform Mlw

• Recently developed conversiions were used:

- Utsu (2002) and Sipkin (2003) to convert ML (local magnitude scale), Ms (surface-wave magnitude scale), and mb (body-wave magnitude scale) to Mw for the 2008 NSHMP

- Arabasz (2013) determined several conversions using the University of Utah's earthquake catalog

- Zuniga and Castro (2005) found that a correction of +0.1 units to the RESNOM catalog's Md (duration) scale is equivalent tomb, which can then be combined with Sipkin's (2003} mb equation to obtain Mw.

• Not enough events in catalo;g to test these relations, so instead, the ANSS catalog of the. western U.S. was used for the evaluation. Orthogonal. regressions were performed on. the ANSS catalog data.

3) Use magnitude conversion equations to estimate a uniform Mw Mw vs. NEIC Mb Mw vs. Nevada Mc

., 7 65 6.5 6 II s.s S.S

• s

• 0.La POlllU

- Onholl~u  ! s

• O-..UPotnu onho Regress

• .s - Sip n 12001) *s - USGS(M\11 *- MQ

- Arab.au mb POU - Atabas.r W.c UUl Mw

• US(0.16) *Mt..O 73(0.91) 3.S H l

l ...s 6 7 l

l H *s 5 ss .,

Mc (kH)

Mw vs. Utah Ml ELS 6

s.s

• 0.IA Poi:nta

~ s - <lf't...,R...,._s

- Utsu (lCO.t)

• .S

-Anbav Ml UUl

' Mw - 0. M(0.07)

• Ml.0. 60(0.86)

.l.S PVNGS SSC SSHAC Figures 6-4, 6-5, & 6-6 l JS

' as s ss &..S ,

ML (UU)

3) Use magnitude conversion equations to estimate a uniform Mw Mw vs. ML (Cl, NEIC, NN, UU) 7

~

6 s.s

• Dita Po1nu J s

~

- Ortho Recress 4.S - utsu IZ00.21

- Arabasl ~* L UUl 4

Mw

• 1.11 *ML-0.57 ll 3

3 3.S .. *.s s s.s Ml.

Arabasz vs. G&R (1956} Intensity Relations 8

7.5 7

6.5 6

~ 5.5 - (G&Rl9S6) s - AtabaulO<V 4.S - Atabllz. 10> v 4

3 .5 PVNGS SSC SSHAC 3 Figures. 6-7 & 6-8 MMI

3) Use magnitude conversion equations to estimate a uniform Mw l\lagnimde or Ioteosity T~11e 1 C 00,-ecion Equ:atiom 1 1 Source h.fw = 1_67

• i(lfL -2.60) (for ~fi. > 16_5) 1

),<~ rv1I_GS, Ji.illY AS~ ~ll.SLC Utm(2002) eke 1,ifw = J..iL

~-~_res (RESNOM) D1',. = Mtii_:fes _._ O_l (use mb eq11a.tiom) Zuniga amd Castro (200j) l\oid ~=1\(a Petei_-sen e-t al (2000)

),*fw_ :M"wHR..\l Mw=Mw 1\iiatMnmtical Identi~r l\~~- = 1.46*Ulb-2.42 (fm- mb > 5.3) else m, 2 me.GS Sjpkin (2003)

Mv.* = ll1J:.

1Ylc Ma.= l\iic Perersen et al. (200&)

NllYfi ~=Mt = {2/3)*~ +I Gutenberg md Richter tn9j6J Unt; n 1\-fw=Unk Petersen e-t at (2008)

}.!~ l\ifn. = l\>ih Peteri:sen et al (2008) 1\'I ~lw=~f Petersen e-t al (2008)

Mu,= 0_7:5 * (M~+L93) (fur~<5. 8):

1vlt Mw = l _50

• Qt-I.. - 2J.l0) (forM; > 7.8)~ Utsn (2002)

PVN ( S SSC SSHAC Table 6-3 ehle ~1. = M.

The procedure used to. cre<3te the PVNGS catalog is :

1) Identify and obtain regi ional and national seismicity catalogs, and then standardize the catalog entry formats.

2) Merge the catalogs and remove duplicate events.

3) Use magnitude conversion equations to estimate a uniform magnitude measure (Mw) for each earthquake.

4) Identify independent e,vents through declustering analysis (tiime-space window declustering approach l[USGS, 2007]).

5) Account for magnitude uncertainty.

6) Assess the overall catalc)g completeness.

4) Declustering analysis

• Tested alternative approaches on a catalog of seismicity. located w ithin the southern B&R. This reduced dataset is an appropriate sensitivity sample because it contains the closest and, hence, most crucial portion of earthquakes of the PVNGS catalog.

• Grunthal {1985) classifies 25% of events as dependent.

• Reasenberg (1985) conversely, does not appear to be well suited for the limited data available; as only a handful of dependent events were removed .

• The Gardner and Knopoff (1974) algorit hm, which reduces the sub-catalog by 20%, is bracketed by the other two methodologies, and thus deemed by the Tl Team to be a reasonable approach for declustering the composite catalog.

Independent EYents Gardner Basin and and Range Grunthal Knop off Reasen berg Events (1985) (1974) (1985) 303 228 -?4'"\;) 284

- 75% 80% 94°/o PVNGS SSC SSHAC Table 6-4

4) Declustering analysis Applying the Gardner and Knopoff declustering algorithm to the composite catalog reduced. the record count from 1,941 to 1,048 events . . This "'50%. reduction is in line.

with that of Gardner and Knopoff (1974), who found that approximately 2/3 of the events in the more completely recorded SoCA instrumental catalogs were identified as aftershocks.

1?0" W 11rw 11<<" W Ill w IOll'W tJ TA 11 PVNGS Catalog z E(M) l!I

~

~

~2 70 and < 3 00 l!I

.)< 315*N n  ! 3 00 and < 4 00 0

\

~ 4 00 and < 5 00

\

• ! 5 00 *nd < 6 00

~ 6 00 and < 7 00

!700 n*N note onty independent f\'tn11 lihown 30'N 0 100 PVNGS SSC SSHAC Figure 6-9

-==--

. _ _ km ml 0 100

The procedure used to. cre<3te the PVNGS catalog is :

1) Identify and obtain regi ional and national seismicity catalogs, and then standardize the catalog entry formats.

2) Merge the catalogs and remove duplicate events.

3) Use magnitude conversion equations to estimate a uniform magnitude measure (Mw) for each earthquake.

4) Identify independent e\,ents through declustering analysis (time-space wir1dow declustering approach [USGS, 2007]).

5) Account for magnitude uncertainty.

6) Assess the overall catalc)g completeness.

5) Magnitude uncertainty

• Following the NSHMP approach, standard errors of 0.1 magnitude unit for earthquakes occurring after 1971, ().2 for 1932-1971, and 0.3 for 1850-1931 were assumed.

• Estimates of catalog cc>mpleteness and recurrence parameters are computed using the same methodolog~y as in CEUS SSC via uniform moment mag1nitudes, E[M] and the equivalent counts, N* [N*= exp(~ 2 a 2 /2)].

The procedure used to. cre<3te the PVNGS catalog is :

1) Identify and obtain regi ional and national seismicity catalogs, and then standardize the catalog entry formats.

2) Merge the catalogs and remove duplicate events.

3) Use magnitude conversion equations to estimate a uniform magnitude measure (Mw) for each earthquake.

4) Identify independent e\,ents through declustering analysis (time-space wir1dow declustering approach [USGS, 2007]).

5) Account for magnitude uncertainty.

6) Assess the overall catalog completeness.

6) Catalog completeness

• Completeness. time interval is defined as the time period over whicl1 earthquakes of a specified magnitude rcinge are believed to be complete.

• Following figures sho" Stepp-style plots (Stepp,. 1972) of subsets of the PVNGS catalog.

based on authoritative~ zones.

• Table list magnitude rc1nges and event completeness for varic>us time periods, as indicated by the arrov\ts in the figures.

Ste pp Plot, S. California, Baja (Zones 1 & 5) Stepp Plot, E. talifomia (Zones 2, 6 & 7) 1 1

• 1..?0 <QM)< l.l7

- UO<EiM)<l.37 c

• 4 71 <E{Ml < !>38 c

• l.37 < QMJ < *-~

0 "I) of "O 0.1 4 1t < l{MI < S.ll S S1<£lMl<60S i..., - 137 < E{MJ < 4.04

• 4.0t < QMJ < 4..71

~

l.,,

- S.JI< l{Ml < 605 - 4J)il < E{Ml < 4.71 01 1 -

• 60S c l(Ml < 6.72 6.0Sc ljM] <6.72

• U1 < C[MJ < s..31

• 71<EIMJ < S.31

• 672<llM)<7.39

• S.38 < !IMI < 6.05

- 6.'72<E.1Ml < 7.Jt - S3*< ElUJdOS G.OS c £!Ml < 6. n 0.01 U!5< E!MI < '*7J 1 10 100 1000 1 10 100 1000 Ytat* befMt Jal\. 1. 2013 Yu n kfor* Jin. L ZOLS lt t f UtSS P'~GS t: GS lli P'~GS r scs 1910 1960 4 71 1933 1910 1963 4 -1 5-38

• 5 38 605 1920 18.,0 1900- 1933 1900 °""" l t P"W ..... 1910 1930 - 1963 1910 1930 6.05

• 672 1850 1850 1910 1850 6.....2 7.39 1850 1850 ,. .. 1850 1850 739 8 06 1850 1850 1850 1850 s-3 1850 1850 l)UnrledCA 2)ANSS 3)SCSN 1) Umfed CA 1850 1850 4)AISN 2)AISN 5)RESNOM 3 )JANSS 4)SCSN S)RESNOM I

1)U111fitdCA 2)SCSN I I

I 33* 04 l)AfliSS I 4)RESNON 5)AISN I '

,1 PVNGS SSC SSHAC Table 6- 1) UnlhclCA 2)RESNOM 1)Uni6odC4

))ANSS 2)RESNOM 4)SCSN 3, 6-10, & 6-11 0 JOO S)AISN 3)ANSS

-=- ...

4)SCSN S)AISN 0 1oa

Stepp Plot, Arizona (Zone 3) Stepp Plot, Sonora Mexico (Zone 4) l 1

• 2.70 < EIMJ < U7 2.10< E(M) < l.37

- 2.?0<E{M)< :UJ c 2.'10 < l[M) < 3.37 0 c

'P

• l.l7 < E[M) < .t.(M 0

'P

• 3.37 <UM) < 4.G4

- 3.J7 < £[MJ < .t.04

..,~.. - 3.37 < lfMl < 4.1)4 0.1 * .t.()1<£tM)<4 .11

..,~.. 0.1

• 4.04 < E(Ml<<Ul

~

- 4.0I < t[M) < 4.11 i - 404<l[M}<4.11 0 4..71 < l[MJ < ~33 .,;

o 4.11 < E(Ml < S.31

- 4.71 < ElMJ < !..38

- 4.1l<E[M)<S.3&

• S.38 < E!MJ < 6-0':i

• S.la<l(M]<~OS

- BS<llMJ<6.0S

- S.l8<E!MJ<6.0S 0.01 l 10 100 1000 1 10 100 1000 Yun bt!tore JML 1, 20ll Years before Jan. l, 2013 P°'~GS 'CSG S

~ :P\~GS t:-SGS l970 l960 c' 4.04 1990

<" 4.71 1920 1963 < 4 71 1960 1963

< 538 1870 1930-1963 1 9~ 1930 - 1963 6.05 1870 1930 °'w 1930 1930 6.72 1850 1850 ---- -- - -

VTAl1

<., 6.72 1850 1850 1850 1850 1850 "

'",. ,.... 739 1850 1850 1850 n

<) 8.06 1850 1850 1850 1850 1) Unified CA \

2)ANSS \ 8.73 1850 1850 3)SCSN 1) Unified CA

<l)AISN 2)AISN 5)RESNOM 3 3)ANSS 41 SCSN S)RESNOM

• l

,' 33'Jl I

I '

11 PVNGS SSC SSHAC Table 6-3, 6-12, & 6-13 0 100 0 100

SSHAC Chapter 6:

PVNGS SSC Earthquake Catalog

• The PVNGS catalog ranges frorr1 1852. through 2012 and. contains 1,048 independent events.

• Duplicate records were remove~d from the catalog when they occurred within a minute of each other and were spatially located within 0.1 degree in latitude and longitude, with exceptions. If only the time criterion was met, a visual inspection was made. Removal of duplicates was done according to the highest priority catalog.

• Used various relations for magnitude conversions to obtain Mw.

• Used Gardner and Knopoff {19~74) algorithm for declustering.

• Completeness, Western CA is complete since 1932 down to a 1

uniform moment magnitude (E[M]) of 4.7; Eastern CA and AZ are complete to E[M] 4.0 since 1930. Northern Sonora, Mexico also appears to be complete to E[M] 4.0, but only since 1970.

PPRP Comment 1.3 on Seismicity data (PPRP Letter #3: PVNGS SSC WSl)

PPRP Comment 1.3:

In consideration of the limitations of the current earthquake monitoring networks in Arizona, it might be reasonable for [APS], on behalf of its PVNGS, to consider installing one or more seismic monitoring stations for specific data targets relating to the current project and for future use regarding seismic hazards related to licensing. As noted in the above comments, the paucity of seismic monitoring data in the low-seismicity environment of PVNGS has both positive and negative* aspects. Here are several possible deployments that could be useful in the short term (the current project) and in the long term (future licensing matters).

a) Given the location of surface bedrock within several miles of the PVNGS site, a broadband station comparable to the vandalized TA station could be installed at a reasonably secure location with data telemetered to a central recording site (potentially operated by the AISN).

b) Several additional short-period seismographic stations could be installed to form a small array around the central station to improve the detection and location of occurring earthquakes. These data would be used to refine the seismicity model used for areal sources.

c) It has been suggested that a strong-motion station be installed within the site perimeter to collect data on kappa for the site. It could be useful to operate similar strong-motion instruments along with the stations described in items (a) and (b) above.

These possible seismic monitoring installations would best be considered in the context of both the near-term application of the data for the current project {a few years) and the longer-term interests of APS with regard to the role of seismic issues in future operational considerations at PVNGS.

Tl Response to PPRP Comment 1.3 (7/15/13 Letter "Response to observations and comments from the PPRP on Workshop #1")

Tl Response to PPRP Comment 1.3: The Tl Team agrees that installation of a seismograph or seismographs at or near the PVNGS site would provide useful data both for the current project and for the longer-term interests of [APS]. Specifically, these data would provide improved earthquake monitoring and reduction of uncertainty on site kappa and other ground motion parameters. The installation of new instrumentation, however, is beyond the scope of the current project. We are in current discussions with APS and Westinghouse Electric Company (WEC) regarding the possibility of obtaining additional budget to install, operate, and maintain this new instrumentation and to determine who would receive and support interpretation of any new data. To. maximize benefits to the current project, the Tl Team understands that any new instrumentation should be installed as soon as possible to maximize the number of earthquakes recorded in this low-seismicity environment.

PPRP Comment on Seismicity Monitoring PPRP Letter #4: PVNGS SSC WS2:

Seismicity Monitoring; This proposal focuses on procuring and installing new broadband and strong-motion instrumentation at the PVNGS site in a borehole drilled for this purpose to bedrock, a depth of about 500 feet beneath the site. The purpose of the instrumentation is to collect data for {l) detection and improved location of earthquakes ("'Ml and larger events) in. the central part of the southern [B&RJ province including near the Palo Verde site, and (2) refining the value of kappa at the site using primarily weak ground motions from local or regional earthquakes. The instrumentation would be operated initially as a freestanding system recording in a triggered mode, prior to establishing more permanent power and Internet data communications for longer-term operation .

The PPRP endorses and strongly supports the funding and implementation of these work items as soon as possible. APS has indicated that funding may be available for new work if it is well justified. The PPRP urges that a high priority be placed on implementing these work items at the earliest possible dates in order that the data may be obtained in a timely manner.

Tl Response to PPRP Comment Tl Response to PPRP Comment on Seismicity Monitoring: [APS], LCI, and the Tl team are working together to implement ... newly proposed work items, including ... installation ~Jf a downhole seismograph array at the site, and collection of Spectral Analysis of Surface Waves (SASW) data at the site. In a recent teleconference with LCI, APS indicated their intention to fund thie... new proposals. During that call, however, APS indicated that funding for the new work largely will not be available until early in 2014, with the exception that procurement of seismograph instrumentation is underway so that it will be available for installation as early as possible in 2014.

NRC. Review of Tl Response to PPRP Comment 1.3 & WS2 Comment on Seismicity Monitoring: There's no mention of installation of a seismograph(s) at or near the PVNC)S site in the PVNGS 50.54f submittal or the SSHAC report, so this PPRP comment is unresolved.

PPRP Comment on Seismicity PPRP Letter #6: PVNGS WS3, Comment #4:

Some uncertainty in the nature of the M>4.65 seismic events. mapped on the west side of the Southern [B&R] province just east of the Gulf of California was noted in the meeting. Conducting a review of the earthquakes comprising these events should be considered to determine if they are located on land or are associated with faulting within the Gulf. If they are pre-instrumental (or otherwise poorly located), efforts could be made to reposition the events.

PPRP Comment on Seismicity Tl Response to PPRP Letter #6: PVNGS WS3, Comment #4: The Tl Team reviewed the portion of the project earthquake catalog directly east of the Gulf of California, where an approximately triangular wedge of seismicity appears to taper off into the Southern [B&R].

In order to assess the likelihood that: (1) the project catalog correctly reflects a region of elevated seismicity rate along the western border of the Southern Basin and Range; and {2) the catalog correctly locates Mw > 4.65 earthquakes in this region, the Tl Team reviewed the age, location uncertainty, and magnitude type of these earthquakes .

62 earthquakes in the area, only 6 have magnitudes Mw > 4.65 {i.e., 1935 Mw 5.0, 1952 Mw 5.1, 1958 Mw 4.9, 1963 Mw 4.7, 1969 Mw 4.8, and 1981Mw4.9) . The location and magnitude for the 1935 earthquake are based on felt intensity reports and therefore may be highly uncertain. The locations and magnitudes of the 1952 and 1958 earthquakes are based on instrumental data but are reported only to the nearest half-degree, reflecting a high degree of uncertainty. The Tl team assu1Ties that the more recent 1963, 1969, and 1981 earthquakes are relatively well located, howe?ver, and should not be repositioned. Given this assumption, it is difficult for the Tl team to justify repositioning the 1935, 1952, and 1958 earthquakes. Therefore, the Tl tearn does not plan to reposition any of the earthquakes in this area.

NRC Review of Tl Response to PPRP WS3, CC>>mment #4: Resolved.

T ABLE OF CO~TE\!S PVNGS SSC SSHAC Report, HID:.

1.0 INTRODUCTION

......... ........ ... . . ... ................................................................................... .. F-3 2.0 OVER.\'IEW OF SEISWC SOURCES ****************-*-*******-*********************-*-*****************-***-******* F-3 21 Areal Soucces ........................................................................................................................... F-3 22 Fau1t Sour~'-*-*-*-*-*****-*-*-*---*-*-*- .................- ................- ..- -..- ........... F-3 CHARACTERISTICS OF AREAL SOL"R.CES .......-.........................- .....................- .......- ....... F-5 Attal So~~ --****--**-**-*-***-***-*-******* -*******--*-*--*** -*-*--***-*-*-* **

• F-5 Rupture !\iecNin*sms for F~ ~ --* ....................- .......................... - .......- ....... F-5 Rupture Oo.enlatlOOS fOI' Funn ~s ........ . _ --****-************* ................. ***-*- F-6

~~ Dip$ for F111Ure EartlJq"aans -*******-****-****-*--*****--*-****-*-****-**********--*-*-**-****-* F-6 R<<uaen~ offunn Earthquakrs m Arw Sourr~ ............. -*-** ...... ****--*** *-** .. F-6 4.0 CHARACTER.ISTICS OF FAULT SOURCES .............................*-******-*-******-**-*-**-******-*** F-7 4.1 Fault Source Probability of Acti'\'lty ***- *****-************-*-*-**-**************-**-*** ..****-*-*******.. **-*** F-8 4.2 Rupture Length*******************************-************************************-************************..*****-**************** F-8 4.3 RuptllteA.rea ............................................................................................- .............................. F-8 44 Displacement pei-Eveot ................................................................................................... .. F-8 4.5 Seiunogeoic Thiclaiess ...................................................................................................- ...... F-8 4.6 Chamcteristic Magnitude .......................................... - ............................- .............................. F-8 4.7 Earthquake~ for Fault Sources ................................................................................. F-9

5.0 REFERENCES

.... .................................................................................................................. . F-10 ATIA~ .................................................................................................. ..F-50 Attachment A: heal Source Coordwates (electronic attachment)

Attachment B: Fault Source Coard.mates (electromc attachment)

Anaclnnen.t C: UCERF33 Rupture Sets (electronic attKhment)

Attachment D: A.BSMOOIB Omput (el-ectronic attachment)

AnacbmHJt E: S\VUS GMS Re~o~ for Fault Sources (electronic attachment)

Append1x F F-2 PVNGS SSC, Rev 0

SSHAC HID 3.0: Characteristics of Areal Sources Areal source boundaries All leaky Rupture mechanism, orientation, seismogenic Assessed on a source-by-source basis thickness, maximum magnitude..

Strike-slip faults dip 80 + 10° Reverse faults dip 45 + 15° Rupture dips Normal faults. dip 50 + 15°.

Dip direction for. all non-vertical faults is random .

All modeled in the areal sources are allowed Top of rupture to rupture up to the ground surface (i.e.,

depth= 0 km) .

Different rate cases for eastern and western Rate cases sources.

All treated as. a truncated exponential.

Recurrence distribution (Gutenberg-Richter) with spatially variable parameters.

Fo r the 2-Zone Areal Sou rces :

Source Bound;n1* Rupture Rupture Ruptun Top of Seismogenic ~Imax Rate l\lagnitude Dip Rupture Thickness Recw*r ence

~fame Type ~lechanism 01ientation p lw) Cases (degrees) (km) (km) ~lodel Strike-slip 70° (20%)

(800'o) 80° (20%)

90° (60%)

6.8 (0.1)

N35°W (20%) 30° (20%) 12 (0.2) 7.0 (0.25)

Re"\*erse West Leaky (10%) N45°W (60%) 45° (60%) 0 15 (0.6) 72 (0.4)

N 55°\V (20~o) 60° (20°'o) 18 (02) 7.5 (0.2) 7.9 (0.05) 35° (20%) 1 (0.4)

Nonual 50° (60%) 2 {0.4) G-R (1.0)

( lO~o) 65° (20%) 3 (02) 35° (20%)

Normal N20°E(10%) 50° (60%) 6.8 (0. 15)

(80°0)

N-S {lOO'o) 65° (20%) 12 (0.2) 7.0 (0.25)

East Leaky N20°W (400'o) 0 15 (0.6) 7.2 (0.35)

~4-0°W (200'o) 70° (20%) 12 (0.2) 7.5 (0.2)

Strike-slip Random (200A>) 7.9 (0.05) 80° (200/o)

(20°0) 90° {60°'o)

PVNGS SSC SSHAC Table 9-1

Top of Seismogenic .:\lagnitude Sour"Ce Boundary Rupture Rupture Rupture .:\Imax Rate Rupture Thickness R ecwTence For the l'inme Type  :\lecbani-;m Orientation Dip (km) (Jan)

[.\In-) C a-;es

\fode.J Seismotecton ic Stnke-slip 700 (200*)

80°(20%) 6.8 (0.15)

(900/e) N35°W (200:.) 12 (0.2) 7.0 (0.25)

Areal Sources: SCABA Leaky N45oW{60%)

900 {60"/o) 0 15 (0 6) 12 (04)

N55°W{20%) 30°(20%) 18 {0.1) 7.5 (0.15)

Reverse 45° (600/o) 7.9 (0.05)

{IO~:.)

600 (200/o)

Strike-slip 700 (20%)

(7000) 800 (200/o) 6.8 (0.05)

N35°W(200o) 900 (600/o) 12 (03) 7.0 (03)

GULF Lcaky N45°W(60%) 0 14 {0.6) 7.1(03)

Normal N55°W{200o) 35° (20~*) 16 (0.1) 7.5 (03)

(30%) 500 {600*) 7.9 (0.05) 65° (200/o) 1 (0.4) 2 (0.4) G-R (l.O) 35° (20%) 3 (0.2)

~onnal 500 (600/o) 6.8 (OJ)

(80%) N-S{Ne) 65° (200/o) 12 (0.2) 7.0 (0.25)

'.\"200W (~*}

SBR Leaky 0 15 {0.6) 7.2 (0.4)

N400W (200'.) 70° (200/o)

Strike-slip 18 {0.2) 7.S (02)

Random (2~*) 800 (20%)

(2000) 7.9 (0 05) 900 (600*)

35° (20%)

Noonal 500 (600'.) 6.8 (0.05)

(800'.) N~S~:.)

65° (200/o) 12 (0.1) 7.0 (025)

N'lO"W (409/o)

MR Leaky 0 15 (0.6) 7.2 (035)

N4WW (200/e) 70° (200/o)

Strike-slip Random GOO o) 18 (0 3) 7 5(03) 800 (20%) 7.9 (0.05)

(2000) 900 (60"/o) 35°(20%)

Nonnal 6 8(0.2) 50° (600*)

(700'.) N20°E (2S~o) H {0.2) 7.0 (0.25) 65° (209/o)

TZ Uaky N20°W (25° o) 0 17 (0.6) 72 (03) 70" (200/o)

Strike-slip Random (500*) 20 (0.2) 7.5 (0.2) 80" (200/o)

{3000) 7.9 (0.05) 900 (6001.)

35° (200:.)

~onnal 6.5 (02) 500 (600/e)

(80"/o) 15 (0.2) 7.0(0.3) 65° (200/e)

CP Leaky Random (1 00~:.) 0 20 (0.6) 7.2 (025)

PVNGS SSC SSHAC Strike-slip 70" (2000) 25 (0.2) 7.S (0.2) 800 (2000) 7.9 (0.05)

(200/o)

Table 9-2 900 (600/o)

PVNGS SWUS (jMC Chapter 5:

Ground Motion Databases a1nd Candidate Models for the Median and Aleatory' Standard Deviation"

• Chapter 5 describes

- ground motion database used to evaluate the alternative GMPEs. The d,atabases were also used to develop new models for the aleatory. variability. using the partially non-ergodic ,a pproach (single-station sigma and single-path region sig;ma) .

- selection of the candidate~ GMPEs. for median ground motion. From the evaluated set of 19 GMPES published between 2004 and 2014, 6 candidate GMPEs were selected for PVNGS.

Table EX-2: Se lected Cand idate GMPEs for the median ground motion GMPE DCPP DCPP Distant PVNGS - Greater PVNGS - Distant Sources Arizona Sources CA & MEX Sources Abrahamson et al {2014) x x x x Boore et al {2014) x x x x Campbell and Bozorgnia {2014) x x x x Chiou and Youngs {2014) x x x x Idriss {2104) x x x Zhao et al {2014) x Zhao and Lu {2011) adj ustment x to magnitude scaling Akka r et al (2014a, 2014b) x x Bind i et al (2014a, 2014b) x

PVNGS SWUS (jMC Chapter 5:

Ground Motion Databases a1nd Candidate Models for the Median and Aleatory' Standard Deviation"

• For PVNGS, 4 empirical ground. motion databases were used by the Tl Team for evaluation of the alternative:

• PEER NGA-West2 database (Ancheta et al., 2014) - Use of the database was restricted to strike-slip and normal 1faulting earthquakes that control the hazard at PVNGS

• Reference Database of Seismic Ground Motion in Europe (RESORCE) described in Akkar et al. (2014c) - Use of the database was restricted to strike-slip and normal faulting earthquakes that control the hazard at PVNGS

• PEER Arizona database (Kishida et al., 2014a) - recordings in AZ from earthquakes in AZ and recordings in AZ from earthquakes in CA and Mexico.

Used to (1) evaluate path effects from median ground motion from CA earthquakes (2) evaluate kappa for rock sites in AZ (3) develop aleatory variability models for earthquakes in CA and Mexico recorded in central AZ (single-path region sigma models)

• Lin et al. (2011) - consisting of ground motion residuals from M4 to M6 earthquakes

• Ground motions from the M6.0 2008 Wells, Nevada and the M6.7 2011 Fukushima-Hamadori normal-faulting earthquakes were also evaluated

Table 5.1-1: Primary Emp irical Data Sets Distance Mechanisms Magnitude Vs30 Range Dip Range DATASET Range by Range (m/s) (degrees)

(RJs in km) Earthquake PEER NGA-West2 3.0 - 7.9 0 - 1532 89 - 2100 10- 90 57% SS 17% NML 26% REV Akkar Subset of Reference 4 .0 - 7 .6 0-200 92-2165 2-90 38% SS database of Seism ic Ground 47% NML Motion in Europe (RESORCE) 15% REV PEER Ari zona {Regions 1, 2, and 3) 4.3 - 7.2 145 - 649 398 - 1312 40- 86 93% SS 7% NML 0% REV PE ER Ari zona 1.2 - 3.4 9 -301 398 -1237 Not Not

{Centra l Arizona) Avai lable Avai lable Lin et al. {2011) 3.9 - 7.6 0.6 - 208 166 - 760 10-90

Table EX-1 : Ground motion databases and their application for the SWUS project.

NGA-West2 RESORCE PE ER-Arizona Lin et al Fin ite-Fault (2011) Simu lations DCPP Median SS and RV SS and RV DCPP complex & splay SS and RV ruptures PVNGS Med ian Greater AZ SS and NML PVNGS Kappa for Ar izona rock Earthqua kes site in Arizona PVNGS Median for CA/Mex sources CA/Mex eqk 200-400 km DCPP & PVNGS x Single-Station Sigma x x PVNGS CA/Mex eqk Sigma for CA/Mex 200-400 km sources DCPP & PVNGS x HW scaling

Table 5.1-l: Data Sets Used for the Evaluation of the Median Ground Motions DATA BASE Of DATASET SUBSET USE Of DATASET ORIGIN NGA-W2oc-Mm PEER NGA- M 25.0 Evaluation of Ule medtan West2 No HW sites* ground motion mode l (for Vno2 250 m/ s OCPP) for base FW model NR£Jeqk ~ 3 Adjusted to Vno=760 m/s NGA-W2py.ym PE ER NGA- M~_O Evaluation of the median West2 -70km s Rx5 70 km for both SS ground motion mode l ( for andNML PVNGS)

VSJtl 2 250 m/s NR£Jeqk ~3 Adjusted to Vm=760 m/s EURPY-M'ED Reference M~S _O EvaJuation of the median database of R~70km ground motion mode l (for Seismic Ground for both SS and NM L PVNGS)

Motion in vn;J 2 250 m/s Europe NRedeqk ~3 (RESORCE) Adjusted to V90=760 m/s PEER-AZuni PEER Arizona Earthquakes from NGA-West2 in Estimation of tile median path Regions 1 and 2&3 recorded at tenns for Regions 1 and 2&3 stanons in Anzona (for PVNGS)

NREc/eqk 23 N11.Ec/slation ~ 5 SIMoc-vm SCEC !iimulations SS: M55, M6_0, M6_6, and M7.l Evaluation of tile median using the broad ground motion mode l (for band platform REV: MS-5, M6_0, and M6.5 OCPP)

SIMH.w SCEC simulations REV: MS.S, M6_0, and M65 Eval uation of the scaling of using the broad Dips: l0, 20, 30, 45, 60 the HW effect for magnitudes band platform ZTo : 2 .5, 7 _5, 12 km between MS and M6.5, and for Zltl11 sca ling {for OCPP)

• Includes the followrng* 0 ~Rx 2 -70 km for both SS and REV; 0 $Rx S 70 km & R3 S 10 km for SS; and 0 s RJC ~ 70 km for SS, & Dip 2: 80 deg.

Table 5.11-3~' Data Sets Us;ed for the Evalu atio111 c f ttlle Kaippai and Ground Moti0ns ffrom Splay a nd Conn p lex Ru ptu res DATABASE OF DA11iASET SUBSET USE OF DATASET OR1GIN PEER-AZM.?i"A PEIE R Ariza nai Earthquakes in Estim a,t ion of kappa for staitioru in Arizona record ed 1 c:entr.a1 Arizcma (for PVNGS) at statton:s in

,Arizona SIM:;l*T SOEC sim ulatlions Mam n !Evaluation of the methods to 1r:omp ute ll.lsingth e SS: M7.0-M7A ground motiof!ls for sp lay ruptllr e:s (fl()r broadbaiind REV ~ MJ.O - M7-4 DCPP)

RJllatfo rm S?la1f SS: M6.0-M6-4 REV ~ M6.4 SIMc:o:npto. SCiEC simulations SS: M6L7- M7_4 IEva[uation o f the m et hods. to compute llllSBlngthe ground motiorms for complex rupt ures tilro a cfba rn d REV ~ M6.4-M7.0 (f or OCPP) platfo rm

Tabte 5.1-4: Data Sets Used for the Evaluation of the Aleatory Standard Deviation DATABASE OF There are multiple versions of the DATASET SUBSET USE OF DATASET ORIGIN dataset depending on which of the EURMS& Refttence database M~S 1. Computation of residuals of Seismic Ground DIST :S SO km 2. Development of single- NGA-West2 GMPEs is used for the Motion in Europe Nucfeqk ~ 3 muon sigma models based on (RESORCE) Nw:/site ~ 3 European data for application residuals, because different subsets to PVNGS sources in Greater Arizona ( ;,~-rr'lt model) of the NGA-West2 dataset were G lO BA4.~ PEER NGA-West2 M ~ S 1. Use of residuals from GMPE used by the different developers of GLOBAL-=....m.u. and Un et aJ (2011) DIST S SOkm developers GLOBA~-cn* N1m;/eqk ? 3 2. Development of single- the GMPEs. For these cases, the GLOBA~14 N11Et/'site ? 3 station sigma modeJs based on the g1obal data tor application subset name includes the reference Selection applied to to both OCPP and PVNGS the subset used by sources in Greater Arizona. GMPE as well.

the GMPE ~ ~ o.t.i - 30 model) developer NGA-W2~ AS'-lol PEER NGA-West2 M?55 L Use of residuals from GMPE NGA-W 21.Mt-.**UU.ll.. 200 S Dist s 400 k:m developers NGA-W2~M" N.uc/eqk ~ 3 2. Development of single-Nuc/site ~3 stauon sigma model based on large distance data for Selection applied to application to PVNGS sources in the subset used by regions 1, and 2&3 (

theGMPE ;ss-.JJIU! -w modeJ) developer NGA*W ~ PEER NGA-West2 California L Use of residuals from GMPE NGA-W2CA~ earthquakes with M developers NGA-W2c.A~ NGA- ?5 2. Development of single-w2~" DIST :S SO k:m station sigma models based on Nuc/eqk ? 3 CA data for application to OCPP Nuc/site ?3 (-S>-<.>-i & ~.t-~ models)

Selection appl~ to the subset used by the GMPE developer PEER-AZ,.\~"UL PffR Arizona Earthquakes from L Compurauon of residuals PEER-AZ,,..~CSA1A NGA-West2 in 2. Development of singl~th PE ER-AZ,4")K'l't.a Region land sigma for application to PVNGS PEER-AZ,i.7H-CB1A Regtons 2&3 from eanhquakes in Region 1 recorded al. stations and Regions 2&3 { - Rlll in Anzona model)

PEER NGA-WE~st2 Database (PVNGS: Used_to_determiine single station sigma)

• Expanded the previous PEER (2008) NGA ground-motion database to include worldwhde ground-motion data recorded from shallow crustal earthquakes in active tectonic regimes after 2003 and range of magnitudes included in the database was extended down to M3.

• Each NGA-West2 developer selected their own subsets from the full NGA-West2 data set, such as, remove recordings that :

- had missing key metadata,

- data that were judged to be unreliable (metadata or ground motion data), _

- Were not considered applicable to shallow crustal earthquakes in active tectonic regions, and

- class 2 earthquakes (aftershocks).

e-------------

Normal Strike* Slip

&-r--;r--,r-;--~.,.--,,........,~

Reverse PEER NGA-West2 f 6 -c~4.~r-+i Is ~~~liililliiiiiili Database 3 -L----+-~+""

01 10 100 3

1000 0 1 10 100 1000 3-+--~~

01 10 100 1000 (PVNGS: Used to RJ8 (\ml RJS llaTI) RJ5 ()om)

IOOOOO-r:==T""T"1~~i"."'T"?~t:=i determine single station sigma) 01 1 10 20 0.1 I 10 20 Majority of the earthquakes are either Period(wc) Period (sec~

10000 . - - - - - - - - strike slip (57%) or reverse and reverse J 1000 t~--=~...;...-=*'""""'........~=;1 0

J oblique {26%). Normal and normal.

~ 100 ;--~~~--:::-~-:-:::-1 l oblique earthquakes. make up 17% of

~

d 10-----..--- the earthquakes but only 8% of the 10 100 RQcoiU.,g~ Plr Earthciualll tOOO recordings.

B20000----~--......--~~

rim) i I=::o~.VS3~50~ I For M > 5, the distribution of "6

Ir 1000 * -* *-

j 1oo ~i...........+-.-_ ,.__,,,........._-t earthquakes is similar: 49% strike-slip, J 10 31% reverse {31%), and 20% normal.

~

a ,......_-+--------~

50 100 1000 3000 l/530(n"O'Sl Figure S.L l -1: Summary of the data distribution of the NGA-West2 database using the .subset of reliable data selected by ASK14.

Arizona Ground Motion Database(Kishida et al.,2014a)

(PVNGS: Used to determine AZ site kappa, fv'ledian GMPE & sigma for CA/Mex sources)

• PEER compiled a database of ground motions recorded by 15 stations in AZ produced b'V 26 earthquakes that occurred in AZ, CA, or Mexico after 20 07. 1

• The closest station to PVNGS (Z14A) is located 8 km away.

• 13 AZ recording stations aroLJnd the PVNGS site were part of the USArray, and 2 were ~>ermanent stations managed by the USGS/CalTech Southern California Seismic Network.

• Dataset consist of 12 small (rVI_< 3.5) earthquakes in AZ with hypocenter distances of 9 to 300 km, and 14 earthquakes in CA and Mexico with RJB distances between 150 and 600 km.

• 14 CA and Mexico earthquakes had recordings in CA that were included in NGA-West2 database, but the ground motions in AZ were not included in the NGA-West2 database.

1 Normal Strike-Slip Unknown ia IJ Arizona Ground i* 3 4

3~~~..;.;.;;..14-+~+-~~ Motion Database 2 2 l L--===~~=3 {PVNGS : Used to determine AZ 10 100 1000 0.1 10 100 1000 R.lB i.n-} RJB(kll"I) site. kappa, Median GMPE & sigma 1:<<00 """"------.....-....,...-..,,...-.,..-,..,-~

t*y()O for CA/Mex sources) l a:

• coo b *()()

~

} IO .._.,_,_..~~~:--"rl--1 1r, .:>.'

• ()()

~

1:-

~ 10 L

?

~

. :ro 0

10 100 n..........,. r.,. :.11'"

r~ 1()00 0

..t*. 100

~

ID

-~  !'i() 1()()

Figure 5.1.2-1: Summary of the data distribution of the PEER-Ar izona database.

V530 for the Arizona Sites (Chapter 3 of Kishida et al., 2014a)

• SWUS sponsored a study to measure the site conditions at the AZ seismic stations.

• For 10 of the 15 stations in the PVNGS region, spectral analysis of surface wave (SASW) dispersion technique was used to determine the detailed site velocity profile, average velocity in the upper 30 m of the profile (VS30), average velocity for th e entire profile (VS,Z), and NEHRP site classification were derived.

• 3 independent inversion techniques were employed .

• Results showed that th ere were 2 typical site types:

deep stiff soil {alluvium) sites - 8 stations, V530 is typically in t he range of 370 to 690 m/s, with usually gentle monotonic increasing velocity with depth thin soil over rock sites - 2 stat ions, V530 is in the range of 970-1240 m/s, with greater variance in t he fie ld dispersion data and greater variabi lity between the inverted profiles t han for th e deep alluvium sites.

Table 5.1.2-2: Ariz.ona Data Set V:30 Values (from Table 3-1 of Kishida et al., 2014a).

Max N umber of Magnitude Distance Inversion Used for Station Vno (m/s)

Recordings Range Range (km) Depth of Vs Kappa profile (m}

Z14A 11 1.2 - 3.4 50 - 206 490- 524 108 Yes U SA 9 1.2-3.4 59-301 424-460 99 Yes Y16A 4 1 .5 - 1.5 158 - 159 970-1028 40 Yes YlSA 8 1.5-3.4 119 -189 499 -566 40 Yes Z15A 2. 1.5-3.1 69 -251 373 -464 39 Yes 113A 7 1.5 - 3.4 96 - 2.50 1140-1237 38 Yes Y14A 9 1 .5-3.4 82 -147 473 - 526 50 Yes Y13A 5 2.0-3.4 42-135 532- 560 so Yes 114A 7 1.2 -2.4 28-183 380-404 so Yes ZHA 5 1.5 -2.4 88-111 652 -689 50 Yes

Akkar et al. Subset fro m. RESORCE Database 1

(PVNGS: Used to determine Median GMPE in Greater AZ (SS+ NML) & single station sigma)

• RESORCE is a pan-European earthquake strong-motion databank and is one of the products of the Seismic (3round Motion Assessment (SIGMA) project. Database accelerograms were processed using a uniform methodology.

• Subset of the RESORCE database used by Akkar et al. (2014c) was used by the SWUS GMC project and it excluded recordings from the full RESORCE datadase that:

- had no measured VS30 values,

- earthquakes with magnitudes less than 4.0,

- earthquakes with unknown style-of-faulting,

- hypocentral depth greater than 30 km, recordings at RJB distances greater than 200 km, and

- events with only one recording.

• Akkar et al. subset from RESORCE database includes

- coverage for distances from 5 to 200 km

- magnitudes from 4.5 to 6.0 normal and strike-slip mechanisms

- sampled periods for the response spectral values between 0.01 and 4 sec

- 1,041 3-component recordings frorn 221 earthquakes recorded at 322 strong-motion stations

Normal Strlke*Slip Reverse

• ------- Akkar et al. Subset from ie......_...~.._..:-~~

~

~5 *++~~M RESORCE Database (PVNGS: Used to determine Median GMPE 10 100 1000 in Greater AZ (SS+ NML) & single station R..e (lim) sigma) t 1(JCX)()

J 1COO t--~~~..,._.,:.- Majority of earthquakes O 100 and accelerograms are from strike-slip I£ 10--1-..,_,.........,....,.__.............~......

- SS events (38% of events and 36% of

- - RV 1

01 i=====~..:,_:_.!..!.;..-J_~

I 10 20 10 20 recordings) and normal events (47% of PttiUd l11Cl t*ooo events and 51% of recordings).

f w 100 ~...-........--~....................,~

'& The number of reverse-slip events and J

i recordings in the database are small s'

I --~--__............._.............._

10 100 1000 100 compared to the other style-of-faulting classes (only 15% of the events and 13%

R"'llD'Cll"QIB 1-'er !:At'~

~20000 *r-------..-----:---.

of recordings are from reverse events).

i 10000 - All>lat el al v

er 1000 -

M>6, R<70

\1~50""'9

• in

'a Most of the sites have VS30 values in the I~

• oo range of 250 to 750 m/s. There is sparse i 10 coverage for V 530 values greater than

~ , _.____.__~~ 750 m/s.

50 !00 1000 VSJO (Miii Figure 5.1.3-1: Summary of the data distnbution of the European database (Aldcar et al, 2014c) using the subset of re'1able data selected for the development of me Akkar et al. (2014a and 2014b) model.

Lin et al. Database a----------a-.----.-----

Normal Strike*SUp Reverse (PVNGS: Used to determine t6-H-"-tr~~fM

~ 5 ..,_...,.,...,..,;-~ ---~

. . . .crititt 6 iE!~~-a.a 5 -+~...,+-,n-,.W 6

~....,..,,.,+--+**:;~...,.,.

single station sigma) 3 _._~_._------1 3 -+---i--~-- 3 r 1 - - + -......-+---"-f 0.1 10 100 1000 0.1 10 100 1000 0.1 10 100 1000

• ,::km-r::>~~~~~ R..ll~(lc111) 1 Database was d eve Io ped to study the 1000 - r r - - RJQ C\On :---i-.........................__........

components of the aleatory  : ~~ ~ *=

variability ( site, path, and source j x

100 ..............__ __.

t HHF terms) using the extensive data set of 'O

.8 10 ground motions from Taiwan. z s

• Because of the large number of 1_._----~

0.1 I 10 20 I 10 20 aftershocks from the 1999 Chi-Chi ..

v1000 ~---~~

Pwiod Cwc:I Pono:l l&OC) earthquake, there are many sites  !

with large numbers of recordings per ~ 10014~g~~~

site. j i 10 .......___.............--.-;.~,...............

• For the objective of evaluating the 1 components of variability, Lin et al. ] 1- t - - - - - _ . . . . . . - - 1 10 100 1000 100 restricted their data set to sites with RooorttirGS Pat Eanl'lquake

~ 20000-.---......-----.---,

at> 10 recordings per site. i '0000

• Data set only used for the evaluation CI:

?>

I 1000 100....__ _ _- - - i -_ _ _ _,

of the single-site within-event z

~

standard deviation, cPss* r J

~

3 1 ~------

50 100 1000 3000 VS30 ltl\'1)

Figure .5 .1.4-1: Summary of the data distribution of the Lin et al (2011) Taiwan database.

Additional Data from Normal Style-of-Faulting Earthquakes 2008 Wells, Nevada Earthquake

• Located "'10 km NE of the town of Wells, NV * *

• Moment magnitude of 6.0

• Occurred on a previously unmapped fault (USGS, 2014)

• B&R earthquake not part of the PEER NGAWest2 database

• Ground motion and metadata from this earthquake were compiled and are summarized in Appendix I

• Earthquake occurred during the time in which the USArray was deployed in AZ

• 8 stations were within 100 km of the

• epicenter (Figure 5.1.5-1), but only 7 had records that were not clipped

• VS30 values were inferred based on the surface geology from the USGS (2007) database and the correlation of VS30 and geology given by Wills and Clahan (2006)

Figure 5.1.5-1: Eplcentnl location of me Wells (NV) event. Also Shown are me muons within 100 km that recorded the event.

Additional Data from Normal Style-of-Faulting Earthquakes 2011 Fukushima-Hamadori, Japan ~

Earthquake . ~1

• M6. 7 occurred in eastern Tohoku, 11 April l 1!

2011 and was apparently triggered by the , Ji 11 March 2011 Tohoku (M9.0) ~

earthquake. ~- ij 37°

• 598 records from K-net and KIK-net *Y'-'

stations within 800 km from the § epicenters were collected by PEER and ~ .~> . ,',.~

• processed following the same procedures r~ ~ ....,

as the NGA-West2 database. ~

• 1/

• Earthquake consisted of a complex rupture involving several faults. According j Vertital orr,~1 c L ~ 'J to Shiba and Noguchi (2012}, the source 1m was comprised of two rupture planes . -~,.---

• The source parameters for the two rupture planes were derived by source Sllp 1m 1 0 10 inversion using empirical Green's functions. The seismic moment is o..-------============::::===---m:l J 40.6.. 1 40 .8 ~

36. 0 partitioned between the faults.

Figure 5.1.5-2: Map showing the shp d5tributlon and the vertical offset associated to the 2-011 ApriJ 11 Fukushima-Hamadori inland earthquake (Figure from Shiba and Noguchi, 2012).

Kappa for the AZ Sites Normal o ...-----~

Strike-Slip e.---r-~~~

Unknown s-r-~=-r~~

7 1 -t--=~.,;-t-t--~~

e s ~~~

Cl

• Previous. region-specific estimates. '! 11-+----'--'.;+t-0-~~

i*

11 4 ~~~~~

of kappa for rock sites in AZ not 3 3 2 -+----4--1---~

available. +----___._~,___~ I  ; I~ , -----~-..;-~

10 100 1000 0.1 I 10 100 1000 0.1 1 10 100 1000

• Data used for kappa consists of 12 AJB tkrr) A.I! (tun) 100000-.---.__,_...,..,.~

fUI (lcmJ

• SS earthquakes with hypocenter *

• SS

~10000 - NMI.. ~,~I"!"!-"!

locations in AZ.

NI.IL UN J 1000 - .______.1- .........-- 1

• Hypocenter locations were i 100 ---£---t-'t-'~t--;*-t-;o4.H1;t"

£ 10 -- -- -*

provided by Jeri Young of the AZ Earthquake Information Center I

.. 1000-.----.-----........

Perico (NIC}

10 20 *-------i OI I Period (MCI 10 20 (2013, personal communication).

~

I LU 100

• Events range in magnitude 1S f~

between 1.2 and 3.4, and were id 10 I~~~~

recorded by stations located at  :>

e I 10 100 1000 epicentral distances between 9 Roc:ordln!P Per urflqu~~o and 300 km.

11 out of the 12 events occurred 1=

'g 1000 T r . + : 1 = = = - r . + = = I 1 - -~~ 1 in 3 distinct clusters.

t z

100 "tt.~........~;t--"":::-1 I 10 ~--+--t-++-1'~

s 1 r-------------1 50 100 1000 3000 VS30 Cmls)

Figure 5.1 . 7-1: Summary o f the rustribuuon of data for the PEER-A4.mo..a. dataset.

Kappa for the AZ Sites

• The zero-distance kappa (K0 ) values for the recording sites estimated using 3 different methods:

- the acceleration spectrum approach (Anderson and Hough, 1984)

=

[KAs1 resulting in K0 33+14 msec;

- displacement spectrum approach (Biasi and Smith, 2001) [K 05,]

resulting in Ko.= 50 msec (set as upper bound);

- broadband approach (EPRI, 1993, and Silva et al., 1997) [K 88] resulting in K0 = 33 msec with a +one standard deviation range of 20-54 msec.

• All of the recordings are from broadband velocity instruments with a sampling rate of 40 samples/sec and a Nyquist frequency of 20 Hz. The high-frequency lir11it is about 16 Hz. The limited high-frequency bandwidth for the USArray data severely limits the resolving power for K.

• Estimates of site kappa values are sensitive to the assessment of site amplification. In Arizona Ground Motion. Database site amplification is included for all of the. kappa estimation methods.

Finite-Fault Simul<~tion Database for Me<:Jian

• A data base of ground motic,ns from finite-fault simulations was developed lJsing multiple simulation methods implemented on t~1e SCEC Broadband Platform (BBP) (Maechling et al., 201~5).

• The scenarios for simulations were selected to address four issues:

- magnitude and distance scaling of near-fault ground motions,

- rules. for estimating ground rnotions. from complex ruptures, .

- rules. for estimating ground rnotions. from splay ruptures,. and.

- magnitude scaling for HW effects for moderate magnitudes.

(MS to M6).

Subsets for PVNGS Median for Greater AZ Table 5.1-2: Data Sets Used f or the f valuatK>n of the M edian Ground Mooons Sources DATABASE OF DATASET SUBSET USE OF DATASET ORIGI N NGA-W2oc:-t.to PEER NGA- M ~5.0 Evaluation of the median West2 No HWsites* ground motion model (for VSJQ~ 250 m/s DCPP) tor base FW model NuJeqk ~ 3 Adjusted to V s.J1J =760 m/s NGA-W2"'1m> PEER NGA- M ~S .O Evaluation of the median West2 -70km S RXS 70 Ian for both SS ground mooon model (for and NML PVNGS)

VSJO ~ 250 m/ s NnJeqk~3 Adjusted to Vno=- 760 m/s EU~ Ref ere nce M ~.O Evaluation of the median database of R_s70km ground mooon model (for Seismic Ground for both SS and NML PVNGS)

Motion in VSJIJ ~ 250 m/s Europe N11£Jeqk ~ 3 (RESORCE) Adjust ed t o Vna= 760 m/s PEER-AZ,.,,,. PEER Arizona Earthquakes from NGA-West2 ln Estimation of the medliln path Regions 1 and 2&3 recorded at terms for Regions 1 and 2&3 m t fons in Arizona (for PVNGS)

NAE<J'eqk ~ 3 N11Ec/stat1on ~ 5 SI Moc-Mii> SCfC simulations SS: M S.5, M 6.0, M 6.6, and M 7 2 Evaluation of the median using the broad ground mooon model (for band platform REV M5.S, M 6.0, and M 6.S OCPP)

SIMHW SCfC simulations REV. MS.S, M 6.0, and M 65 Eval uation of the scaling of using the broad Dips* 10, 20, 30, 45, 60 the HW effect for magnitudes band platform zlOJI. 2.5, 75 , 12 ltm between M5 and M 6.5, and for ZTD' scaling (for DCPP)

• includes the f ollowing: 0 ~ R.~ -70 ltm for both SS and REV; 0 s R~ s 70 km & Ra~ 10 km for SS; and 0 s Rx ~ 70 km for SS, & Dip ~ 80 deg.

Normal Reverse NGA-W2Pv-MED dataset

!8+-..,._,_~'f-:,....,._,.~_.,~M

~ G-t-..-...........,.~_,,_....'""'""~..,......f.'-l PEE R NGA- M ~5_0 3

0 1 1 10 100 0.1 10 100 0.1 10 100 1000 West2 -70km::: Ft~ JO km for b oth SS RJB~ml RJ8(kll'I) fUI (km) and NML 100000 r-'

v'ii!J ~ 250 m/ s NREc/e(!k ~. 3 1000-Adjusted to Vsr716 0 m /cs 0 100 ~

l C'

2 10 1

I 10 20 0.1 I 10 20 Percd (9eC) Peflod (MC)

)1000 1000

& I Iii t

IM 100 0

'l!

f J*

i 10 3 10 1!!

u

~

i tO 100 1000 FloCO!dil'QO Por E11rlt)QuaM t=

llS 1000 z

f !00 t 10

~

d

!00 1000 3000 V$30 (llY$)

Figure 5.3.2-1: Summary of the data distnbutton of the NGA-W2PY.Mm dataset.

Nonna! Strike-Slip Reverse I * - -,__--,__,____,,....----. I 7 7

~6

~

6 EURPV-MED dataset

!6 5 5

.. " 4 Ref erence M~5J) 3 10 100 3

0.1 10 100 1000 3

01 10 100 100 dat.aba ~e of 1Rre*~70 km RJB(rcmJ RJll (km) FUJ(km)

Seismic Groun d fo r bottl SS and N.M L 100000 Motion in Vs;o 2:: 250 m/ s r'oooo [Eu rope NRKle ~k ~ 3 J

l5 1000 (RESORCE) Adj usted to 'V~*7160 m/ s 100 I

1: 10 14-......;...~..,._""4-__;_~..,._...._--f 1 1-----,! -----1---1 0I , 10 20 0 I I 10 20 Period (&ftll) PMod~

-~ ~:-J.,...~--;-...-.--r-r;~-.....-1 j 1000 ~.l;i;.-..I-.;........;.._;_:_:~ _ _ : _j

- EUR PV*MED

't>

tz 100.t~~~-

~ 10 ii

~

3 I00 I000 3000 VS30 1mJs) figure 5.3.2-2: Summary of the data distribution of the EUR~ dataset.

Normal Strike-Slip Rovorse 8 a.._.~-.-~~~--.~--.

!6-t-~--;-~~~--,+-t PEER-AZPATH dataset 3 6-1--~4-__.__.___,..4"-1 3 1----+------- 3 - --~ ........~_...~--~.......

10 100 1000 0.1 10 100 1000 01 10 100 1000 R.iB lkm) RJB (km) R..e (knl) 100000*n------------..-!"""l'"'!"!'"!'l'!'~"""I PEER Arizona Eanhquakes from NGA-West2 in Estimation of the median patt Regions 1 and 2&3 re-corded at tenns for Regions 1 and 2&3 stations in Arizona (for PVNGS)

NR.Ec/eqk ~ 3 110 NR.Ec/station ~ 5 z

l *--~-.-~........-1-~-,-~

01 I

.10. . . 20

--1 Ot Perkid ($eel Ptlt IOd ($0CJ i1000....--,-,.---...--,-,-=,-.-..,.-,-,...,..=-o

~ ..

~

!l> 100 0

~

I 1o;=~~:m-=~~rrt==r::R'.ii:it1

} 10 - --=....,__:..:.....:.....--:...-;..--<1

~

~

I 10 100 Recordings Pet Eart'1Quake d

- PEEA* AZ PATH 100 1000 VS30 (rnhl)

Figure 5.3.3-1: Summary of the data distribution of the PEE.R-AZ,~~ dataset.

Data Sets for the <l>SS and <l>SP-R Mo~~els

• There are 4 types of the single-station sigma.

(<PSS) models :

1. short-distance global n1odel based on RESORCE,

2. short-distance global n1odel based on NGA-West2 and Lin et al. (2011),

3. long-distance global m odel based on NGA-West2, 1

4. 2 short distance modells based on CA data in the NGA \6Jest2 database-~ DCNPP, not PVNGS

5. a subset of the AZ dataset was compiled to derive the magnitude-independent <PsP-R and path-adjustment models for PVNGS.

Data Sets for the <l>SS and <l>SP-R Models

1. short-distance global model based on RESORCE

• A subset of the. Akkar. et al. . (a subset of RESORCE) data set was selected to be used in developing the single-station sigma models for application to PVNGS for the Greater Arizona sources.

• M ~ S.0 and RJ B < SO km

• At least 3 recordings/earthquake and at least 3 recordings/site

• magnitude limit (e.g. all magnitudes are used to constrain the site term, but only a subset of the site corrected residuals with M ~San d RJB ~SO km are used to compute the cpSS values) .

• Subset consists of 223 recordings from 73 earthquakes (3S normal, 2S strike-slip, and 13 reverse events) recorded at 79 stations

• Ground-motion data at periods greater than 4.0 seconds are not available for the European dataset.

Normal Strike-Slip Reverse 8 8 7

~.

• - ~

h.

1~

I 7

~

I I I I "~ ""

- .. I i . ,..*l 6 - ~t 6

I I

(\

(\

I I 5

I

- 5 I

I 1

I

'V1 4

I I

I

-I- -h I

4 rt I I I

~ ,_,_ ,.._

3 I 3 r t' I ' "

0.1 1 10 100 0.1 10 100 1000 0.1 10 100 1000 RJB (km) RJB (km) RJB {km)

Figure 5.4.2-1: Magnitude-distance distribution of the EURPH1ss dataset. (Note: the m inimum of 3 recordings per earthquake and per site is applied to the full data set. This plot only shows the subset for distance less than 50 km and magnitudes great er t han 5.0)

Data Sets for the <l>SS and <l>SP-R Models

2. short-distance global model based on NGA-West2 and Lin et al.

• Combination of the NGA-West2 data and the Lin et al (2011) data

• Within-event residuals for Taiwan (Lin et al., 2011) were combined with the NGA-West2 residuals after removing residuals from common recordings from *raiwan earthquakes

• Idriss 2014 (ld14) residuals are not used for the SS <P evaluation because Idriss did not separate his residuals into between-event and within-event terms.

• A subset is developed for each <)f the four NGA-West2 GMPEs that separated the within-event and between-event residuals.

• M ~ S.O and. RJB. <SO km

• At least 3 recordings/earthquake and at least 3 recordings/site

• magnitude limit (e.g. all magnitudes are used to constrain the site term, but only a subset of the site corrected residuals with M ~ S and RJB ~SO km are used to cornpute the <PSS. values).

Normal Strike-Slip RevetSe Data Sets for the <l>SS and <l>SP-R Models

• 3 0 I 10 100 0. I 10 100 1000

,~---------.;i...;.~

01 10 100 tOOO

2. short-distance global model based A.II I""' R.81.,...) RJ8 (1cm) figure SA.1-1: MagnltudHllstance distribution of the GLOB~. dataset.

NoonaJ Stnke-Slip Reverse on NGA-West2 and Lin et al. 8 7

!e Table 5.4.1-1: Number of recordines and eantiquakes in the "°bal dataset (M ~ 5, R < 501<m) for four o f J,

the NGA*West2 models for the short-distance ;xr .

• 3 l 3...-..--...........-----~

01 10 100 1000 0. 1 10 100 1000 0, 10 100 1000 ASK14 8SSA14 C814 CY14 RJ8 foanl RAJ t""') RJ!I (1cml Reeion Nb Recs NbEqks Nb Recs Nb Eqk.s Nb Recs Nb Eqks Nb Recs Nb Eqk.s Figure 5.4.1-2; MainitudHlistance distnbUtJOn of the GlOBAi...-.-.u.. dataset.

CA 672 54 630 48 342 38 349 41 TalWcln 846 28 846 28 846 28 846 28 Normal Reverse Japan 65 3 65 3 0 0 63 3 * *

• Italy 69 15 62 11 6 4 0 0 7 7 Ouna 10 2 102 26 17 4 0 0 Total 1,662 102 1,705 116 1,211 74 1.258 72 Je

• t ta 5

• * ~

3 , ,

01 10 100 1000 0.1 10 100 *000 0 I 10 1CIO 1000 lllB ...... R.8 (*"') RJ8 (llnll rifUre SA.1-3: Maznitude-dlstance distnbutJOn of the GLOB~ dataseL CA dataset comprises about 30% to 40% of the recordings and 40 to 60% of the earthquakes 7 Normal I

Str1ke-S1lp Reverse

I

• in the global dataset, while the Taiwanese data  :.. I I represent 50% to 70% of the number of I

' I

• . ~

3 ,_,________.._____.......,.

recordings and 25% to 40% of the number of 3 . . 10 100 3-r---.__--.................................

1000 0. I 10 100 1000 0I 10 100 ICOO 01 RJ8 OMll IUllli"') IUl()it!IJ earthquakes . FiJure 5.4.1-4: \1aenitude-distance distnbutJOn of the GLO~ datase-t.

Data Sets for the <l>SS and <l>SP-R Models

3. long-distance global model based on NGA-West2

• Taiwan and CB14 sets were excluded from the <PSS analysis, because dataset with magnitude greater than or equal to M 5.5 and distances of 200 to 400 km lacks Taiwanese data in that range of interest and CB14 used mixed-effects regression to derive t he anelastic attenuation term from data with RRUP > 80 km, but allowed the source ter ms to vary from those with a maximum RRUP distance of 80 km.

• :. 3 sets of NGA-West2 residuals were used to develop the <PSS model for PVNGS - Distant California and Mexico sources.

• Global dataset in this magnitude and distance range of interest (M ;::: 5.5, distance 200-400 km) consists of 264-415 recordings from 4 to 23 earthquakes (mostly from Japan).

Table 5.4.4-1: 'N umber of recordings and earthqu akes in t h e global dataset {M ~ 55, R = 200 to 400km) for three of the NGA*West2 models ASK14 BSSA14 CY14 Region Nb Recs Nb Eqlts Nb Recs Nb Eqlts Nb Recs Nb Eqk s CA 133 1 209 4 160 2 Japan 131 3 157 4 129 4 China 0 0 49 15 0 0 Total 264 4 415 23 289 6 ASK14 BSSA14 CY14 8 B 8 75 7

75 7

75 7

I L- ()()(>

• l,. 6.5 . *i.... 6.5 0

"i.. 6 .S - *""-.x;

~ _J ~ ~

6 6 ~ ~ 6 -- ....

55 SS s.s 5 s 1 5 200 250 300 350 400 200 250 300 350 400 200 l50 300 350 Rtup (km) R!up (km) R:rup (km)

CA Jap4n < CA China Japan ) CA lap;in

Data Sets for the <l>SS and <l>SP-R Models

5. a subset of the AZ dataset was c:ompiled to. derive the magnitude-independent <l>SP-R and path-adjustment models for PVNGS

• Dataset consists of earthquakes in CA and 15 1 -

Mexico that have been 6.s - u recorded at the 9 stations ~ 6 .

in the vicinity of PVNGS. ii 5.5 - u

• 49 records from 11 s- ._ . . il C~ o ~)

earthquakes with rupture 4.5 l 66 ~ b.

distances that range from I '¢ ~

'¢ 8 V¢(.}

<ll. o

• 200 to 500 km are used to 0 I

100 I

200 300 400 500 600 compute cpSP-R for 3 Rrup (kmt regions: Region 1 (4 Region l . Region 2 Region ~

earthquakes), Region 2 (3 Figure 5.45-1: Magnitude-distance distribution of the PEER-AZPAJK dataset.

earthquakes), and Region 3 (3 earthquakes)

Proponent Models for Median Ground Motions Table S.5.1-1: Extrun& GMPEs Considered for the Development of Median GrounO-Mouon Models (contmues on the following paeeJ Candidate few Candidate few Candidate for Candidate for Candidate for Candidate for GMPE Comments PVNGS Greater PVNGS Distant GMPE Comments PVNGS Grea ter PVNGS Distant OCPP DCPP Arizona Sources California Sources Aritona Sourus California Sources Abrahamson et Update of Yes Yes Yes GtMzer (2014) NGA We.st 1 database No, not published No, nonnal fault No, not published al (2014) Abrahamson and Sliva plus 2004 Partrfield in a peer- not speoflcally in a peer-(2008) and 200S San Simeon revteWed Journal studied reviewed 1oumal Reeional for Turtey No, supers~ed No, superseded No, superseded Hermkes et al. RESORCE No, experimental No, expenmental No, expenmental Akkarand (2013) Expenrnental Model Cainan (2010) by pan- by pan- by pan-Idriss (2014) Update of ldnss Yes, not used for No, normal fault Yes Europe/Middle Europe/Middl e Europe/Middle (20081 R- < 3krn not speofically East ACR GMPEs East ACR GMPEs East ACR GMPEs studied plus non Kanno et al Used only depth for No, no clear No, not relevant No, no clear Cahfomla/Wester (2006) separation of event separation of ACR to tectonics sepamion of ACR n Arizona type from SZ interface from SZ interface attenuation earthquakes earthquakes, plus Akkuetal Update of Akkar and Yes Yes No, non non (2013a. 2014) Bommer (2010) Cahfom1a/Wester Cllhfomta/Wester nAnzona n Arizona anenuat1on anenuation B1nd1et al. Update of 81ndi et at No, extrapolation Yes, M > 7 not a No, e.nnipolauon McVerry et al R~nal for New No, speafic to No, specrfK to No, speafK to (2014a. 2014b) (2011) above M7 st&ntficant above M7 (2006) Zealand New Zealand, New Zealand, New Zealand, problematic a1 contributor to problematic at superseded by superseded by su~edby some pertods hazard some penods Clobal models &IObal models i)obal models Boore et al. Update of Boore and Yes Yes Yes that use recent that use recent that use recent Atkmson (2008) New Zealand NewZealllnd NewZeaJand (201 4) earthquake data earthquake data earthquake data Bora etal. RESORCE No, upenmental No, expenmental No, expenmenral (2013) E.menmental Model Pankow and Update of Spudictl et No. not relevant No. superseded No, not relevant Pechmann al (1999) to tectonKS by more recent to tectonacs Bradley (2013) Modification of ChlOu No, re&ional No, rqional No, reitonal (2004) models (e.e. NGA-et al. (2010) for New adjustment to adJUstment to adjustment to West2)

Zealand other model other model other model Zhao and Lu Proposed chanee 1n Yes No, not relevant No, non tnduded m study included tn study &neluded in study (2011) maenitude saline to tectonics Clllrfomia/ Wester Campbeft and Update of Campbell Yes Yes Yes above-M7.l nArizona Bozor'1)ta and Bozorcnia (2008) anenuation (2014) Zhao et al. Mostly .lapan data, Yes No, not relevant No, non Chiou alld Update of Chiou and Yes Yes Yes (2006) AGR and SZ wittt to tectonics califomia/Wester Youngs (2014) Youngs (2008) and separate factors nAmona Chiou et al (20101 attenuation Derras et al. RESORCE No, experimental No, expenmental No. experimental (2013) Expenmental Model facool1 et al. Global data, primanly No, single Linear No, smile lmear No, sin&le linear (2010) Japan magnitude scaJm.g mainitude scalane magnitude scaling over enure range over ent1re ranee over entire range

Selection of Candidate Models Subset of candidate models was selected based on the following seven criteria:

1. More recent published GMPEs by the same development team were selected over older GMPEs on the basis that the newer models would have benefited from more data and refinements to the approach. In the case of the modified magnitude scal ing suggested by Zhao and Lu (2011), the Zhao et al. {2006) GMPE is selected along with a modified form based on Zhao and Lu {2011). The basis is that Zhao and Lu have not developed a full GMPE to replace Zhao et al. {2006).

2. Models that represent an adjustment of another model to fit data from a specific region which is not CA or western AZ were not selected (e.g. Bradley, 2013). The basis for rejecting these models is that they have been adjusted from one region to another and should not be adjusted back to the original region or to a third region.

3. Models that do not extrapolate well beyond the magnitude-distance range over which they were developed were not selected. For example, models that have only a single linear magnitude sca ling term were not selected, as eva luat ions by many investigators of data sets containing a large range in magnitude have shown that a single linear magnitude sca ling term does not capture the magnitude sca ling over the range of magnitudes from MS to M8.

4. Models that do not clearly separate sha llow crustal earthquakes from those occurring as a part of subduction were not selected. The basis for rejecting these models is that the magnitude and distance sca ling from subduction zone earthquakes is different than from crustal earthquakes.

5. Models developed as research. tools.were not selected ..The basis for rejecting these. models. is that they have not developed to the point where they could be used for engineering application.

6. Models developed for a relatively small specific region different from the ones of interest (e.g. Italy) were not selected. The basis for rejecting these models is that the data from a single region may be too limited to capture scal ing for the full range of magnitude aind distance of interest and the specifics of the regional behavior may be different from CA and western AZ.

7. Models that have not been peer reviewed or vetted. by the larger scientific community were not selected (Graizer 2014).

Host Kappa Values for Selected Candidate Models at VS30

• For PVNGS, the planned site response analysis accounts for differences in the kappa implied for the candidate GMPEs and the kappa for rock sites in central AZ. The kappa implied by the spectral shape of the GMPEs is called the "Host" kappa .

• Host kappa values were derived for normal-faulting scenarios with a dip angle of 50 degrees, for M 5.0, 6.0, and 7.0 and Rx distances of 5, 10, and 20 km on the footwall.

• Under the direction of the Tl Team, Dr. Al-Atik estimated the host kappa values for the 7 candidate GMPEs selected for PVNGS using the IRVT approach :

Table 5.5.3-1: Host kappa va lues for the seven candidate GMPEs for PVNGS Greater Arizona sources for a reference V530 of 760 m/sec.

GMPE ASK14 BSSA14 CB14 CY14 ASB14 Bi14 ZH06 Host Kappa {sec) 0.045 0.038 0.037 0.041 0.042 0.045 0.042 ASK14 =Abrahamson et al. {2014), BSSA14 = Boore et al. {2014),

CB14 =Campbell and Bozorgnia {2014), CY14 =Chiou and Youngs {2014), Bi14 = Bindi et al.

{2014a, 2014b), ASB14 = Akkar et al. {2014a, 2014b), and ZH06 =Zhao et al. {2006).

Single-Station Sigma Approach -

Rodriguez-Marek et al., 2014

• PVNGS used the RodrigL1ez-Marek et al., 2014 single-station sigma approach:

- Partially non-ergodic app roach 1

- Removes the systematic site response effects from the traditional ergodic withirt-event standard deviation

- Avoids double counting c>f the epistemic uncertainty of the site response that ca1n occur if the traditional ergodic sigma is used, and the site response also addresses the epistemic 1uncertainty.

Single-Station Sigma Approach. - Rodriguez-Marek et al., 2014 Swttzeriand Rodriguez-Marek et al., :1 2014 approach compiled f s 4>

~.

8 .a

~2' s

~.

ground motion data with :L0 100 200 300 3

2 0 100 200 300 multiple recording per 0 1$lance (km) Dist nee(~)

I T~tw.n Tu~ey 81 site from 5 regions. . ¥:[ *1 .-

J T

is ~ ...

~ 5 c 5 .;

i'

~ 4 i 41 I

3 3 Distribution of the data in 2 0 100 200 300 J 2-0 100 200 300 DIStanc:e (km) Distance Ckm)

Japan All Sites terms of magnitude and 8 7 - ....

• distance for each region q;;

-g

~5 8 ' Cll "O

3 c;

Cit e

5 i, Q

t

• 3 3 100 200 300 100 200 300 01s1*nc:e (kml 0 tance CkmJ Figure 5.6~2-1 : Distribution of data for smgl~station sigma as compiled by Rodriguez-Marek et at.

(2013)_ (Figure from Rodriguez*Marek et al., 2013.)

Single-Station Sigma Approach - Rodriguez-Marek et al., 2014

• Average cpSS values for M > 4.5 and R < 200 km for each region.

• The lower plot shows the traditional ergodic cp values and the upper plot shows the partially non-ergodic cpSS. values.

• cpSS values are mainly between 0.4 and 0.5 In units and are much more consistent across regions than the cp values.

08,---------------,

07 08

...*05l b <l r* ~ *

<J 0*1 + + + ~

03 j 02 PG~ 01 020 3 0 5 10 30 Penod (s) *

+

08 c '"'*'t 07*

06 Q

6

+.*

• x 0

7 for 5 Hz. This feature of the Bi14 model is not seen for all spectral frequencies, but for simplicity of application, the Tl Team decided not to use the Bi14 model for M > 7 for all spectral frequencies.

Figure 6.2.2-1: Magnitude. scaling of the candidate. GMPEs at 5.0 Hz for strike-slip event with RX distance 5 km.

1 .50 ~--.-----.-----.-----..----.------..-----.------..---.------.------.-----.--..-----.------.----.---~----r----.----.-----.-----.----.----....----.---..----.-----.-----.--:.

T = 0.2 1.00 SS Rx=5 0.70 r---1 Cl

\.......,J

<( 0.50 en a..

• ASK14

• ASB14 0.30

• Bi14

• BSSA14

• CB14 0.20 f

• CY14 5.0 5.5 6.0 6.5 7.0 7.5 8.0 M

PV~JGS

• The regional Q (quality_factor) structure can have a significant impact on attenuation of the seismic waves with distance. Several of the proponent GMPE models (Section 5.5) have used data out to distances of 400 km to constrain the distance scaling, but these GMPEs did not include ground-motiorl data from sites in Arizona

• In general, the Q values av eraged over the paths from 1

California to Arizona indicate that events from central California have higher Q, events from Baja California have lower Q, and events from southern California and the Transverse Ranges hav1e intermediate Q.

• Although there are differences in the. Q between California and Arizona, these differences do not lead to a significant discrepancy in the average distance attenuation at distance of 200-400 km in central Arizona as compared to California. The differences in Q between California and Arizona \Nould have a larger effect for larger distances (>. 400 km), so the conclusion by Kishida et al. {2014) that the average attenuation is similar for southern California and central Arizona only applies to distances less than 400 km.

• Based on the evaluation of the candidate GMPEs for application to earthquakes in California and ME~xico recorded in central Arizona given in Kishida et al. {2014), the Tl Team judged that the NGA-West2 GMPEs are suitable for estimating path terms for the paths from California and Mexico to central Arizona.

Summary of Kishida et al. 2014

• This report summarizes the products and results of a study on the collection, processing, and analysis of earthquake ground-motions recorded in Arizona at several recording stations within 200 km from the Palo Verde Nuclear Generating Station in central Arizona.

• Additionally, "kappa" a measure of energy dissipation in the top 1 to 2 km of the crust, was estimated by three methodologies. The average KO (kappa at. zero-kilom,eter. distance) was. estimated from all. sites as 0.033 sec.

• Response spectra of the recorded ground motions in Arizona were compared with those predicted by the NGA-West2 ground motion prediction equations at large distances in Arizona. The comparison showed that overall the recorde~d 5% damped response spectral ordinates were over predicted by the NGA-West2 models by a range of 0-0.35 natural log units for events occurring in Central California, and by a range of 0.2-0.7 natural log units for events occurring in Southern California and the Gulf of California.

TA Data Limitations

• The Nyquist frequency for all TA recordings is 20 Hz, because the TA has <3 low sampling rate of 40 Hz. An anti-alias filter w las applied to the TA data at about 80% of the Nyc1uist frequency with a corner frequency near 1.6 Hz.

• These observations are interpreted to indicate that the noise from microseisms and other sources are dominant at frequencies less than about 0.5 Hz.

• After differentiation vel,Jcity records high-frequency limit becomes even lower than 8 Hz.

• SAs based on TA data <3re most likely severally affected. by low sampling rate. of 40 samples/sec and necessity to. differentiate velocity recordings.

• Examples of similar effects are demonstrated in Graizer 2012 (lSWCEE) paper.

Observations

• Q-values in Arizona arE~ inconsistent with conclusion about samE~ attenuation as predicted by the NGA-West2 GMPEs.

• Kishida 2014 conclusic~n (The comparison showed that overall the recorded 5% damped response spectral ordir1ates were over predicted by the NGA-West2 models) can not be justified by TA recordings .

Difrancesco, Nicholas

Subject:

FW: DEDO/ET Weekly Fukushima Status Meeting Location: 0 -13D20 Start: Tue 05/05/2015 2:00 PM End: Tue 05/05/2015 3:00 PM Show Time As: Tentative Recurrence: Weekly Recurrence Pattern: every Tuesday from 2:00 PM to 3:00 PM M eeting Status: Not yet responded Organizer: NRR_JLD Resource

Wayne, DEDO/ET Weekly Status Brief.

Plan to suggest bullets similar to these for my management:

WUS Seismic Screenino Results Letter l* Group 1 - Columbia, u1ao10 t_;anyon; Group 3 - Palo Verde

• June 2017 Seismic Probabilistic Risk Assessment Consistent with Industry Endorsed Timelines [i.e. no relien

• Interim Evaluations for Diablo Canyon are adequate without completion of the Expedited Approach [i.e.

LTSP provides safety basis]

• No immediate safety issues - interim evaluations support time to complete SPRAs

• Communication plan developed to support public issuance on May 12. 2015 Thanks ,

Nick

---

Original Appointment-----

From: NRR_JLD Resource Sent: Tuesday, October 14, 2014 2:48 PM To:. NRR_JLD Resource;.Johnson, Michael;. Dean, Bill;. Uhle,. Jennifer; NRR_ET_Activity Resource; Davis,. Jack; Franovich, Mike; Bowen,. Jeremy;. Proffitt, Andrew; Inverso,.Tara;. Evans, Michele Cc: McHale, John; NRR_LT_Calendar Resource; Mohseni, Aby; Kokajko, Lawrence; DprNrrCal Resource; Oesterle, Eric

Subject:

Copy: DEDO/ET Weekly Fukushima Status Meeting When: Tuesday, May 05, 2015 2:00 PM-3:00 PM (UTC-05:00) Eastern Time (US & Canada).

Where: 0 -13D20 10/14/2014 - Requested by JLD Management; Scheduled by Esther Cho (2239) 1

Difrancesco, Nicholas From: DiFrancesco, Nicholas Sent: Tuesday, May 05, 2015 11:32 AM To: Dapas, Marc; Kennedy, Kriss; Uselding, Lara;. Walker, Wayne; Miller, Geoffrey; Dricks, Victor Cc: Shams, Mohamed; Alexander, Ryan

Subject:

POP for. DEDO/ NRR/R-IV JLD Status Brief (2pm Eastern)

Attachments: POP - JLD Status (05.05.15).docx

Folks, Below and attached is the POP for the JLD status briefing including the WUS Seismic Screening anp, 6

Prioritization letter. The briefing is scheduled for 2pm Eastern (Bridgeline: 888-455-0567; Passcodef,_(b-)(- )_ __ ,

Please let me know if there are any questions or concerns .

Regards, Nick OiFrancesco Sr. PM - Seismic Reevaluations (301 ) 415-1115 PURPOSE Update NRR ET on status of JLD activities EXPECTED OUTCOMES .

• wus -

' t5' Screening & prioritization letter - targeting issuance 05/12/15 o All 3 plants screen in for sPRA. no immediate safety issues o Columbia & Diablo Canyon - Group 1

• sPRA due 06/30/17

• Diablo Canyon - Separate letter on ESEP* SP provides safety basis 1

• Public Meetings o Diablo Canyon (04/28/15) o Columbia 06 04/15) o Palo Verde - Group 3

• sPRA due 12/31 /20

• Public Meetina 06/09/15 Non Responsive 2

Difrancesco, Nicholas From: DiFrancesco, Nicholas Sent: Wednesday, May 13, 2015 2:56 PM To: Shams, Mohamed; Jackson, Diane; Wyman, Stephen; Vega, Frankie Cc: Devlin-Gill, Stephanie; Munson, Clifford; Heeszel, David; Seber, Dogan; Stieve, Alice

Subject:

Summary of R2.l Licensing Calls (Related to WUS Hazard Reviews, CEUS Hazard and ESEP Reviews)

Folks,.

Completed a number of licensing calls today for awareness or follow-up.

Diablo Canyon:

Licensee expressed interest in timing of ESEP decision and feedback on licensee SFP Evaluation Approach NRC Suggested Resolution: Target 2 months for ESEP letter- SFP approach should be considered within the guidance development over the summer (JLD and DSEA)

Licensee Action: Timing of information request response early next week (will be PG&E formal letter or public website)

Columbia :

Licensee interested in Public Meeting NRC Request: For ground motion questions only general topics earlier than May 21, if possible to support consultant schedule (Dogan/DiFrancesco)

NRC Request: Request for 2nd bridgeline for licensee staff (Frankie)

Palo Verde:

Licensee interested in Public Meeting and conditional screen-out process NRC Request: Information to support Palo Verde Public Meeting by May 26.

Non Responsive Thanks ,

Nick 1

Senior Project Manager - Seismic Reevaluation Activities U.S. Nuclear Regulatory Commission Office of Nuclear Reactor Regulation Japan Lesson Learned Project Division.

nicholas.difrancesco@nrc.gov I Tel: (301) 415-1115 2

Hill, Brittain From:Hill, Brittain Sent:22 Apr 20 15 11 :49:34 -0400 To:Munson, Clifford ;Flanders, Scott;Kock, Andrea;Jackson, Diane

Subject:

FYL Diablo seismic hazards plot For your reference, a rather messy summary plot for Diablo that shows the

- new GMRS (red),

- Double Design and Hosgri design-basis earthquakes (green),

- 84'h percentile ground motions from 1988 long-term seismic program (long dash),

- 1.35x margins curve derived from the LTSP (short dash) ,

- 201 1 Shoreline report, 84"' percentile ground motion for San Luis Bay fault, ergodic approach (highest deterministic accelerations) (blue dashed)

- 201 1 Shoreline report, 84"' percentile ground motion for San Luis Bay fault, single station approach (highest deterministic accelerations) (Blue)

- 2014 report, 84'h percentile ground motion, linked Shoreline+Hosgri+San Simeon faults, at turbine building (highest deterministic. accelerations) (Purple)

'~efutes arguments that PG&E has been lowering seismic hazards by using "new and improved methods

- facts show calculated hazard has increased.

Shows that implementing probabilistic approach has resulted in higher hazards than 2011-2014 deterministic approaches.

Demonstrates Hosgri design basis was robust and that well-established (i.e., SSER Rev 34) LTSP margins provide assurance plant is safe to operate.

3.0 I Diablo Canyon , ,-- ......

2.5

/ \

- GMRS \

\

- - DOE \\

- HE \

-E1 c

2.0

- - LTSP84%ite

---

LTSP x 1.35

\

\

\

\

\

0 \

+;;

~ 1.5 Q)

- - 11 SLB Ergo ' \

(i) - 11 SLBSStn ' \

\

8 - 14 Sh+HE+SS \

\

< '--i--------

1.0 0.5 0.0 -~-------.,...---------------------

0.1 1.0 10.0 100.0 Frequency {Hz)

Brittain E. Hill, Ph.D.

Sr. Technical Advisor US Nuclear Regulatory Commission MS T7-F03, NRO/DSEA Washington, DC 20555-0001 Ph+1 301 415-6588* Fax+1 (301)415*5399; Mobil (b)( 6l mail: Brittain.Hill@nrc.gov

Columbia - Hazard Curves by Seismic Source Figure 10.44 SSHAC Report

Seismic Sources Characterized in the SSC Model -

SSHAC Figure 8.1

• Zone B

• Zone C

• Zone D

• YFTB Background

• YFTB Faults 12ZW 120W E[M]

1.85 to 2 0 5 to 6 -

• Fault Sources 0

0 2 to 3 3 to 4 0 6 to 7 c:J Source Zones 0 4 to 5 0 >= 7 2

Seismic Sources Characterized in the SSC Model -

SSHAC Figure 8.1

• Zone B

• Zone C

• Zone D

• YFTB Background

• YFTB Faults 3

Ta ble 8.5. Fault soW'Ce~ including fault M!gments.

Fault So1irce Abbreviation 20 Fault Sources Ahtamun Ridtte Arlmgtou Cleman Mountain Columbia HiUs AR AF CM CH Columbia Hills-Centrnl-East CH-C-E Columbia Hills-East CH*E

" ~l")(r'Q'\¥ QI' ' tr.)QO'W Colm11b111 Hills-West CH-\\"

FH*E SFZ Columbia Hills-Ce11m1I CH-C Columbia Hills-Ce111ral-W~1 CH-C-W

~--

+~

Freuchmilll Hills FH Hom Rapids Fault HR

>>f""

Table 8.5. (contd)

~.

Fault Source Abbreviarion Horse Hea\en Hills HHH Horse Heaven Hills-Central HHH*C Horse H~\'eu Hill>-Cenn*aJ-East HHH-C*E CCf>' 0 100 200 Kilometers Horse Heaven Hills-Central-West HHH-C-W Horse Heaven Hills-\Vei;t HHH-W Laurel 1.F Lw13 Bune LB

VlanaMa>h Ridge MR Mauastasb Ridge-Cenrral MR*C Mauastasb Rid{te-East MR-E Manastash Ridge-West MR-W Mau pm MF Ranle$ of tbe Ranlesnake-Wallula Ahgmueru RA\\"

Ranlesn:ike Rills RH Rau lesnake !\fountain RM Snd<Ue Mountains S~I Saddle Mow1tn1M-E.'lst SM-E Saddle Mo1unains-Wes1 SM-W Sean I~ Fault SFZ Selah Bune SB Topptubh Ridge TR TOJlpemsh Ridge-Ehl TR-E Toppeni>IJ Ridge-West TR-W Umtauum Ridge UR tJnu:mmn Ri~e..Southeast Anticline UR*SA Umtauuru Ri~e-Oable Mountain UR-GM Umtamun Ridge.('eorrnJ UR-C Umtamuu Ridire-East UR-E Um1a111u11 Ridge-W~r UR-W Figure 8.43. Fault sources and fau lt segments. Teeth are shown on the hanging wall of the faults and Wallula Fault WF Yakima Ridge YR squares define the segment boundaries. Acronyms for fault sources are given in Table 8.5.

Yakwia Ridge-Ea>t YR*E The insert shows the location of the Seattle fault relative to the Hanford Site and to other Yakima Ridge-West 4.'R-W fa ult sources. Yakinia Ridge-Southeast YR-SE

Subduction Zone Sources - CSZ and JDF (used sources from BC Hydro Report - modified?)

CSZ - Cascadia Interface JDF - Cascadia lntraslab Source Downdip seismic limit ,. *.

Near c oa tal cities closest approach Plate Interface in Black red labeled ""-"----=z~one Seismogenic zone = ~~!!:...~----~ Arrows.=

Intra Slab Cascadia Interface Source (Licensee's (Licensee's CSZ Source)

JDF Source)

Figur~ 8.8. Diagrammatic depiction of lhe seismic sources and other elements related to the CSZ (Hyndman 2013). The plate interface source is shown m red and labeled "'sei.smogemc zone." The mtraslab source is shown by the do\\-n-gomg blatl arrows withm the Juan de Fuca plate. Episodic tremor md slip (ETS) earthquakes are labeled ~-ETS zone." The landward extent of the plate mterface. which is a ke)' aspect of the SSC model. is labeled "clowndip seisnnc limit" (Hyndman 2013) The location of Hanford Site lies east (to the right) of the Cas.cade volcanic arc.

5

SiteC mean TOT

--* -*- CSZ

..... ..lf.."*o* JDF

--0-*- ZONES

....-.9 ... . - ZON EC

--*&- YFTB

- ..-& ..... ZONED AF ARH CH CM FH HHH HR LB LF MF MR RAW RM SB SFZ SM TR UR YR WF 1o*8 L-'----1.__._._..............J_..1-..1..._.__._................_,,L..........L-..L...A.....L..u..i..L:--'"-.1.-._........_i~~~!.J&.w..~.......___.__._~ o 10"6 10*5 10-4 10*3 10*2 10*1 10 6 T1 Osec Spectral Acceleration [g] f = 1/period = 1/10 sec = 0.1 Hz

% Source Contribution - 0.1 Hz

% Contribution 0.1 Hz at lE-04

% Contribution 0.1. Hz at lE-05 UR RM UR RAW 1% RM SM HHH HHH HR No Data No Data 1%

11% 11%

CM or YR 1 5%

3%

3%

Note: Sources that contribute less than 1% and w ith no data are not labeled 7 Hand digitized, uncertainty= +/-10%

SiteC

- - meanTOT

• --* -- CSZ

• -***-*.. JDF

- &-- ZONEB

.....- &-..- ZONEC

- ..**0 - YFTB

~-a--.- ZONED

- - AF

- - ARH CH

- - CM

- - FH

= = HHH

- - HR

- - LB

- - LF MF

- - MR

- - RAW RM SB

- - SFZ

- - SM

- - TR UR

--YR

- - WF 1o*s L.....__.___.__._.a....LL.&l_...__._..................i...u.L___.__.__...............~~......._,,...............uu;up.~~~ o ....Jt.al-..__._.~ 1 10-5 10-4 10*3 10*2 10*1 10 10 8

T1 .Osec Spectral Acceleration [g] f = 1/ period =1/ 1 sec = 1 Hz

% Source Contribution - 1 Hz

% Contribution 1 Hz at lE-04  % Contribution 1 Hz at lE-05 No Data\

SM 3% \

1%

No Data UR SM 8% 6%

UR RAW 2% LF. RM.

6%

5%

HHH RAW 2%

HR 4%

CM or YR 1 6%

ARH 2%

Note ~ Sources that contribu te. less than 1% and with. no data. are not labeled. 9 Hand digitized,. uncertainty= +/-10%

SiteC

- - meanTOT

- -+**-- CSZ

................. J OF 10-1

-*-&*- .. ZONES

....- 9 ........ ZONEC Cl) -*-&-..... YFTB 0

c "

0

-a . .... ZONED cu - - AF "t:J - - ARH Cl)

Cl) - - CH 0 - - CM

>< 10-3 w - - FH

....0 HHH

- - HR

~ 10 4

- - LB c - - LF Cl)

J - - MF C"' - - MR Cl> 1o-5 - - RAW

~

LL RM cu SB

J - - SFZ c 10-6 c - - SM

<( - - TR UR 10-7 --YR

- - WF 10~.____.__..___._.. . . . . . . ...___._..._._. . . . . . . ......____.__.__._.....................____...~.....__..........................._...._.......................__............_............_.............,

4 3 2 1 1 2 10 10- 10- 10- 10° 10 10 T0.1 sec Spectral Acceleration [g] 10 f = 1/period = 1/0.1sec = 10 Hz

% Source Contribution - 10 Hz

% Contribution 10 Hz at lE-04  % Contribution 10 Hz at lE-05 No Data -4 RAW UR No Data I RM 3%\ 3%

1%. I 4%

~

CM or YR 1 4%

HHH 3 Note : Sources that contribute less than 1% and with no data are not labeled 11 Hand digitized, uncertainty= +/-10%

SiteC

- - meanTOT

• - ---*- CSZ

--***-- JDF

- &-- ZONEB

.......e-*- ZONEC

- &-- YFTB

• • -*0-- ZONED AF

- - ARH CH

- - CM

- - FH

- - HHH

- - HR

- - LB

- - LF

- - MF

- - MR

- - RAW RM SB

- - SFZ

- - SM

- - TR UR

--YR

-~ WF 10*B .__......___._._._..............._....__.__._._................_....__._........................._~_.........._.......U-1.1.i.~........................~...__.__.__._~

10-4 10-3 10*2 10* 1 10° Peak Ground Acceleration (g] 12 f = 100 Hz

% Source Contribution - PGA

% Contribution PGA at lE-04  % Contribution PGA at lE-05 No Data No Data 12%

17%

SM UR 5%

UR 5%

SM 2% J RAW 1%

RM 7%

RAW CM.or. YR.1 4%

2%

CM or YR 1 ARH 3% 1%

Note: Sources that contribute less than 1% and with no data are not labeled 13 Hand digitized, uncertainty= +/-10%

Results, Questions, and Notes

• At 0.1 Hz CSZ, but JDF subduction zone source doesn't contribute to hazard at lE-04 and lE-05 hazard levels

• At 1, 10, and 100 Hz the YFTB background source dominates the hazard, followed by fault sources in the YFTB (CSZ contributes a 1 and 10 Hz).

• Top 5 sources at lE-04 and lE-05 for 0.1, 1, 10, and 100 Hz are:

- RM - Rattlesnake Mountain

- CM or YR 1- Cleman Mountain or Yakima Ridge

- UR-. Umtanum. Ridge

- HHH - Horse Heaven Hills

- HR - Horn Rapids Fault

- ARH - Ahtanum Ridge (AR)+. Rattlesnake Hills (RH)? 14

Results, Questions, and Notes (continued)

• Ask Licensee?

- Provide data not available for curves (outside plot area) - those sources less than lowest percent?

- Can't read color difference between CM and YR (black). Hazard for black curve (either CM or YR) matters at 10 Hz where it is 15% at lE-04 and 20% at lE-05 and at 0.1 Hz and lE-05 (second highest contributing source = 5%)

- Is ARH for the hazard curves AR + RH on the map? How do they decide when to split and combine faults?

- Double counting deformation YFTB? - fault sources+ YFTB background seismicity- do they interpret them to be connected?

• For Site B - there is an odd kink in TR curve for 10 Hz (O.lsec) between 0.1 and 1 Hz (Hazard Curves at 10 Hz add up to >100% for sources with data)?

Over estimated hazard? Why does. this. kink exist?~ not seen for site C

• Note x and y axis extents on hazard curve plots changes between plots

• Need to check - Do these results match with deaggregation plots?

15

Figure 8.43 . Fault sources and fault segments. Teeth are shown on the hanging wall of the faults and squares define the segment boundaries. Acronyms for fault sources are given in Table 8.5.

The inse1t shows the location of the Seattle fault relative to the Hanford Site and to other 16 fault sources.

5/19/2015 Columbia R2.1 Review - Questions for Public Meeting from Lisa SSHAC Report Section 8.4.3.4 - Structural Relief to Net Slip Conversions In SSHAC Report Section 8.4.3.4, the licensee indicated that "the dip angle of thrust or reverse faults beneath the YFB anticlines was derived using a simple model to relate the fault geometry to the plan dimensions of the folds." The licensee indicated that a range of alternative dips and seismogenic depths were. used to model. the deformation of surface topography from faulting .. The licensee then used the parameters from the range of models that best fit the observed topography above each fault and the measured surface relief to predict the net fault slip.

1. Limited information on structural relief to net slip conversion provided -

Staff acknowledges that variation and fault dip angle and depth of faulting will create variations in the surface topography above the fault and that with limited information the exact fault geometry could be unknown . . Please provide additional information on the. following

i. Provide the depth of faulting and dip value parameters used. to derive the slip rates with their models and the range of values used in alternative models to estimate the uncertainty ii. Show all observed topographic profiles in cross-section and their location in map view and how they compare to modeled topography predicted by model used for fault slip estimat ion iii. Provide a table with structural relief, displacement estimate, and slip rate iv. Show profiles with varying depth of faulting and dip angles and how the associated modeled topography compared to the observed topography

2. Listric or complex geometries not fully considered or discussed-

1. Observations show that some of the faults have a back thrust, why isn't this incorporated into the three geometries indicated by the licensee? In one of the models, the. licensee uses a steep fault that the. licensee indicates. gives a slight concavity to the resulting topographic profile .. Field. and geologic mapping in the region indicates faulting is more complex (i.e. presence of back thrusts and listric geometries). Are these geometries described in the geologic mapping sections of the SSHAC report? They are cited in published literature for the YFTB. These geometries don't appear to be incorporated into the modeling used to estimate the net slip.

2.. Please provide a more in depth discussion of why the. thin-skinned model. was not selected, do not have access to zacharanin report.

3. Previous research by Watters et al. of similar styles of faulting in basalt-like material indicated that using listric geometry over a simple fault could produce net slip rates 1.5

- 2 or more t imes greater. If they have evidence for listric geometry or backthrust for the faults in the YFTB, why aren't they used in their models, especially if it wou ld produce a higher net slip rate? I still need to read more in the geology sections to see what the licensee describes about the faulting from field and geophysical observations.

4. I need to look up the reference, but I know that there were geophysical survey's done in the YFTB that show complex fault geometries beneath the anticlines in the YFTB.

5/19/2015 Ideally, those geometries should be used in the net fault slip estimation rather than a simple flat dipping fault.

5. Average not maximum relief used to derive slip rates -

a. The licensee stated that the average, not maximum, structural relief along faults used to estimate fault slip (See Appendix E page 5.10). Is this valid? With uncertainty, why not use maximum measured relief? Is the maximum relief considered in the logic tree as a possibility for rupturing?

b. Rattlesnake Mountain unfaulted quaternary deposits= not maximum estimated slip rate? Or timing of slip stopped before quaternary?

Columbia R2.1 Review - Questions for Public Meeting from Lisa Seismic Source Characterization Structural Relief to Net Slip Conversions Please show additional details of modeling used to. convert structural. relief to net slip, including

1. Please show comparisons of the actual observed topographic profiles with topographic profiles predicted from elastic dislocation modeling used for slip calculations for each fault source in the YFTB

2. Please provide a table with measured structural relief from topographic profile, displacement estimate from. modeling, and slip rate for each fault source in the. YFTB

3. What range of depth of faulting and dip value parameters were used in the modeling? How were. these alternative models used to predict uncertainty?

4. Was a model considering a listric geometries or complex geometries (i.e. presence of a backthrust) considered?

5. Please explain why the average, rather than the maximum, structural relief along faults was used to estimate fault slip (See Appendix E page 5.10).

Thin- vs. Thick-skinned Tectonic Environments.

6. Please provide additional background information on the findings in Zachariasen et al. (2006) and supporting evidence. that led your selection of 100% Thick-skinned over Thin-skinned tectonics. If the models.were weighted equally in previously studies, what is the new evidence.

that establishes that weight should be given solely to the thick-skinned model?

7. If the regional tectonics is solely thick-skinned, can that type of tectonic environment readily explain t he nearly regu lar spacing of faults in the YFTB that could be explainable by faulting on a decollement via the thin-skinned model? (i.e. Schultz and Watters, 1995)

8. If you used the thin-skinned tectonics model in your structural relief to net slip calculations, how much would it change the predicted structural relief, thus net slip and resulting hazard curves at lE-04 and lE-05 at 1 and 10 Hz for the GMRS?

1.

Columbia - Source% Contribution - 0.1 Hz.

SlteC

- meanTOT

- + - CSZ JOF ZONEB ZONEC

- YFTB

-E ZONED

- AF

- - ARH CH

- - CM

- FH HHH

- - HR

- LB

-LF MF

- - MR

- - RAW RM

- - se

- - SF2

- - SM

- - TR UR

- - YR

- - WF 10*.___..__...._...___..___.,......____.___..............__~_,__...__,_,,~...............-....................

10* 10~ 10~ 10-3 10 2 10*1 10° T10sec Spectral Acceleration [g]

RJ6 Contribution 0.1 Hz at lE-05 UR 2%

HHH c~

ARH 1%

YFTB 3%

1

Columbia - Source % Contribution - 1 Hz SiteC

- meanTOT csz

- JDF ZONEB ZONEC YFTB ZONED

- AF

- - ARH CH

- - CM

- FH HHH

- - HR

- LB

--LF

- MF

- - MR

- RAW RM SB

- SFZ

- - SM

- - TR UR

- - YR

- WF 10~--~............-----...... --~--.___...__._._~...._.,.._.........................._..._....

10"5 10"' 10*3 10 2 10*1 10° 10 1

T1 .Osec Spectral Acceleration [g)

No % Contribution 1 Hz at lE-04 UR  % Contribution 1 Hz at lE-05 Data~ 6 8%

UR

%~ata

% \

SM 6% RAW 2% 5%

RAW 2%

HR 9%

6% 2%

2

Columbia - Source % Contribution - 10 Hz SiteC 1a6 ----~----~~----~--~~~~---~~---~---.-. -------------~ - meanTOT 10*

1Ir------- csz JOF ZONEB

--&- ZONEC YFTB

-< ZONED

- - Af

- ARH CH

- - CM

- FH HHH

--HR

- - LB

--Lf MF

- -MR

- - RAW RM SS

- - SFZ

--SM

- -TR UR

- - YR

- - WF 10~ .__--~........._ ____~......._____~......._............._............ua......._._..~------_...,

10.. 10 l 10~ 10** 10° 101 101 T0.1sec Spectral Acceleration [g)

% Contribution 10 Hz at lE-04  % Contribution 10 Hz at lE-05 No Data RAW UR No Data

-4 RM 3% 3%

4% I 4%

3

Columbia - Source% Contribution 100 Hz {PGA)

SiteC 10° ., ., T I I l

- meanTOT l csz

__...,__ JOF 1

Cl>

10*

.r ZONEB ZONEC

-& VFTB 0 ZONED c - AF ca "O - - ARH Cl>

Cl> CH u - - CM

)(

w - - FH 0

HHH

- - HR

~ 10"4 - - LB c --LF CD j r - MF CT - - MR

£ 10.st - - RAW RM ca j

J

- sa 10""1 - SFZ c - - SM c

ct - TR UR 10*7 - - YR l - - WF f

io""

10.. 10-$ lO~ 10 I 100 10 1

10 1

Peak Ground Acceleration [g)

% Contribution PGA at lE-04  % Contribution PGA at lE-05 UR UR 5% SM %5 SM """'\. RM 2% \ 6%-!!:__1 RAW J HR l% \

4%

HHH 2%

CM or HHH 4% YR 1 ARH 4% 1%

3% 1%

4

Seismic Sources Characterized in the SSC Model -

SSHAC Figure 8.1

• Zone B

• Zone C

• Zone D

• YFTB Background

• YFTB Faults 4

5

20 Fault Sources ......_... a.Mie<

~

l1CllUllM-11u1

~ .. ma.

Alt U'

CM (1J C<lhllllli* llllk<ftllllli.~ C1H'-.t

('.......... Htll>-Ealc ru-r

.:*"WWW ......... ('o-*Hdi..WN Ol-W f1oi-E SFZ ('olulltiit Rollt-C""""I Ol*<

('o,....*lloll..C-.1.\\ ..1 rn..c.w

~- rn

+~

FJtD..ii:1.,. lhlh Hom llapab r ..11 Hit 0

w

~. 100 2 0 0 -....

11..... ~.... 11.n.

t i - ffn\'9 lltlM'-

• I H.cltoe Ko\ t'i6 lu..a.-t ftllft.IS,..hlll 11......-llillo.c_..1-.......

1toc.. lb,.. HJ.11o.\\n1 llHH HID!.(

HlW.('.f HHll-<:'-1\

llHlf.\\

Ulftl Lf

~-llullt ~

\1.W1iu.i.lt,o(h;< \Qt

~i.w....-.o>l>~t><°<<UlL>I \IR-('

M-Rldf:t*E.obl MR*£

~l->J,tatb lbd;*-\\'"" \lfl.\\'

M.w...  !\IT R.wtle.ofo11t~i.....i........u.,.~ RA'f.

IW.i.-l:t HlU* IUI

&Mllc>A.l*t Moc.1:1111 RM s.JJlc M-*la S1'j Wiit~....&" SM*t

"-"l.lk\i.-.....w... '~M\ "

'W"Nl l~f<lc~ '<Fl

!><lab 8'11!0 ~a T"l'1-,b~ rR fqll-!Wp-C.'4 fk-l 1.,,.....r1 !W~w.., 111-\I t_!W,,. lTR

""'"""" Rllbe-'io111....... Antl<fu>o IJJl. A

'-'--~1.,1 ......... L'l\-G.\f L*,.~- IW*e-<'n111ll U"R.C 1*,......, RJ4li ...r_... Vll-1'.

l\i.

V*ro*u R.Mltt"-\\ '~1 l'll-1\

Figure 8.43. Faull source and fault segments. Teeth are sllowu on me haugiug wall of the faults and \\anu1araodl WT squares define the segment bow1druies. Acronyms for fault sources are given in Table 8.5. \ 'ol..... \R

\ "ol.11. . R*.ljle-i:... \R.(

Tite insen shows the locariou of 1he Seanle fanh relative to the Hanford Sue and to 01her \" llJ.o111R~\\ ... 5,R*V.

fault ources. Yal:rono Rl4tt-~- nt.!>J:

6

Subduction Zone Sources - CSZ and JDF (used sources from BC Hydro Report- modified?)

CSZ - Cascadia Interface JDF - Cascadia lntraslab Source Near eoa tal cities Down dip closest approaeh Vole .

seismic limit 1* * ~

Y*ncouver Seattle Pord8nd Plate Interface in era ck red labeled S.!!isnioge.nic zone= ~~--H-_....--- Arrows=

Intra Slab lntraslab Source Source (Licensee's (Licensee' s CSZ Source)

JDF Source)

Fi:ure 8.8 Diagrmuna:tic dep1ctiou of the SfiSDDC sowces and other elements related to the CSZ (Hyndman 2013)_ The plate interface source 1$ shown m Rd and labded ..~lSmogemc zone .. The mtraslab source is shown by the down-gomg black arrows wllhm the Juan ~

fuea plate Ep1$od.te tremor and slip (ETS) ewquakes are l~beled ..ETS zone... The landward extent of~ plate mterface. which is a key aspect oflhe SSC model. 1S labeled

downdip seismic li.mlt.. (Hyndman 2013) ~location of Hanford Site lies east (to the nght) of the Cascade volcanic arc..

2 7

~--

+ ~ +

....... 0

~ 100 200 Kilome~

Figul'e 8.43. Fault sources and faulr segmeuts. Teeth are shown ou the hanging wall of the faults and quares define the egmenc botmdarie . Acronyms for fault sources are given in Table 8.5.

The insert shows the location of the Seattle faulc relative 10 the Hanford Site and to other 11

,.. *~

8

115*w 0 0 00 O<fj O 0 Q) 0

'IC"t>

0 0 0 0 o 0

0 0 0 z

o'f' ' I~ ~ z I

- - USGS Quaternary Fault Database Earthquakes C!: Magnitude 1.85 IU 0 Licensee Cataog

• Crustal

• 11124/1866 to 413012013 z 0 0 11> 0 Licensee Catalog - Subduction M1

• 1112411866 to 4130/2013

" 0 Licensee Catalog* Subduction M2 - 11/24/1866 to 4/30l2013 o\ o i

'~ Coordinate System: World G-wdetic Sysl'l!m 1984 ~

0 NRC Confirmatory Calalog. ANSS 5/112013 to 6/1/2015 (post-licensee catalog)

~I \'~ Projec\ion: Transverse Mercator, UTM Zone 11N 0 NRC Confirmatory Catalog* ANSS 1/1/1898 to 4130/2013 Esri, ~b::>~rne, GEBCO, NOAA NGOC, and other contributors 12s*w 11s*w

Digitized from Columbia SSHAC Report Figure 10.44 (electronic files not available)

Opt1Hz_10sec Color. of Line SA value at 1 OOE-04 for all is 0 0 % Contrib SA value at 1 OOE-05 for all is 0 037 % Contrib mean TOT 1.00E-04 mean TOT 1.00E-05 csz 0.000075 75 csz 0.0000075 75 subduction zone sources JDF 0.00000002 0.02 JDF NO DATA NO DATA ZONEB 0.00000008 0.08 ZONEB 0.000000015 0.15 four seismic source zones in the SSC mode ZONEC 0.00000003 0.03 ZONEC NO DATA NO DATA red dashe YFTB 0.0000026 2.6 YFTB 0.0000003 3 ZONED NO DATA NO DATA ZONED NO DATA NO DATA AF NO DATA NO DATA AF NO DATA NO DATA ARH 0.0000007 0.7 ARH 0.00000005 0.5 CH 0.00000002 0.02 CH NO DATA NO DATA black (can CM or YR 0.0000028 2.8 CM or YR 0.0000005 5 FH 0.000000015 0.015 FH NO DATA NO DATA HHH 0.0000016 1.6 HHH 0.000000075 0.75 HR 0.0000007 0.7 HR 0.000000045 0.45 LB NO DATA NO DATA LB NO DATA NO DATA LF NO DATA NO DATA LF NO DATA NO DATA MF NO DATA NO DATA MF NO DATA NO DATA MR 0.000000012 0.012 MR NO DATA NO DATA RAW 0.00000065 0.65 RAW 0.000000045 0.45 RM 0.0000028 2.8 RM 0.0000002 2 SB NO DATA NO DATA SB NO DATA NO DATA SFZ 0.00000003 0.03 SFZ NO DATA NO DATA SM 0.00000055 0.55 SM 0.00000003 0.3 TR 0.00000009 0.09 TR NO DATA NO DATA UR 0.0000016 1.6 UR 0.00000016 1.6 black (can YR or CM. NO DATA NO DATA YR or.CM NO DATA NO DATA WF 0.0000001 0.1 WF NO DATA NO DATA NO DATA HAZARD CURVE OFF CHART AND CANT EXTRACT VALUE

SSC model Source AFE PGA 1E-4 %

csz NO DATA NO DATA JDF NO DATA NO DATA MR AF NO DATA NO DATA TR LB NO DATA NO DATA 0.2%

LF NO DATA NO DATA SB CH MF NO DATA NO DATA 0.04%

SFZ NO DATA NO DATA No Data (CSZ, JDF, AF, LB, LF, YR or CM NO DATA NO DATA MF, SFZ, YR or 1 YFTB 0.000055 CM2) 2 RM 0.000007 16.9%

3 UR 0.0000048 4 HHH 0.000004 5 CM or YR 0.0000032 3.2 6 HR 0.0000024 2.4 7 SM 0.000002 2 8 RAW 0.0000016 1.6 9 ARH 0.0000012 1.2 10 ZONEC 0.00000045 0.45 11 WF 0.00000038 0.38 12 ZONES 0.00000035 0.35 13 CH 0.00000018 0.18 Columbia-13 FH 0.00000018 0.18  % Contribution 14 TR 0.00000016 0.16 by Seismic Source 15 ZONED 0.00000012 0.12 (PGA, lE-4) 16 MR 0.00000008 0.08 17 SB 0.000000035 0.035 No Data (CSZ, JDF, AF, LB , I 16.865

Source AFE (PGA%

ZONED NO DATA #VALUE!

TR csz NO DATA NO DATA ZONEC ZONES WF JDF NO DATA NO DATA AF NO DATA NO DATA CH NO DATA NO DATA FH NO DATA NO DATA RAW LB NO DATA NO DATA 1.3%

LF NO DATA NO DATA MF NO DATA NO DATA MR NO DATA NO DATA SB NO DATA NO DATA SFZ NO DATA NO DATA YR or CM NO DATA NO DATA 1 YFTB 6.SE-06 65 2 RM 5.5E-07 5.5 3 UR 4.5E-07 4.5 4 CM or YR 3.9E-07 3.9 5 HR 3.5E-07 3.5 6 HHH 1.8E-07 1.8 7 RAW 1.3E-07 1.3 Columbia -

8 SM 9.5E-08 0.95  %. Contribution 9 ARH SE-08 0 .8 by Seismic Source 10 ZONEC 1.1E-08 0.11 (PGA, lE-5) 11 ZONEB 1E-08 0 .1 12 TR 9.5E-09 0 .095 13 WF 9E-09 0.09 No Data (Zone D, CS 12.355 meanTOT 1.00E-05

Source AFE 110Hz. Double fault sources onlv  %  % double !Source AFE  % I csz NO DATA NO DATA NO DATA NO DATA ZONES 1.30E-08 0.13 JDF NO DATA NO DATA NO DATA NO DATA ZONEC 1.25E-08 0.125 AF NO DA TA NO DATA NO DATA NO DATA YFTB 0.000008 80 LB NQDATA NO DATA NO DATA NO DATA ZONED NO DATA NO DATA LF NO DA TA NO DATA NO DATA NO DATA AF NO DATA NO DATA HR, 3.8 MF NQDATA NO DATA NO DATA NO DATA ARH 4.SE-08 0.48 SB NO DATA NO DATA NO DATA NO DATA CH NO DATA NO DATA SFZ NQDATA NO DATA NO DATA NO DATA CM or YR 4E.Q7 4 YR or CM NODA TA NO DATA NO DATA NO DATA FH NO DATA NO DATA YFTB 0.000075 0.000075 75 75 HHH 8E*08 0.8 ZONED 2. 1E--08 0.00000002 1 0.021 0.021 HR 3.4E-07 3.4 RM 0.0000075 0.000015 7.5 15 LB NO DATA NO DATA UR 0.000005 0.00001 5 10 LF NO DATA NO DATA CM or YR 0.0000042 0.0000084 L 4.2 8.4 MF NO DATA NO DATA HR 0.0000038 0.0000076 ( 3.8 7.6 MR NO DATA NO DATA HHH 0.000003 0.000006 3 6 RAW SE--08 0.8 RAW 0.0000016 0.0000032 I 1.6 3.2 RM 3.SE-07 3.5 SM 0.0000015 0.000003 1.5 3 RM, ?.S SB NO DATA NO DATA TR 0.0000012 0.0000024 1.2 2.4 SFZ NO DATA NO DATA ARH o.ooooo~J 0.000002 1 2 SM NO DATA NO DATA ZONEC 3.SE-07: 0.00000035 0.35 0.35 z~!ll.taia TR 4.SE--08 0.48 WF 9E--08 0.00000018 0.09 0.18  % Contiibution UR 3.4E*07 3.4 ZONES 1.3E-07 0.00000013 0.13 0.13 by Seismic Source YR or CM NO DATA NO DATA FH 4.2E-08 0.000000084 0.042 0.084 (10Hz Hz, 1E*4) WF NO DATA NO DATA CH 3.9£-08 0.000000078 0.039 O.D78 CSZ NO DATA NO DATA MR 1.2E-08 0 000000024 0.012 0.024 JDF NO DATA NO DATA No Data -4.484 0 104.484 UR, 10.0 Columbia -

% Contribution b y Selsmlc Source (10Hz Hz, lE-4)

• fault sources doubled

Source AFE 110H; % !source AFE (10H; %

csz 7E-07 7 ZONED NO DATA NO DATA JDF NO DATA NO DATA AF NO DATA NO DATA ZONEB 2.8E-08 0.28 FH NO DATA NO DATA ZONEC 1.7E-08 0.17 CH NO DATA NO DATA YFTB 0.000006 60 LB NO DATA NO DATA ZONED NO DATA NO DATA LF NO DATA NO DATA AF NO DATA NO DATA MF NO DATA NO DATA ARH 1E-07 1 MR NO DATA NO DATA CH NO DATA NO DATA SB NO DATA NO DATA CM or YR 9E-07 9 SFZ NO DATA NO DATA FH NO DATA NO DATA SM NO DATA NO DATA HHH 1.8E-08 0.18 YR or CM NO DATA NO DATA HR 5E-07 5 WF NO DATA NO DATA LB NO DATA NO DATA 1 CSZ NO DATA NO DATA LF NO DATA NO DATA 2 JDF NO DATA NO DATA MF NO DATA NO DATA 3 YFTB 0.000008 80 MR NO DATA NO DATA 4 CM or YR 4E*07 4 RAW 1.5E-07 1.5 5 RM 3.SE-07 3.5 RM 5E-07 5 5 UR 3.4E-07 3.4 SB NO DATA NO DATA 6 HR 3.4E-07 3.4 SFZ NO DATA NO DATA 7 RAW 8E-08 0.8 SM 7.5E-08 0.75 8 HHH 8E-08 0.8 TR 1.4E-08 0.1 4 9 ARH 4.8E-08 0.48 UR 6E-07 6 10 TR 4 .8E-08 0.48 Columbia.-

YR or CM NO DATA NO DATA 11 ZONES 1.30E-08 0.13 % Contribution WF NO DATA NO DATA 12 ZONEC 1.25E-08 0 .125 by Seismic Source No Data 2.885 (lOHz Hz, lE-4)

-...~~~l!!L"D!."W"~- *ll C!lll!Lo

...O l f l t - U . h W...* - -...... ~*-.........._.. _ _ ....._elWI-

.*~

1-QbJ~- . i *"*

g -;._--

. . :. \ *. *:

5. ii-~-il~ ~ '9 I* . ~

• ]**

=

1111 SSE RG 1.60 Frequency SA [g] Frequency SA [g]

0.4 0.12 0.1 0.019 2.05 0.6 0.25 0.118 6.1 0.6 2.5 0.783 18.9 0.25 9 0.653 100 0.25 33 0.25

freq 5% Damped.Spectral Acceleration [g]

0.500 0.283 0.980 0.608 1.937 0.881 3.214 1.007 4.770 1.034 6 .861 0.963 9 .720 0.810 19.985 0.522 29.709 0.396 49.765 0.395

Columbia Generating Station Site Response (Part 2)

Vs Measurements Shear Wave Velocity (ft/s) 0 2000 4000 6000 8000 10000 12000 0

I I I I

"'"'IL_

200 II I

~ - WNP-2 I

400 I - WNP-1 Crosshole

- WNP-1 Downhole I

E' .

I

- 600 I - WNP-4 Crosshole

.s=

0 Q.

Q) 800 I

I I - WNP-4 Downhole

---

* I I

- WTP Downhole

.......... I 1000 I I I

I ---- WTP P-S suspension I

I I

1200 .... .:

a 1400

Licensee Vs Profile Shear Wave Velocity (ft/s) 0 2000 4000 6000 8000 10000 12000 0 ...... I I I I I Zone 1 200 -

Zone 2 400 -

Zone 3

.I E'

.c:

+J

~

600 -

I I

I

~ 800 -

---

. I I

1000 - I* ----

I Zone 4 I

I t

1200 -

• --*I 1400

- -vs P1 (Wt= 0.67) ----Vs P2 0Nt = 0.33)

Vs Profiles Shear Wave Velocity (ft/s) 0 2000 4000 6000 8000 10000 12000 0 ---,_,=---~~*~~~~-*~~~~-*~~~~~~~~~~~~~~

• ~ -

200

,_I _. . .I I I I r.mm _ _,

- Licensee Pl (Wt =0.67}

I 400 I I - - - - Licensee P2 (Wt= 0.33)

I I I I I * - - NRC BC

--.... 600 ~~~~~~~~~~~~~~---

I

• ~--

I I I

' .I

! ---* NRC LBC/UBC

~

.c c:::::::::::::::::::::.:::::;;;:::=~-~. -.

I cr1n = 0.15

~ I I I Q) I I I 0 I I I 800

..-_~...................*. ......~-----L - -r --*l I I I I I I 1000 t, ~------------------llllllil----11111111-----

11 : ---,--*

I I I I I I I t I I I I 1

~~~~~~~~~~~~~~~~*~~------~*~

1 1200 r,_

I

-..m11111 ii

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . __ ~~--~

I

_~--~~~

Vs Profile Questions

• Is it. adequate to only lJse aleatory uncertainty in upper 525 ft?

• Does use of the two Vs profiles and associated weights for the SMB stack adequately capture the epistemic uncertainty at CGS?

• Should random profiles consider limited.

lateral extent of interb eds (i.e. zero thickness 1

interbed)?

Damping

• Suprabasalt Sediments

- Strain. dependent dam1:>ing

• SMB stack

- Q model used to determine small strain damping

- Constant strain in. basalt layers.

- Strain dependent dam1:>ing in interbeds

Development of Q Model Used for SMB Stack Damping 0.08 y =1/0.0345*x (d}

Q = Yl1s 0.07 R2 =0.5033 u

Q) 0 .06 0 .05

• Kd amping 'H*

= L,. t. v;Si ~*t

.e> 0 .04

~

IV

~ 0.03 0.02 0 .01 0

0 0 .0005 0.001 0.0015 0.002 SUM(Hl/Vsll\2) (sec2 /m)

• Data points Fit Data points represent Kdamping for stations HAVvA, E07 A, E08A, F07 A, D08A, E09A and associated Vs profiles.

Damping in SMB Stack Table 7.2-'. Damping propert ies in the S:MB stack .

Profile 1 Profile 2 Uuit V s (km/sec) s V s (km/sec) ~

Ice Harbor (Martindale flow rap 1) 1.4 1 1.03% 1.77 0.82%

Ice Harbor (Ma1tindale flow rop 2) 1.67 0.87% 2.11 0.69°0 Ice Harbor (Manindale flow rap 3) 1.93 0.75% 2.44 0.59%

Ice Harbor (Manindale flow) 2.3 1 0.63% 2.9 1 0.50%

Levy Inrerbed 0.85 1.71% 0.85 1.71%

Elephant Mountain (flow top 1) 1.41 1.03% 1.77 0.82%

Elephant Motmtain (flow top 2) 1.67 0.87% 2. 11 0.69%

Elephant Mountain (flow top 3) 1.93 0.75% 2.44 0.59%

Elephant Mow1tain 2.31 0.63% 2.9 1 0.50%

Rattlesnake ridge Interbed 0.8.i 1.73% 0.83 1.74°0 Pomona (flow top 1) 1.43 1.01% 1.77 0.82%

Pomona (flow top 2) 1.70 0.85% 2.11 0.69%

Pomona (flow top 3) 1.97 0.74% 2.44 0.59%

Pomona 2.52 0.58% 3. 12 0.46%

Selah Interbed 0.88 1.65% 0.97 l.49°'o Esquatzel (flow rop 1) 1.49 0.97% 1.74 0.83%

Esquatzel (flow top 2) 1.80 0.81% 2.11 0.69%

Esquatzel (flow top 3) 2.00 0.72% 2.35 0.62%

Esquatzel 2.52 0.58% 2.95 0.49%

Cold Creek Interbed 0.82 1.76% 0.76 1.9 1°'o

Baserock K Kbasero ck = Ksit e-Kstack Uncertainty in SMB stack damping absorbed into Kbaserock Note: At this point, Kstack and Ksit e only considers damping portion Table 7.25. Target Kbasernck at the five hazard calculation sites at Hanford.

Profile 1 Profile . . .

Site Case All Sed. Half of Sed. No Sed. All Seel. Half of Sed. o Sed.

Lower Km\* 0.0297 0 .0189 0.0058 0.0378 0.0233 0.0058 Central 1'.:inv 0.0441 0.0285 0.0096 0.0556 0.03-l8 0.0095 Upper Kin\* 0.0491 0.0319 0.0110 0.0618 0.0388 0.0109 Site C Lo\ver KAH 0.0164 0.0100 0.0022 0.0240 0.0144 0.0028 Central 1'.:AH 0.0286 0.0182 0.0055 0.0370 0.0228 0.0056 Upper K ..\H 0.0485 0 .0315 0.0108 0.0563 0.0352 0.0097

Shear Modulus and Damping Curves SSHAC Example Licensee

• Sands (H2)

• Pasco Gravel (Sand with

- EPRI and Menq Scattered Gravel)

• Gravels - EPRI and Peninsular

- Rollins et al. and Menq

• Ringold

• Ringold - EPRI and Peninsular

- Peninsular

• lnterbeds

• lnterbeds - Darendeli and Stakoe

- Darendeli and Stakoe

G/Gmax for Pasco Gravel {Sands) 0.8

- MenqCu=5

- Menq Cu= 15

>< 0.6 cu - MenqCu=25 E

(!)

- - EPRI 20 - 50 ft

(!) 0.4

- - Peninsular 0 - 50 ft

- -

• Darendeli and Stokoe 35 ft 0 -+-___._-'-~~+--~~~--1~~~~--+-~~~~~

0.0001 0.001 0.0 1 0.1 1 Shear Strain (o/o)

Damping for Pasco Gravel (Sands) 10 - MenqCu=5 C)

- MenqCu= 15 c

Q.

- MenqCu=25 E

C'CS c - - EPRI 20 - 50 ft

- - Peninsular 0 - 50 ft

- -

• Darendeli and Stokoe 35 ft 0 -+-~----'--~~r-----~~~---+~~~~~~~~~~

0.0001 0.001 0.01 0. 1 1 Shear Strain ( 0/o)

G/Gmax for Ringold 0 +--~~~~~~~~~~~~~~~~~~~~-----!

---c:--, ,,,, ,

0.0001 0.001 0.01 0.1 0.8 ' '

Shear Strain(%) ' '

\

)( 0 .6

\

\\

\

' \

\

ca - - EPRI 120

• 250 ft E \ \

<!> 0.4

\

\

\

\

\

\

\

- - Peninsular 50

• 500 ft

- Darendefi and Stokoe 185 ft

\

0.2

0 ~~~~~~~~~--,~~~~~-.-~~~~4 0.0001 0.001 0.0 1 0.1 Shear Strain (%)

Damping for Ringold I

I I

I I

I 10 I I

I I - - EPRI 50

• 120 ft I

I - - Peninsular 50

• 500 ft I

,,I - Darendeli and Stokoe 85 ft 5

-- <Il !  ;

0 +-~~~~~~~~~~~~~~~~~~~~----<

0.0001 0.001 0.01 0.1 Shear Strain(%)

10 O>

c - - EPRI 120

• 250 ft

• a.

E - - Peninsular SO

• 500 ft ns c - Oarendeli and Stokoe 185 ft 5

0 -t-~~~~--.-~~~~--,.~~~~~..--~~~~~

0.0001 0.001 0.01 0.1 Shear Strain(%)

Strain Levels in lnterbeds Strain Profile (Target PGA=O.Sg)

Maximum Shear Strain (%)

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0

200 400

~

600

~

800 a.

~ 1000 1200 1400 1600 _.____ _____.___ ___..___ _ _ _____.__ _ _ _ _ _______,

- ~ Nonlinear Rock Behavior o Linear Rock behavior

Cold Creek lnterbed Shear Modulus Degradation

-NRC 0 .9 --* Licensee Max. Strain Range Target PGA = 0.5 g 0 .7 0 .6 -l---+---_,_..__-~_,__..__--.-1 _ _ _ _._..

>< 0 .5 _,__ _ _

ca ~ _ ____,______.___ __..

E a o.4 C) 0 .3 -!---------'-----+---~

0 .2 0.1 --i---~-----+-----.-+----i--1 0 +------.----------+----~---____,

0.0001 0.00 1 0.01 0. 1 1 Shear Strain ( /o) 0

Cold Creek lnterbed Damping Curve 20 I

---Licensee Max. Strain Range

-NRC Target PGA = 0.5 g 1---.. .........

I 15 -- I

~

I I

C>

D = b ( GG max y- 1 DMassing + Dmin I

'/

1 c

I . ,*

I

c. 10 I E ,

I I

cu c

5 ~

~

v --- I ,

I I

0 0.0001 0.001 0.01 0.1 1 Shear Strain (0/o)

Damping Equations G )0.1 D = b(G D Massing + Dmin max where:

- y- y, 1n(Y+YrJ -

0

_ 100 Yr _1 (Vo) - fI 4 D Afasing.a=l.0 y -' -

y +yr c1 = -1.1 143a 2 + 1.86 18a + 0.2523 I vr =(¢1 +¢2

• PI

• OCR¢3)* a 0 h 2

c 2 = 0.0805a - 0.0710a - 0.0095 a = ¢s J

c 3 = - o.ooosa - + o.0002a + 0.0003 Dmin = (¢6 + ¢7

• PI

• OCR ~)* a 0

'~ *[l + ¢10

• h1(fi~q)]

b = ¢11 + ¢12

• ln(1V)

Questions on G/Gmax and Damping

• Does. the use. of EPRI and Peninsular adequately characterize the epistemic uncertainty for the Pasco Gravel?

• Does. the use. of a single sand curve (Darendeli and Stakoe) adequately capture the epistemic uncertainty. in the san<:Jstone. and claystone interbeds?

- Should curves with less degradation and near constant damping be considered?

Randomization

• Layer Thickness

- Uniform distribution

• Velocity

- Suprabasalts: Lognormal, USGS B Correlation Model

- SMB Stack: Lognormal, LJncorellated

• G/Gmax and Damping

- Lognormal distribution at y = 3.16E-2

- (Tin( G ) = 0.3, CT1n(D) = 0.15 Gm ax

- cr1n(D) = 0.4 for strain inclependent damping ratios

Questions on Randomization

• Why did you randomize the damping for the basalt layers in the SMB stack when uncertainty for small strain damping in the SMB stack was incorpc>rated into the base rock kappa?

Conditional Mean Spectra 2.5 , . . . - - - - . . . - - - - - . . . . - - - - - r - - - - - - - . . - - - - - - . - - - - - - .

- - Predicted median spectrum, M-7.03, R- 12.2 km

,. - - *Predicted median+ 2o spectrum

--en c

2 I \'

- - Individual record sp ctrum

- - Conditional Mean Spectrum 0

'+:; 1.5 \

'° QJ I

CV u

u

<(

'°'-

tJ CV Q.

V'l 0.5 o.__~~---~~---i.~~~--~~__._~~~..._~~~

0 0.5 1 1.5 2 2.5 3 Period (s)

Application of CMS at CGS 101 CMS 1.0 Hz. AEF: l.Oe-04 Site C (CGS): 10-4 MAFE: T=1.0 sec SA "4 C<>ntnbutlon 25%

20'lfo

\S'tl o;

vv u io* 1 1()q(,

~u

~

Vl ta.

E 10-l IO 0

~

11"1 UHRS OISllln<:e (~m) \f : 5.6, R: 7 km , Wt: 0.15

\I* : 6.5, R: 11 km, Wt 0 39

\f : 7.2, R: 20 km. Wt: 0.25 II : 9 0. R: 336 km , Wt : 0 21 Frequency [ Hz]

CGS Site Response Logic Tree CMS2 (Wt= 0.39)

Profile 1 (EPRI Curves)

(Wt= 0.33)

CMS3 (Wt= 0.25)

Profile 2 (EPRI Curves)

(Wt= 0.167)

Profile 3 (Peninsular Curves)

(Wt= 0.33)

Profile 4 (Peninsular Curves)

(Wt= 0.167

• Note: CMS weights change with MAFE (Jln,T = Ii Wi [ (µln,i - µln,T )

2

+ ((Jln,i) 2

]

Minimum Amplification b .--__,-...-.....-....-_.........-...-...-...----.__,_....__.......__-...-...-...__,_,..__..............__.....,__,__,__,...,_.,.___,..,.,

b w,

u. 0

'\

' \

\

\

\

\

\

\

\

\

\

'9

~ I--__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,....,

Bo$erock '\

Soil M tt "'I Min mum Amr; O1 '

sou n% arn1 9:ilh't . !'Annum Arnp: o I '

Soi Meari. \4inlmurn Amp 0 5 \

son StM~ an:f 95th%. Mnlrnum Amp c 0.5 I

'b --~__,__,__,__,__,__,__,__,__,__,__,__,~~~---~~~~~--........~---~~~--..._.....

\ .-1....

-0 (10 0.01 O. t 1 1()

PGA (g)

Minimum Amplification

• SSHAC

- "The effect of imposing th*~ EPRl(2013) minimum amplification is to change the shape of the soil hazard curve such that it parallels the baserock curve at large PGA values. Evaluation of an a1p propriate minimum level of site amplification is an im1lortant assessment."

• Licensee

- "The 0.5 limit is not imposied. here. in the calculation. of the surface hazard because thE~ intended. purpose. of this report is to obtain. the best estimate of the mean and.

fractile levels of the seismic response for plant risk assessment with no added conservatism."

Questions on Minimum Amplification

• The example site respc)nse calculations from the SSHAC report shovv the impact of reducing the minimum site am~~lification from 0.5 to 0.1 and states that evaduation of an appropriate minimum level of site amplification is an imp~ortant assessment.

Please describe the assessment you made on the effect of not implementing a minimum amplification function *?

Columbia Generating Station Site Response 1

CGS Vs Model Shear Wave Velocity (ft/s) 0 2000 4000 6000 8000 10000 12000 0 ...... I Zone 1 200 -

Zone 2 L...-

400 -

Zone 3

.I

~ 600 -

.r:.

I ---..........

---

* - I

~ I

a. I

~ 800 - ........... I

____, I 1000 -

,. ---. Zone4 I

I 1200 - __ ..I I

1 1400

- vs P1 (Wt = 0.67) ---- Vs P2 (Wt = 0.33) 2

Shear Modulus and Damping Curves

• Sand Gravels, 0 - 525 ft

- EPRI and Peninsular

• Basalts, 525 - 1290 ft

- Linear

• lnterbeds, 525 - 1290 ft

- Darendeli

• Note: Effective strain is magnitude dependent

- ELSRAP uses a constant factor of 0.65 3

Kappa

• Kappa at top of SMB stack provided by SSHAC.

study

• Kappa above SMB stac:k based on low strain damping from damping curves

- Computed using site re~sponse

- Was damping from SMB stack double counted?

4

Vs Randomization

• USGS B. correlation model above SMB stack Thickness S-Wave Velocity Stratum Unit Variatlon Uncertainty, 01nvs P-1 Pasco Gravel +/-22% 0.25 P-2 Pasco Gravel +/-22°/o 0.25 P-3 Pasco Gravel +/-22°/o 0.25 P-4 Pasco Gravel +/-22°/o 0.25 MR-1 Middle Ringold +/-10°/o 0.25 MR-2 Middle RinQold +/-10°/o 0.15 MR-3 Middle RinQold +/-10°/o 0.15 LR-1 Lower Ringold +/-15°/o 0.15 LR-2 Lower Ringold +/-15% 0.30 LR-3 Lower Ringold +/-15°/o 0.20

• SMB stack profiles from SSHAC report 5

Input Motion

• Based on conditional rnean spectra provided in SSHAC report

- 4 events for each of th~e 20 frequencies and 27 mean annual frequenc y of exceedances (2160 1

spectra)

• Duration

- Estimated using magnitude and distances from CMS text files along with WUS duration model.

6

Vs-Kappa Factors Vs-Kappa Correction Cmnp..tte h<l!>t G} IPE FCSJ><.ll16'! ~CU'J. Ill abon dt.lil..~.s unJ on ~IT 9:lil e>r rook t \~pU1c r.!!1(1\l!'I.,~ spect1.t<C:Omj1'3uhl~ FAS us*~ IR \.T

+

D1\ 1J.-1.h~ FAS bylh.l ho.l.-r~10tt ..1tt amphlirotJM fxtob

.i E~im~ hOlft o. IJl~ fr~ I.he fti¥h fr<1qucnc) ~lopt vfU-.: FAS u11J 11vc1~

hOt!l a; <W a 111 cc:>!l5Jdcrcd '<X'ruinos I

l***~(i>-

+ l{O(ll1.":'

I I I Yes No

', '~

Ditfinr! .I lll<W h.....t FAS >UCh thJl ti~ lu~- '\J'lllr t. aailmg tiy muhlplyir:g dic ~ f \S by ttt<jlll'tlC)' ~Of'~ follows K scalin:g exp ( -rrf("car11.i - "11o!J1))

,'lf'pl)*;.. .calln;l bv mu!upJyu1s the lust FAS by erp (-11f(~uu11 tt - K110si))

.1 Jt'.1c-i1Calcd FAS tru:.:al F/\S Id hJSh frC<lllffiCl~I.. force th.? t.-scaltd FAS Co ti.? e.:iwJ to the ltUlJal o:MJ'E FAS Jt th.?!>c: tttqu.:nc1cs y C'<ltW<lrt ... ~FA~ le> N,,.~ lf1'.'Ct1a usms R\'T t+-

Cumput.l ' a:\lhni f~* ~ Ji~ 1J1.. lh<.. A. *"'11lk!J r~*n.t ~Ciro.I

\*lllu.!~ by th.! OMPE rup.ir.se "J'l'ctraJ v:l!Ub Figur~ 9.21. Steps for dennng kappa sca1.tng factors nsing the CRVT approach.

Host Kappa ompute host Glv1PE response spectra at short distances and on stiff soil or rock Compu te r sponse sp ctra-compatible FAS using IRVT Di ide the FAS by the host-region site amplification factors Estimate host!<. values fro1n the high-fn~quency slope of the FAS and average host K over all considered scenarios

Host Kappa 3.5 ------------~~~-------

• Calculate Host.

GM PE/FAS 9 magnitude/distance pairs

• Divide Host FAS by 0.1 1 10 100 Host. Amplification. - Profile Al aoi o -

Freq uency (Hz)

Profile 81 aoi o - Profile Cl aoi o

- Profile 01 - aol O - Profile El - aol O - - -

• Profile A2 . aoi O

- - -

• Profile 82 -aol o - --* Profile 0- aoi o ---*Profile 02 - aoi o

---

ProfileE2-.ioiO - Hm lWUS760 Figure 9.23. QWL linear site amplification fac1ors for the host WUS Vs profile with V530 of 760 m/sec compared to target site amplification factors at the Hanford hazard calculation sites.

Host Kappa (a) 10*1 ~~-~---~~

- - Cale FAS

- - 25perFAS - k =0.039

• Frequency Range for Kappa Calculation Method 1

( /'J (b) 10*1 . . - - - . - - - . . - - - - - - - - . - - - . .

- - Cale FAS

• fl at 25% below peak - - 25perFAShigh - k =0.041

• f2 at 1.5 x PGA but no greater than 20 Hz \"/

1o*1 .-------.------r-----r-----.-----.

Method 2 - - Cale FAS

- - 25perFASlow - k =0.036

• fl at 25% below peak

• f2 at fl plus 10 Hz Method 3

• f2 at 1.5 x PGA but no greater than 20 Hz 10-4

• fl at f2 minus 10 Hz 0 10 20 30 40 50 Freauencv (Hz)

Host Kappa Calculation 0.050 I I I I t I I I I I I I I I I I I I I I

---

r---.,----.,----,---- ----,----,----7----T----

1 I I f

t I

I I I I

I I I

I I

---

T*---}----~----~----

I I I

I I I I t I

• I I I '

I I

I I

I  !* I I I t

---

~----t-----1----i- -- ---1----i----i----t---*

' t t I I I I t-1

-f --~----t---- ----~----+-

I J  :  :  :

---1-- -

I

---

L--- . ----'----.J----

I I I I I I I f

---

~----:-----:----~----

' I f I

• ---1----1----;----t---- ----1---t*---~----r--- 1 I I I

I I

I I' I II II I I I I

• I I

-*-* I I I t 0.045

---

~---.L.---

1 I

I I I

f

... ---~---

I f

I I ...... __......:

I I

I

---

~----~----~----~----

I I

I

' I I

I I

I I

I I

I

• --~----L**--~----

I I I

l I I

---..4---~----~----

I a

I I

I I

I I

I I

I I

I iI I

I I

I I

~-- T' I

I I t I I  : I

___ .,... ___ I I I I I f I I I I I

---

~---~~----~----*---- ---~---~---~----~----

---

~---.,._ ~---

cu

' I I t

I

• ' *I ' I I I

.. ---~---- '...

I I I a.

I

---

~--_....,.

I f

___ ...,.I ___ I I I

~----

I I

• ---~----4----*----*----

I I I - I I

_.. _..'I ____II..,.

---

I *

---

:---~---1----1----*

a. I I I I I

---

~---...-----.----1----

I I

I I

. *-1---....

I I

~

I I I cu

• I I

I I

I t

I I

I I

I

!  ! O Lower

~ 0.040 I I o

I I I I

I I I I I t I t t

---

l-~-...,..--~~--4----

I I I I I I I I 0 ----1-----:-----:----4---- ---1---1 -~---t---~---

• Central a=  : I  !

--:----~

I

---

1-- -.l I

---

~

-~-

t' -~--

I 1I

• f' 4-I z I-I ---:--

t I f I t1 I j

-- 1 & Upper

---

~----r---i----~----

I I I I I

---

r----t-----1----i----* - ~--r----*----~----t--~

.J.I I I I i - Equality

-+---- ----t-I I I  : I  :

---

L----'----..1...*

I I J I .L. --~----~ ----:------:----~-- t t I

I I

I I I  : I  : 1 I

I I

I

• ~

t 0.035

.. I I I

_' J' __..._

I I I I I

I I

4 I

L I

I I

-'---..1-I 1

I

--~---

____ .,'____ ., ...... I' I I I I  : l I I I

' ----~- --~ -~ ---+---.. . .---~----~----

I I I ___ .,. ____ ..___. _

I I I I I

• I I I I

I I

1 I I I

"'4---

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

I

I I

I I

- -i----.!-- ~----J - -

I lt I

--..-----r---~---- ----1----1----,----~---- ----r----~----~----~---- ---~---~---~----~----

1 I I 1 I I ' I I I I I I' II II I I I  :  !

I I I I I 0.030 0.03 0.035 0.04 0.045 0.05 Licensee Kappa

Kappa Scaling Target K < Host K?

Yes No

~ ' ,,

Define a new host FAS such that the high- Apply K scaling by mu ltip lying the host FAS by fre quency slope follows K scaling exp ( - nf( Ktarge l - Khost) )

H*

Apply K scaling by mulliplying the host FAS by exp ( -nf ( Ktarger: - Khosr:) ) ]

~

If K-scaled FAS < initial F .AS al high :frequen cies, force the K-scaled FAS to be equal to the initial G:MPE FAS at these foequencies

,.... Convert K-scaled FAS to response spectra using RVT f.E-Compute K scaling factors by dividing the K-scaled response spectral values by the GMPE response spectral alues

Logic Tree for Vs-Kappa Correction Hose K II Target K I

Target Profile Approach to Estimate y Within-Approach Profile Depth I U ncettainty I l\'.uiv

• 1. 1 (0. 1)

H igh Profile 1 Inversions Kev All Sub-Basal! Sed (0.3) (0.67) (0.5)

(0.67)

(0.33)

Knv / 1.4 Central (0.4)

(0.-l)

(0.34) lCAJi*exp{l.6cr)

(0.2)

Profile 2 Anderson & H ough No Sub-Basalt Sed Low (0.33) KAH

( 1984)

(0.33)

(0.3) (0 ~ ~) ( 0.6)

KAH /exp( l.6a")

(0.2)

Target Kappa Table 7.25. Target Kooserock at the fiye hazard calculation ites at Hanford.

Profile 1 Profile 2 Site Case All Sed~ Half of Sed. ~o Sed. All Sed. Half of Sed. No Seel.

LO\Yer Kmv 0.0297 0.0189 0.0058 0.03 - 8 0.0233 0.0058 C entt*al Kmv 0.0-Wl 0.0285 0.0096 0.0556 0.03-J8 0.0095

. O.O-J91 0.0319 0.0110 0.0618 0.0388 0.0109 pper Kim*

Site C Lo\Yer KAH 0.0164 0.0100 0.0021 0.0-40 0.01-W 0.0028 Central K.oIB 0.0286 0.0182 0.0055 0.03 .,0 0.0228 0.0056 Upper KAH ~

O.O-l85 0.0315 0.0108 0.0563 0.0351 0.0097

Vs-Kappa Factors NRC Calculation

- - *SSHAC Branch 1

- - -SSHAC Branch 2

- - -sSHAC Branch 3

- - -ssHAc Branch 4

- - *SSHAC Branch 5 SSHAC Branch 6

- - -SSHAC Branch 7 10* 1 ._________.____...__.__...__..__.._._..........._______.._____...___.__.__.__.._._............________.____.___ _.__.__.___._......_.~

10* 1 Frequency (Hz)

Reviewing a SSHAC

" ... effective implementation of a good elicitation process cannc~t NUREG-1563 guarantee acceptance of the BTP on the Use of technical conclusions; howeve~r, Expert Elicitation use of a flawed process or (1996) improper implementation of a good process cannot help but cast serious doubt on the quality of the conclusions."

Simply put, NRG needs to C(Jnclude that the SSHAC process developed a PSHA that appropriately represents the center, body, and range of the informed technical community.

Reviewing a SSHAC NUREG-2117, Sect. 4.10.5: PPRP review of documents Process Considerations

• Have all of the essential ste~ps of a SSHAC process been followed and documented?

• Is the data evaluation proce~ss sufficiently justified?

• Is there clear documented E3vidence that the views of the larger technical community have been considered?

• Has the integration process been sufficiently documented such that the c~enter, body, and range of technically defensible inter~>retations are well justified?

Reviewing a SSHAC NUREG-2117, Sect. 4.10.5: PPRP review of documents Technical Considerations

• Have all data used in the assessment been identified and documented?

• Have all elements of the mc)del been defined in sufficient detail?

• Have the model elements alnd expressions of uncertainty (e.g., logic-tree branches and their weights) been technically justified?

• Are there any technical issues that have not been sufficiently addressed?

Reviewing a SSHAC Practical Considerations based on staff experience

• Were contrarian views given appropriate (e.g., unbiased) presentation at workshops?

• Was some potentially relevc:int information excluded from presentation at workshops?

• Does documentation providle a clear and traceable basis to support selection of mod,els and data used in PSHA?

• How engaged was PPRP in SSHAC process?

• How were PPRP review co1mments resolved?

WUS SSHAC Review Guidance Goal of Review: Establish confidence that the center, body, and range of the informed technical community have been considered appropriately in the PSHA.

Three Perspectives for SSHAC Reviewers

- Hanks et al. (2009): "It is simply not possible to verity that the center, body, and range of the full ITC have been successfully captured without repeating the entire process of expert interaction with a different group of experts and perhaps different Tiff Fis as well. As a matter of practical reality, this has not occurred."

- NUREG-1563 (1996): "effective implementation of a good elicitation process cannot guarantee acceptance of the technical conclusions ; however, use of a flawed process or improper implementation of a good process cannot help but cast serious doubt on the quality of the conclusions."

- The acceptability of a SSHAC ultimately depends on the transparency and traceability of its documentation.

Review Approach: To determine if the SSHAC process was acceptable,. reviewers should address the following 7 key questions that focus on the important attributes of an acceptable SSHAC. Not every sub-topic needs to. be addressed, or answered affirmatively, to conclude that the SSHAC process is acceptable. Nevertheless, the information provided in the SSHAC documentation should provide a traceable basis to answer these 7 key questions, and many of the. sub-topics as well. Staff will. use information from these review. questions to develop assessments on the acceptability of the SSHAC process in the hazard reevaluations.

SSHAC Review Considerations Organized Around Considering 7 Key Questions

1) Did the SSHAC process reasonably follow guidance in NUREG/CR-6372 and NUREG-2117?

- Was the potential for cognitive bias discussed early in workshops?

- Did Tl lead identify or address potential cognitive bias issues in workshops?

- Did the scope of the workshops support logical development of a PSHA?

2) How effective was the Participatory Peer Review Panel?

- Was the PPRP engaged throughout the SSHAC?

- Were early comments/concerns by PPRP addressed in later workshops/meetings?

- How were PPRP review comments resolved?.

- Are there unresolved PPRP review comments that might affect results significantly?

3) Has applicable data been considered?

- Was a common database developed?

- Was the database easily accessible to participants (including PPRP)?

- Were updates to database made, and were all participants aware of updates?

- Were local experts engaged in identifying potentially relevant data?

- Is there indication that some potentially relevant data was not considered?

- Are the data summarized and presented sufficiently to support use in the PSHA?

4) Were data uncertainties identified and considered?

- Were uncertainties in original measurements maintained in compilations?

- If data were collected in different studies, were issues of calibration/bias addressed?

- Were data quality issues considered by the T l team and PPRP?

- How have data uncertainties been propagated in resulting models/calculations?

5) Was an appropriate range of potentially applicable models considered?

- How was the range of potential models developed/documented?

- Did an appropriate range of model proponents participate in workshops?

- How well did the T l team engage with the proponents?

- Were contrarian views given appropriate representation?

- Are there indications that potentially relevant models were not considered?

6) How were models selected and weighted in the analyses?

- Is the basis for inclusion or exclusion of models well supported?

- Is there appropriate documentation for how model weights were developed?

- Were models tested or supported by independent information?

- Were there interactions to openly discuss inclusion or exclusion of models?

- If new models were developed, are limitations in existing models documented?

7) How were models and data assembled into the PSHA?

- How were models abstracted into logic trees and weighted?

- How were sensitivity studies used to refine. logic trees or weights?

- How was uncertainty partitioned into aleatory and epistemic components?

- How were these uncertainties propagated into ensemble results?

- Is. it clear how. ensemble results were generated?.

Kishida et al. (2014) report provides estimates of kappa values for Arizona seismic stations based on the data recorded by TA array. As stated in the above mentioned report, these. data have high-frequency range limited by the sampling rate. The earthquake recordings used for kappa estimates are mostly from earthquakes with magnitudes not exceeding M 3.4.

Considering the relatively high corner frequencies of these events and limited recording instrumentation frequency range please comment on the. reliability of kappa estimates.

Kishida et al. (201 4) report concludes that the comparison showed that overall the recorded 5%

damped response spectral ordinates were over predicted by the NGA-West2 models developed for California. It also provides estimates of kappa in average similar or lower than the average kappas in the NGA-West2 GMPEs. The estimates of 0-values in Arizona (Phillips et al. , 2014) are generally higher than in California for which GMPEs were developed Considering the estimated values of Q and kappa please discuss seismological effect contributing for potentially same or faster attenuation of seismic radiation with distance in Arizona relative to California.

The approach to ground-motion models for PVNGS from sources located in central and southern California and Mexico is based on as-published NGA-West2 GMPEs with path-specific adjustment factor to take advantage of the available ground-motion data in Arizona from these sources.

Please provide assessment of the influence of this path term correction effect on hazard calculations from California and Mexico sources.

Please clarify if any path corrections were performed to the 6 models used for Greater Arizona sources.

Section 5.5 of the SWUS GMM report provides recommendation of using Bindi et al. (2014) and Akkar et al (2014) attenuation models for greater Arizona sources in addition to the models developed for California.

The rationale for adding Bindi model is that this GMPE is based on the ground motion database that contains significant number of normal fault events typical for Arizona. This model also satisfies other requirements/criteria specified for model selection including publication in the peer-review journal. Bindi model is characterized by a specific magnitude scaling differing from other models that may be considered as alternative to other models representing the range of opinions.

Please clarify the reasoning for the limitation on magnitude in applying this model to PVNGS.

Please discuss potential effect of this limitation on hazard calculations.

Four NGA-West2 GMPEs are. applied to both distant California & Mexico. and greater Arizona sources. These NGA-West2 GMPEs include basin effect dependency as a depth to the 1.0 or 2.5 shear-wave velocities.

Please clarify if the. basin effect was accounted in using GMPEs. for distant and/or Arizona sources.

Diablo Canyon ESEP Discussions, 6/9/15 3.0 r.===========-------------,

IDiablo Canyon , ..

2.5 - 2015PSHA I

- DE I I ' \

\

- DOE

- HE

' \

\

\

- LTSP \

- - LTSP margin \

0.0 .....__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __,

0.1 1.0 10.0 100.0 Frequency (Hz)

- PG&E (3/11/15) states:

"PG&E has determined that it is not necessary to perform an expedited seismic evaluation process as PG&E's interim evaluation provides reasonable assurance that it is safe to operate DCPP while the updated/upgraded seismic PRA is developed." and "However, as discussed in Section 5.0, the DCPP GMRS is bounded by other previous seismic evaluations, including the design/licensing basis 1977 HE evaluations and the 1988 L TSP evaluation. Therefore, there are no additional benefits in performing this activity in parallel with the more robust risk evaluation associated with updating/upgrading the SPRA PG&E will devote the critical skilled resources to expediting the update/upgrade of the SPRA in order to gain additional risk insights in a timely manner." (p. 51).

- Nevertheless, EPRI (2013, Evaluation Guidance section 3 .2) states:

"In responding to EA 12-049, each plant will have defined an essentially indefinite coping capability for scenarios involving an extended loss of alternating current (AC) power condition .... Plant-specific evaluations for FLEX will determine the specific equipment and strategies to be employed in these three phases. The scope of the ESEL is limited to installed plant equipment and FLEX equipment connections."

- In responding to Order EA-12-049, what plant-specific equipment did PG&E identify for ESEP?

- Is. that plant-specific equipment evaluated in the. LTSP analyses?

Palo Verde Site Response My understanding ...

• Scale input ot appropriate input amplitude (11)

• Apply FAS Correction (9)

• Calculate site responsE~ using corrected FAS

• Amplification function is surface response/original bedrock response

- iRVT does not recover IHF

Site characteristics

• Geology is relatively uniform

- Exception if Fanglomerate

• Extensive Geophysical s1urveys

- Initial siting

- Downhole (R2.1)

- SASW across site (R2.1)

• Three profiles are easily within the randomization of the base case profile

- Does their model of epistemic really represent what is meant by. epistemic unce~rtainty?

Confirmatory proposal

• Develop site. profile based on updated information

- Single profile

- Three profiles represer1ting geologic differences under reactors

• Damping model is oka,y and consistent with SPID approach

• Reduce Aleatory uncertainty to be in line with the epistemic proposed by the SHSR (Table 3)

Amplification 4

3.5 3

- Licensee 1 Hz 2.5 - Licensee 10 Hz c:

0

• - Licensee PGA

"'u

E 2 - ShallowlHz a.

E - Shallow 10 Hz

<(

1.5 - Shallow 100 Hz

- DeeplHz 1 - Deep 10 Hz Deep 100 Hz 0.5 0

0.001 0.01 0.1 1 10 Spectral Acceleration (g)

GMRS

- Licensee GMRS (g) 1.4

- Confirmatory GMRS (Shallow) 1.2 1

~

c 0

IV (ii 0.8 Qi 8

ct IV

~ 0.6 Q,J a.

Ill 0.4 0.2 0

0 1 10 100 Frequency (Hz)

Palo Verde SSHAC Level 3 s~;c 09/01/2015 PVNGS 1

120° w 117° w 114° W 111° w 1os 0 w 36° N 33° N 30° N

>e1smotecron1c areas a1ternat1ve

120'W 117'W n**w 1tl'W 1oa*w USGS Quaternary fault and fold database z lasi updated 11-3-2010 m

~

Shp Rate (mm/yr) 36' N Less lhan O 2 0 2 to 1 0 1Oto5.0

..' Greater than 5.0 "I

Unknown or not reported

.* Earthquake Locations from PfO]ecl Catalog (MW 2.7

• 7.3) 33' N SBR

\

JO"N 0 80 1887 rupture trace

- mt m::i- km 0 80 Figure 9-7. Certain Seismotectonic sources are defined based on trends in the density and orientation of Quaternary fau lts. The CP and TZ sources are differentiated by changes in crustal thickness (Figure 9-6), as well as the lack of Quaternary faults in the interior of the physiographic Colorado Plateau. The 1'1H and SBR sources are differentiated based ou interpreted cmstal twckness (Figure 9-7), as well as the lleiglJtened deusity of Quatemary faults in the southeastern portion of the southern Basin and Range.

ElMI

. ~ )'

Figure 9-23. ~tap mowing the PVNGS catalog (color coded by magnitude bin) and Sei~t:ectoruc sources

120° w 117° W 114° w 111° w 10s* w

-- UT~

z m

~

' ~*

m 35* N

x-()

' \ 10

\

l

\

' I

• __r~ 33* N

)

\ I

(

\ J' It

///1:\~)

- ,, { 30° N 0 80 mi km 0 80 Fault sources (168) in PVNGS SSC

Slip rate mm/yr

>3.0

=: 1and<3 0

?. O 11nd < 1

!001and<0 1

<001 30* 111 U169 0

From Figure 10-1: High slip rate plate boundary faults (red) modeled with UCERF 3 and layered alternative.

Basin & Range and low slip rate CA modeled with characteristic fault alternative (blue and green)

Sand Tank and Ballenas have their own special detailed logic tree (greater uncertainty)

120°w 111°w 114° w 111° w 108"W 0.., 1'~

( /~ "':-1()

- ... - UTAH

.... ..... z

.... rn Pisgah- o-? ~ ~

Bullion Mtn- 1'/ / ' '

' ' l~

Mesquite lk ..,/ 36° N

/

San Andreas Calico-Hidalgo ----

\

I

/

/

/

B_ig Chin.o- ~

Little Chmo ,_ Wlliamson Valley

\

\

()

0

\

\

grabens I \

Plomosa East l Horseshoe \

1

' Carefree

• )

Sand Tank 33° N

\

Laguna Salada ________ ,

I

\ I I

/

/

' /

/

' /

30° N 0 80

- c:=--* mi

• km 0 80 Final fault sources (18) in PVNGS SSC

Hazard Curves in general.

Total Hazard is driven by the Areal sources

@ 10 and 1 Hz, Areal sources ( 2 zone and seismotectonic (6) zones)> all fault sources Closer look @ Areal sources for 10 Hz: seismotectonic zones> 2 zone> CA faults (mostly)> AZ faults for 1 Hz: seismotectonic (mostly) >CA faults> 2 zone> AZ faults Separating the areal sources:

for 10 Hz: SBR > East> TZ >everything else for 1 Hz: SBR mostly> East mostly> TZ mostly> everything else By fault:

for 10 Hz: SAF >Cerro Prieto >San Jaciinto >everything else, exception SD TK For 1 Hz: SAF >San Jacinto> Cerro Prieto > everything else, exception SD TK

10 Hz Site Specific Rock Hazard at Palo Verde, by Source Type 1E*3 QI v

c QI "O

QI 1E*4 QI v

)(

QI 0 - Mean v> lE*S c:

QI

~

- Area er QI IV c

~

1E*6

- Fault c:

1E*7 +---------....._._ __,________._____--'-+_ _ _ _ ____..___ __

0.01 0.1 1 10 10 H l cno,.tr::>I ,,.,.,..,1or::>nn n l a\

1 Hz Site Specific Rock Hazard at Palo Verde, by Source Type 1E*3 QI v

c QI "O

QI lE-4 QI v

)(

QI 0 - Mean

~ lE-5 c

QI

~

- Area

-r:::r QI IV

~

c 1E*6

- Fault c

lE-7 -<--------~-----------------------~

0.01 0.1 1 10 1 Hz spectral acceleration (g)

-f.a--llA.....,. ____

c=i Olt1 & .,..._ _ RC'l_ dlt &Qr~ ~ wl'l1u"140'*...,*'"' c::::J Cif*-*~--

llll)OiL' Cl- ~\NT c::::J r*v-Figure 37. Geologic cross-section showing the shallow and deep stratigraphy at the PVNGS site; modified from UFSAR Figure 2.4-27. The map in plan view on the right shows the cross-section line, as noted by A-A'; map is modified from PSAR Figure 2.5-13. Note that Units 1-3 are west of the crosssection line, between borings PV-21 and PV-24. Source: Figure 8 from LCI (LCI, 2015d).

PPRP identified issues

• SSC site region is mostly. in S B&R and as such is devoid of Q faults. Is this valid? are we missing some Q faultsi?

• Geodetic rates are gre,a ter than fault slip rates in central and souther1n AZ

• There are no modern cjetailed geologic maps of the PVNGS site area1 and vicinity.

• There are 3 faults near the site that have uncertainty about characteristics.

'E

' )<

' ti 0

' \

.. ' \

r \

~

\

\

\

I

'I M' I

\

i:*u 0* - "1!1

...... 4(1 Ftgun 7-1. Qu,atenwy fanlts m Arizona and neighboring sta1es from d:te USGS Quaternary Fault and Fold Databa!.e (USGS. '.!006. last updated 2010). Arrow points 10 the Sand Tanlc fault ~ast oftbe PVNGS sae (black stat) .

l *O' W

' \

' \

'~

' \

\

I

'I a4' "

'I l:' 11

)O' H fiprt -1. Quatemmy fanlts in Arizona am neighboring '1ates from me USGS Quatemiuy Fault and Fold Database (USGS. 2006. last updated 1010). Arrow points to the Sand Tank fault ~asl oftbe PVNGS sue {black star).

\SW 11""\V lll' W 11~*w l WW 110' YI 1CJ6" W IS\

ov

'I

"+&~ *

,ii,.

1) ~*ti

it"' I

.SI I fl/

"""' 0

I I

I

~A1i1GE

~~

12' N

}, I ~) i

..* (

.. (, ' ' 'l

\

l' \

\

31" 4 0

-==-*40km 40 ni

..._~___......__...._____~~~~~~--.~-'-l-1__~~~~--~-~----~~~ ' ......._..___.____~-'--~--------~--~----'10" U Figure 7-7. Green lint ~ nre po1em~11 Quatemary ac ti~e faulL' in northern $(mora i<k'Tltirled b)' the! n Team b~ a1'Pl) tn~ Appro~h I. Purple li11-e'

' ho" faulh t1lnt appear"" "iet'\ icio C"ioologico Me\ic.ano (SCiM) geolosic mar, of fl( 11hem Soll(lrn 1ha1 are \\ithin or bounJ Quatemar; unit . Rod hrn: an: Qu~uemary foulh f1vrn I he USC. ' Qu~1enrnry ra ull nud fold D111nbu..'e (U ' C..S. 2006. I~"' ' up<l41ted 2010).

PHOF.N X NORTH Figurr -11. Location m.ip show~ extents of ~tailed geologic: mapping oft.be SIU! Acea (8-km radius) aod recOllDiili.sanc~level mapping of the Site Viciwty (40-bn radim). .Pwple lines sho\V extents oftbe Pboemx North and Pboeni.""t South 30- x 60-minute geolopc map sheets. Green shadi.ag show:. area of ma.pplng m'ised by Phil P~ (AZGS) for this pcoject. Orange shading s.bo\\'s ~a mapped by Jeri Youog (AZGS) Btne 'lhading shows areas mapped by LCL The Saud Tani.: fault is represented by solid aod dashed black lines northwest of the SU TOUlkMountains. Modified aftecLO (2014).

IU' W Sand Tan k fau lt Elf!Yation

~IOMl\rO Feui~CIYP n 065 Tonal lmeamenl 0

0

--=z=-

-==-- * - 8 Ftgure-l-LExteut ofSand Tank fwlt scarp (sohd bladd1oe) and toe.al lineament {dashed blackhne) modified after I>cimey and Pearthrtt (1990). Sc-aip hes outsi~ of the 40-bn P\lNGS S1te Vicinity.

113.is* w 113* \V n?1s*w t%5* VY J:7S k Piedmont Alluvium Yat1rgtr Uni

~

ft OdcJUnts

'tr

~

l

~

tl

~

(\#

>>S'N

..,?

~....

%l'..,

Ri;1el Gi\O 33 25' ~

0

-c=:i- -mi 0- - 8 km 8

Figure 7-1 J. Simplified QU'1tt:ntaf') geotugic map vf lhl! Site: Vicuttt) ( 10-"m mdiu!>) Yellow .,,how di 1ribu1ion of )ot111gc.:r Jt'pO"\it ~ 12 ka

{Qy), orange ~ow' olderQuillcrn!l1') umt' >SO kn <Qa3. Oil (.)i i. and (.)o). gra~ -.llO\H. area.c; of Pliocene and older hedmck. Modified atler I.

(2014).

UU7S"W Explanation

• Bonng UFSAR June 2001 Revision 11 Appendix 2F H Profile PSAR July 1975 Ch 2 5 Part 2 Figure 39 Piedmont Alluvium Oy2 Late HoJocene aauv1um Oy1 Holocene alluvium Qy Holocene a~uvlaJ depoS1t5, und1fferenuated 013 Late Pleistocene alluvium Oi2 Middle lo late P1e1stocene alluvlum Qtc Quatemary htllslope iaws and colluvlum Bedrock Units Tbu Upper basalt Tbl Lower basalt Basatt undllferenbated 0 2,000 ft m

33 37$' N 0 400 Figure 7-15. Portion of the 1:24,000-scale geologic map of the Wintersburg quadrangle (Pearthree et al. , 2006). The unnamed fault of Pearthree et al. (2006) is shown as a black dashed and queried line. Blue line shows location of line of geologic cross-section shown in Figure 7- 16, whereas b lue circles indicate key borings that constrain the subsurface geology in the cross-section. Star indicates the location of operating nuclear Unit 2 at the PVNGS.

l' ig ure 7-24. Lran ect <lfi?eodciically "" ed ea-.t-\\e ~ e'ICl'biori rates tin mm*~ rt acm..s Arvona and \\>C"temmo'I New M~x1co tmm Kreemer

(\\ or~tll'p I p~~mat_ion. t:\! A~ndix B }.

F igure 9-t7. C-0mpanson of calculated deformation rates across the PVNGS region using the Seismotectonic model and Case 1 magnitude ~s (red-nlues) to GPS-measured extension rates by Kreemer (\11hles in black). ~!odified from Kreemer (Woikshop 1 presentmon.. see Appendi."t B).

Magnitude-Range Source Type Weighting Zone Model Sources Fault Model West N/A Two-zone 0.2 East Case 1 So. Caflfornia and Baja 0.4 SCABA NA Areal sources Case2 0.4 Gulf of California GULF Case3 So. Basin and Range (SBR) 0.2 Seismotectonic 0.8 Transition Zone TZ)

UCERF 3 I

Colorado Plateau (CP)

Fau lt Model: Mexican Hi hlands MH Ru ture Sets

• UCERF 3 0.8 San Andreas fault

• Layered Layered

• Characteristic fault 0.2 Ru ture Sets San Jacinto fault 0.8 La ered 0.2 18 faults in final SSC model Ru ture Sets Elsinore fault 0.8 La ered 111 0.2 .i::.

Algodones Fault sources N/A N/A b.O I

Big Chino-Little Chino Cerro Prieto fault La ered Carefree South of the .

Ballenas transform fault Layered }

Horseshoe Border Plomosa East 13 fau lts - - . . All other faults Characteristic Sand Tank . :===========-~ 7inAZ Wi lliams Valley grabens Figure 8-1. Master logic tree for the PVNGS SSC model.

Slip Rate, Rupture~

Model Objective Fault Model Resetting MRE? EPR Distribution Fault~

tMRE for EPR See Table 104 36,139' point sources from ~an liernard1no (0.25)

South to lmperb l ApprmcimatP I 1r~ ru:~ Rupture event set 25 mm/yrJ UCERF3. sections rupture behavior from UCERF3 294yrs No See Table 1o-4 (Appendix G) NA (0.8) (O.S)

See Table l D-4 (0.25)

Allow for fault se~menta tlon and variations in slip rate L.ayered fault model See Figure 10-Sb (0.2)

Figure I 0-Sa. San Andreas fault source logic tree.

Seismogenlc Slip rate, Rupture Slip rate Style of Dip/ Resetting EPR Rupture Mchar Recurrence Thickness layer tMRE for length (mm/yr) Faulting Direction MRE? Distribution Area (km2 ) Basis Model (km) EPR (km)

See Table 10-4 Full rupture Hi h I.a er 1 (0.25) area (HBOS) Char EQ (0.2) (0.5)

Slip rate, See See 15 90 d Layer 2 tMRE Table 10-4 Table 10-4 RL RA (O.S)

Low La er3 See Char EQ (0.2) Table 10-4 I.a er4 (0.25)

• Hal f rupture area is only applied to Layer 2. All other layers use full rupture area with a weight of 1.0.

Figure 10-Sb. Layered fault model branch of the Sao Andreas fault sourre logic tree. Layer-specific sUp rates and magnitudes are provided in Table 10-

3. Parameters fer EPR are provided in Table 10-4.

Selsmogenlc Magnitude Slip Rate EQ Recurrence Fault Style Dip/ Direction Geometry Magnitude M ethod M char Thickness (km) Regression(s) (mm/yr) M odel Sand Tank Fault

-0.2 0.001 WC94max.

displ. (a ll} Char. EQ (0.8) (1.0)

N/A Per. even (0.2)

WC94 avg.

~ -0.2 (0.2) displ. (all) 7.2 (0.2) (0.6}

+0.2 (0.2}

Avg. ofWC94 RA (all)i Stirling

- 0.2.

Ru pt. et al. 2002 Cl; Wesnousky (0.2)

Long dimensions (32 (lineaments) km length) 15km 2008 SRL (all) 6.9 Char. EQ 0.2) (1.0) (0.6} (1.0)

+0.2 (0.2)

Scarp height more reliable indicator of mag than length Distribution broader Uncertainty and distance than other normal faults from range front Figure 10-30. Logic tree for the Sand Tank fault. Mchar is calculated from scarp height (upper branch) and fault dimensions (lower branch), owing t<

uncertainty related to the full length of this fault source.

Tabk 7-2. Preseniatlon and degradation of f.ault scarps in the desert landscape.

modrlied after dePolo and Anderson (2000).

RI Degrade RI Degradt Decradt time Tbnt Tina RI (kyr) timt Timt Slip nm {)QT) (kp) tim t

'TrO Tr 0 4m WO (mm yr) lm (k.yr)

-c:;u']>

m {JcyT) ic:;up ~ c:arp

(\.:r)

K AJ'p

,cup lm sc:u *p ?m 4m 0001 l,000 25 gs* . 2.000 100 95*. -l,000 400 90-.

0.005  :!00  ::!5 ss*. 400 100 75*. 800 400 50-t 0.01 100 25 15*.  :!00 100 5~* 400 400 ..

o**

0.02 50 25 5~9 100 100 o**

• • 200 400 o*.

0.05 20 25 O!o 0 100 o~. so 400 o~ .

0.1 5 25 ()! . 10 100 o,. 20 ..$00 ()! *

\ot~:

RI: Average recuaence interval. in thousa.ods ofyean (kyr).

Degrade Tune: scarp degradation tune, m thousands of years (l-yr).

PVNGS SSC Areal (Background)

Sources SSC SSHAC Report Section 8.2 and Chapter 9

CHAPTE R 8: SEIS:\llC SOL'RCE CHARACTERIZATIO~ OVER\"IE\Y.................................... 8-1 PVNGS SSC SSHAC Report: .

8.1 Types of Seismic Sources Identified and Characterized in the SSC Model ........................................ 8-1 8 ..2 Areal Seismic Sources ......................................................................................................................... 8-.2 8.2.l Maximum Magnitude (Mmax) Assessment for Areal Sotu*ccs ...................................................... 8-2 8.2 2 Zone Boundaries for Areal Sotu*cc-s ............................................................................................... 8-3 8.2.3 Ma~n1de Freguency :M odel for Areal Sources ............................................................................ 8-3 8.2.4 Earthquake Recurrence Assessment for Areal Sources ................................................................. 8-3 8.2.4.1 Smoothing to Represent Spatial Stationarity ........................................................................... 8-4 8.2.4.2 Formulation of Penalized-Likelihood Model for Recurrence Parameters ............................... 8-5 8.2.4.3 Modeling the Joint Distribution of Recurrence Parameters and Development of Altemative Recun*ence ~laps ................................................................................................................... 8-13 8.3 Fault Sow*ces ...................................................................................................................................... 8- 14 8.3. 1 Model Approaches for Fault Sources ........................................................................................... 8-14 8.3.2 Slip Rates for fault Sotu"Ces ......................................................................................................... 8-15 8.3.3 Magnitude Scaling for Fault Sow*ces ........................................................................................... 8-15 8.3.4 Magnitude Frequency Model {Characteristic Earthquake) for Fault Studies ............................... 8-17 8.3 .5 Earthqua ke Recurrence Models (Poisson and Time Dependent) for Fa ult Sources .................... 8-19 8 .3. 5.1 Time-Independent Ea1thquake Recurrence ............................................................................ 8-19 8.3.5.2 Time-Dependent Eanhquake Recurrence ............................................................................. 8-20 8.3.5.3 Equivalent Poisson Ratios ...................................................................................................... 8-.23 8.3.5.4 Ocher Considerations.............................................................................................................. 8-25 8.3.5.5 Sununa ................................................................................................................................ 8-25

PVNGS SSC SSHAC Re ort: .

CHAPTER 9: SSC MODEL! .!IU:..U SOl"R.CI S 9 1 Critma forDetinm.g Areal Soun:~ and Data Used.

9.U Future ~uab ClwtdmStics __ ****-**

9 .l.1 Sel.SDlOgemc Tbic~s **--**************--******* ....................*-***-*****-**-**--**********-*-*-*****-****9-2

.. 9-~

9-3

... 9-3 9.3.4.l Basis for Defining Soun:e ______ ........._ .................... _ .. ______ ....... _ ...... _......... ..9-9 91.~ Futun~~~~~~~~~~~~~~~=======~*c!:'. 9-~ .5 9.3 4.2 Basis fur Source Mmax ......... ..........------*-*--***-....--........- ......-....... 9-9 9.2..5 &sis for Mmax m the East Somc::e.= . :;;**:***

• • • = *:::.: ***:*::=*= ==========:::.:*::.:.

• • • .:. **9:....-::,;

.5 9.3 4J future ~alee Characteristics ....... _ .....................- ....... _ ....................... _. ~..... ..9-10 9 2.6 futuR Eartbqulle Chanictensbcs m~the =:...Eut =!...:S:.::~~ce::...= ** ::::*========= 9-6 93 .5 Southem Cahfonua and Baja Cahforaia ....--------- *--*-... - ... *- - * *- 9-10 93 Se:ismot~mc Altematl\"e ***********- **--*-****-- - -*--*****- *-*****-*..- *- -...-.------- 6 9.3.5.1 Basts for Defimng Sowu. ____... ____ ,______________ .. _____ .. _______________ 9.10 9.3.1 Soulhem B.tsul and Range ..........- ...................- ....--..**-*-**-..... _ ..........................._._.......... 9-6 9.3.5.2 Daw for~ Source Mmax . ......- -.. ----..- -..*---*--**-**-*-**-* .9-10 9.31 1 Basis for Ddining Source ... ***--*--..-**-*--..- -........ --....*****--***--..---............ -* 9-7 93.5J future Earthquake Charactmsttcs ..... -----...- ..- ............ _. _______ ,, ____., ______ ,,,.,._...... 9-10 9.3J .2 Basu for Source Mma:t.....-*-**................._______ ..........- .... - ..............- ............_,, __,.. 9- 7 93.6GulfofCahfonm ... - - * - - * -..- - - - - - - - - - - * - * * - - - - * * - 9-11 9.3 U Future Earthqwke Cbaracterutics .......,_ ---.. - - * - * * - - - - * * - - - ____ ,,_ 9-7 9J.6.l Basis for Ddimng Source .....- .............. _ ..___ ,.................. ________ ,,_______..........._ .51-11 9.3.2 Colorado Plateau_ .....- ...*.. -**---*------*..***-..-*-*---....------*----*----.. *-*-*-*----9-7 93.6..2 Bms for Source Mmax... . ** _ -*****-**..**----*--**--*********-....._..____ .... ..9-11 9.3 2 1 Basts for0e1inmg Source .... _ .......... _, __ ,.____ ....................,.._________ ,,....,_, ____ ... 9-7 93 63 fulUR Earthquake Clw-.c:1rrutia **-*-*-------*-**----*-**----*-----....*----9-11 9.3.2.2 Bws for Source Mm.ax .........- ...........------*-*--*---*-*-*-*-------**..*****----***--*-**---*9-8 9.3.7 Seislnogemc Thickness ............ - .....- .........................................................- ........................... 9-11 9.323 Future Earthquake Cba.rac::tensti~ __ ,..... _.. _, __ ..___ ,________,,,_ ........... _ .... _ .. ____ ,, ___ ,9-8 93.3 Trmsrtion Zone .......................- ............. __ ,, ____ ....... - -............................. - .........- .... 9-8 9.3 3.1 Buis for.Defining Source ......................- .......- ...-*-****-*..**-**...................- ................ 9-8 9.4.1 l R.emo\-al of On.fault E\-ents from the P\J"NGS E~e Catalog .....- ............................ 9-13 9.332 Ba.us for Source Mmu ____ ,,,,.......- ....*------*-**-*--..**-**-***-**--*****---**-**-*9-8 9.-U.1 Cakulanons.._____ *-- -* ---**----****--**---****-*-**---*-*-**---*-*9-13 9.3 3.3 F11ture Eanhquake Characteristics.. . _ .... __ *---.... - *** - -.. *-*-..- .. 9-9 9.3.4 t.lexican Hig)ilands ........ _ ............_ ............. - .........- ....................- ..................... _. _____ ,,_..... 9-9 9 4.1 4 Coonderanon of Constant I>-\'al~ Kemtl l\J'l1f:o.aches ..........- -..* - . *-*

9.-t U C~on io USGS ~aps for the Basm and~ ~gi,~o::;; n ;.;;;

-*;* ====.;;.;.;.;.; "*..-.

9-"""1"""

8 9 4 l 6 Stram-Rate !izq>licanons of RKmrmce Puuneten .. 9-I &

PVNGS PPRP-TI Team Correspondence: .

l.rUh Co1nul1..11nh lnttn1a1ion11l. la(.

1.t.1 1 luurlt<.'j Kl"' u *r *~11

\.du..:Ja. CA QI 355 IMlt ~ll~-'/90't; l.n 16611 ~ '1-W<JU Apnl 17, 2015 Mr. Ronald Gaydos Pro1ect Manager Engineered Equipment & Major Pr()fects Westinghouse Electric Company 1000 Westinghouse Dnve CWH03-410M Cranbeny Townshtp, PA 16066

Subject:

Additional Documentation for the PVNGS SSC Report Mr. Gaydos, Letbs Consultants International, Inc. (LCI) 1s pleased to submit lhtS additional documentation associated with the Palo Verde Nuclear Generabng station (PVNGS) Seismic Source Characterization (SSC) Report (Revision 0, dated February 2015). This additional documentabon satisfies dehverable requirements for Task 1, as descrtbed in Project Impact Notice (PIN) No. 8 for Scope Changes lo the AriZona Public Service (APS) 2.1 Seismic Hazards Evaluation (SHE) Project The U.S. Nuclear Regulatory Commtssion (NRC) requested that APS provide additional information detailing the interactions between the Participatory Peer Review Panel (PPRP) and the Technical Integrator (Tl) Team. The requested information 1s provided m the three attachments that accompany this letter:

• Attachment 1: PPRP Comments on the Protect Plan and TI Team Responses.

• Attachment 2: Formal Correspondence Between the Tl Team and the PPRP.

• Attachment 3: PPRP Comments on the Draft SSC Report and Tl Team Responses Please do not hesrtate to contact us with any questions.

Sincerely, Lettis Consuttants International. Inc.

"?-, '\). \-4~-'-~

Ross Hartleb Project Manager

PVNGS SSC Areal (Background) Sources Areal sources. are Future earthquakes in the characterized with a areal sources are modeled defined: with rupture characteristics

• geometry such as:

• seismogenic thickness

• geometry

• rate of earthquake

• rate occurrence

• slip sense

• Mmax

• magnitude-frequency distribution function

SSHAC Section 8.2: .

PVNGS Areal (Background) Sources

• Two alternative depictions of PVNGRS areal sources

1. Two-Zone (0.2)

2. Seismotectonic (0.8) 1n*w lll'W 117'W 114'W 111"W 108'W

,,,z

l1

~)( 3&'N 0

0 33' N PVNGS SSC SSHAC Figures ES-2 & ES-3

10 Hz Site Specific Rock Hazard at Palo Verde, by Source Type lE-3 ~

~ - ~

-~

Cll v

c Cll

"'O lE-4 f'.. -

~ =

Cll - -- ~ "'!:::

Cll v \f :'.E f: .+

-x Cll 0

v>

c Cll lE-5

- --r--

- ~~

~ -X=+/-

-+-

r"N K ~ ~

,~

- Mean

s - Area

• ' -. -~Iii.

- CT

-- t 1t4- - ....

Cll

~

- Fault 1: .'"" \

IO

s c

lE-6

= = ~ T .r_ = p  := lt lt t T

~-

~ = 1: = I=

ct c

-~ -- - ....

lE-7

~ ~

0.01 0.1 1 10 10 Hz spectral acceleration (g) 1 Hz Site Specific Rock Hazard at Palo Verde, by Source Type lE-3 Cll v f ~~

c

• ~ i Cll

"'O Cll lE-4

,,, t;. + -= -=t= ~

I I

-c:::=

! fl FtiK Cll v

- x Cll 0

~

c lE-5

,- , ~ ~

...._.... _,... ~

...,..... _ - Mean Cll 1: - Area

s .

er I QI

~

IV lE-6 J l --\ ., .t' "1 - Fault

s c

c *-

..._~

ct -\ ~

I  :~

lE-7

. I

\. '\..

0.01 0.1 1 10 1 Hz spectral acceleration (g)

10 Hz Site Specific Rock Hazard at Palo Verde, by Source Sub-category lE-3

~~

~

Cl.I u +~

c:

Cl.I "O

l E-4 '- I~ ~ I

- Mean Cl.I

~ ._,__ '\ ~

Cl.I u

)(

....0 Cl.I

- ,._ ~ ""

i\

... '.- --~

""' ~~~ ~

- seismotectonic area

~

c:

lE-5

- Two-Zone area

-~~

Cl.I

,.._~

J CT ......

Cl.I '-" I l J Greater AZ faults

'\. - ~--1 .... -

"'c:

J lE-6

.. ~- .... , ..-

- California-Mexico c:

ct '"-'-" ' '~~ faults lE-7 0.01 Ql ~"" 1 10 10 Hz spectral acceleration (g) 1 Hz Site Specific Rock Hazard at Palo Verde, by Source Sub-category lE-3

... -- -~,:- -

~") "

Cl.I i - - _,_ ,__

u c: - ,_

Cl.I - Mean "O

Cl.I l E-4 -... - --

"~

~

Cl.I u

)(

....0 Cl.I

-'\~ *~ ~ -~* - Seismotectonic area

> - Two-Zone area

• 111 u l E-5

- '\. ~

c:

Cl.I

J

...... ~~

CT '

1n 111rn Cl.I Greater AZ faults

"' \,,

"' lE-6 - - ...

~ ~

J c:

- - Ca lifornia-Mexico

' ~ ~~

c: I faults ct ........

I l E-7 * '*

' \. l" J 0.01 Ql 1 10 1 Hz spectral acceleration (g)

10 Hz Site Specific Rock Hazard at Palo Verde, by Area Source lE-3 QI Total u - Mean c

QI

'tJ - 5BR QI lE-4 QI u

)(

East QI TZ 0

u lE-5

- GULF c

QI - MH

~

C" QI - west lE-6 - SCABA "c'

~

- CP c

<(

lE-7 +----...:....._____;:......_...:....._...:.........;:......:.._..,---~~-~...:....._...:....._......,.__,__..,-- _ __,,,,.___,......_-'--...:.........;:......:..---l 0.01 0.1 1 10 10 Hz spectral acceleration (g) 1 Hz Site Specific Rock Hazard at Palo Verde, by Area Source lE-3

~ *

~ :*~"

~

QI Total u - Mean c

QI

'tJ QI lE-4

- - SBR

- ~~~~

QI u - East

)(

QI

-~

• TZ 0 -

~ lE-5 - GULF

~~ ~ ~

c QI

~

-- - MH C"

QI -

west SCABA lE-6

"'c - '~~ ~~~ ~-

~

-f~ *,~ ~ ~

~

c

--* - cP I

<(

lE-7 0.01 0.1 1 10 1 Hz spectra l acceleration (g)

PVNGS Areal Source!;: Mmax Assessment Distribution of Mmax values and their weights for the areal sources are based on th1e expert judgment of the Tl Team and informed by

• regional geologic, geophysical, and seismic information,

• previous PSHA studies performed for the PVNGS (REI, 1993; LCI, 2012),

• regional PSHA studies performed by the USGS for the NSHMP (Frankel et al., 199 6, 2002; Petersen et al.,

1 2002,2008,2014),and

• PSHA studies performed fc>r the Yucca Mountain site in southern Nevada (e.g., Wo ng and Stepp, 1998).

1

PVNGS Areal Source!;: Mmax Assessment

• The CEUS-SSC project (EPRI et al., 2012) utilized two alternative approaches

. . for estimating the Mmax distributions for areal se1sm1c sources,

- the Bayesian approach

• Requires a robust database of earthquakes in analogous ("tectonically comparable") crust. The Tl Tearn is not aware of such a database. for actively extending crust and plate. margins that would be applicable ta. the PVNGS. model region, thus the. Bayesian approach was not used in the PVNGS. SSC to estimate Mmax distributions for areal seismic sources.

- the Kijko (2004) approach

• Relies on observed seismicity w ithin a region to provide a direct (or posterior) assessment of Mmax . This approach, however, does not provide stable results when the number of observed earthquakes is low (Kijko, 2004). The Tl Team investigated the applicability of the Kijko approach, but determined that there have been an insufficient number of earthquakes throughout much of the model region to apply the Kijko approach to estimate Mmax distributions for areal seismic sources.

PVNGS Areal Source!;: Mmax Assessment Key uncertainties with the assessment of Mmax for areal seismic sources is the possibility,. and perhaps likelihood,. that Mmax varies. spatially throughout each areal source. In the PVNGS SSC, the broad distributions of Mmax for each areal source. are assumed to be entirely epistemic and applicable throughout the source, such that the diistribution applies. to all locations within that source.

The Tl Team considered whether it is realistic to assume that a single Mmax distribution developed for an. areal source. applies to all locations within. that source, especially given. the. large extents of these sources and the broad range of magnitudes in the Mmax distributions. The Tl Team utilized the available data to the extent possible to identify areas of varying expected Mmax. Spatial variation in Mmax is one of the primary criteria used by the Tl Team to differentiate areal seismic sources, such that if there is a basis for identifying spatial variation in Mmax, that information was used to identify a separate seismic source.

PVNGS Areal Source!;: Mmax Assessment

• Mmax distributions are based on the judgment of the Tl Team, informed by the evaluation of published data, rather than quantitative estimates such as the Bayesian method or the Kijko (2004) approach.

• Mmax distribution for all areal sources ranges from M6.8 to M7.9, except CP ranges from M6.5 to M7.9.

• The uppermost values in the PVNGS Mrnax distribution (M7.5 and M7.9) are representative of the largest known on-fault earthquakes within the western U.S.:

the M7.5 1887 Sonoran earthquake (Suter, 2008a), which is also the largest known normal fault earthquake worldwide (based on rupture length), and the M7.9 1857 Fort Tejon earthquake (e.g., Townley, 1939; Sieh, 1978; Agnew and Sieh, 1978).

Fo r the 2-Zone Areal Sou rces :

Rupture Top of Seismogenk l\ la gnitu de Source Boundary Ruptw*e Ruptw*e ~Imax Rate Dip Rupture Thickness Recurrence

~fame Type ~lechanism 01ientation (-\Iw) Cases (degrees) (km) (km) ~lodel Strike-slip 70° (20%)

(80%) 80° (20%)

90° (60%)

6.8 (0.1)

N35°W (20%) 30° (20%) 12 (0.2) 7.0 (0.25)

Reverse West Leaky (lO~o)

N45°W (60%) 45° (60"0) 0 15 (0.6) 7.2 (0.4)

N55°\V (20%) 60° (20° ~) 18 (0.2) 7.5 (0.2) 7.9 (0.05) 35° (20%) 1 (OA)

Normal (lO~o) 50° (60%~ 2 ~0.4) G-R ( LO) 65°(20% 3 02) 35° (20%)

Normal N20°E (100/o) 50° (60%) 6.8 (0. 15)

(80°0)

N-S (100~) 65° (20%) 12 (0.2) 7.0 (025)

East Leaky N20°W (400/o) 0 15 (0.6) 7.2 (0.35)

N40°W (200~) 70°(20%) 22 (0.2) 7.5 (0.2)

Strike-slip Random (200/o) 7..9 (0.05) 80° (20%)

(20°~)

90°(60%)

PVNGS SSC SSHAC Table 9-1

Top of Seismogenk  :\Iagnitude Source Boundary Rupture Rupture Rupture  :\hnax Rate Rupture Thickness R ecmTence For the l'\ame Typ e )lechanism Orientation Dip (km) (km)

()Ill") C ases llodel Seismotecton ic Strilre-shp 700 (200/o) 800 (20%) 6.8 (0.15)

(90%) N35°W (20"~) 12 (0.2) 7.0 (O 25)

Areal Sources: SCABA Leaky N45oW (60°'.)

90° (60"/o) 0 15 (0 6) 12 (0.4)

N55°W(20'lo) 300 (20" o) 18 (0.2) 7.5 (0.15)

Reverse 45° (60%) 7.9 (0.05)

(10%)

600 (20"/o)

Strike-slip 700 (20"'.)

(W~.) 800 (20%) 6.8 (0.05)

N35°W (20'lo) 900 (60" '.) 12(0.3) 7.0 (03)

GULF ual")' N45°W (60"*) 0 14 (0.6) 7.2 (03)

Normal N55°W {20'lo) 35° (20"'.) 16 (0.1) 7.5 (03)

(30%) 500 (60"/o) 7.9 (0.05) 65° (20"/o) l {OA) 2 (0.4) G-R ( l.O) 35° (20"/o) 3 {0.2)

~ormal 500 (60"/o) 6.8 (0.1)

(80%) ~S(Ne) 65° (20'l'O) 12 (0.2) 7.0 (0.25)

N2()0W (400/o)

SBR Leaky 0 15 (0.6) 7.2 (0.4)

N400W (200'.) 70° (20"/o)

Strike-slip 18 (0.2) 7.5 (02)

Random (2001.)

800 (20%) 7 9 (005)

(20'lo) 900 (60"'.)

35° (20"/o)

Normal 50° (60"'o) 6.8 (0.05)

(80%) N-S('.?OO*)

65°(20%) 12 (O.l ) 7 0 (0.25)

N20"W (400.4)

MR Leaky 0 15 (0.6) 1.2(03 5)

N4<>°W (2001.) 700 (20010)

Strikr-slip 18 (0.3) 7 5 (03)

Random <2°'*) 800 (20"'0)

(20%) 7 9 (0.05) 900 (60%)

35° (20'"1o)

Normal 6.8{0.2) 50° (600fo)

(700,'.) N20°E (25~o) 14 {0.2) 7.0 (025) 65° ('.?00/o)

TZ Leaky N20°W (25°~) 0 17 {0.6) 7.2 {03) 700 (2001.)

Strike-slip Random (50"/o) 20 (0.2) 7.5 (0.2) 800 (200'.)

(300*) 7.9 (0.05) 900 (60"!.)

35° (2~.)

~ormal 6.5 (0.2) 500 ( 6Qo,.)

(80"'o) 15 (0-2) 7.0(03) 65° (2~*)

CP uaky Random (1OO'lo) 0 20 (0.6) 12(0.25)

PVNGS SSC SSHAC Strike-slip 700 (200/o) 25 (0.2) 7.5 {0.2) 800 (200 o) 7.9 (0.05)

(200/o)

Table 9-2 900 (600/o)

PVNGS Areal Source~s: Zone Boundaries.

Primary criteria for defining and differentiating areal seismic sources include changes in Mmax potential, seismogenic tt1ickness, and/or future rupture characteristics (fault orientation, fault style, etc.) between volumes of crust.

All areal source are mod eled as leaky," modeled 1

earthquakes that initiate~ in one areal source are allowed to rupture beyo1 nd the boundary into the adjacent areal sourc~e or sources.

PVNGS Are~al Sources:

Frequency Model & Earthquake Recurrence

• Recurrence of future earthquakes for all areal sources in the PVNGS SSC is treatecJ as a truncated exponential distribution (Gutenberg-Richter) with spatially variable parameters (a- and b-valuE~s).

• As discussed in the earthquake catalog development, each individual earthquakE~ in the catalog is. expressed in. the form of the expecte(J magnitude, E[M], and an equivalent count, N*.

• Using. these. two quantities in the recurrence calculations are used to pnoduce earthquake rates (a) and b-values.

PVNGS Areal Sources: Smoothing Three conceptual models for the spatial variation of v recurrence rate per unit area (v) within an areal source.

~ uniform sei:smicity; not used by PVNGS.

v

~variable, but continuous and relatively smooth seismicity; used by PVNGS. Became standard model of current practice (e.g., CEUS SSC). two general approaches have been used :

1. penalized maximum likelihood approach (EPRI ,1988 and CEUS SSC) where epistemic uncerta inty can be readily incorporated into the PSHA; used by PVNGS v 2. Gaussian "kernel" function to calculate the rate at any grid point as 1 *

• a distance-weighted sum of the earthquake counts within the areal

• t source (Frankel 1995)

~ assumes earthquakes can occur only at some discrete geographic locations within the large areal source; not used by PVNGS.

PVNGS SSC SSHAC Figure 8-2

Section 8.2.4.2 - Fornnulation of Penalized-Likelihood Model for R~ecurrence Parameters

1. Divide source zone into 0.25 degree cells

2. Formulate Poisson likelih(Jod function in each cell (depends on vi, bi, and earthquakes within cell)

3. Introduce penalty function that discourages large changes in v or b betwee11 adjacent cells

4. Introduce prior distributic)n of b (discourages solutions where b value dliffers from regional b)

5. Generate many realizations of joint penalized likelihood function ~ joir1t distribution of vi, bi, and smoothing parameters for all cell in zone

6. Generate representative 1maps from penalized likelihood functions

Section 8.2.4.2 - Fornnulation of Penalized-Likelihood Model for R~ecurrence Parameters

• Inputs to Recurrence c:alculations

- Weights to magnitude bins

- Prior distribution for b

- Priors for smoothing parameters

- Number of iterations

PPRP WS3 Comment #3: Pre~paration of a succinct narrative and more complete documentation of the analytical tool called "smoothing" during the meeting should be considered so that reviews of the analyses can be carried out in an informed and efficient manner.

Tl Response to PPRP WS3 Cc>mment #3: The Tl team acknowledges the hazard si~~nificance of the. decision. to calculate earthquake recurrE~nce parameters for areal source zones using a smoott1ed seismicity approach.

The Tl team also understancls that the soothing approach used in the project is a complex procedure that may not be well understood by all readers of the SSC report. For these reasor1s, a complete and thorough documentation o1f the smoothing process and assumptions will be pr<>vided in the SSC report.

PPRP WS3 Comment #4: Some. uncertainty in the nature of the M>4.65 seismic events mapped on the west side of the Southern Basin and Range province just east of the Gulf of California was noted in the rneeting. Conducting a review of the earthquakes comprising these events sho1uld be considered to determine if they are located on land or are associated with faulting within the Gulf. If they are pre-instrumental (or otherwise poorly located), efforts could be made to reposition the events.

Tl Team Reponse to PPRP WS3 Comment #4: The Tl Team. reviewed the portion. of the project earthquake. catalog. directly east of the Gulf of California, where. an approximately triangular wedge of seismicity appears to taper off into the Southern Basin. and Range. In order to assess the lik1elihood that: (1) the. project catalog correctly reflects a region of elevated seisrnicity rate along the western border of the Southern Basin and Range; and (2) the catalog correctly locates Mw > 4.65.

earthquakes in this region, the Tl Team reviewed the age, location uncertainty, and.

magnitude type of these earthquakes.

"'62 earthquakes. in the. project catalog are~ identified in this area, 8 occurred prior.

to 1950. The majority of these 8 earthquakes are based on instrumental data, such that only 2 earthquakes are reported with intensity-based (MMI) magnitudes. The Tl team assumes that the more recent 1963, 1969, and 1981 earthquakes are relatively well located, however, and should not be repositioned . Given this assumption, it is difficult for the Tl team to justify repositioning the 1935, 1952, and 1958 earthquakes. Therefore, the Tl team does not plan to reposition any of the earthquakes in this area. From these observations, the Tl Team assumes that the project catalog correctly reflects a region of elevated seismicity along the western border of the Southern Basin and Range.

PPRP WS3 Comment #6:.

(1) Usefulness should be reconsidered of the two-zone model (2) Reexamine. the positions of zone boundaries in the seven-zone model

• Include a possible extension of the transition zone to the west to include faults. and seismicity that are concentrated between the 320- and 400-km radii.

• Absorb the ETR zone. into the SCAB~~ zone

• Result would be 6-zone model that 'Nould be either the only model. or. the dominantly weighted model Tl Team Response to PPRP WS3 Comment #6:. The Tl team generally agrees that the geologic data suggest the two-zone model may be unrealistic, but this alternative is included to capture the range of technically defensible interpretations. Going forward, the Tl team will further evaluate the need for the two-zone alternative.

The Tl team agrees that the Eastern Transverse Ranges (ETR) zone is unnecessary, based on discussions with the PPRP and on hazard sensitivity results presented at Workshop #3. The Tl team likely will combine the ETR zone into the adjacent Southern California and Baja (SCABA) source zone.

The mathematics of the P'VNGS approach are the same as those used in CElJS SSC and are contained in PVNGS SSC SSHAC Section 8.2.4 and CUES SSC Sections 5.3.2, 6.4, and 7.,5.

Details on the application of this approach to the PVNGS are_contained in P~VNGS SSC SSHAC Section 9.4, where they discuss results of the recurrence calculations and the specific recurrence parameters for the areal seismic sourc:es. in the PVNGS SSC.

9.4.1.1: Removal <)f On-Fault Events from the PVNGS Earthquake Catalog

• Areal sources model the occurrence of future earthquakes not associated with identified active fault sources, so the recurrence in each areal source relies on a modified version of the PVNGS catalog in which on-fault earthquakes have been removed. Avoids double counting.

Earthquakes of M>5.5 were judged to be on-fault events if they produced surface rupture, or if published studies associated the. earthquake with a fault based on other data (e.g., seismological data, historical accounts).

Exception AZ earthquakes: Proponent Expert, Philip Pearthree, at WS2 asserted that no earthquakes within AZ can be definitively associated with known faults, with the possible exception of the 1992 ML 5.8 St. George earthquake (Christenson, 1995). The 1992 St. George earthquake occurred in southernmost Utah near the active Hurricane and Washington faults and may have been generated by slip on one of those faults (Pearthree and Wallace, 1992; Black et al., 1995). The St. George earthquake did not rupture the surface, however, so the causative structure is not known with certainty (Black et al., 1995).

Identified a total of 21 earthquakes that could be associated with identified active fault sources. Only the West, SCABA, and GULF areal sources are affected by this modification to the PVNGS catalog.

11e*w ns* w 9.4.1.1: Removal of On-Fault Events from the PVNGS Earthquake Catalog 20'014'4

7. 19 **

19~11?13 1 1 ~34112131 71 6.4& 191511112~ . "' * " ..

s 6 tq101 t 2n t 961> 811

~ 79 6.;3 f igutt 9-10. Eaitbquakrs remc>\-ed from~ P\~GS catalog for I.be~ of calculating re.curmice U1 areal soutces (Section 9.4 1) Earthquake epicenters are represented by red dots. associated surfucc ruptures (fromFigtn +14) att tngbliglued in red along !he causative &ult. The ma~ ofthr 193.i 11131 Cerro Pneto earthquake is taken from.Andet'son and BodU1 (1987).

9.4.1.2 Calculations - Recurrence Maps Tablr 9-4. !vlaguin1de-dependent "'*eights for east0FB (SCABA. GULF. and West) source zones.

western

\12.67- M 3.33-  :\1 4.0-  :\14.67- M 5.33-  :\1 > 6.0 Casr 3.33 4.0 4.67 5.33 6.0 1 (0.4) 0 0.8 1 1 l 1 2 {0.4) 0 0.3 1 1 1 1 3 (0.2) 0 0.2 0.5 1 1 1 Tablr 9-5. Magnitude-dependent weights for western (SBR. TZ. CP.1IH. and East) source zones.

eastern M2.67-  :\13.33- M4.0- :M 4.67- M 5.33-  :\I > 6.0 Ca st>

3.33 4.0 4.67 5.33 6.0 1 (0.4) 0 0 0 1 1 1 2 (0.4) 0 0 0 0.5 1 1 3 (0.2) 0 0 0 0.3 0.5 1

• 3 cases that were selected for the weights to the magnitude bins

• Weights are different for Eastern and Western seismic sources, b/c the catalog is very different in magnitude coverage and completeness between these two regions.

• Reasons for 0 weights:

• Stepp completeness analysis showed large spike in activity during passing of TA array suggesting low magnitude bin not complete before then

• magnitude-recurrence behavior of the data may deviate from exponential

• magnitude-conversion models or completeness models may be less reliable for lower magnitudes

9.4.1.2 Calculations - Recurrence Maps

• The mean value for the prior distribution of~ was set to 1.84 for all sources; corresponding to the regional bprior =0.85 (based on USGS)

• Strength of the prior distributiion of~ is specified by the standard deviation, where a large value indicates a weak prior.

ab:

- Eastern sources

• Case 1 crb=0.6

• Case 2 & 3 crb=0.7

- Western sources, crb=0.4

• Smoothing parameters were determined such that for each source zone, and for each of the three cases, the. smoothing parameters o 11v and a 11 ~ were allowed to vary, allowing the catalog data to determine the optimal range of values (described by the mean and standard deviation).

9.4.1.2 Objectively determined smoothing (roughness penalty) parameters+ one-sigma range Eastern Sources Case 1 (0.4) Case 2 (0.4) Case 3 (0.2) f f

I .,- -!- -

I i I f i

"i!

E

~ 0.01 b

aflv 0.001 ._____.__ _ . __ _ _- - I 1 o - - - - - - - - - - 4 .___.__ ____.__-.----------

r:

' ~ ~  ! ~

Source Zone Source Zone Sourcrt Zone 0.1 . , . - - - - - - - - . .

f

~ t t f i t II I

001 i 1~

f

...,s0 1~

1-1 0.001

~

0.0001 1 - - . . . . . . - - - - - . - - - . . - - - - -

l.L 0

.L

l 1

Ill Source Zone Sourco Zono Source Zone smaller ~ larger PVNGS SSC Figures 9-11 to 9-13

9.4.1.2 Objectively determined smoothing (roughness penalty) parameters+ one-sigma range Western Sources Case 1 (0.4) Case 2 (0.4) Case 3 (0.2)

"C" 1"' 0.1 "l

Iii E

~

,. 0.01

~ '

a~v 0.001 ii

~ ~

~ '.J I

J :l 8

0 Source Zone "' Source Zone Source Zone 0.1

.s::.

t 8E I\

0.01 t i Cl)

~

.!?.

a 0.001

~

O~p 0.0001 a u

~ ~

~

~

rn

~ (5  ::>

Cl I Source Zone Source Zone "' Source Zone smaller ~ larger PVNGS SSC Figures 9-14 to 9-16

' ' ~* 1::111 9.4.1.3 Resulting Race(M>5.0)/deg1/yr b value

_...__..__.___,_.....____.__.___. Recurrence Parameters Ca e 1 ( .4) and Maps Maps of mean recurrence rate (M>S) and b-value for Seismotectonic sources.

...... The right panel also displays the M>2.67 earthquakes in the PVNGS catalog.

Ca e 2 ( .4)

JI~ "

.2)

PVNGS SSC Figures 9-17 to 9- 19, 9-23

W *A I .If:

R:ite{M>5.0)/deg1/yr

' .. I to

• 9.4.1.3 Resulting

,.......__._----~~----1 Recurrence Parameters Ca e 1 ( .4) and Maps 1"'°' ...

C' Jt.:'

Maps of mean recurrence rate (M>S) and b-value for Two-zone sources.

t1'1 '"'

The right panel also displays the M>2.67 l~S ,, ' ~_.____.____.___.__.._____.._ . earthquakes in the PVNGS catalog.

Ca e 2 ( .4)

'=-' 1: ~

" ;llT1'o Ca e 3 0.2)

PVNGS SSC Figures 9-20 to 9-23

... 11..... """ """ u " llG'" ........

'- ~ " I *~ lJ :tu 1 9.4.1.3 Resulting

,__..__....____._...____.___._._----I ,... .., ,_....____.____.__.._____.____......__.___ Rec urre nce Pa ramet ers Cas 1 ( .4) and Maps Map of coefficient of variation (CoV) in earthquake recurrence rate (M>2.67) and sigma b-value for Seismotectonic sources (statistical uncertainties high Cas 2 ( .4) where there are few earthquakes) ur*w tu*w

.. ~'

Cas 3 ( .2)

• ~ '6~

PVNGS SSC Figures 9-23 to 9-26

Ir U1 9.4.1.3 Resulting CoVf R:m:(M>2.67)1 Sigm:i(h v.llue)

Recurrence Parameters Ca e 1 .4) and Maps Map of coefficient of variation (CoV) in earthquake recurrence rate (M>2.67) and sigma b-value for Two-zone sources (statistical uncertainties high where 1.\"'-oj '8'-S there are few earthquakes)

Ca e 2 .4)

Ca e 3 0.2)

PVNGS SSC Figures 9-23, 9-27 to 9-29

. .. ... 1\1'11 '" I ~ t 111 IJO 9.4.1.3 Resulting Rate(M>5.0)/<lcg2/yr b value Recurrence Parameters

.., I '

and Maps Two realizations of rate (M>S) and b-value for Seismotectonic sources. Magnitude weights according to Case 1.

PVNGS SSC Figures 9-23, 9-30, 9-31

l* I , ,. I 1, I u >it ! n,n tJilll II b value

• • 110 9.4.1.3 Resulting Rate(M>5.0)/dcg:/yr Recurrence Parameters and Maps t11it

~* lh'11 llA ll II' ' II ll Two realizations of rate (M>S) and b-value for Two-zone sources. Magnitude weights according to Case 1.

PVNGS SSC Figures 9-23, 9-32, 9-33

Comparison of model-predicted earthquake 1000 counts for SBR source zone (host zone) r------------~~=====:i ~--------------~

• C;Uog Case 1 (0.4) - Reaoi:nbon I Case 2 (0.4) - -1

-2

-3

- Roelt:auon 4

-~'

- Rooli:ntal s - ReolizabJR 5 i' - -s - ~e

~

~ 10 -.t--------~~-----i - ~7

- -7

~

ro

- -e RllGllclllon8 cr

..r:

t!

ro w

0 0

z 0.1 - - - - - - . - - - - - - - - - . . . . . . - - - - - - - - . - - - - - . - - - - - - - - - - . - - - - - - . - - - - . . . . . . - - - - - - - . - - - - - - . - - - - - - - - !

2.9 3.6 4.3 5 5.7 6.4 7.12.9 3.6 4.3 5 5.7 6.4 7.1 Magnitude Magnitude 1000 ~--------------~

• CilalOO Case 3 (0.2) - Reolcatm 1 The error bars are the 16%-84%

100 uncertainty associated with the data.

- Reablon4

~

- Reaizallcn 5 Predicted counts are obtained by summing

- Rleabltlon 6

~

~

~ 10 - ~7 the rates by bin for all cells within the Ill

- Rlmbtlcn9 er source taking into account the ro w

z 0

0 completeness times for each cell.

0.1 + - - - - - - - - . - - - - - - . - - - - . . . . . - - - - - - - - - . - - - - - - - . . - - - -

2.9 3.6 4.3 5 5.7 6.4 7.1 Magnitude

Comparison of model-predicted earthquake counts for East source zone (host zone)

Case 1 (0.4) 1000 , , . . . - - - - - - - - - - - - - - - . . , - - - - - - - - - - ,

• CainlOQ ,

. , , - - - - - - - - - - - - - - --=i Case 2 (0.4

- ~1 C>lalcg 100 -3 - Reallzatlon 3

- R-4 - Reammn 4

E

- -s - RIWzlllion 5

- Reali:alm 6

~

10 - -7 - Rmlclllcn 7 ca

J r:r

- -8

.c t::

ca w

0 0

z 0.1 ------.---------.-----.--------.-----4 --- - - - - - - - - - - - . - - - - . . . . - - -

2.9 3.6 4.3 5 5.7 6.4 7.1 2.9 3.6 4.3 5 5.7 6.4 7.1 Magnitude Magnitude 1000 ~--------------~

• Czlloo

- -1 The error bars are the 16%-84%

ReallZalOn 2 100 uncertainty associated with the data.

i' Predicted counts are obtained by summing

~

10 - Reaiz:lboo 7 the rates by bin for all cells within the

.x ca

J r:r

.c source taking into account the t:

ca completeness times for each cell.

w 0

0 1

z 0.1 ------.----------~-~--~------!

2.9 3.6 4.3 5 5.7 6.4 7.1 Magnitude

Rate {M>5)/deg2/yr 0

0 0

0 0 U)

~ ~

~

I I I

• I I I I USGS IKerael 1 r+

~

~

U1 I I II c(./') m n

• I usus Rates WitJiJlf loor G)

(./')

OJ 0

• I I I I I 11 QJ (J) 3 USGS Rates ~ith Fl'oo112 (,S-CA~IZ filoor /2) -* -0

l QJ QJ

l (J) c.. 0 I I I I I

Sei shi\otecton i c SBR c*ase 1, mean 1 1 l  :::0 QJ

l r+

0

- I I

SeismotectoniGSBR case 2 *me,an 1

I II :l C1Q m

c V>

I I I I  :::0 G)

• :s.e*i~ot1et"toni~ SBR Cas~ 3 nte,an1

-0 zG) m V>

C1Q s:

T *- ---*

0-*

(./')

• I I I III (./')

(./')

n QJ

~Z1ooe 1 E~sl Case 1 mean :l -0

• - ~-

II (J)

• I I I Z..Z!o oe E:ast !Case 2 mea~

~

I I II

,0

• I I 2*Z1ooe East 1Case 3 rmeani

PVNG Implementation of SSC SSHAC -

2 simplifications to area seismic sources (1) collapsing the rupture orientation branch to the central value, and (2) modeling fault dips as vertical

• APS stated

- For non-host sources, these simplifications were used, b/c non-host sources are minor contributors to the total 10-4 hazard at 1 Hz SA and the insensitivity of hazard to rupture orientation and fault dip was confirmed by performing a rupture orientation sensitivity using the SBR source {LCI, 2015a).

- Host sources are. sensitive. to. dip. and crustal thickness,. so. the difference.

in ground motion between the SSC fault dips and seismogenic thicknesses, and a vertical dip and single crustal thickness, was taken into account in the host sources by adjusting the ground motion for a vertical fault to the ground. motion for a non-vertical fault with multiple.

down-dip widths. The adjustment was calibrated to achieve accurate hazards at MAFEs of 10-4 and 10-6 {LCI, 2015a).

Simplification: (1) collapsing the rupture orientation branch to the central value 10 Hz Reference Rock SBR Hazard Sensitivity to Fault Orientation lE-3 al u

- - ~

c ""-........._

al

"'t:J t

u lE-4 '-

x al

~

- N-S 0

> r- ~ -....

~

al

J D"

al lE-5

_,, - N20W LL.

- N40W ni lE-6

\.

s *~

c 1 c 'Ill

<(

lE-7

\

0.01 0.1 1 10 10 Hz spectral acceleration (g)

Simplification: (2) modeling fault dips as vertical 10 Hz Reference Rock SBR Sensitivity to Fault Dip/Thickness

- 35 degrees 12 km Cl> - - 35 degrees 15 km u

c

~ lE-4 - t - - - r - - - . - - - - . - - - . - - , - ; - - . - r - - t - - - - ; - --rollk---r-7-i--r-:--t----r---r----7---i--:-r--.--rl - - -* 35 degrees 18 km Cl>

u

)(

Cl> - so degrees 12 km 0

u lE-5 -t---T--t----;--t---t-i--+-1---t---T--i---.--i--:--+~

c Cl>

J C- ---* 50 degrees 18 km C>

"' - 65degrees12 km E 1e-6 -j--~=====t==t=:i=i::::t=;=t -----:t==:t==::;=:t='f~=tt:-~~-~=Fr:R9=1:=T.I c -

• 65 degrees 15 km ct

- --* 65degrees18 km lE-7

• 0.01 0.1 l 10 - 90 degrees 15 km 10 Hz spectral acceleration (g)

Corrected for in the host sources by adjusting the 90° normal rupture ground motion to a dipping rupture ground motion, using: A2 = A1

• exp(C1+ C2

• ln(A1 )) (Eq'n 11-2) in which, A1 is the predicted median acceleration, A2 is the adjusted median acceleration, and C1 and C2 are adjustment coefficients.

Simplification: (2) modeling fault dips as vertical 10 Hz Reference Rock SBR Hazard, Modeled vs SSC Model 5

I:

1!-A Ii~

- SBRSSC

~ Mod'et i-o Wetstited lE*S

~

c cu

J CT - SBR: Modeled

-GI I;;

I c

1£*6 c

1<

~l 1 10 10 H-z spectral acceleration (g)

Corrected for in the host sources by adjusting the go0 normal rupture ground motion to a dipping rupture ground motion, using: A2 = A1

• exp(C 1+ C2

• ln(A1 }) (Eq'n 11-2) in which, A1 is the predicted median acceleration, A2 is the adjusted median acceleration, and C1 and C2 are adjustment coefficients. The coefficients were calibrated by performing sensitivities for the expected difference in hazard at MAFEs of 10-4 and 10-6 between the SBR source mean hazard calcu lated using dips described in the HID (Appendix F) and the corresponding hazard calculated using a goo dip. These adjustment coefficient s were used in the final hazard calculations.

PVNGS - Recurrence Rates &

Smoothing

• recurrence of future earthquakes in each area source is treated as a truncated exponential distribution (Gutenberg-Richter) with spatially variable parameters based on the smoothing of observed seisn1icity

• smoothing approach used is the penalized maximum likelihood. approach. that was implemented by the CEUS-SSC Project (EPRI et al., 2012).

• Activity rates and b-values calculated for area sources using assumptions on spatial smoc~thing of parameters and on interpretations of historical earthquakes. This process resulted in activity rates (for M>S) and b-values for each 0 .25 degree cell, for each area source used in the hazard calculations

PPRP Commen1ts and Tl Team Resolution wrt WSO (Kick-off Meeting)

PPRP. Comment #3: We concur with the benefits of having a recent SSHAC Le\lel 2 seismic source characterization for the P\/NGS site. However, care will need to be taken to a\loid the occurrence of anchoring (e.g., cognitive lbias). The Project Plan (p.

2) provides a procedure intended to address this subject using self-evaluatiions, but does not clearly include independent persJJectives that could identify a condition of bias. We would appreciate your informing us of how 'VOU plan to obtain independent views of any possible bias on the Team's part.

PPRP Commen1ts and Tl Team Resolution wrt WSO (Kick-off Meeting)

Tl Team Response to PPRP Comment#~!:

• The SSHAC Level 3 Tl team includes members who did not participate in development of the SSHAC Level 2 SSC. Specifically, PTI William Lettis and Tl team member Ross Hartleb did not participate in development of the SSHAC Level 2 SSC and thus bring fn~sh perspectives to the ongoing SSHAC Level 3 SSC effort. Gabriel Toro acted as a hazard analyst for the SSHAC Level 2 PSHA, but he did not participate in the development of the SSHAC Level 2 SSC. As such, Gabriel Toro also brings his fresh perspective to the SHAC Level 3 SSC Tl team.

• Discussions of cognitive bias will be included at the start of each workshop and working meeting by the PTI or Tl Lead. Moreover, if apparent cognitive bias arises at any point during a workshop or working meeting, the Tl Lead or other Tl team members or staff will be responsible for alerting the Tl team.

• Continual review of SSC development will be performed by the PPRP for the duration of the project. The Tl t~~am expects that the PPRP will alert the Tl team of any perceived cognitive bias at any point during the project.

PPRP Commen1ts and Tl Team Resolution wrt WSO (Kick-off Meeting)

PPRP. Comment #4: It was helpful for the PPRP to have participated via conf~~rence call with the

[GMC] team members during the SSC meeting.

There was the appearance~ of a gap in communications regardin~~ GMC-SSC interface items that came out in Norm Abrahamson's discussion. It is advantageous to have Tlhomas Rockwell of the SSC-PPRP also serving on the GMC-PPRP for the Project to help assure good coordination. Even so, we would appreciate your informing us of how you intend to maintain an effective interface between the GMC and SSC aspects of the PSHA.

PPRP Commen1ts and Tl Team Resolution wrt WSO (Kick-off Meeting)

Tl Team Response to PPRP Comment #4:

An effective interface between the SSC and GMC efforts will be maintained by the following:

• In addition to his role as PPRP member for the PVNGS SSC project, Thomas Rockwell also serves as a member of the PPRP for the SWUS GMC project. As such, he will b1e able to provide information and coordination between the SSC and GMC projects.

• The Project Plan defines the role of t he PTI as a technical expert responsible for ensuring coordination and compatibility between the SSC and GMC Projects. William Lettis is the PTI for the PVNGS SSC project and, therefore, is responsible for maintaining effective communication between the SSC and GMC projects.

• Members of the PVNGS SSC project also serve as members of the SWUS GMC project. Specifically, PVNGS SSC hazard analyst Robin McGuire is the Pa lo Verde PTI for the SWUS GMC project. Thus, he wi ll attend all PVNGS SSC and SWUS GMC workshops, be informed of SWUS GMC Tl team deliberations, and provide an important interface between the PVNGS SSC and SWUS GMC projects. Likewise, PVNGS SSC hazard analyst Melanie Walling serves as the Palo Verde hazard analyst for the SWUS GMC project. In this role, she attends all SWUS GMC workshops and working meetings. Thus, Melanie Walling wi ll be able to provide to the PVNGS SSC Tl team her first-hand knowledge of discussions and activities of the SWUS GIVIC project , and vice versa.

• PTI William. Lettis and Tl Team Lead Scott Lindvall. attended and presented at SWUS GMC Workshop. #1, which was held on March 19-21, 2013. This workshop also was attended by PVNGS SSC hazard analysts Robin M cGuire and Melanie Walling.

• Hazard analyst Melanie Walling and Tl team member and Project Manager Ross Hartleb participate in weekly status conference calls for the Palo Verde Seismic Hazard Evaluation Project. These conference calls also include participants from [APS] and Westinghouse Electric Company. The purpose of these calls is to discuss project progress, schedule, and SSC-GMC interfaces.

Backup slides PVNGS Area Source - Two-Zone Alternative 120" w 117'W ,, ... w 1n* w 1ot*w

PVNGS Area Source - Seismotectonic Alternative.

120* w 117°W 114* w 111° W 1oa* w 36° N 33* N 30° N

Section 8.2.4.2 - Formulation of Penalized-Likelihood Model for Recurrence Parameters n(m ~ M < m+dm ) =ATvj3e-f3(m-moJ [Aki, 1965}

n =.#of earthquakes b/t magnitude m & m+dm A= area of source T = duration of complete catalog (years) v = rate/unit time/unit area for earthquakes with m~m 0 (PVNGS m 0 = 3.0)

~=slope of the exponential magnitude-recurrence law (i.e., the b-value times ln[lO])

Section 8.2.4.2 - Formulation of Penalized-Likelihood Model for Recurrence Parameters Likelihood function takes the form:

f(v l 1l

/3) = IJ(v ATpe-/J(mi-mo))exp - vA T/Je-P(m-m,i)dm J i 1 v =rate/unit time/unit area for earthquakes with m>m 0 (PVNGS m 0 = 3.0)

B=slope of the exponential magnitude-recurrence law (i.e., the b-value times ln[10])

A= area of source T = duration. of complete catalog (years)

N = number of such earthquakes Likelihood function indicates the degree of consistency between the parameters that one. wants to estimate (in this case,. v and B) and the available data (in. this.

case, the number of earthquakes and their magnitudes during a time period T)

Section 8.2.4.2 - Formulation of Penalized-Likelihood Model for Recurrence Parameters N

£(v,f3) == (vAT) Ne-vAT x IT (f3e- /J(m;-mo) )

i ==l Apply simplification, because T depends on magnitude, and separate the likelihood function into a product of factors that depend on the rate v and factors that depend on the exponential slope~

Section 8.2.4.2 - Formulation of Penalized-Likelihood Model for Recurrence Parameters N

VML =

AT The separability implies that the maximum-likelihood estimates of v and ~are decoupled. In particular, take the logarithm of the previous expression, differentiating with respect to each parameter, making the result equal to 0, and solving for the parameter, the results are well-known eq'ns for maximum-likelihood estimates of v and~ [Aki (1965) and Utsu (1966)].

9.4.1.1: Removal of On-Fault Events from the PVNGS Earthquake Catalog YE'ar .Month Day FauJts RupturE'd Ea11bquake :Name Mw 1890 2 9 San Jacinto or Elsinore fault 6.8 1892 2 14 Laguna Salada fau lt Laguna Salada 7.3 1899 12 25 San Jacinto fault Christmas day 6.7 1906 4 19 Brawley fault 6.2 1915 11 21 Laguna Salada fault 6.6 1918 4 21 San Jacinto fault San Jacinto 6.8 1934 12 31 Cerro Prieto fauh 7.1

• 1940 5 19 Imperial fauh El Centro 6.9 1947 4 10 Manix fault Mani,-x. 6.5 1954 3 19 San Jacinto fault Arroyo Salada 6.4 1956 2 9 San Miguel fault San *Miguel 6.8 1966 8 7 CeITo Prieto fault 6.3 1968 4 9 San Jacinto fault BoITego Mountain 6.6 1976 12 7 Cerro Prieto fault 5.8 1979 3 15 Homestead Valley and Homestead Valley 5.5 Kickapoo faults 19 9 10 15 Imperial Brawley. and Rico Imperial Valley 6.51 faults 1987 11 24 Elmore Ranch fault Elmore Ranch 6.5 1992 6 28 Johnson Valley. LandeJs. Landers 7.28 Homestead Valley. Emerson.

and Camp Rock faults 1999 10 16 La,ic Lake and Bullion faults Hector Mine 7.12 2010 4 -t Sierra Cucapah fault system El Mayor-Cucapah 7.3 2010 6 15 Elsinore or Laguna Salada 5.8 fault

PVNGS SSC SSHAC Report: . c nAPTER 6: c ATAL oG oF L'I>EPE1'-i:DE.'i-r EARrnQc AKEs FOR PsHA ............................ 6-1 6.1 Catalog Region and Paranieter Lwllts ................................................................................................. 6-1 62 Earthqu ake Data Sourc:es ..................................................................................................................... 6-2 vi PVNGS SSC, Rev 0 6.2. 1 AISN Catalog ............................. ******-* ............................. **-**-- ---**********-*** ...................................... 6-2 6.2.2 SCSN Catalog ................................................................................................................................ 6-3 6.2.3 RESNOM Catalog.****--*-*********-********-**-*******-*******************-********************-*******-*********-*-*** .................. 6-4 6.2.4 Unified Earthquake Catalog of California (Unified CA) ............................................................... 6-5 6.2.5 ANSS Catalog ................................................................................................................................ 6-6 6.2.6 UCERF3 Catalog ........................................................................................................................... 6-7 6.3 Regional and National Catalog Selection and Standardization............................................................ 6-7 6.4 Remo\*al ofDuplicates ......................................................................................................................... 6-7 6.5 Standardize Earthquake Size Measure ................................................................................................. 6-8 6.6 Catalog Declustering ............................................................................................................................ 6-9 6.7 Magnitude Uncertainty ...................................................................................................................... 6-11 6.8 Catalog Co1Dpleteness ........................................................................................................................ 6-11

SSHAC Chapter 6: PVNGS SSC Independent Earthquake Catalog I::>" W 11r w t H' W 111' W 11111' W PVNGS Catalog

~ ,. .,., UT"H

~\ ~/,. z m E(MJ

~,,,' *""i>"o"f ~

... \ ~ c  ! 2 70 and < 3 00

~\ ~ r:I

)< 36' N

'Q. \

\

n 0 *  ! 3 00 *nd < 4 00

? ' 00 and< 500

\

~ 5 00 and < 8 00

\

\

! 6 00 and < 7 00

• 33' N e ?700 note ont-11ndepenclef\I events shown PVNGS SSC SSHAC Figure 6-9

• Eastern portion of the model region, earthquakes included in catalog if M > 2.7.

• Southern CA, N Baja CA, and Mexico earthquakes included in catalog if M > 4.7.

• Black, dashed "East-West divide" line shows line over which the criteria for inclusion in the catalog changes ..

SSHAC Chapter 6:

PVNGS SSC Earthquake Catalog

• PVNGS catalog is in SSC SSHAC report Appx E

• Tl Team used

- the catalog to calculate recurrence parameters for areal sources (SSC SSHA(: report Chp 8 & 9)

- individual inputs to the catalog that retained foreshocks and aftershocks to evaluate the distribution of seismicit~' in the model region to look for the presence of seisnnicity lineaments that might suggest activity along an unmapped fault in the southern B&R province (SSC SSHAC report Section 4.2.1)

PPRP Comment 83 from PPRP Comments on the Draft SSC Reoort and Tl Team Resoonses Location of Date Location ln N.o. PPRP Comment Comment Summ:iry of Revisions to Report Received Report' Revision runctlon of time. network operator at AZGS maintains tnat M4.0 are oomote.te since 2007 and MS.Osince 1970.

83 211712015 Chapter 6 The tlUe and subsequent terminology should Throughout As dascussed In the Chapter 4 comments and be clearer. The objecUve is to have a catalog or Chapter 6 responses. there IS no version of the catalOg that events that are statis\Jcally independent of 11dudes earthquakes below M2.7 In that case, each other. thereby not Including roreshocks, our *complete* catalOg pnor to dec1uster1ng anershoc.ks, or swarm-like occurrences. Since cannot meanangfuly be used to search for there also needs to be a complete se1sm1clty setsmicrty lrieaments The pre-dedustering catalog that Includes the events excluded from catalOg alSo has no utility in terms of PSHA. II is the statistically independent catalog, it needs a therefore not presented meaningfUI name, such as "historical catalog" In chapter 4, ANSS and AZGS catalogs aown to or *composite catalog*. There are several MO are used to search tor spatial trends in add1Uona1 names ror the "tndependenr catalog m1croselsm1c1ty.

In Chapter 6, such as "project catalOg," -rmal The catalog is now oonSistent.ly referred to as catalog", *oata catalog*, "final PVNGS catalog; "the PVNGS catalog*

• comprehensive composite catalog", etc.-

please pick two names ror the two catalogs used In the SSC study, and donl use any others.

It would be helpful to the reader to explain the characteristics of the two catalogs in this lntroductlon. pointing out that the Independent catalog Is used for eva1ua11ng recurrence of PVNGS SSC. SSHAC Report future earthquakes for earthquakes magnitude 2.7 and larger. While the

• more complete*

catalog (name not established yet) contains all CHAPTER6 reported earthquakes in the region (needs to be specified) Including earthquakes smaner than M2.7. In particular It is necessary to have a *complete* seismtclty map within the 320-l<m-radius region around the PV srte to support a discussion or the association of seismicity With CATALOG OF INDEPENDENT EARTHQUAKES FOR PSHA oeotoo1c structure and faults.

This chapter describes the development of the catalog of independent earrl1quakes for PSHA for the P\'NGS SSC (hereafter refe1red to as the PVNGS catalog). Developin,g a comprehensive catalog of independenc events is crucial to the tmdersranding of futw*e earthquake hazard. panicularly in regions where rbe causarive mechanisms for earthquakes are not well known and se1smicity rates are low. such as in central Arizona. The PVNGS catalog is provided in Appendix E.

Iu constrncting the catalog of PVNGS catalog. foreshocks aud aftershocks were removed. Tbe TI Team used the P\'NGS catalog to calculate recurrence parameters for areal sources (Chapters 8 and 9). The TI Team also used mdi\idual mp11rs to the catalog (Section 6.2) that retained dependent and smaller evems 10 eviiJuate the dtsmburiou of se1suuc11y w the model 1eg1011. ')pecillcally lookmi? fo1 lhe reseuce of se1sm1c1ty lineaments 1bat might suggest acti\ ity along au munapped faull (Figwe .t-8)

PVNGS model region extends to 400 km to include major faults in SoCA_and NW Mexico 1ZO-W u,.~ ioe*w Seismic Netwol't(

"' Authoritahve Regions Antona Int gratetf

"' se1siii1e ~etwort< (AJSN) l6* N La Red Slsl'Tllca def Noroe&te de MflUCO (RESNOMl

- NevadaS mic

\ NolWOfk tNN)

\

\ . . Southern Ca11f0tl'Ma Seis c Net~*o"k (SCSN)

I mJ Utah S.1$mograph Netlf.'Otk (lJU)

• I D'N Ul\lfi d CA J

I '

30'N 0 100

- mi PVNGS SSC SSHAC Figure 6-1 -==-*

0 100 km

The procedure used to cre<3te the PVNGS catalog is :

1) Identify and obtain regi ional and. national.

seismicity. catalogs, and. then standardize the catalog entry formats.

2) Merge the catalogs and remove duplicate events.

3) Use magnitude conversion equations to estimate a uniform magnitude measure (Mw) for each earthquake.

4) Identify independent e\,ents through declustering analysis (time-space wir1dow declustering approach [USGS, 2007]).

5) Account for magnitude uncertainty.

6) Assess the overall catalc)g completeness.

1) Identify and obtain regional and national seismicity catalogs, and then standardize the catalog entry formats.

Catalog Source AISN Arizona Geological Sun'ey (AZGS)

SCSN Southen1 California Earthquake Data Center (SCEDC)

RE NOM Center for cienrific Research and Higher Education at Ensenada (CICESE)

Unified CA University of California, Los Angeles (UCLA)

ANSS Advanced National Seismic System (a USGS Consortiu111)

UCERF3 \J\Torking Group on Califon1ia Earthquake Probabilities (\VGCEP) 1~*w nrw 11 ~ *w 111'W 1orw PVNGS Catalog Sources ACIV..nceG NllllOl\dl Set$1111C System (ANSS) 36" N e Anzuna Integrated Setsmtc Network (Al Sff) e La Red Sfsrn1ca de!

Noroeste de. M6iuco (R.ESNOM)

Souahern Celtfo11ua Setsmtc Netwarl< (SCSN)

Cal.tomla Undied Catalog (Urned CA) ll' N 0 100

- ITll 0- 100km

AISN Catalog

• Network: Arizona Integrated Seismic Network {comprises two regional.

networks operated by the AZGS and N. AZ University {NAU).

• Dates of operation : AZGS adopted eight legacy Transportable Array stations in 2008, 7 of which are currently working. Earthquake monitoring in northern AZ can be traced back to 1961 when a station was installed at Flagstaff. In 1986, formation of the Arizona Earthquake Information. Center {AEIC) and the northern Arizona Seismic network, which originally consisted of 3 stations at Flagstaff, Williams and the Grand Canyon. Currently, this network includes 8 stations (5 running) stretching from the Utah boarder to the southern edge of the Colorado Plateau.

• Mission: Monitor AZ earthquakes at a lower threshold and with more uniform coverage than USGS, and generate a n1eaningful earthquake catalog that can be used to determine seismicity rates and elucidate tectonically active seismic areas.

• Catalog

Description:

3,489 events in the AISN catalog dating back to 1852. "'60 to 100 events are added/year. Estimates of completeness as a function of time :

M4.0 since 2007; M5.0 since 1970.

PPRP Comment 1.1 on Seismicity data (PPRP Letter #3: PVNGS SSC WSl)

PPRP Comment 1.1:

It was clear from the presentations and discussions that earthquake monitoring within Arizona has been given a low funding priority historically and is operating in a fragile manner through the dedication of a few individuals. This severe restriction of resources may have led to operational practices that could have impacted the quality of the data being relied on for catalog developnnent. We suggest that the Tl Team consider performing a friendly "quality assurance" review of the earthquake. monitoring. data analysis procedures. used. for the ABN and NASN. The national standard for seismic network operations is established by the US Geological Survey's Advanced National Seismic System (ANSS); the seismic networks to the west (California) and north (Nevada and Utah) are. members of ANSS.

Tl Response to PPRP Comment on Seismicity (7/15/13 Letter "Response to observations and comments from the PPRP on Workshop #1")

Tl Response to PPRP Comment 1.1: To the extent possible, the Tl Team will perform the recommended "friendly quality assurance review" of the data analysis procedures of the Arizona Broadband Network (ABN) and.the. Northern Arizona Seismic Network (NASN), which collectively comprise the Arizona Integrated Seismic Network (AISN). To our knowledge, operational procedures for these networks are not published and are not readily available. The Tl Team will contact Jeri Young (Arizona Geological Survey) and David Brumbaugh (Northern Arizona University) to see if such procedures are available for the ABN and NASN, respectively, for Tl Team1 review. The Tl Team will use this information to evaluate quality and uncE~rtainty of the data to inform our judgment when weighting alternatives in the SSC, but will not reevaluate earthquakes in those earthquake catalog~.

NRC Review of Tl Response to PPRP Co1nment 1.1: There's no mention of "quality assurance" review in the PVNGS 50.54f submittal or the SSHAC report, but it may have been incorporated, as the Tl team stated in its comment response, into evaluation of quality and uncertainty of the data to inform catalog priority and weighting of alternatives in the SSC.

SCSN Catalog

• Network: Southern California Seismic Network (SCSN)

• Dates of operation: Been running in some form since 1921. Starting in 1932 it had 6 stations. Today the SCSN records data from 4:~4 stations.

• Catalog

Description:

As of 1/1/2015, the SCSN catalog has 608,908 local events (events for which SCSN is authoritative, excluding q1Jarry blasts and other non-tectonic events). On avera1ge the catalog is complete for Mc~ 3.2 since 1932, and Mc~ 1.8 since 1981, excluding early hours or days of large aftershock sequences and regions near the edge of the network.

RESNOM Catalog

• Network: Red. Sfsmica del Noroeste. de Mexico.

• Dates of operation: RESNiOM has been operating for more than 30 years. 9 shc>rt period instruments started operating in 1978, this number expanded. to 13 in 2013. The first broa~dband stations were installed in 2001. 14 new stations were installed during 2011 and 2012. At present, RESNOM operates 19 broadband stations.

• Catalog

Description:

7,0510 earthquakes of magnitude 2.7 or greater in the RESNOM catalog through December 2012 ,with 100s of events added per year. The minimum rr1agnitude of completeness for current reporting is 2.4.

Unified Earthquake Catalog of California (Unified CA)

• Mission: Construct and test certain hypotheses of earthquake occurrence, Wang et al. (2009) compiled a new catalog covering the vvhole of California, which lists all known earthquakes at magnitude 4. 7 and above.

• Catalog

Description:

Unifie~ d CA lists all known regional earthquakes at magnitude 4.7 and above, from 1800-2006. 23 existing catalogs 'Nere examined for this effort, enabling more accurate magnitude and location estimation; different magn1itude types were also converted to moment mag~nitude. Catalog. ends. on.

December 31, 2006.

ANSS Catalog

• Compiler: Northern. Cadifornia Earthquake.

Data Center, Universit~f of California at Berkeley Seismologica l Laboratory

• Catalog

Description:

Cjomposite national earthquake catalog th<3t is created by merging the master earthquake~ catalogs from.

contributing ANSS institutions and then removing duplicate soll utions for the same event.

UCERF3 Catalog

• Compiler: USGS

• Catalog.

Description:

Tl1e UCERF3. earthquake.

catalog is an update of the catalog compiled for UCERF2. Only eight, post 2006 CA earthquakes(> M4.7) ,were. selected from this catalog. It was necessc1ry to add these events because the primary cla talog for the Unified CA ended on DecembE~r 31, 2006.

PPRP Comment 1.2 on Seismicity data (PPRP Letter #3: PVNGS SSC WSl)

PPRP Comment 1.2:

The PVNGS site is within a part of the Southern [B&R] Province that is characterized by very low seismicity. For example, within 50 miles of the site, there is only one M3+ event in the current catalog and only one event of M<2 recorded during the three years of the TA operation. We think that it would be useful to further quantify the seismicity rate by searching for recordings of earthquakes that are large enough to be detected but too small to have enough stations to locate. One could start by looking at the three-year

[transportable array (TA)] database using the nine TA stations roughly centered. on the station closest to the PVNGS site. These data could. be helpful in providing a more refined subdivision of the seismicity patterns associated with the Southern [B&R] Province, and. thereby used for refining the areal sources used in smoothing seismicity for areal source rates. Some of the initial results for computing rates for areal sources. using the base case model, as.

discussed by Melanie Walling, were startlingly high given the observed low seismicity within 50 miles. of the. PNVGS site.

Tl Response to PPRP Comment 1.2 (7/15/13 Letter " Response to observations and comments from the PPRP on Workshop #1")

Tl Response to PPRP Comment 1.2: Mlany of the earthquakes recorded in the PVNGS study region during the three-year window of the [TA] have magnitudes that are below the lower rnagnitude cutoff (Mw 2.7) for inclusion in the project earthquake catalog. The Tl Team agrees, however, that the TA earthquakes may be usefu l, in particular for evaluating seismicity rates and patterns within the study region and possibly to provide additional information or insights on seismicity rates. As such, the Tl Team will continue to investigate the TA earthquake data to assess alternate ways to capture uncertainty in the SSC. These investigations likely will include sensitivity analyses intended to assess the impacts of various modeling decisions on seismic hazard at the PVl\JGS site.

NRC Review of Tl Response to PPRP Comment 1.2: Discussed in SSHAC.

report Section 4.2.1 regarding lineaments or patterns that would reflect activity of unmapped faults in Southern B&R.

The procedure used to. cre<3te the PVNGS catalog is :

1) Identify and obtain regi ional and national seismicity catalogs, and then standardize the catalog entry formats.

2) Merge the catalogs andl remove duplicate events.

3) Use magnitude conversion equations to estimate a uniform magnitude measure (Mw) for each earthquake.

4) Identify independent e\,ents through declustering analysis (time-space wir1dow declustering approach [USGS, 2007]).

5) Account for magnitude uncertainty.

6) Assess the overall catalc)g completeness.

2) Merge the catalogs and remove duplicate events.

• 7 authoritative zones were drawn and prioritized (figure on next slide)

• Duplicate records were identified and removed from the composite catalog when they occurred within a minute of each other and were spatially located within 0.1 degree in latitude and longitude (about 9 to 11 km) . If only the time criterion was met, a visual inspection of the records in question determined if removal was required . When duplicates were identified, the record from the highest priority catalog was retained .

Table 6-1. Catalog priority within authoritative zones. Catalogs are described in Section 6.2.

Authoritative Zones are hown in Figure 6-2. (Priority ranking: 1 is high. 5 is lo\v and 0 does not apply).

Zone Location l:nified CA scsx ANSS AIS .. '" RE S~Ol\I 1 W. California l 0 0 0 0 2 E. California l 2 .) 4 5 3 Arizona l 4 3 2 5 4 r . Sonora 1 4 3 5 2

. Baja 5 California 1 4 3 5 2 6 NW. Sonora I 2 ~

.) 5 4 7 S. Nevada 1 3 2 4 5

2) Merge the catalogs and remove duplicate events.

120* w 114*w t \I" w 108' W

- z m

~

-x 0

Ci 36 N

1) Unified CA \

2) ANSS

3) SCSN 1) Un1f1ed CA

4) AISN 2) AISN

5) RESNOM 3 3) ANSS

4) SCSN

5) RESNOM

1) Unified CA

, f 33' N

2) SCSN I

3) ANSS I 4 ) RESNON I

5) AISN I 4

1) Unified CA

2) RESNOM

3) ANSS 1) Unified CA 4)SCSN 2) RESNOM

5) AISN 3) ANSS

4) SCSN

5) AISN 0 PVNGS SSC SSHAC Fi ure 6 3

The procedure used to. cre<3te the PVNGS catalog is :

1) Identify and obtain regi ional and national seismicity catalogs, and then standardize the catalog entry formats.

2) Merge the catalogs and remove duplicate events.

3) Use magnitude convers;ion equations to estimate a uniform magnitude rr1easure (Mw) for each earthquake.

4) Identify independent e\,ents through declustering analysis (time-space wir1dow declustering approach [USGS, 2007]).

5) Account for magnitude uncertainty.

6) Assess the overall catalc)g completeness.

3) Use magnitude conversion equations to estimate a uniform Mlw

• Recently developed conversiions were used:

- Utsu (2002) and Sipkin (2003) to convert ML (local magnitude scale), Ms (surface-wave magnitude scale), and mb (body-wave magnitude scale) to Mw for the 2008 NSHMP

- Arabasz (2013) determined several conversions using the University of Utah's earthquake catalog

- Zuniga and Castro (2005) found that a correction of +0.1 units to the RESNOM catalog's Md (duration) scale is equivalent tomb, which can then be combined with Sipkin's (2003} mb equation to obtain Mw.

• Not enough events in catalo;g to test these relations, so instead, the ANSS catalog of the. western U.S. was used for the evaluation. Orthogonal. regressions were performed on. the ANSS catalog data.

3) Use magnitude conversion equations to estimate a uniform Mw Mw vs. NEIC Mb Mw vs. Nevada Mc

., 7 65 6.5 6 II s.s S.S

• s

• 0.La POlllU

- Onholl~u  ! s

• O-..UPotnu onho Regress

• .s - Sip n 12001) *s - USGS(M\11 *- MQ

- Arab.au mb POU - Atabas.r W.c UUl Mw

• US(0.16) *Mt..O 73(0.91) 3.S H l

l ...s 6 7 l

l H *s 5 ss .,

Mc (kH)

Mw vs. Utah Ml ELS 6

s.s

• 0.IA Poi:nta

~ s - <lf't...,R...,._s

- Utsu (lCO.t)

• .S

-Anbav Ml UUl

' Mw - 0. M(0.07)

• Ml.0. 60(0.86)

.l.S PVNGS SSC SSHAC Figures 6-4, 6-5, & 6-6 l JS

' as s ss &..S ,

ML (UU)

3) Use magnitude conversion equations to estimate a uniform Mw Mw vs. ML (Cl, NEIC, NN, UU) 7

~

6 s.s

• Dita Po1nu J s

~

- Ortho Recress 4.S - utsu IZ00.21

- Arabasl ~* L UUl 4

Mw

• 1.11 *ML-0.57 ll 3

3 3.S .. *.s s s.s Ml.

Arabasz vs. G&R (1956} Intensity Relations 8

7.5 7

6.5 6

~ 5.5 - (G&Rl9S6) s - AtabaulO<V 4.S - Atabllz. 10> v 4

3 .5 PVNGS SSC SSHAC 3 Figures. 6-7 & 6-8 MMI

3) Use magnitude conversion equations to estimate a uniform Mw l\lagnimde or Ioteosity T~11e 1 C 00,-ecion Equ:atiom 1 1 Source h.fw = 1_67

• i(lfL -2.60) (for ~fi. > 16_5) 1

),<~ rv1I_GS, Ji.illY AS~ ~ll.SLC Utm(2002) eke 1,ifw = J..iL

~-~_res (RESNOM) D1',. = Mtii_:fes _._ O_l (use mb eq11a.tiom) Zuniga amd Castro (200j) l\oid ~=1\(a Petei_-sen e-t al (2000)

),*fw_ :M"wHR..\l Mw=Mw 1\iiatMnmtical Identi~r l\~~- = 1.46*Ulb-2.42 (fm- mb > 5.3) else m, 2 me.GS Sjpkin (2003)

Mv.* = ll1J:.

1Ylc Ma.= l\iic Perersen et al. (200&)

NllYfi ~=Mt = {2/3)*~ +I Gutenberg md Richter tn9j6J Unt; n 1\-fw=Unk Petersen e-t at (2008)

}.!~ l\ifn. = l\>ih Peteri:sen et al (2008) 1\'I ~lw=~f Petersen e-t al (2008)

Mu,= 0_7:5 * (M~+L93) (fur~<5. 8):

1vlt Mw = l _50

• Qt-I.. - 2J.l0) (forM; > 7.8)~ Utsn (2002)

PVN ( S SSC SSHAC Table 6-3 ehle ~1. = M.

The procedure used to. cre<3te the PVNGS catalog is :

1) Identify and obtain regi ional and national seismicity catalogs, and then standardize the catalog entry formats.

2) Merge the catalogs and remove duplicate events.

3) Use magnitude conversion equations to estimate a uniform magnitude measure (Mw) for each earthquake.

4) Identify independent e,vents through declustering analysis (tiime-space window declustering approach l[USGS, 2007]).

5) Account for magnitude uncertainty.

6) Assess the overall catalc)g completeness.

4) Declustering analysis

• Tested alternative approaches on a catalog of seismicity. located w ithin the southern B&R. This reduced dataset is an appropriate sensitivity sample because it contains the closest and, hence, most crucial portion of earthquakes of the PVNGS catalog.

• Grunthal {1985) classifies 25% of events as dependent.

• Reasenberg (1985) conversely, does not appear to be well suited for the limited data available; as only a handful of dependent events were removed .

• The Gardner and Knopoff (1974) algorit hm, which reduces the sub-catalog by 20%, is bracketed by the other two methodologies, and thus deemed by the Tl Team to be a reasonable approach for declustering the composite catalog.

Independent EYents Gardner Basin and and Range Grunthal Knop off Reasen berg Events (1985) (1974) (1985) 303 228 -?4'"\;) 284

- 75% 80% 94°/o PVNGS SSC SSHAC Table 6-4

4) Declustering analysis Applying the Gardner and Knopoff declustering algorithm to the composite catalog reduced. the record count from 1,941 to 1,048 events . . This "'50%. reduction is in line.

with that of Gardner and Knopoff (1974), who found that approximately 2/3 of the events in the more completely recorded SoCA instrumental catalogs were identified as aftershocks.

1?0" W 11rw 11<<" W Ill w IOll'W tJ TA 11 PVNGS Catalog z E(M) l!I

~

~

~2 70 and < 3 00 l!I

.)< 315*N n  ! 3 00 and < 4 00 0

\

~ 4 00 and < 5 00

\

• ! 5 00 *nd < 6 00

~ 6 00 and < 7 00

!700 n*N note onty independent f\'tn11 lihown 30'N 0 100 PVNGS SSC SSHAC Figure 6-9

-==--

. _ _ km ml 0 100

The procedure used to. cre<3te the PVNGS catalog is :

1) Identify and obtain regi ional and national seismicity catalogs, and then standardize the catalog entry formats.

2) Merge the catalogs and remove duplicate events.

3) Use magnitude conversion equations to estimate a uniform magnitude measure (Mw) for each earthquake.

4) Identify independent e\,ents through declustering analysis (time-space wir1dow declustering approach [USGS, 2007]).

5) Account for magnitude uncertainty.

6) Assess the overall catalc)g completeness.

5) Magnitude uncertainty

• Following the NSHMP approach, standard errors of 0.1 magnitude unit for earthquakes occurring after 1971, ().2 for 1932-1971, and 0.3 for 1850-1931 were assumed.

• Estimates of catalog cc>mpleteness and recurrence parameters are computed using the same methodolog~y as in CEUS SSC via uniform moment mag1nitudes, E[M] and the equivalent counts, N* [N*= exp(~ 2 a 2 /2)].

The procedure used to. cre<3te the PVNGS catalog is :

1) Identify and obtain regi ional and national seismicity catalogs, and then standardize the catalog entry formats.

2) Merge the catalogs and remove duplicate events.

3) Use magnitude conversion equations to estimate a uniform magnitude measure (Mw) for each earthquake.

4) Identify independent e\,ents through declustering analysis (time-space wir1dow declustering approach [USGS, 2007]).

5) Account for magnitude uncertainty.

6) Assess the overall catalog completeness.

6) Catalog completeness

• Completeness. time interval is defined as the time period over whicl1 earthquakes of a specified magnitude rcinge are believed to be complete.

• Following figures sho" Stepp-style plots (Stepp,. 1972) of subsets of the PVNGS catalog.

based on authoritative~ zones.

• Table list magnitude rc1nges and event completeness for varic>us time periods, as indicated by the arrov\ts in the figures.

Ste pp Plot, S. California, Baja (Zones 1 & 5) Stepp Plot, E. talifomia (Zones 2, 6 & 7) 1 1

• 1..?0 <QM)< l.l7

- UO<EiM)<l.37 c

• 4 71 <E{Ml < !>38 c

• l.37 < QMJ < *-~

0 "I) of "O 0.1 4 1t < l{MI < S.ll S S1<£lMl<60S i..., - 137 < E{MJ < 4.04

• 4.0t < QMJ < 4..71

~

l.,,

- S.JI< l{Ml < 605 - 4J)il < E{Ml < 4.71 01 1 -

• 60S c l(Ml < 6.72 6.0Sc ljM] <6.72

• U1 < C[MJ < s..31

• 71<EIMJ < S.31

• 672<llM)<7.39

• S.38 < !IMI < 6.05

- 6.'72<E.1Ml < 7.Jt - S3*< ElUJdOS G.OS c £!Ml < 6. n 0.01 U!5< E!MI < '*7J 1 10 100 1000 1 10 100 1000 Ytat* befMt Jal\. 1. 2013 Yu n kfor* Jin. L ZOLS lt t f UtSS P'~GS t: GS lli P'~GS r scs 1910 1960 4 71 1933 1910 1963 4 -1 5-38

• 5 38 605 1920 18.,0 1900- 1933 1900 °""" l t P"W ..... 1910 1930 - 1963 1910 1930 6.05

• 672 1850 1850 1910 1850 6.....2 7.39 1850 1850 ,. .. 1850 1850 739 8 06 1850 1850 1850 1850 s-3 1850 1850 l)UnrledCA 2)ANSS 3)SCSN 1) Umfed CA 1850 1850 4)AISN 2)AISN 5)RESNOM 3 )JANSS 4)SCSN S)RESNOM I

1)U111fitdCA 2)SCSN I I

I 33* 04 l)AfliSS I 4)RESNON 5)AISN I '

,1 PVNGS SSC SSHAC Table 6- 1) UnlhclCA 2)RESNOM 1)Uni6odC4

))ANSS 2)RESNOM 4)SCSN 3, 6-10, & 6-11 0 JOO S)AISN 3)ANSS

-=- ...

4)SCSN S)AISN 0 1oa

Stepp Plot, Arizona (Zone 3) Stepp Plot, Sonora Mexico (Zone 4) l 1

• 2.70 < EIMJ < U7 2.10< E(M) < l.37

- 2.?0<E{M)< :UJ c 2.'10 < l[M) < 3.37 0 c

'P

• l.l7 < E[M) < .t.(M 0

'P

• 3.37 <UM) < 4.G4

- 3.J7 < £[MJ < .t.04

..,~.. - 3.37 < lfMl < 4.1)4 0.1 * .t.()1<£tM)<4 .11

..,~.. 0.1

• 4.04 < E(Ml<<Ul

~

- 4.0I < t[M) < 4.11 i - 404<l[M}<4.11 0 4..71 < l[MJ < ~33 .,;

o 4.11 < E(Ml < S.31

- 4.71 < ElMJ < !..38

- 4.1l<E[M)<S.3&

• S.38 < E!MJ < 6-0':i

• S.la<l(M]<~OS

- BS<llMJ<6.0S

- S.l8<E!MJ<6.0S 0.01 l 10 100 1000 1 10 100 1000 Yun bt!tore JML 1, 20ll Years before Jan. l, 2013 P°'~GS 'CSG S

~ :P\~GS t:-SGS l970 l960 c' 4.04 1990

<" 4.71 1920 1963 < 4 71 1960 1963

< 538 1870 1930-1963 1 9~ 1930 - 1963 6.05 1870 1930 °'w 1930 1930 6.72 1850 1850 ---- -- - -

VTAl1

<., 6.72 1850 1850 1850 1850 1850 "

'",. ,.... 739 1850 1850 1850 n

<) 8.06 1850 1850 1850 1850 1) Unified CA \

2)ANSS \ 8.73 1850 1850 3)SCSN 1) Unified CA

<l)AISN 2)AISN 5)RESNOM 3 3)ANSS 41 SCSN S)RESNOM

• l

,' 33'Jl I

I '

11 PVNGS SSC SSHAC Table 6-3, 6-12, & 6-13 0 100 0 100

SSHAC Chapter 6:

PVNGS SSC Earthquake Catalog

• The PVNGS catalog ranges frorr1 1852. through 2012 and. contains 1,048 independent events.

• Duplicate records were remove~d from the catalog when they occurred within a minute of each other and were spatially located within 0.1 degree in latitude and longitude, with exceptions. If only the time criterion was met, a visual inspection was made. Removal of duplicates was done according to the highest priority catalog.

• Used various relations for magnitude conversions to obtain Mw.

• Used Gardner and Knopoff {19~74) algorithm for declustering.

• Completeness, Western CA is complete since 1932 down to a 1

uniform moment magnitude (E[M]) of 4.7; Eastern CA and AZ are complete to E[M] 4.0 since 1930. Northern Sonora, Mexico also appears to be complete to E[M] 4.0, but only since 1970.

PPRP Comment 1.3 on Seismicity data (PPRP Letter #3: PVNGS SSC WSl)

PPRP Comment 1.3:

In consideration of the limitations of the current earthquake monitoring networks in Arizona, it might be reasonable for [APS], on behalf of its PVNGS, to consider installing one or more seismic monitoring stations for specific data targets relating to the current project and for future use regarding seismic hazards related to licensing. As noted in the above comments, the paucity of seismic monitoring data in the low-seismicity environment of PVNGS has both positive and negative* aspects. Here are several possible deployments that could be useful in the short term (the current project) and in the long term (future licensing matters).

a) Given the location of surface bedrock within several miles of the PVNGS site, a broadband station comparable to the vandalized TA station could be installed at a reasonably secure location with data telemetered to a central recording site (potentially operated by the AISN).

b) Several additional short-period seismographic stations could be installed to form a small array around the central station to improve the detection and location of occurring earthquakes. These data would be used to refine the seismicity model used for areal sources.

c) It has been suggested that a strong-motion station be installed within the site perimeter to collect data on kappa for the site. It could be useful to operate similar strong-motion instruments along with the stations described in items (a) and (b) above.

These possible seismic monitoring installations would best be considered in the context of both the near-term application of the data for the current project {a few years) and the longer-term interests of APS with regard to the role of seismic issues in future operational considerations at PVNGS.

Tl Response to PPRP Comment 1.3 (7/15/13 Letter "Response to observations and comments from the PPRP on Workshop #1")

Tl Response to PPRP Comment 1.3: The Tl Team agrees that installation of a seismograph or seismographs at or near the PVNGS site would provide useful data both for the current project and for the longer-term interests of [APS]. Specifically, these data would provide improved earthquake monitoring and reduction of uncertainty on site kappa and other ground motion parameters. The installation of new instrumentation, however, is beyond the scope of the current project. We are in current discussions with APS and Westinghouse Electric Company (WEC) regarding the possibility of obtaining additional budget to install, operate, and maintain this new instrumentation and to determine who would receive and support interpretation of any new data. To. maximize benefits to the current project, the Tl Team understands that any new instrumentation should be installed as soon as possible to maximize the number of earthquakes recorded in this low-seismicity environment.

PPRP Comment on Seismicity Monitoring PPRP Letter #4: PVNGS SSC WS2:

Seismicity Monitoring; This proposal focuses on procuring and installing new broadband and strong-motion instrumentation at the PVNGS site in a borehole drilled for this purpose to bedrock, a depth of about 500 feet beneath the site. The purpose of the instrumentation is to collect data for {l) detection and improved location of earthquakes ("'Ml and larger events) in. the central part of the southern [B&RJ province including near the Palo Verde site, and (2) refining the value of kappa at the site using primarily weak ground motions from local or regional earthquakes. The instrumentation would be operated initially as a freestanding system recording in a triggered mode, prior to establishing more permanent power and Internet data communications for longer-term operation .

The PPRP endorses and strongly supports the funding and implementation of these work items as soon as possible. APS has indicated that funding may be available for new work if it is well justified. The PPRP urges that a high priority be placed on implementing these work items at the earliest possible dates in order that the data may be obtained in a timely manner.

Tl Response to PPRP Comment Tl Response to PPRP Comment on Seismicity Monitoring: [APS], LCI, and the Tl team are working together to implement ... newly proposed work items, including ... installation ~Jf a downhole seismograph array at the site, and collection of Spectral Analysis of Surface Waves (SASW) data at the site. In a recent teleconference with LCI, APS indicated their intention to fund thie... new proposals. During that call, however, APS indicated that funding for the new work largely will not be available until early in 2014, with the exception that procurement of seismograph instrumentation is underway so that it will be available for installation as early as possible in 2014.

NRC. Review of Tl Response to PPRP Comment 1.3 & WS2 Comment on Seismicity Monitoring: There's no mention of installation of a seismograph(s) at or near the PVNC)S site in the PVNGS 50.54f submittal or the SSHAC report, so this PPRP comment is unresolved.

PPRP Comment on Seismicity PPRP Letter #6: PVNGS WS3, Comment #4:

Some uncertainty in the nature of the M>4.65 seismic events. mapped on the west side of the Southern [B&R] province just east of the Gulf of California was noted in the meeting. Conducting a review of the earthquakes comprising these events should be considered to determine if they are located on land or are associated with faulting within the Gulf. If they are pre-instrumental (or otherwise poorly located), efforts could be made to reposition the events.

PPRP Comment on Seismicity Tl Response to PPRP Letter #6: PVNGS WS3, Comment #4: The Tl Team reviewed the portion of the project earthquake catalog directly east of the Gulf of California, where an approximately triangular wedge of seismicity appears to taper off into the Southern [B&R].

In order to assess the likelihood that: (1) the project catalog correctly reflects a region of elevated seismicity rate along the western border of the Southern Basin and Range; and {2) the catalog correctly locates Mw > 4.65 earthquakes in this region, the Tl Team reviewed the age, location uncertainty, and magnitude type of these earthquakes .

62 earthquakes in the area, only 6 have magnitudes Mw > 4.65 {i.e., 1935 Mw 5.0, 1952 Mw 5.1, 1958 Mw 4.9, 1963 Mw 4.7, 1969 Mw 4.8, and 1981Mw4.9) . The location and magnitude for the 1935 earthquake are based on felt intensity reports and therefore may be highly uncertain. The locations and magnitudes of the 1952 and 1958 earthquakes are based on instrumental data but are reported only to the nearest half-degree, reflecting a high degree of uncertainty. The Tl team assu1Ties that the more recent 1963, 1969, and 1981 earthquakes are relatively well located, howe?ver, and should not be repositioned. Given this assumption, it is difficult for the Tl team to justify repositioning the 1935, 1952, and 1958 earthquakes. Therefore, the Tl tearn does not plan to reposition any of the earthquakes in this area.

NRC Review of Tl Response to PPRP WS3, CC>>mment #4: Resolved.

T ABLE OF CO~TE\!S PVNGS SSC SSHAC Report, HID:.

1.0 INTRODUCTION

......... ........ ... . . ... ................................................................................... .. F-3 2.0 OVER.\'IEW OF SEISWC SOURCES ****************-*-*******-*********************-*-*****************-***-******* F-3 21 Areal Soucces ........................................................................................................................... F-3 22 Fau1t Sour~'-*-*-*-*-*****-*-*-*---*-*-*- .................- ................- ..- -..- ........... F-3 CHARACTERISTICS OF AREAL SOL"R.CES .......-.........................- .....................- .......- ....... F-5 Attal So~~ --****--**-**-*-***-***-*-******* -*******--*-*--*** -*-*--***-*-*-* **

• F-5 Rupture !\iecNin*sms for F~ ~ --* ....................- .......................... - .......- ....... F-5 Rupture Oo.enlatlOOS fOI' Funn ~s ........ . _ --****-************* ................. ***-*- F-6

~~ Dip$ for F111Ure EartlJq"aans -*******-****-****-*--*****--*-****-*-****-**********--*-*-**-****-* F-6 R<<uaen~ offunn Earthquakrs m Arw Sourr~ ............. -*-** ...... ****--*** *-** .. F-6 4.0 CHARACTER.ISTICS OF FAULT SOURCES .............................*-******-*-******-**-*-**-******-*** F-7 4.1 Fault Source Probability of Acti'\'lty ***- *****-************-*-*-**-**************-**-*** ..****-*-*******.. **-*** F-8 4.2 Rupture Length*******************************-************************************-************************..*****-**************** F-8 4.3 RuptllteA.rea ............................................................................................- .............................. F-8 44 Displacement pei-Eveot ................................................................................................... .. F-8 4.5 Seiunogeoic Thiclaiess ...................................................................................................- ...... F-8 4.6 Chamcteristic Magnitude .......................................... - ............................- .............................. F-8 4.7 Earthquake~ for Fault Sources ................................................................................. F-9

5.0 REFERENCES

.... .................................................................................................................. . F-10 ATIA~ .................................................................................................. ..F-50 Attachment A: heal Source Coordwates (electronic attachment)

Attachment B: Fault Source Coard.mates (electromc attachment)

Anaclnnen.t C: UCERF33 Rupture Sets (electronic attKhment)

Attachment D: A.BSMOOIB Omput (el-ectronic attachment)

AnacbmHJt E: S\VUS GMS Re~o~ for Fault Sources (electronic attachment)

Append1x F F-2 PVNGS SSC, Rev 0

SSHAC HID 3.0: Characteristics of Areal Sources Areal source boundaries All leaky Rupture mechanism, orientation, seismogenic Assessed on a source-by-source basis thickness, maximum magnitude..

Strike-slip faults dip 80 + 10° Reverse faults dip 45 + 15° Rupture dips Normal faults. dip 50 + 15°.

Dip direction for. all non-vertical faults is random .

All modeled in the areal sources are allowed Top of rupture to rupture up to the ground surface (i.e.,

depth= 0 km) .

Different rate cases for eastern and western Rate cases sources.

All treated as. a truncated exponential.

Recurrence distribution (Gutenberg-Richter) with spatially variable parameters.

Fo r the 2-Zone Areal Sou rces :

Source Bound;n1* Rupture Rupture Ruptun Top of Seismogenic ~Imax Rate l\lagnitude Dip Rupture Thickness Recw*r ence

~fame Type ~lechanism 01ientation p lw) Cases (degrees) (km) (km) ~lodel Strike-slip 70° (20%)

(800'o) 80° (20%)

90° (60%)

6.8 (0.1)

N35°W (20%) 30° (20%) 12 (0.2) 7.0 (0.25)

Re"\*erse West Leaky (10%) N45°W (60%) 45° (60%) 0 15 (0.6) 72 (0.4)

N 55°\V (20~o) 60° (20°'o) 18 (02) 7.5 (0.2) 7.9 (0.05) 35° (20%) 1 (0.4)

Nonual 50° (60%) 2 {0.4) G-R (1.0)

( lO~o) 65° (20%) 3 (02) 35° (20%)

Normal N20°E(10%) 50° (60%) 6.8 (0. 15)

(80°0)

N-S {lOO'o) 65° (20%) 12 (0.2) 7.0 (0.25)

East Leaky N20°W (400'o) 0 15 (0.6) 7.2 (0.35)

~4-0°W (200'o) 70° (20%) 12 (0.2) 7.5 (0.2)

Strike-slip Random (200A>) 7.9 (0.05) 80° (200/o)

(20°0) 90° {60°'o)

PVNGS SSC SSHAC Table 9-1

Top of Seismogenic .:\lagnitude Sour"Ce Boundary Rupture Rupture Rupture .:\Imax Rate Rupture Thickness R ecwTence For the l'inme Type  :\lecbani-;m Orientation Dip (km) (Jan)

[.\In-) C a-;es

\fode.J Seismotecton ic Stnke-slip 700 (200*)

80°(20%) 6.8 (0.15)

(900/e) N35°W (200:.) 12 (0.2) 7.0 (0.25)

Areal Sources: SCABA Leaky N45oW{60%)

900 {60"/o) 0 15 (0 6) 12 (04)

N55°W{20%) 30°(20%) 18 {0.1) 7.5 (0.15)

Reverse 45° (600/o) 7.9 (0.05)

{IO~:.)

600 (200/o)

Strike-slip 700 (20%)

(7000) 800 (200/o) 6.8 (0.05)

N35°W(200o) 900 (600/o) 12 (03) 7.0 (03)

GULF Lcaky N45°W(60%) 0 14 {0.6) 7.1(03)

Normal N55°W{200o) 35° (20~*) 16 (0.1) 7.5 (03)

(30%) 500 {600*) 7.9 (0.05) 65° (200/o) 1 (0.4) 2 (0.4) G-R (l.O) 35° (20%) 3 (0.2)

~onnal 500 (600/o) 6.8 (OJ)

(80%) N-S{Ne) 65° (200/o) 12 (0.2) 7.0 (0.25)

'.\"200W (~*}

SBR Leaky 0 15 {0.6) 7.2 (0.4)

N400W (200'.) 70° (200/o)

Strike-slip 18 {0.2) 7.S (02)

Random (2~*) 800 (20%)

(2000) 7.9 (0 05) 900 (600*)

35° (20%)

Noonal 500 (600'.) 6.8 (0.05)

(800'.) N~S~:.)

65° (200/o) 12 (0.1) 7.0 (025)

N'lO"W (409/o)

MR Leaky 0 15 (0.6) 7.2 (035)

N4WW (200/e) 70° (200/o)

Strike-slip Random GOO o) 18 (0 3) 7 5(03) 800 (20%) 7.9 (0.05)

(2000) 900 (60"/o) 35°(20%)

Nonnal 6 8(0.2) 50° (600*)

(700'.) N20°E (2S~o) H {0.2) 7.0 (0.25) 65° (209/o)

TZ Uaky N20°W (25° o) 0 17 (0.6) 72 (03) 70" (200/o)

Strike-slip Random (500*) 20 (0.2) 7.5 (0.2) 80" (200/o)

{3000) 7.9 (0.05) 900 (6001.)

35° (200:.)

~onnal 6.5 (02) 500 (600/e)

(80"/o) 15 (0.2) 7.0(0.3) 65° (200/e)

CP Leaky Random (1 00~:.) 0 20 (0.6) 7.2 (025)

PVNGS SSC SSHAC Strike-slip 70" (2000) 25 (0.2) 7.S (0.2) 800 (2000) 7.9 (0.05)

(200/o)

Table 9-2 900 (600/o)

PVNGS SWUS (jMC Chapter 5:

Ground Motion Databases a1nd Candidate Models for the Median and Aleatory' Standard Deviation"

• Chapter 5 describes

- ground motion database used to evaluate the alternative GMPEs. The d,atabases were also used to develop new models for the aleatory. variability. using the partially non-ergodic ,a pproach (single-station sigma and single-path region sig;ma) .

- selection of the candidate~ GMPEs. for median ground motion. From the evaluated set of 19 GMPES published between 2004 and 2014, 6 candidate GMPEs were selected for PVNGS.

Table EX-2: Se lected Cand idate GMPEs for the median ground motion GMPE DCPP DCPP Distant PVNGS - Greater PVNGS - Distant Sources Arizona Sources CA & MEX Sources Abrahamson et al {2014) x x x x Boore et al {2014) x x x x Campbell and Bozorgnia {2014) x x x x Chiou and Youngs {2014) x x x x Idriss {2104) x x x Zhao et al {2014) x Zhao and Lu {2011) adj ustment x to magnitude scaling Akka r et al (2014a, 2014b) x x Bind i et al (2014a, 2014b) x

PVNGS SWUS (jMC Chapter 5:

Ground Motion Databases a1nd Candidate Models for the Median and Aleatory' Standard Deviation"

• For PVNGS, 4 empirical ground. motion databases were used by the Tl Team for evaluation of the alternative:

• PEER NGA-West2 database (Ancheta et al., 2014) - Use of the database was restricted to strike-slip and normal 1faulting earthquakes that control the hazard at PVNGS

• Reference Database of Seismic Ground Motion in Europe (RESORCE) described in Akkar et al. (2014c) - Use of the database was restricted to strike-slip and normal faulting earthquakes that control the hazard at PVNGS

• PEER Arizona database (Kishida et al., 2014a) - recordings in AZ from earthquakes in AZ and recordings in AZ from earthquakes in CA and Mexico.

Used to (1) evaluate path effects from median ground motion from CA earthquakes (2) evaluate kappa for rock sites in AZ (3) develop aleatory variability models for earthquakes in CA and Mexico recorded in central AZ (single-path region sigma models)

• Lin et al. (2011) - consisting of ground motion residuals from M4 to M6 earthquakes

• Ground motions from the M6.0 2008 Wells, Nevada and the M6.7 2011 Fukushima-Hamadori normal-faulting earthquakes were also evaluated

Table 5.1-1: Primary Emp irical Data Sets Distance Mechanisms Magnitude Vs30 Range Dip Range DATASET Range by Range (m/s) (degrees)

(RJs in km) Earthquake PEER NGA-West2 3.0 - 7.9 0 - 1532 89 - 2100 10- 90 57% SS 17% NML 26% REV Akkar Subset of Reference 4 .0 - 7 .6 0-200 92-2165 2-90 38% SS database of Seism ic Ground 47% NML Motion in Europe (RESORCE) 15% REV PEER Ari zona {Regions 1, 2, and 3) 4.3 - 7.2 145 - 649 398 - 1312 40- 86 93% SS 7% NML 0% REV PE ER Ari zona 1.2 - 3.4 9 -301 398 -1237 Not Not

{Centra l Arizona) Avai lable Avai lable Lin et al. {2011) 3.9 - 7.6 0.6 - 208 166 - 760 10-90

Table EX-1 : Ground motion databases and their application for the SWUS project.

NGA-West2 RESORCE PE ER-Arizona Lin et al Fin ite-Fault (2011) Simu lations DCPP Median SS and RV SS and RV DCPP complex & splay SS and RV ruptures PVNGS Med ian Greater AZ SS and NML PVNGS Kappa for Ar izona rock Earthqua kes site in Arizona PVNGS Median for CA/Mex sources CA/Mex eqk 200-400 km DCPP & PVNGS x Single-Station Sigma x x PVNGS CA/Mex eqk Sigma for CA/Mex 200-400 km sources DCPP & PVNGS x HW scaling

Table 5.1-l: Data Sets Used for the Evaluation of the Median Ground Motions DATA BASE Of DATASET SUBSET USE Of DATASET ORIGIN NGA-W2oc-Mm PEER NGA- M 25.0 Evaluation of Ule medtan West2 No HW sites* ground motion mode l (for Vno2 250 m/ s OCPP) for base FW model NR£Jeqk ~ 3 Adjusted to Vno=760 m/s NGA-W2py.ym PE ER NGA- M~_O Evaluation of the median West2 -70km s Rx5 70 km for both SS ground motion mode l ( for andNML PVNGS)

VSJtl 2 250 m/s NR£Jeqk ~3 Adjusted to Vm=760 m/s EURPY-M'ED Reference M~S _O EvaJuation of the median database of R~70km ground motion mode l (for Seismic Ground for both SS and NM L PVNGS)

Motion in vn;J 2 250 m/s Europe NRedeqk ~3 (RESORCE) Adjusted to V90=760 m/s PEER-AZuni PEER Arizona Earthquakes from NGA-West2 in Estimation of tile median path Regions 1 and 2&3 recorded at tenns for Regions 1 and 2&3 stanons in Anzona (for PVNGS)

NREc/eqk 23 N11.Ec/slation ~ 5 SIMoc-vm SCEC !iimulations SS: M55, M6_0, M6_6, and M7.l Evaluation of tile median using the broad ground motion mode l (for band platform REV: MS-5, M6_0, and M6.5 OCPP)

SIMH.w SCEC simulations REV: MS.S, M6_0, and M65 Eval uation of the scaling of using the broad Dips: l0, 20, 30, 45, 60 the HW effect for magnitudes band platform ZTo : 2 .5, 7 _5, 12 km between MS and M6.5, and for Zltl11 sca ling {for OCPP)

• Includes the followrng* 0 ~Rx 2 -70 km for both SS and REV; 0 $Rx S 70 km & R3 S 10 km for SS; and 0 s RJC ~ 70 km for SS, & Dip 2: 80 deg.

Table 5.11-3~' Data Sets Us;ed for the Evalu atio111 c f ttlle Kaippai and Ground Moti0ns ffrom Splay a nd Conn p lex Ru ptu res DATABASE OF DA11iASET SUBSET USE OF DATASET OR1GIN PEER-AZM.?i"A PEIE R Ariza nai Earthquakes in Estim a,t ion of kappa for staitioru in Arizona record ed 1 c:entr.a1 Arizcma (for PVNGS) at statton:s in

,Arizona SIM:;l*T SOEC sim ulatlions Mam n !Evaluation of the methods to 1r:omp ute ll.lsingth e SS: M7.0-M7A ground motiof!ls for sp lay ruptllr e:s (fl()r broadbaiind REV ~ MJ.O - M7-4 DCPP)

RJllatfo rm S?la1f SS: M6.0-M6-4 REV ~ M6.4 SIMc:o:npto. SCiEC simulations SS: M6L7- M7_4 IEva[uation o f the m et hods. to compute llllSBlngthe ground motiorms for complex rupt ures tilro a cfba rn d REV ~ M6.4-M7.0 (f or OCPP) platfo rm

Tabte 5.1-4: Data Sets Used for the Evaluation of the Aleatory Standard Deviation DATABASE OF There are multiple versions of the DATASET SUBSET USE OF DATASET ORIGIN dataset depending on which of the EURMS& Refttence database M~S 1. Computation of residuals of Seismic Ground DIST :S SO km 2. Development of single- NGA-West2 GMPEs is used for the Motion in Europe Nucfeqk ~ 3 muon sigma models based on (RESORCE) Nw:/site ~ 3 European data for application residuals, because different subsets to PVNGS sources in Greater Arizona ( ;,~-rr'lt model) of the NGA-West2 dataset were G lO BA4.~ PEER NGA-West2 M ~ S 1. Use of residuals from GMPE used by the different developers of GLOBAL-=....m.u. and Un et aJ (2011) DIST S SOkm developers GLOBA~-cn* N1m;/eqk ? 3 2. Development of single- the GMPEs. For these cases, the GLOBA~14 N11Et/'site ? 3 station sigma modeJs based on the g1obal data tor application subset name includes the reference Selection applied to to both OCPP and PVNGS the subset used by sources in Greater Arizona. GMPE as well.

the GMPE ~ ~ o.t.i - 30 model) developer NGA-W2~ AS'-lol PEER NGA-West2 M?55 L Use of residuals from GMPE NGA-W 21.Mt-.**UU.ll.. 200 S Dist s 400 k:m developers NGA-W2~M" N.uc/eqk ~ 3 2. Development of single-Nuc/site ~3 stauon sigma model based on large distance data for Selection applied to application to PVNGS sources in the subset used by regions 1, and 2&3 (

theGMPE ;ss-.JJIU! -w modeJ) developer NGA*W ~ PEER NGA-West2 California L Use of residuals from GMPE NGA-W2CA~ earthquakes with M developers NGA-W2c.A~ NGA- ?5 2. Development of single-w2~" DIST :S SO k:m station sigma models based on Nuc/eqk ? 3 CA data for application to OCPP Nuc/site ?3 (-S>-<.>-i & ~.t-~ models)

Selection appl~ to the subset used by the GMPE developer PEER-AZ,.\~"UL PffR Arizona Earthquakes from L Compurauon of residuals PEER-AZ,,..~CSA1A NGA-West2 in 2. Development of singl~th PE ER-AZ,4")K'l't.a Region land sigma for application to PVNGS PEER-AZ,i.7H-CB1A Regtons 2&3 from eanhquakes in Region 1 recorded al. stations and Regions 2&3 { - Rlll in Anzona model)

PEER NGA-WE~st2 Database (PVNGS: Used_to_determiine single station sigma)

• Expanded the previous PEER (2008) NGA ground-motion database to include worldwhde ground-motion data recorded from shallow crustal earthquakes in active tectonic regimes after 2003 and range of magnitudes included in the database was extended down to M3.

• Each NGA-West2 developer selected their own subsets from the full NGA-West2 data set, such as, remove recordings that :

- had missing key metadata,

- data that were judged to be unreliable (metadata or ground motion data), _

- Were not considered applicable to shallow crustal earthquakes in active tectonic regions, and

- class 2 earthquakes (aftershocks).

e-------------

Normal Strike* Slip

&-r--;r--,r-;--~.,.--,,........,~

Reverse PEER NGA-West2 f 6 -c~4.~r-+i Is ~~~liililliiiiiili Database 3 -L----+-~+""

01 10 100 3

1000 0 1 10 100 1000 3-+--~~

01 10 100 1000 (PVNGS: Used to RJ8 (\ml RJS llaTI) RJ5 ()om)

IOOOOO-r:==T""T"1~~i"."'T"?~t:=i determine single station sigma) 01 1 10 20 0.1 I 10 20 Majority of the earthquakes are either Period(wc) Period (sec~

10000 . - - - - - - - - strike slip (57%) or reverse and reverse J 1000 t~--=~...;...-=*'""""'........~=;1 0

J oblique {26%). Normal and normal.

~ 100 ;--~~~--:::-~-:-:::-1 l oblique earthquakes. make up 17% of

~

d 10-----..--- the earthquakes but only 8% of the 10 100 RQcoiU.,g~ Plr Earthciualll tOOO recordings.

B20000----~--......--~~

rim) i I=::o~.VS3~50~ I For M > 5, the distribution of "6

Ir 1000 * -* *-

j 1oo ~i...........+-.-_ ,.__,,,........._-t earthquakes is similar: 49% strike-slip, J 10 31% reverse {31%), and 20% normal.

~

a ,......_-+--------~

50 100 1000 3000 l/530(n"O'Sl Figure S.L l -1: Summary of the data distribution of the NGA-West2 database using the .subset of reliable data selected by ASK14.

Arizona Ground Motion Database(Kishida et al.,2014a)

(PVNGS: Used to determine AZ site kappa, fv'ledian GMPE & sigma for CA/Mex sources)

• PEER compiled a database of ground motions recorded by 15 stations in AZ produced b'V 26 earthquakes that occurred in AZ, CA, or Mexico after 20 07. 1

• The closest station to PVNGS (Z14A) is located 8 km away.

• 13 AZ recording stations aroLJnd the PVNGS site were part of the USArray, and 2 were ~>ermanent stations managed by the USGS/CalTech Southern California Seismic Network.

• Dataset consist of 12 small (rVI_< 3.5) earthquakes in AZ with hypocenter distances of 9 to 300 km, and 14 earthquakes in CA and Mexico with RJB distances between 150 and 600 km.

• 14 CA and Mexico earthquakes had recordings in CA that were included in NGA-West2 database, but the ground motions in AZ were not included in the NGA-West2 database.

1 Normal Strike-Slip Unknown ia IJ Arizona Ground i* 3 4

3~~~..;.;.;;..14-+~+-~~ Motion Database 2 2 l L--===~~=3 {PVNGS : Used to determine AZ 10 100 1000 0.1 10 100 1000 R.lB i.n-} RJB(kll"I) site. kappa, Median GMPE & sigma 1:<<00 """"------.....-....,...-..,,...-.,..-,..,-~

t*y()O for CA/Mex sources) l a:

• coo b *()()

~

} IO .._.,_,_..~~~:--"rl--1 1r, .:>.'

• ()()

~

1:-

~ 10 L

?

~

. :ro 0

10 100 n..........,. r.,. :.11'"

r~ 1()00 0

..t*. 100

~

ID

-~  !'i() 1()()

Figure 5.1.2-1: Summary of the data distribution of the PEER-Ar izona database.

V530 for the Arizona Sites (Chapter 3 of Kishida et al., 2014a)

• SWUS sponsored a study to measure the site conditions at the AZ seismic stations.

• For 10 of the 15 stations in the PVNGS region, spectral analysis of surface wave (SASW) dispersion technique was used to determine the detailed site velocity profile, average velocity in the upper 30 m of the profile (VS30), average velocity for th e entire profile (VS,Z), and NEHRP site classification were derived.

• 3 independent inversion techniques were employed .

• Results showed that th ere were 2 typical site types:

deep stiff soil {alluvium) sites - 8 stations, V530 is typically in t he range of 370 to 690 m/s, with usually gentle monotonic increasing velocity with depth thin soil over rock sites - 2 stat ions, V530 is in the range of 970-1240 m/s, with greater variance in t he fie ld dispersion data and greater variabi lity between the inverted profiles t han for th e deep alluvium sites.

Table 5.1.2-2: Ariz.ona Data Set V:30 Values (from Table 3-1 of Kishida et al., 2014a).

Max N umber of Magnitude Distance Inversion Used for Station Vno (m/s)

Recordings Range Range (km) Depth of Vs Kappa profile (m}

Z14A 11 1.2 - 3.4 50 - 206 490- 524 108 Yes U SA 9 1.2-3.4 59-301 424-460 99 Yes Y16A 4 1 .5 - 1.5 158 - 159 970-1028 40 Yes YlSA 8 1.5-3.4 119 -189 499 -566 40 Yes Z15A 2. 1.5-3.1 69 -251 373 -464 39 Yes 113A 7 1.5 - 3.4 96 - 2.50 1140-1237 38 Yes Y14A 9 1 .5-3.4 82 -147 473 - 526 50 Yes Y13A 5 2.0-3.4 42-135 532- 560 so Yes 114A 7 1.2 -2.4 28-183 380-404 so Yes ZHA 5 1.5 -2.4 88-111 652 -689 50 Yes

Akkar et al. Subset fro m. RESORCE Database 1

(PVNGS: Used to determine Median GMPE in Greater AZ (SS+ NML) & single station sigma)

• RESORCE is a pan-European earthquake strong-motion databank and is one of the products of the Seismic (3round Motion Assessment (SIGMA) project. Database accelerograms were processed using a uniform methodology.

• Subset of the RESORCE database used by Akkar et al. (2014c) was used by the SWUS GMC project and it excluded recordings from the full RESORCE datadase that:

- had no measured VS30 values,

- earthquakes with magnitudes less than 4.0,

- earthquakes with unknown style-of-faulting,

- hypocentral depth greater than 30 km, recordings at RJB distances greater than 200 km, and

- events with only one recording.

• Akkar et al. subset from RESORCE database includes

- coverage for distances from 5 to 200 km

- magnitudes from 4.5 to 6.0 normal and strike-slip mechanisms

- sampled periods for the response spectral values between 0.01 and 4 sec

- 1,041 3-component recordings frorn 221 earthquakes recorded at 322 strong-motion stations

Normal Strlke*Slip Reverse

• ------- Akkar et al. Subset from ie......_...~.._..:-~~

~

~5 *++~~M RESORCE Database (PVNGS: Used to determine Median GMPE 10 100 1000 in Greater AZ (SS+ NML) & single station R..e (lim) sigma) t 1(JCX)()

J 1COO t--~~~..,._.,:.- Majority of earthquakes O 100 and accelerograms are from strike-slip I£ 10--1-..,_,.........,....,.__.............~......

- SS events (38% of events and 36% of

- - RV 1

01 i=====~..:,_:_.!..!.;..-J_~

I 10 20 10 20 recordings) and normal events (47% of PttiUd l11Cl t*ooo events and 51% of recordings).

f w 100 ~...-........--~....................,~

'& The number of reverse-slip events and J

i recordings in the database are small s'

I --~--__............._.............._

10 100 1000 100 compared to the other style-of-faulting classes (only 15% of the events and 13%

R"'llD'Cll"QIB 1-'er !:At'~

~20000 *r-------..-----:---.

of recordings are from reverse events).

i 10000 - All>lat el al v

er 1000 -

M>6, R<70

\1~50""'9

• in

'a Most of the sites have VS30 values in the I~

• oo range of 250 to 750 m/s. There is sparse i 10 coverage for V 530 values greater than

~ , _.____.__~~ 750 m/s.

50 !00 1000 VSJO (Miii Figure 5.1.3-1: Summary of the data distnbution of the European database (Aldcar et al, 2014c) using the subset of re'1able data selected for the development of me Akkar et al. (2014a and 2014b) model.

Lin et al. Database a----------a-.----.-----

Normal Strike*SUp Reverse (PVNGS: Used to determine t6-H-"-tr~~fM

~ 5 ..,_...,.,...,..,;-~ ---~

. . . .crititt 6 iE!~~-a.a 5 -+~...,+-,n-,.W 6

~....,..,,.,+--+**:;~...,.,.

single station sigma) 3 _._~_._------1 3 -+---i--~-- 3 r 1 - - + -......-+---"-f 0.1 10 100 1000 0.1 10 100 1000 0.1 10 100 1000

• ,::km-r::>~~~~~ R..ll~(lc111) 1 Database was d eve Io ped to study the 1000 - r r - - RJQ C\On :---i-.........................__........

components of the aleatory  : ~~ ~ *=

variability ( site, path, and source j x

100 ..............__ __.

t HHF terms) using the extensive data set of 'O

.8 10 ground motions from Taiwan. z s

• Because of the large number of 1_._----~

0.1 I 10 20 I 10 20 aftershocks from the 1999 Chi-Chi ..

v1000 ~---~~

Pwiod Cwc:I Pono:l l&OC) earthquake, there are many sites  !

with large numbers of recordings per ~ 10014~g~~~

site. j i 10 .......___.............--.-;.~,...............

• For the objective of evaluating the 1 components of variability, Lin et al. ] 1- t - - - - - _ . . . . . . - - 1 10 100 1000 100 restricted their data set to sites with RooorttirGS Pat Eanl'lquake

~ 20000-.---......-----.---,

at> 10 recordings per site. i '0000

• Data set only used for the evaluation CI:

?>

I 1000 100....__ _ _- - - i -_ _ _ _,

of the single-site within-event z

~

standard deviation, cPss* r J

~

3 1 ~------

50 100 1000 3000 VS30 ltl\'1)

Figure .5 .1.4-1: Summary of the data distribution of the Lin et al (2011) Taiwan database.

Additional Data from Normal Style-of-Faulting Earthquakes 2008 Wells, Nevada Earthquake

• Located "'10 km NE of the town of Wells, NV * *

• Moment magnitude of 6.0

• Occurred on a previously unmapped fault (USGS, 2014)

• B&R earthquake not part of the PEER NGAWest2 database

• Ground motion and metadata from this earthquake were compiled and are summarized in Appendix I

• Earthquake occurred during the time in which the USArray was deployed in AZ

• 8 stations were within 100 km of the

• epicenter (Figure 5.1.5-1), but only 7 had records that were not clipped

• VS30 values were inferred based on the surface geology from the USGS (2007) database and the correlation of VS30 and geology given by Wills and Clahan (2006)

Figure 5.1.5-1: Eplcentnl location of me Wells (NV) event. Also Shown are me muons within 100 km that recorded the event.

Additional Data from Normal Style-of-Faulting Earthquakes 2011 Fukushima-Hamadori, Japan ~

Earthquake . ~1

• M6. 7 occurred in eastern Tohoku, 11 April l 1!

2011 and was apparently triggered by the , Ji 11 March 2011 Tohoku (M9.0) ~

earthquake. ~- ij 37°

• 598 records from K-net and KIK-net *Y'-'

stations within 800 km from the § epicenters were collected by PEER and ~ .~> . ,',.~

• processed following the same procedures r~ ~ ....,

as the NGA-West2 database. ~

• 1/

• Earthquake consisted of a complex rupture involving several faults. According j Vertital orr,~1 c L ~ 'J to Shiba and Noguchi (2012}, the source 1m was comprised of two rupture planes . -~,.---

• The source parameters for the two rupture planes were derived by source Sllp 1m 1 0 10 inversion using empirical Green's functions. The seismic moment is o..-------============::::===---m:l J 40.6.. 1 40 .8 ~

36. 0 partitioned between the faults.

Figure 5.1.5-2: Map showing the shp d5tributlon and the vertical offset associated to the 2-011 ApriJ 11 Fukushima-Hamadori inland earthquake (Figure from Shiba and Noguchi, 2012).

Kappa for the AZ Sites Normal o ...-----~

Strike-Slip e.---r-~~~

Unknown s-r-~=-r~~

7 1 -t--=~.,;-t-t--~~

e s ~~~

Cl

• Previous. region-specific estimates. '! 11-+----'--'.;+t-0-~~

i*

11 4 ~~~~~

of kappa for rock sites in AZ not 3 3 2 -+----4--1---~

available. +----___._~,___~ I  ; I~ , -----~-..;-~

10 100 1000 0.1 I 10 100 1000 0.1 1 10 100 1000

• Data used for kappa consists of 12 AJB tkrr) A.I! (tun) 100000-.---.__,_...,..,.~

fUI (lcmJ

• SS earthquakes with hypocenter *

• SS

~10000 - NMI.. ~,~I"!"!-"!

locations in AZ.

NI.IL UN J 1000 - .______.1- .........-- 1

• Hypocenter locations were i 100 ---£---t-'t-'~t--;*-t-;o4.H1;t"

£ 10 -- -- -*

provided by Jeri Young of the AZ Earthquake Information Center I

.. 1000-.----.-----........

Perico (NIC}

10 20 *-------i OI I Period (MCI 10 20 (2013, personal communication).

~

I LU 100

• Events range in magnitude 1S f~

between 1.2 and 3.4, and were id 10 I~~~~

recorded by stations located at  :>

e I 10 100 1000 epicentral distances between 9 Roc:ordln!P Per urflqu~~o and 300 km.

11 out of the 12 events occurred 1=

'g 1000 T r . + : 1 = = = - r . + = = I 1 - -~~ 1 in 3 distinct clusters.

t z

100 "tt.~........~;t--"":::-1 I 10 ~--+--t-++-1'~

s 1 r-------------1 50 100 1000 3000 VS30 Cmls)

Figure 5.1 . 7-1: Summary o f the rustribuuon of data for the PEER-A4.mo..a. dataset.

Kappa for the AZ Sites

• The zero-distance kappa (K0 ) values for the recording sites estimated using 3 different methods:

- the acceleration spectrum approach (Anderson and Hough, 1984)

=

[KAs1 resulting in K0 33+14 msec;

- displacement spectrum approach (Biasi and Smith, 2001) [K 05,]

resulting in Ko.= 50 msec (set as upper bound);

- broadband approach (EPRI, 1993, and Silva et al., 1997) [K 88] resulting in K0 = 33 msec with a +one standard deviation range of 20-54 msec.

• All of the recordings are from broadband velocity instruments with a sampling rate of 40 samples/sec and a Nyquist frequency of 20 Hz. The high-frequency lir11it is about 16 Hz. The limited high-frequency bandwidth for the USArray data severely limits the resolving power for K.

• Estimates of site kappa values are sensitive to the assessment of site amplification. In Arizona Ground Motion. Database site amplification is included for all of the. kappa estimation methods.

Finite-Fault Simul<~tion Database for Me<:Jian

• A data base of ground motic,ns from finite-fault simulations was developed lJsing multiple simulation methods implemented on t~1e SCEC Broadband Platform (BBP) (Maechling et al., 201~5).

• The scenarios for simulations were selected to address four issues:

- magnitude and distance scaling of near-fault ground motions,

- rules. for estimating ground rnotions. from complex ruptures, .

- rules. for estimating ground rnotions. from splay ruptures,. and.

- magnitude scaling for HW effects for moderate magnitudes.

(MS to M6).

Subsets for PVNGS Median for Greater AZ Table 5.1-2: Data Sets Used f or the f valuatK>n of the M edian Ground Mooons Sources DATABASE OF DATASET SUBSET USE OF DATASET ORIGI N NGA-W2oc:-t.to PEER NGA- M ~5.0 Evaluation of the median West2 No HWsites* ground motion model (for VSJQ~ 250 m/s DCPP) tor base FW model NuJeqk ~ 3 Adjusted to V s.J1J =760 m/s NGA-W2"'1m> PEER NGA- M ~S .O Evaluation of the median West2 -70km S RXS 70 Ian for both SS ground mooon model (for and NML PVNGS)

VSJO ~ 250 m/ s NnJeqk~3 Adjusted to Vno=- 760 m/s EU~ Ref ere nce M ~.O Evaluation of the median database of R_s70km ground mooon model (for Seismic Ground for both SS and NML PVNGS)

Motion in VSJIJ ~ 250 m/s Europe N11£Jeqk ~ 3 (RESORCE) Adjust ed t o Vna= 760 m/s PEER-AZ,.,,,. PEER Arizona Earthquakes from NGA-West2 ln Estimation of the medliln path Regions 1 and 2&3 recorded at terms for Regions 1 and 2&3 m t fons in Arizona (for PVNGS)

NAE<J'eqk ~ 3 N11Ec/stat1on ~ 5 SI Moc-Mii> SCfC simulations SS: M S.5, M 6.0, M 6.6, and M 7 2 Evaluation of the median using the broad ground mooon model (for band platform REV M5.S, M 6.0, and M 6.S OCPP)

SIMHW SCfC simulations REV. MS.S, M 6.0, and M 65 Eval uation of the scaling of using the broad Dips* 10, 20, 30, 45, 60 the HW effect for magnitudes band platform zlOJI. 2.5, 75 , 12 ltm between M5 and M 6.5, and for ZTD' scaling (for DCPP)

• includes the f ollowing: 0 ~ R.~ -70 ltm for both SS and REV; 0 s R~ s 70 km & Ra~ 10 km for SS; and 0 s Rx ~ 70 km for SS, & Dip ~ 80 deg.

Normal Reverse NGA-W2Pv-MED dataset

!8+-..,._,_~'f-:,....,._,.~_.,~M

~ G-t-..-...........,.~_,,_....'""'""~..,......f.'-l PEE R NGA- M ~5_0 3

0 1 1 10 100 0.1 10 100 0.1 10 100 1000 West2 -70km::: Ft~ JO km for b oth SS RJB~ml RJ8(kll'I) fUI (km) and NML 100000 r-'

v'ii!J ~ 250 m/ s NREc/e(!k ~. 3 1000-Adjusted to Vsr716 0 m /cs 0 100 ~

l C'

2 10 1

I 10 20 0.1 I 10 20 Percd (9eC) Peflod (MC)

)1000 1000

& I Iii t

IM 100 0

'l!

f J*

i 10 3 10 1!!

u

~

i tO 100 1000 FloCO!dil'QO Por E11rlt)QuaM t=

llS 1000 z

f !00 t 10

~

d

!00 1000 3000 V$30 (llY$)

Figure 5.3.2-1: Summary of the data distnbutton of the NGA-W2PY.Mm dataset.

Nonna! Strike-Slip Reverse I * - -,__--,__,____,,....----. I 7 7

~6

~

6 EURPV-MED dataset

!6 5 5

.. " 4 Ref erence M~5J) 3 10 100 3

0.1 10 100 1000 3

01 10 100 100 dat.aba ~e of 1Rre*~70 km RJB(rcmJ RJll (km) FUJ(km)

Seismic Groun d fo r bottl SS and N.M L 100000 Motion in Vs;o 2:: 250 m/ s r'oooo [Eu rope NRKle ~k ~ 3 J

l5 1000 (RESORCE) Adj usted to 'V~*7160 m/ s 100 I

1: 10 14-......;...~..,._""4-__;_~..,._...._--f 1 1-----,! -----1---1 0I , 10 20 0 I I 10 20 Period (&ftll) PMod~

-~ ~:-J.,...~--;-...-.--r-r;~-.....-1 j 1000 ~.l;i;.-..I-.;........;.._;_:_:~ _ _ : _j

- EUR PV*MED

't>

tz 100.t~~~-

~ 10 ii

~

3 I00 I000 3000 VS30 1mJs) figure 5.3.2-2: Summary of the data distribution of the EUR~ dataset.

Normal Strike-Slip Rovorse 8 a.._.~-.-~~~--.~--.

!6-t-~--;-~~~--,+-t PEER-AZPATH dataset 3 6-1--~4-__.__.___,..4"-1 3 1----+------- 3 - --~ ........~_...~--~.......

10 100 1000 0.1 10 100 1000 01 10 100 1000 R.iB lkm) RJB (km) R..e (knl) 100000*n------------..-!"""l'"'!"!'"!'l'!'~"""I PEER Arizona Eanhquakes from NGA-West2 in Estimation of the median patt Regions 1 and 2&3 re-corded at tenns for Regions 1 and 2&3 stations in Arizona (for PVNGS)

NR.Ec/eqk ~ 3 110 NR.Ec/station ~ 5 z

l *--~-.-~........-1-~-,-~

01 I

.10. . . 20

--1 Ot Perkid ($eel Ptlt IOd ($0CJ i1000....--,-,.---...--,-,-=,-.-..,.-,-,...,..=-o

~ ..

~

!l> 100 0

~

I 1o;=~~:m-=~~rrt==r::R'.ii:it1

} 10 - --=....,__:..:.....:.....--:...-;..--<1

~

~

I 10 100 Recordings Pet Eart'1Quake d

- PEEA* AZ PATH 100 1000 VS30 (rnhl)

Figure 5.3.3-1: Summary of the data distribution of the PEE.R-AZ,~~ dataset.

Data Sets for the <l>SS and <l>SP-R Mo~~els

• There are 4 types of the single-station sigma.

(<PSS) models :

1. short-distance global n1odel based on RESORCE,

2. short-distance global n1odel based on NGA-West2 and Lin et al. (2011),

3. long-distance global m odel based on NGA-West2, 1

4. 2 short distance modells based on CA data in the NGA \6Jest2 database-~ DCNPP, not PVNGS

5. a subset of the AZ dataset was compiled to derive the magnitude-independent <PsP-R and path-adjustment models for PVNGS.

Data Sets for the <l>SS and <l>SP-R Models

1. short-distance global model based on RESORCE

• A subset of the. Akkar. et al. . (a subset of RESORCE) data set was selected to be used in developing the single-station sigma models for application to PVNGS for the Greater Arizona sources.

• M ~ S.0 and RJ B < SO km

• At least 3 recordings/earthquake and at least 3 recordings/site

• magnitude limit (e.g. all magnitudes are used to constrain the site term, but only a subset of the site corrected residuals with M ~San d RJB ~SO km are used to compute the cpSS values) .

• Subset consists of 223 recordings from 73 earthquakes (3S normal, 2S strike-slip, and 13 reverse events) recorded at 79 stations

• Ground-motion data at periods greater than 4.0 seconds are not available for the European dataset.

Normal Strike-Slip Reverse 8 8 7

~.

• - ~

h.

1~

I 7

~

I I I I "~ ""

- .. I i . ,..*l 6 - ~t 6

I I

(\

(\

I I 5

I

- 5 I

I 1

I

'V1 4

I I

I

-I- -h I

4 rt I I I

~ ,_,_ ,.._

3 I 3 r t' I ' "

0.1 1 10 100 0.1 10 100 1000 0.1 10 100 1000 RJB (km) RJB (km) RJB {km)

Figure 5.4.2-1: Magnitude-distance distribution of the EURPH1ss dataset. (Note: the m inimum of 3 recordings per earthquake and per site is applied to the full data set. This plot only shows the subset for distance less than 50 km and magnitudes great er t han 5.0)

Data Sets for the <l>SS and <l>SP-R Models

2. short-distance global model based on NGA-West2 and Lin et al.

• Combination of the NGA-West2 data and the Lin et al (2011) data

• Within-event residuals for Taiwan (Lin et al., 2011) were combined with the NGA-West2 residuals after removing residuals from common recordings from *raiwan earthquakes

• Idriss 2014 (ld14) residuals are not used for the SS <P evaluation because Idriss did not separate his residuals into between-event and within-event terms.

• A subset is developed for each <)f the four NGA-West2 GMPEs that separated the within-event and between-event residuals.

• M ~ S.O and. RJB. <SO km

• At least 3 recordings/earthquake and at least 3 recordings/site

• magnitude limit (e.g. all magnitudes are used to constrain the site term, but only a subset of the site corrected residuals with M ~ S and RJB ~SO km are used to cornpute the <PSS. values).

Normal Strike-Slip RevetSe Data Sets for the <l>SS and <l>SP-R Models

• 3 0 I 10 100 0. I 10 100 1000

,~---------.;i...;.~

01 10 100 tOOO

2. short-distance global model based A.II I""' R.81.,...) RJ8 (1cm) figure SA.1-1: MagnltudHllstance distribution of the GLOB~. dataset.

NoonaJ Stnke-Slip Reverse on NGA-West2 and Lin et al. 8 7

!e Table 5.4.1-1: Number of recordines and eantiquakes in the "°bal dataset (M ~ 5, R < 501<m) for four o f J,

the NGA*West2 models for the short-distance ;xr .

• 3 l 3...-..--...........-----~

01 10 100 1000 0. 1 10 100 1000 0, 10 100 1000 ASK14 8SSA14 C814 CY14 RJ8 foanl RAJ t""') RJ!I (1cml Reeion Nb Recs NbEqks Nb Recs Nb Eqk.s Nb Recs Nb Eqks Nb Recs Nb Eqk.s Figure 5.4.1-2; MainitudHlistance distnbUtJOn of the GlOBAi...-.-.u.. dataset.

CA 672 54 630 48 342 38 349 41 TalWcln 846 28 846 28 846 28 846 28 Normal Reverse Japan 65 3 65 3 0 0 63 3 * *

• Italy 69 15 62 11 6 4 0 0 7 7 Ouna 10 2 102 26 17 4 0 0 Total 1,662 102 1,705 116 1,211 74 1.258 72 Je

• t ta 5

• * ~

3 , ,

01 10 100 1000 0.1 10 100 *000 0 I 10 1CIO 1000 lllB ...... R.8 (*"') RJ8 (llnll rifUre SA.1-3: Maznitude-dlstance distnbutJOn of the GLOB~ dataseL CA dataset comprises about 30% to 40% of the recordings and 40 to 60% of the earthquakes 7 Normal I

Str1ke-S1lp Reverse

I

• in the global dataset, while the Taiwanese data  :.. I I represent 50% to 70% of the number of I

' I

• . ~

3 ,_,________.._____.......,.

recordings and 25% to 40% of the number of 3 . . 10 100 3-r---.__--.................................

1000 0. I 10 100 1000 0I 10 100 ICOO 01 RJ8 OMll IUllli"') IUl()it!IJ earthquakes . FiJure 5.4.1-4: \1aenitude-distance distnbutJOn of the GLO~ datase-t.

Data Sets for the <l>SS and <l>SP-R Models

3. long-distance global model based on NGA-West2

• Taiwan and CB14 sets were excluded from the <PSS analysis, because dataset with magnitude greater than or equal to M 5.5 and distances of 200 to 400 km lacks Taiwanese data in that range of interest and CB14 used mixed-effects regression to derive t he anelastic attenuation term from data with RRUP > 80 km, but allowed the source ter ms to vary from those with a maximum RRUP distance of 80 km.

• :. 3 sets of NGA-West2 residuals were used to develop the <PSS model for PVNGS - Distant California and Mexico sources.

• Global dataset in this magnitude and distance range of interest (M ;::: 5.5, distance 200-400 km) consists of 264-415 recordings from 4 to 23 earthquakes (mostly from Japan).

Table 5.4.4-1: 'N umber of recordings and earthqu akes in t h e global dataset {M ~ 55, R = 200 to 400km) for three of the NGA*West2 models ASK14 BSSA14 CY14 Region Nb Recs Nb Eqlts Nb Recs Nb Eqlts Nb Recs Nb Eqk s CA 133 1 209 4 160 2 Japan 131 3 157 4 129 4 China 0 0 49 15 0 0 Total 264 4 415 23 289 6 ASK14 BSSA14 CY14 8 B 8 75 7

75 7

75 7

I L- ()()(>

• l,. 6.5 . *i.... 6.5 0

"i.. 6 .S - *""-.x;

~ _J ~ ~

6 6 ~ ~ 6 -- ....

55 SS s.s 5 s 1 5 200 250 300 350 400 200 250 300 350 400 200 l50 300 350 Rtup (km) R!up (km) R:rup (km)

CA Jap4n < CA China Japan ) CA lap;in

Data Sets for the <l>SS and <l>SP-R Models

5. a subset of the AZ dataset was c:ompiled to. derive the magnitude-independent <l>SP-R and path-adjustment models for PVNGS

• Dataset consists of earthquakes in CA and 15 1 -

Mexico that have been 6.s - u recorded at the 9 stations ~ 6 .

in the vicinity of PVNGS. ii 5.5 - u

• 49 records from 11 s- ._ . . il C~ o ~)

earthquakes with rupture 4.5 l 66 ~ b.

distances that range from I '¢ ~

'¢ 8 V¢(.}

<ll. o

• 200 to 500 km are used to 0 I

100 I

200 300 400 500 600 compute cpSP-R for 3 Rrup (kmt regions: Region 1 (4 Region l . Region 2 Region ~

earthquakes), Region 2 (3 Figure 5.45-1: Magnitude-distance distribution of the PEER-AZPAJK dataset.

earthquakes), and Region 3 (3 earthquakes)

Proponent Models for Median Ground Motions Table S.5.1-1: Extrun& GMPEs Considered for the Development of Median GrounO-Mouon Models (contmues on the following paeeJ Candidate few Candidate few Candidate for Candidate for Candidate for Candidate for GMPE Comments PVNGS Greater PVNGS Distant GMPE Comments PVNGS Grea ter PVNGS Distant OCPP DCPP Arizona Sources California Sources Aritona Sourus California Sources Abrahamson et Update of Yes Yes Yes GtMzer (2014) NGA We.st 1 database No, not published No, nonnal fault No, not published al (2014) Abrahamson and Sliva plus 2004 Partrfield in a peer- not speoflcally in a peer-(2008) and 200S San Simeon revteWed Journal studied reviewed 1oumal Reeional for Turtey No, supers~ed No, superseded No, superseded Hermkes et al. RESORCE No, experimental No, expenmental No, expenmental Akkarand (2013) Expenrnental Model Cainan (2010) by pan- by pan- by pan-Idriss (2014) Update of ldnss Yes, not used for No, normal fault Yes Europe/Middle Europe/Middl e Europe/Middle (20081 R- < 3krn not speofically East ACR GMPEs East ACR GMPEs East ACR GMPEs studied plus non Kanno et al Used only depth for No, no clear No, not relevant No, no clear Cahfomla/Wester (2006) separation of event separation of ACR to tectonics sepamion of ACR n Arizona type from SZ interface from SZ interface attenuation earthquakes earthquakes, plus Akkuetal Update of Akkar and Yes Yes No, non non (2013a. 2014) Bommer (2010) Cahfom1a/Wester Cllhfomta/Wester nAnzona n Arizona anenuat1on anenuation B1nd1et al. Update of 81ndi et at No, extrapolation Yes, M > 7 not a No, e.nnipolauon McVerry et al R~nal for New No, speafic to No, specrfK to No, speafK to (2014a. 2014b) (2011) above M7 st&ntficant above M7 (2006) Zealand New Zealand, New Zealand, New Zealand, problematic a1 contributor to problematic at superseded by superseded by su~edby some pertods hazard some penods Clobal models &IObal models i)obal models Boore et al. Update of Boore and Yes Yes Yes that use recent that use recent that use recent Atkmson (2008) New Zealand NewZealllnd NewZeaJand (201 4) earthquake data earthquake data earthquake data Bora etal. RESORCE No, upenmental No, expenmental No, expenmenral (2013) E.menmental Model Pankow and Update of Spudictl et No. not relevant No. superseded No, not relevant Pechmann al (1999) to tectonKS by more recent to tectonacs Bradley (2013) Modification of ChlOu No, re&ional No, rqional No, reitonal (2004) models (e.e. NGA-et al. (2010) for New adjustment to adJUstment to adjustment to West2)

Zealand other model other model other model Zhao and Lu Proposed chanee 1n Yes No, not relevant No, non tnduded m study included tn study &neluded in study (2011) maenitude saline to tectonics Clllrfomia/ Wester Campbeft and Update of Campbell Yes Yes Yes above-M7.l nArizona Bozor'1)ta and Bozorcnia (2008) anenuation (2014) Zhao et al. Mostly .lapan data, Yes No, not relevant No, non Chiou alld Update of Chiou and Yes Yes Yes (2006) AGR and SZ wittt to tectonics califomia/Wester Youngs (2014) Youngs (2008) and separate factors nAmona Chiou et al (20101 attenuation Derras et al. RESORCE No, experimental No, expenmental No. experimental (2013) Expenmental Model facool1 et al. Global data, primanly No, single Linear No, smile lmear No, sin&le linear (2010) Japan magnitude scaJm.g mainitude scalane magnitude scaling over enure range over ent1re ranee over entire range

Selection of Candidate Models Subset of candidate models was selected based on the following seven criteria:

1. More recent published GMPEs by the same development team were selected over older GMPEs on the basis that the newer models would have benefited from more data and refinements to the approach. In the case of the modified magnitude scal ing suggested by Zhao and Lu (2011), the Zhao et al. {2006) GMPE is selected along with a modified form based on Zhao and Lu {2011). The basis is that Zhao and Lu have not developed a full GMPE to replace Zhao et al. {2006).

2. Models that represent an adjustment of another model to fit data from a specific region which is not CA or western AZ were not selected (e.g. Bradley, 2013). The basis for rejecting these models is that they have been adjusted from one region to another and should not be adjusted back to the original region or to a third region.

3. Models that do not extrapolate well beyond the magnitude-distance range over which they were developed were not selected. For example, models that have only a single linear magnitude sca ling term were not selected, as eva luat ions by many investigators of data sets containing a large range in magnitude have shown that a single linear magnitude sca ling term does not capture the magnitude sca ling over the range of magnitudes from MS to M8.

4. Models that do not clearly separate sha llow crustal earthquakes from those occurring as a part of subduction were not selected. The basis for rejecting these models is that the magnitude and distance sca ling from subduction zone earthquakes is different than from crustal earthquakes.

5. Models developed as research. tools.were not selected ..The basis for rejecting these. models. is that they have not developed to the point where they could be used for engineering application.

6. Models developed for a relatively small specific region different from the ones of interest (e.g. Italy) were not selected. The basis for rejecting these models is that the data from a single region may be too limited to capture scal ing for the full range of magnitude aind distance of interest and the specifics of the regional behavior may be different from CA and western AZ.

7. Models that have not been peer reviewed or vetted. by the larger scientific community were not selected (Graizer 2014).

Host Kappa Values for Selected Candidate Models at VS30

• For PVNGS, the planned site response analysis accounts for differences in the kappa implied for the candidate GMPEs and the kappa for rock sites in central AZ. The kappa implied by the spectral shape of the GMPEs is called the "Host" kappa .

• Host kappa values were derived for normal-faulting scenarios with a dip angle of 50 degrees, for M 5.0, 6.0, and 7.0 and Rx distances of 5, 10, and 20 km on the footwall.

• Under the direction of the Tl Team, Dr. Al-Atik estimated the host kappa values for the 7 candidate GMPEs selected for PVNGS using the IRVT approach :

Table 5.5.3-1: Host kappa va lues for the seven candidate GMPEs for PVNGS Greater Arizona sources for a reference V530 of 760 m/sec.

GMPE ASK14 BSSA14 CB14 CY14 ASB14 Bi14 ZH06 Host Kappa {sec) 0.045 0.038 0.037 0.041 0.042 0.045 0.042 ASK14 =Abrahamson et al. {2014), BSSA14 = Boore et al. {2014),

CB14 =Campbell and Bozorgnia {2014), CY14 =Chiou and Youngs {2014), Bi14 = Bindi et al.

{2014a, 2014b), ASB14 = Akkar et al. {2014a, 2014b), and ZH06 =Zhao et al. {2006).

Single-Station Sigma Approach -

Rodriguez-Marek et al., 2014

• PVNGS used the RodrigL1ez-Marek et al., 2014 single-station sigma approach:

- Partially non-ergodic app roach 1

- Removes the systematic site response effects from the traditional ergodic withirt-event standard deviation

- Avoids double counting c>f the epistemic uncertainty of the site response that ca1n occur if the traditional ergodic sigma is used, and the site response also addresses the epistemic 1uncertainty.

Single-Station Sigma Approach. - Rodriguez-Marek et al., 2014 Swttzeriand Rodriguez-Marek et al., :1 2014 approach compiled f s 4>

~.

8 .a

~2' s

~.

ground motion data with :L0 100 200 300 3

2 0 100 200 300 multiple recording per 0 1$lance (km) Dist nee(~)

I T~tw.n Tu~ey 81 site from 5 regions. . ¥:[ *1 .-

J T

is ~ ...

~ 5 c 5 .;

i'

~ 4 i 41 I

3 3 Distribution of the data in 2 0 100 200 300 J 2-0 100 200 300 DIStanc:e (km) Distance Ckm)

Japan All Sites terms of magnitude and 8 7 - ....

• distance for each region q;;

-g

~5 8 ' Cll "O

3 c;

Cit e

5 i, Q

t

• 3 3 100 200 300 100 200 300 01s1*nc:e (kml 0 tance CkmJ Figure 5.6~2-1 : Distribution of data for smgl~station sigma as compiled by Rodriguez-Marek et at.

(2013)_ (Figure from Rodriguez*Marek et al., 2013.)

Single-Station Sigma Approach - Rodriguez-Marek et al., 2014

• Average cpSS values for M > 4.5 and R < 200 km for each region.

• The lower plot shows the traditional ergodic cp values and the upper plot shows the partially non-ergodic cpSS. values.

• cpSS values are mainly between 0.4 and 0.5 In units and are much more consistent across regions than the cp values.

08,---------------,

07 08

...*05l b <l r* ~ *

<J 0*1 + + + ~

03 j 02 PG~ 01 020 3 0 5 10 30 Penod (s) *

+

08 c '"'*'t 07*

06 Q

6

+.*

• x 0

7 for 5 Hz. This feature of the Bi14 model is not seen for all spectral frequencies, but for simplicity of application, the Tl Team decided not to use the Bi14 model for M > 7 for all spectral frequencies.

Figure 6.2.2-1: Magnitude. scaling of the candidate. GMPEs at 5.0 Hz for strike-slip event with RX distance 5 km.

1 .50 ~--.-----.-----.-----..----.------..-----.------..---.------.------.-----.--..-----.------.----.---~----r----.----.-----.-----.----.----....----.---..----.-----.-----.--:.

T = 0.2 1.00 SS Rx=5 0.70 r---1 Cl

\.......,J

<( 0.50 en a..

• ASK14

• ASB14 0.30

• Bi14

• BSSA14

• CB14 0.20 f

• CY14 5.0 5.5 6.0 6.5 7.0 7.5 8.0 M

PV~JGS

• The regional Q (quality_factor) structure can have a significant impact on attenuation of the seismic waves with distance. Several of the proponent GMPE models (Section 5.5) have used data out to distances of 400 km to constrain the distance scaling, but these GMPEs did not include ground-motiorl data from sites in Arizona

• In general, the Q values av eraged over the paths from 1

California to Arizona indicate that events from central California have higher Q, events from Baja California have lower Q, and events from southern California and the Transverse Ranges hav1e intermediate Q.

• Although there are differences in the. Q between California and Arizona, these differences do not lead to a significant discrepancy in the average distance attenuation at distance of 200-400 km in central Arizona as compared to California. The differences in Q between California and Arizona \Nould have a larger effect for larger distances (>. 400 km), so the conclusion by Kishida et al. {2014) that the average attenuation is similar for southern California and central Arizona only applies to distances less than 400 km.

• Based on the evaluation of the candidate GMPEs for application to earthquakes in California and ME~xico recorded in central Arizona given in Kishida et al. {2014), the Tl Team judged that the NGA-West2 GMPEs are suitable for estimating path terms for the paths from California and Mexico to central Arizona.

Summary of Kishida et al. 2014

• This report summarizes the products and results of a study on the collection, processing, and analysis of earthquake ground-motions recorded in Arizona at several recording stations within 200 km from the Palo Verde Nuclear Generating Station in central Arizona.

• Additionally, "kappa" a measure of energy dissipation in the top 1 to 2 km of the crust, was estimated by three methodologies. The average KO (kappa at. zero-kilom,eter. distance) was. estimated from all. sites as 0.033 sec.

• Response spectra of the recorded ground motions in Arizona were compared with those predicted by the NGA-West2 ground motion prediction equations at large distances in Arizona. The comparison showed that overall the recorde~d 5% damped response spectral ordinates were over predicted by the NGA-West2 models by a range of 0-0.35 natural log units for events occurring in Central California, and by a range of 0.2-0.7 natural log units for events occurring in Southern California and the Gulf of California.

TA Data Limitations

• The Nyquist frequency for all TA recordings is 20 Hz, because the TA has <3 low sampling rate of 40 Hz. An anti-alias filter w las applied to the TA data at about 80% of the Nyc1uist frequency with a corner frequency near 1.6 Hz.

• These observations are interpreted to indicate that the noise from microseisms and other sources are dominant at frequencies less than about 0.5 Hz.

• After differentiation vel,Jcity records high-frequency limit becomes even lower than 8 Hz.

• SAs based on TA data <3re most likely severally affected. by low sampling rate. of 40 samples/sec and necessity to. differentiate velocity recordings.

• Examples of similar effects are demonstrated in Graizer 2012 (lSWCEE) paper.

Observations

• Q-values in Arizona arE~ inconsistent with conclusion about samE~ attenuation as predicted by the NGA-West2 GMPEs.

• Kishida 2014 conclusic~n (The comparison showed that overall the recorded 5% damped response spectral ordir1ates were over predicted by the NGA-West2 models) can not be justified by TA recordings .

Difrancesco, Nicholas

Subject:

FW: DEDO/ET Weekly Fukushima Status Meeting Location: 0 -13D20 Start: Tue 05/05/2015 2:00 PM End: Tue 05/05/2015 3:00 PM Show Time As: Tentative Recurrence: Weekly Recurrence Pattern: every Tuesday from 2:00 PM to 3:00 PM M eeting Status: Not yet responded Organizer: NRR_JLD Resource

Wayne, DEDO/ET Weekly Status Brief.

Plan to suggest bullets similar to these for my management:

WUS Seismic Screenino Results Letter l* Group 1 - Columbia, u1ao10 t_;anyon; Group 3 - Palo Verde

• June 2017 Seismic Probabilistic Risk Assessment Consistent with Industry Endorsed Timelines [i.e. no relien

• Interim Evaluations for Diablo Canyon are adequate without completion of the Expedited Approach [i.e.

LTSP provides safety basis]

• No immediate safety issues - interim evaluations support time to complete SPRAs

• Communication plan developed to support public issuance on May 12. 2015 Thanks ,

Nick

---

Original Appointment-----

From: NRR_JLD Resource Sent: Tuesday, October 14, 2014 2:48 PM To:. NRR_JLD Resource;.Johnson, Michael;. Dean, Bill;. Uhle,. Jennifer; NRR_ET_Activity Resource; Davis,. Jack; Franovich, Mike; Bowen,. Jeremy;. Proffitt, Andrew; Inverso,.Tara;. Evans, Michele Cc: McHale, John; NRR_LT_Calendar Resource; Mohseni, Aby; Kokajko, Lawrence; DprNrrCal Resource; Oesterle, Eric

Subject:

Copy: DEDO/ET Weekly Fukushima Status Meeting When: Tuesday, May 05, 2015 2:00 PM-3:00 PM (UTC-05:00) Eastern Time (US & Canada).

Where: 0 -13D20 10/14/2014 - Requested by JLD Management; Scheduled by Esther Cho (2239) 1

Difrancesco, Nicholas From: DiFrancesco, Nicholas Sent: Tuesday, May 05, 2015 11:32 AM To: Dapas, Marc; Kennedy, Kriss; Uselding, Lara;. Walker, Wayne; Miller, Geoffrey; Dricks, Victor Cc: Shams, Mohamed; Alexander, Ryan

Subject:

POP for. DEDO/ NRR/R-IV JLD Status Brief (2pm Eastern)

Attachments: POP - JLD Status (05.05.15).docx

Folks, Below and attached is the POP for the JLD status briefing including the WUS Seismic Screening anp, 6

Prioritization letter. The briefing is scheduled for 2pm Eastern (Bridgeline: 888-455-0567; Passcodef,_(b-)(- )_ __ ,

Please let me know if there are any questions or concerns .

Regards, Nick OiFrancesco Sr. PM - Seismic Reevaluations (301 ) 415-1115 PURPOSE Update NRR ET on status of JLD activities EXPECTED OUTCOMES .

• wus -

' t5' Screening & prioritization letter - targeting issuance 05/12/15 o All 3 plants screen in for sPRA. no immediate safety issues o Columbia & Diablo Canyon - Group 1

• sPRA due 06/30/17

• Diablo Canyon - Separate letter on ESEP* SP provides safety basis 1

• Public Meetings o Diablo Canyon (04/28/15) o Columbia 06 04/15) o Palo Verde - Group 3

• sPRA due 12/31 /20

• Public Meetina 06/09/15 Non Responsive 2

Difrancesco, Nicholas From: DiFrancesco, Nicholas Sent: Wednesday, May 13, 2015 2:56 PM To: Shams, Mohamed; Jackson, Diane; Wyman, Stephen; Vega, Frankie Cc: Devlin-Gill, Stephanie; Munson, Clifford; Heeszel, David; Seber, Dogan; Stieve, Alice

Subject:

Summary of R2.l Licensing Calls (Related to WUS Hazard Reviews, CEUS Hazard and ESEP Reviews)

Folks,.

Completed a number of licensing calls today for awareness or follow-up.

Diablo Canyon:

Licensee expressed interest in timing of ESEP decision and feedback on licensee SFP Evaluation Approach NRC Suggested Resolution: Target 2 months for ESEP letter- SFP approach should be considered within the guidance development over the summer (JLD and DSEA)

Licensee Action: Timing of information request response early next week (will be PG&E formal letter or public website)

Columbia :

Licensee interested in Public Meeting NRC Request: For ground motion questions only general topics earlier than May 21, if possible to support consultant schedule (Dogan/DiFrancesco)

NRC Request: Request for 2nd bridgeline for licensee staff (Frankie)

Palo Verde:

Licensee interested in Public Meeting and conditional screen-out process NRC Request: Information to support Palo Verde Public Meeting by May 26.

Non Responsive Thanks ,

Nick 1

Senior Project Manager - Seismic Reevaluation Activities U.S. Nuclear Regulatory Commission Office of Nuclear Reactor Regulation Japan Lesson Learned Project Division.

nicholas.difrancesco@nrc.gov I Tel: (301) 415-1115 2

Hill, Brittain From:Hill, Brittain Sent:22 Apr 20 15 11 :49:34 -0400 To:Munson, Clifford ;Flanders, Scott;Kock, Andrea;Jackson, Diane

Subject:

FYL Diablo seismic hazards plot For your reference, a rather messy summary plot for Diablo that shows the

- new GMRS (red),

- Double Design and Hosgri design-basis earthquakes (green),

- 84'h percentile ground motions from 1988 long-term seismic program (long dash),

- 1.35x margins curve derived from the LTSP (short dash) ,

- 201 1 Shoreline report, 84"' percentile ground motion for San Luis Bay fault, ergodic approach (highest deterministic accelerations) (blue dashed)

- 201 1 Shoreline report, 84"' percentile ground motion for San Luis Bay fault, single station approach (highest deterministic accelerations) (Blue)

- 2014 report, 84'h percentile ground motion, linked Shoreline+Hosgri+San Simeon faults, at turbine building (highest deterministic. accelerations) (Purple)

'~efutes arguments that PG&E has been lowering seismic hazards by using "new and improved methods

- facts show calculated hazard has increased.

Shows that implementing probabilistic approach has resulted in higher hazards than 2011-2014 deterministic approaches.

Demonstrates Hosgri design basis was robust and that well-established (i.e., SSER Rev 34) LTSP margins provide assurance plant is safe to operate.

3.0 I Diablo Canyon , ,-- ......

2.5

/ \

- GMRS \

\

- - DOE \\

- HE \

-E1 c

2.0

- - LTSP84%ite

---

LTSP x 1.35

\

\

\

\

\

0 \

+;;

~ 1.5 Q)

- - 11 SLB Ergo ' \

(i) - 11 SLBSStn ' \

\

8 - 14 Sh+HE+SS \

\

< '--i--------

1.0 0.5 0.0 -~-------.,...---------------------

0.1 1.0 10.0 100.0 Frequency {Hz)

Brittain E. Hill, Ph.D.

Sr. Technical Advisor US Nuclear Regulatory Commission MS T7-F03, NRO/DSEA Washington, DC 20555-0001 Ph+1 301 415-6588* Fax+1 (301)415*5399; Mobil (b)( 6l mail: Brittain.Hill@nrc.gov