ML15355A550

Diablo Canyon, Units 1 and 2 - Transmittal of Response to NRC Request for Additional Information Dated October 1, 2015, and November 13, 2015, Regarding Recommendation 2.1 of the Near-Term Task Force Seismic Hazard and Screening Report

ML15355A550
Person / Time
Site:	Diablo Canyon
Issue date:	12/21/2015
From:	Strickland L J Pacific Gas & Electric Co
To:	Document Control Desk, Office of Nuclear Reactor Regulation
Shared Package
ML15362A569	List: DCL-15-154, Response to NRC Request for Additional Information Dated October 1, 2015, and November 13, 2015, Regarding Recommendation 2.1 of the Near-Term Task Force Seismic Hazard and Screening Report ML15355A550
References
DCL-15-154, TAC MF5275, TAC MF5276
Download: ML15355A550 (39)

Text

acilic Gas and Electric Company(/) December 21, 2015 PG&E Letter DCL-15-154 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D.C. 20555-0001 Docket No. 50-275, OL-DPR-80 Docket No. 50-323, OL-DPR-82 Diablo Canyon Units 1 and 2 L. Jearl Strickland, P.E. Director Technical Services Diablo Canyon Power Plant P.O. Box 56 Avila Beach, CA 93424 805.595.6476 E1 H8@pge.com 10 CFR 50.54(f) Response to NRC Request for Additional Information dated October 1. 2015. and November 13. 2015. Regarding Recommendation 2.1 of the Near-Term Task Force Seismic Hazard and Screening Report

References:

1. PG&E Letter DCL-15-035, "Response to NRC Request for Information Pursuant to 10 CFR 50.54(f) Regarding the Seismic Aspects of Recommendation 2.1 of the Near-Term Task Force Review of Insights from the Fukushima Dai-ichi Accident: Seismic Hazard and Screening Report," dated March 11, 2015 (ADAMS Accession No. ML 15071A046) 2. NRC Letter, "Diablo Canyon Power Plant, Unit Nos. 1 and 2 -Request for Additional Information Associated with Near-Term Task Force Recommendation 2.1, Seismic Reevaluations (TAC Nos. MF5275 and MF5276)," dated October 1, 2015 (ADAMS Accession No. ML 15267A774) 3. NRC, "Information Request Related to Diablo Canyon Regulatory Audit of Reevaluated Seismic Hazard," E-Mail from N. DiFrancesco (NRC) toP. Soenen (PG&E), dated November 13, 2015 (ADAMS Accession No. ML 15323A200) 4. NRC Letter, "Diablo Canyon Power Plant, Unit Nos. 1 and 2 -Request for Additional Information Associated with Near-Term Task Force Recommendation 2.1, Seismic Reevaluations (TAC Nos. MF5275 and MF5276)," dated June 29, 2015 (ADAMS Accession No. ML 15153A033) 5. PG&E Letter DCL-15-095, "Response to NRC Request for Additional Information Regarding Recommendation 2.1 of the Term Task Force Seismic Hazard and Screening Report," dated August 12, 2015 (ADAMS Accession No. ML 152248575) A member of the STARS (Strategic Teaming and Resource Sharing) Alliance Callaway

• Diablo Canyon

• Palo Verde

• Wolf Creek Document Control Desk December 21, 2015 Page 2

Dear Commissioners and Staff:

PG&E Letter DCL-15-154 On March 11, 2015, Pacific Gas and Electric Company (PG&E) submitted PG&E Letter DCL-15-035, "Response to NRC Request for Information Pursuant to 10 CFR 50.54(f) Regarding the Seismic Aspects of Recommendation 2.1 of the Near-Term Task Force Review of Insights from the Fukushima Dai-ichi Accident: Seismic Hazard and Screening Report," (Reference 1 ). On October 1, 2015, and November 13, 2015, the NRC Staff requested additional information to complete the review of PG&E's response (References 2 and 3). These information requests were subsequently discussed by the NRC Staff and PG&E representatives during an audit held in Bethesda, MD on December 3, 2015, which included clarifications of the information requests. PG&E's responses to the Staff's questions, including the clarifications identified during the December 3, 2015 audit, are included in the Enclosure to this letter. The updated ground motion characterization information, described in the Enclosure to this letter, supersedes that previously submitted to the NRC on March 11, 2015, (Reference 1) and updates the information provided by PG&E in response to the NRC Staff's June 29, 2015, request for additional information (Reference 4) in PG&E Letter No. DCL-15-095 (Reference 5). This information represents the final seismic hazards and ground motion response spectrum for Diablo Canyon Power Plant, which will be used as input to the screening evaluation in response to the NRC's Request for Information Pursuant to 10 CFR 50.54(f) Regarding the Seismic Aspects of Recommendation 2.1 of the Near-Term Task Force Review of Insights from the Fukushima Dai-ichi Accident. The conclusions from the screening evaluation remain the same as in Reference 1. PG&E makes no new or revised regulatory commitments (as defined by NEI 99-04) in this letter. If you have any questions, or require additional information, please contact Mr. L. Jearl Strickland at (805) 595-6476. A member of the STARS (Strategic Teaming and Resource Sharing) Alliance Callaway

• Diablo Canyon

• Palo Verde

• Wolf Creek Document Control Desk December 21, 2015 Page 2 PG&E Letter DCL-15-154 I have been delegated the authority of Edward D. Halpin, Senior Vice Power Generation and Chief Nuclear Officer, during his absence. I declare under penalty of perjury that the foregoing is true and correct. Executed on December 21, 2015. Director, Technical Services mjrm/50465913-99/4557

Enclosure:

cc: Diablo Distribution cc/enc: Marc L. Dapas, NRC Region IV Administrator Nicholas J. DiFrancesco, NRR/JLD Senior Project Manager Siva P. Lingam, NRR Project Manager Gonzalo L. Perez, Branch Chief, California Department of Public Health John P. Reynoso, NRC Acting Senior Resident Inspector A member of the STARS (Strategic Teaming and Resource Sharing) Alliance Callaway

• Diablo Canyon

• Palo Verde

• Wolf Creek PG&E Letter DCL-15-154 Enclosure Page 1 of 50 Response to NRC Request for Additional Information dated October 1, 2015 and November 13, 2015 Regarding DCPP Seismic Hazard and Screening Report NRC Request dated October 1, 2015 Review of Site Response Evaluation By letter dated August 12, 2015, Pacific Gas and Electric Company (the licensee) sent a response to the U. S. Nuclear Regulatory Commission's (NRC's) June 29, 2015, request for additional information (RAJ) for Diablo Canyon Power Plant, Unit Nos. 1 and 2 (DCPP, Diablo Canyon), which provides an estimate of the site amplification using the analytical site response modeling approach. As shown in Figure 1 of the RAJ response, which compares the DCPP site term as developed from the observed ground motion or empirical approach with the site term from the analytical approach (i.e., SPID3 methodology), there are notable differences in the site term from the two approaches particularly in the 1-3 Hertz, as well as the higher frequency ranges. The licensee attributes these differences to the analytical modeling approach using (1) a shallow velocity model that does not capture the effects of the site-specific deep velocity profile and (2) a broad range of site kappa values that far exceed the range of observed values for the site. Commenting on the second factor, the RAJ response states on page 4, The broad uncertainty range for kappa is included in the response to the questions to be consistent with the SPID methodology, but, based on the high frequency content of the observed ground motions at DCPP, we consider this low kappa value to be not applicable to DCPP. The NRC staff notes that the guidance in Appendix B of the SPID was developed to systematically capture the uncertainty in the properties of the near-surface materials in the site-amplification functions and the subsequent control point seismic hazard curves using a probabilistic methodology. Broad uncertainty ranges for the subsurface material properties are necessary for sites for which the level of detail and scope of geological and geotechnical investigations are limited; however, the DCPP site has abundant subsurface data that can be used to constrain the range of uncertainty for these properties. a) Please provide an updated analytical site response analysis which reflects the uncertainties in the material properties specific to the Diablo Canyon site, with 3 The NRC endorsement of the industry issued SPID Guidance "Screening, Prioritization and Implementation Details (SPID) for Resolution of Fukushima Near-Term Task Force Recommendation 2.1'. Appendix B-contains an approach to develop site-specific amplification factors (Agencywide Document Access and Management System (ADAMS) Accession No. ML12333A170).

PG&E Letter DCL-15-154 Enclosure Page 2 of 50 respect to the shear-wave velocity profiles, low-strain damping or kappa, and capturing the potential differences in the site terms as developed from both the empirical and analytical approaches. The RAJ response cites an updated 3-D velocity model (Reference 34) for development of the base case shear wave velocity profiles. The NRC staff review of the 3-D velocity model provided in Reference 3 indicates that the near-surface shear wave velocities beneath seismic station ESTA27 are higher than previous estimates used to develop the empirical site term for the March 11, 2015, Seismic Hazard and Screening Reporl . b) Please update the SHRS to reflect the empirical site response analysis that incorporates the higher near-surface shear wave velocities for station ESTA27, shown in Reference 3. In addition, provide updated control point seismic hazard curves, uniform hazard response spectra, and ground motion response spectrum that incorporate any changes to the DCPP site term. Also, please provide any updates and refinements to the empirical site response approach in an Appendix to the revised SHSR. NRC Request dated November 13, 2015 In follow-up to the Regulatory Audit conducted on Sept 11, 2015 (Agencywide Documents Access and Management System [ADAMS] No. ML 152448099) NRC staff identified technical information needs and issued a request for additional information dated October 1, 2015 (ADAMS No. ML 15267A774) to supporl reviewing the Diablo Canyon Power Plant's reevaluated seismic hazard. In response to the technical information requests, PG&E made available electronic records for review on the PG&E electronic reading room. In review of those records, the NRC staff has identified the following additional information needs to support understanding of the site response approach: VS-kappa adjustment factors

• Clarify the source(s) of the host-region VS30 760 mlsec profile(s) and provide the profile(s) in tabular format

• Provide the target VS profiles (lower, middle, upper) in tabular format

• Provide the quarter wavelength (QWL) or square-root impedance (SRI) linear site amplification factors (or explain applicable approach) for the host VS30 760 m/sec profile(s) compared to the QWL amplification factors for the target VS profiles

• Provide the magnitudes and distances used to compute the response spectra compatible [Fourier Amplitude Spectrum] FAS using Inverse Random Vibration Theory (or explain applicable approach) 4 Fugro (2015). Updated of the Three-Dimensional Velocity Model for the DCPP Foundation Area, May 2015.

PG&E Letter DCL-15-154 Enclosure Page 3 of 50

• Provide the host kappa values and target site kappa values

• Provide the target reference baserock kappa values where kappabaserock = kappasite -kappaprofile and indicate the depth for the reference baserock horizon

• Provide the final VS-kappa factors used to modify the [Southwest United States] SWUS median [Ground Motion Model] GMMs Analytical Site Response Approach

• Provide in a table: layer description, thickness, density, and VS values for the lower, middle and upper base case VS profiles as well as the scale factor used to develop the lower and upper profiles

• Provide the shear modulus and damping ratio curves and the depth ranges over which each curve is implemented

• Provide the site kappa values for each of the three profiles

• Provide the number of randomizations, and the correlation model used to randomize the VS about each of the three base case profiles

• Indicate whether the damping ratios are constrained to a maximum of 15 percent

• Provide the magnitudes and distances of the earthquakes used for the input kappa corrected spectra and indicate the location where these spectra are input into the site response analysis

• Provide a description of the approach used to develop the site amplification factors, including the incorporation of both the aleatory and epistemic uncertainty.

• Indicate whether the amplification factors are constrained to not fall below 0.5

• Provide a description of the approach used to develop the control point hazard curves, including how the aleatory uncertainty in the amplification factor is incorporated into the hazard integral Empirical Site Response Approach

• Provide a description of any deviations from the approach used to develop the empirical site term as described in Sections 2.3.5 and 2.3.6 of the March 15, 2015 Seismic Hazard Screening Report [SHSR] submittal

• Provide the VS30 values used for [seismic station] ESTA27 and ESTA28 Final Ground Motion Response Spectra (GMRS)

• Provide the bases for developing control point hazard curves that combine the results of both the analytical and empirical site response approaches, including the weighting for the two approaches Pacific Gas and Electric Company (PG&E) Response PG&E Letter DCL-15-154 Enclosure Page 4 of 50 In order to provide a comprehensive response to the above information requests, Pacific Gas & Electric Company (PG&E) has prepared a technical discussion describing the updated site response evaluation for the Diablo Canyon Power Plant (DCPP). This updates the information previously provided in Section 2.3, "Site Response Evaluation," and Section 2.4, "Control Point Response Spectra," of the March 11, 2015, DCPP Seismic Hazards and Screening Report (Reference 3). The conclusions described in Section 4.0, "Screening Evaluation," Section 5.0, "Interim Evaluation," and Section 6.0, "Conclusions," of Reference 3 remain unchanged.

PG&E Letter DCL-15-154 Enclosure Page 5 of 50 1. INTRODUCTION The specific questions from the November 13, 2015, request for additional information (RAI) are listed in Table 1-1 along with the section of this document in which the response to the question is provided. Note that the responses to questions from the October 1, 2015, RAI are implicitly addressed in this enclosure. Table 1-1 -November 13, 2015, RAI Questions and Response Sections Question Response VS-kappa Adjustment Factors Section Clarify the source(s) of the host-region VS30 760 m/sec 2.1 profile(s) and provide the profile(s) in tabular format. Provide the target VS profiles (lower, middle, upper) in 2.2 tabular format. Provide the quarter wavelength (OWL) or square-root 2.3 impedance (SRI) linear site amplification factors (or explain applicable approach) for the host VS30 760 m/sec profile(s) compared to the OWL amplification factors for the target VS Provide the magnitudes and distances used to compute the 2.4 response spectra compatible [Fourier Amplitude Spectrum] FAS using Inverse Random Vibration Theory (or explain applicable approach). Provide the host kappa values and target site kappa values 2.4 Provide the target reference baserock kappa values where 2.5 kappabaserock = kappasite -kappaprofile and indicate the depth for the reference base rock horizon. Provide the final VS-kappa factors used to modify the 2.6 [Southwest United States] SWUS median [Ground Motion Model] GMMs. Analytical Site Response Approach Provide in a table: layer description, thickness, density, and App A VS values for the lower, middle, and upper base case VS profiles, as well as the scale factor used to develop the lower and upper profiles. Provide the shear modulus and damping ratio curves and 3.2 the depth ranges over which each curve is implemented. Provide the site kappa values for each of the three profiles. 2.4 PG&E Letter DCL-1-5-154 Enclosure Page 6 of 50 Table 1-1 -November 13, 2015, RAI Questions and Response Sections (continued) Question Response Analytical Site Response Approach (continued) Section Provide the number of randomizations, and the correlation 3.3 model used to randomize the VS about each of the three base case profiles. Indicate whether the damping ratios are constrained to a 3.2 maximum of 15 percent. Provide the magnitudes and distances of the earthquakes 2.4, 3.1 used for the input VS-kappa corrected spectra and indicate the location where these spectra are input into the site response analysis. Provide a description of the approach used to develop the 3.1 site amplification factors, including the incorporation of both the aleatory and epistemic uncertainty. Indicate whether the amplification factors are constrained to 3.2 not fall below 0. 5. Provide a description of the approach used to develop the 5.2 control point hazard curves, including how the aleatory uncertainty in the amplification factor is incorporated into the hazard integral. Empirical Site Response Approach Provide a description of any deviations from the approach 4.2 used to develop the empirical site term as described in Sections 2.3.5 and 2.3.6 of the March 15, 2015, Seismic Hazard and Screening Report [SHSR]. Provide the VS30 values used for [seismic station] ESTA27 4.1 and ESTA28 Final Ground Motion Response Spectra(GMRS) Provide the bases for developing control point hazard curves 5.2 that combine the results of both the analytical and empirical site response approaches, including the weighting for the two approaches.

2. VS-KAPPA ADJUSTMENT FACTORS PG&E Letter DCL-15-154 Enclosure Page 7 of 50 The hazard calculation was conducted for a reference rock site condition corresponding to a time-averaged shear-wave velocity in the top 30 meters (VS30)=760 meters per second (m/s) for a site with a shear-wave velocity (VS) profile representative of the data used to derive the ground motion prediction equations (GMPEs) used in the Southwestern U.S. (SWUS) study by GeoPentech, "Southwestern United States Ground Motion Characterization SSHAC Level 3" (Reference 6). To adjust the results for the reference rock condition to the site conditions for the control point, the differences between the VS profiles and kappa values for the reference rock condition (called the host profile and host kappa) and the control point (called the target profile and target kappa) are evaluated. 2.1 Reference VS Profile for California for VS30=760 m/s The host profile for the SWUS GMPEs is taken as the generic California profile for VS30=760 m/s developed by Pacific Engineering and Analysis, and described in Kamai et al, "Nonlinear Horizontal Site Response for the NGA-West2 Project" (Reference 7). The layer thicknesses, shear-wave velocities, and densities for the host profile are listed in Table A-1 in Appendix A. 2.2 Control Point Definition and VS Profiles The control point is defined as a hypothetical location with VS profiles representative of the range of site conditions over the power-block and turbine building footprint at elevation 85 feet. This region is shown in Figure 2-1. To define the velocity profile for the control point, the three-dimensional (3-0) velocity model described in the May 2015 version5 of the Fugro Report, "Update for, the Three-Dimensional Velocity Model for the Diablo Canyon Power Plant (DCPP) Foundation Area," (Reference 4) was used. The range of one-dimensional (1-0) profiles extracted from the 3-D model are shown in Figure 2-2 for the top 125 meters (m). The central profile is developed based on the geometric mean VS profile, which approximates the median profile. The standard deviation of the natural logarithm of the VS is depth dependent with a maximum value of 0.21 at a depth of 10m. The lower and upper profiles shown in Figure 2-2 are based on plus and minus (+/-)1.6 standard deviations above and below the median VS. A minimum range of 10 percent was applied (affects the lower part of profile in Figure 2-2). Because the distribution of the velocities is not normal, the +/-1.6 standard deviation range are near the bounds the 1-D profiles from the best 3-0 model. The Fugro Report for the 3-0 model (Reference 4) gives an additional uncertainty of about 0.15 natural log (LN) units due to different tomographic inversions. This additional 5 The 3-D velocity model was updated in November 2015 (Reference 12). The May 2015 and November 2015 velocity models are compared in Appendix A.

PG&E Letter DCL-15-154 Enclosure Page 8 of 50 uncertainty was not included in the range shown in Figure 2-2, but when the broad range of upper and lower profiles shown in Figure 2-2 are combined with the profile randomization, the resulting profiles used in the site response will capture the range of alternative 3-D models due to different inversions. To compute the upper and lower bound shallow velocity profiles, the central profile is scaled by factors shown in Figure 2-4 representing +/-1.6 standard deviations of the LN (VS) values or a factor of 1.1, whichever is larger. This standard deviation did not include the additional epistemic uncertainty due to the tomographic inversion uncertainty. The Fugro Report, 1-D Vp Profile below the DCPP Area (Reference 5) provides an estimate of the VS in the depth range of 125 m to 3000 m. Below that depth, the profiles were extended to a depth of 8 kilometer (km) based on the reference profiles for the NGA-West2 data set provided in Pacific Engineering and Analysis (PEA) Report, "Development of Amplification Factors for the Diablo Canyon Nuclear Power Plant: Site-Wide Profiles," (Reference 8). Figure 2-3 compares the VS profiles for the Host region with the VS profiles for the central, upper, and lower target VS models for the full 8 km depth range. The layer thicknesses, shear-wave velocities, and densities for each of the three profiles are listed in Table A-2 in Appendix A.v The scale factor used to develop the lower and upper profiles are shown in Figure 2-4a and 2-4b, for the shallow and full profiles, respectively. The scale factors are listed in Table A-3 in Appendix A.

E. :; Pacific Gas and Electric Company Diablo Canyon Power Plant OutlineofMajorStructuresand Yard Areas PG&E Letter DCL-15-154 Enclosure Page 9 of 50 VsGrid Points Containment Structure (Unit 1) --c.ontainment Structure (Unit 2) -Turbine Building -Auxiliary Building

• Seismic Instrument ESTA27 e Seismic Instrument ESTA28 = f u Q ESTA27 = * * * * * * *

• 60400 60200 60000 59800 59600 59400 59200 Northing in DCPP Plant Grid (ft.) Figure 2-1 -Locations of 1-D Profiles used to Define the Power-Block and Turbine Building Region 40-.E 60 --PG&E Letter DCL-15-154 Enclosure Page 10 of 50 0 200 400 600 800 1000 1200 1400 1600 1800 Shear-\Nave Velocity (r /s) Figure 2-2 -Range of VS Profiles Under the Power-Block and Turbine Building Region in the Top 125m (The heavy black curves show the central, upper, and lower profiles) (From PG&E Calculation No. GEO.DCPP.15.02 (Reference 9))

0 2,000 4,000 E 6,000 ..s:::. ..... a.. Q) 0 8,000 10,000 12,000 14,000 Vs (m/sec) PG&E Letter DCL-15-154 Enclosure Page 11 of 50 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 -Refe renee 760 -PlantRegion-Central -PlantRegion-Lowe r -PlantRegion-Upper Figure 2-3 -Comparison of the Host VS Profile (labeled Reference 760) and the Central, Upper, and Lower Profiles for the Target (From Reference 8)

£ c.. Q) a 100--Upper Profile Lower Profile PG&E Letter DCL-15-154 Enclosure Page 12 of 50 150 , . , . , , * . *

• 1 * * ; , , * * * * '!' ** 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 Scale Factor Applied to Central Profile Figure 2-4a -Scale Factors used to Develop the Upper and Lower VS Profiles for the top 150m (From Table A-3) 0 '1000 2000 _3000 E *-£4000 a.. (l) a :)ooo 6000 7000 800 I l , r r ............ I I ; f < f i r I I. I l\ I J 1 \ l i I l { \ UpperProfile ----Lower Profile l 1 ij I .: I I > r > ' ' ' ' ' ' l ' ---I PG&E Letter DCL-15-154 Enclosure Page 13 of 50 {o-=c= . ' '., 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 Scale Factor Applied to Central Profile Figure 2-4b-Scale Factors used to Develop the Upper and Lower VS Profiles for the Full Depth Range (From Table A-3) 2.3 Quarter-Wavelength Amplification PG&E Letter DCL-15-154 Enclosure Page 14 of 50 The quarter-wavelength (QWL) method can be used to estimate the effect of the differences in the linear amplification between the host and target VS profiles. The QWL crustal amplification is usually given in terms of the scaling on the Fourier amplitude spectra, not the response spectra. The QWL crustal amplification factors for the three alternative profiles for the control point and the host profile are compared in Figure 2-5. The host profile amplification is similar to the central target profile amplification for frequencies less than 2.5 Hertz (Hz). At frequencies above 5 Hz, the host profile amplification is similar to the lower target profile amplification. 4 I I I 3.5 3 E" 2.5 <t: Q) :!:: 2 Vl -' s cJ 1.5 0.5 0.1 1 10 Frequency (Hz) -Reference 760 -PlantRegion-Central -PlantRegion-Lower -PlantRegion-Upper 100 Figure 2-5 -Quarter Wave-Length Crustal Amplification Factors (Fourier Amplitude Spectra Amplification) for the Host VS profile and the Central, Upper, and Lower Target VS Profiles (From GEO.DCPP.15.03 (Reference 1 0)) 2.4 Target Kappa at Surface The host kappa value was estimated for the both the SWUS DCPP ground motion model and the NGA-West2 GMPEs. The Inverse Random Vibration Theory (IRVT) method was used for both SWUS and NGA-West2 GMPEs. The broadband inversion method was applied only to the NGA-West2 GMPEs.

PG&E Letter DC L 154 Enclosure Page 15 of 50 The broadband inversion of the response spectral shapes was conducted by PEA using the point-source spectrum with kappa being one of the parameters in the point-source model. The broadband inversion fit the spectral shape up to frequencies of 20 Hz. From the broadband inversion, the best estimate of kappa for the NGA-West2 models is 0.03 seconds (sec). An alternative approach is to use IRVT to estimate the Fourier Amplitude Spectrum (FAS) from the response spectral values and then estimate the kappa from the slope of the estimated FAS. The IRVT approach was applied to the NGA-West2 GMPEs and to the SWUS weighted ground motion model. The IRVT evaluation used M6 at rupture distances of 5, 10, and 20 km. The resulting kappa values from the IRVT method are listed in Table 2-1. Best High Low Table 2-1 -Kappa Values Based on IRVT Method (From Reference 1 0) ASK14 BSSA14 CB14 CY14 0.0405 0.0419 0.0294 0.0356 0.0438 0.0430 0.0312 0.0369 0.0361 0.0409 0.0266 0.0335 swus 0.0341 0.0366 0.0309 Based on evaluations of the kappa from the San Simeon and Parkfield earthquakes at DCPP, the target kappa is constrained to a range of 0.03 to 0.05 sec. The resulting alternative kappa values are 0.03, 0.040, and 0.050 sec with weights of 0.2, 0.6, and 0.2 representing the 5 to 95 percent range of the kappa values. 2.5 Host and Target Kappa at Base rock and in the Profiles The kappa at the surface (kappasite) is the sum of the kappa at the base rock (kappabaserock) and the kappa due to the low strain damping as modeled in the shallow layers (kappaprofile). The kappabaserock is the value of kappa input into the point source model. For this application, the baserock is at a depth of 8 km. The low strain damping is only modeled in the top 500 feet of the profile. For layers between 500 feet and 8 km, there is no damping in the layers. The kappabaserock, kappaprofile, and kappasite values for the three target profiles are listed in Table 2-2. For depths greater than 152.4 m, there is no damping in the layers and nonlinearity is not applied.

Base-case Profile Name Name M1P1K1 M1P1K2 M1P1K3 Lower M2P1K1 M2P1K2 M2P1K3 -M3P1K1 M3P1K2 M3P1K3 M1 P1 K1 M1 P1 K2 M1P1K3 Central M2P1K1 M2P1K2 M2P1K3 M3P1K1 M3P1K2 M3P1K3 M1P1K1 M1 P1 K2 M1P1K3 M2P1K1 Upper M2P1K2 M2P1K3 M3P1K1 M3P1K2 M3P1K3 PG&E Letter DCL-15-154 Enclosure Page 16 of 50 Table 2-2. Kappa Values (from Reference 8) Kappa_profile Kappa_baserock Kappa_site (sec.) (sec.) (sec.) Surface to 500 feet 500 feet (152.4 m) at Surface (152.4 m) to 8. 0 km depth depth 0.005 0.035 0.040 0.005 0.045 0.050 0.005 0.025 0.030 0.011 0.029 0.040 0.011 0.039 0.050 0.011 0.019 0.030 0.002 0.038 0.040 0.002 0.048 0.050 0.002 0.028 0.030 0.004 0.036 0.040 0.004 0.046 0.050 0.004 0.026 0.030 0.009 0.031 0.040 0.009 0.041 0.050 0.009 0.021 0.030 0.002 0.038 0.040 0.002 0.048 0.050 0.002 0.028 0.030 0.003 0.037 0.040 0.003 0.047 0.050 0.003 0.027 0.030 0.008 0.032 0.040 0.008 0.042 0.050 0.008 0.022 0.030 0.002 0.038 0.040 0.002 0.048 0.050 0.002 0.028 0.030 2.6 Final VS-Kappa Factors PG&E Letter DCL-15-154 Enclosure Page 17 of 50 In some applications, the VS-kappa correction is first made to develop the site rock motion from the reference rock condition. In a second step, the site response is conducted relative to the adjusted rock motion. For DCPP, the VS-kappa correction and the site response are done in a single step. The VS-kappa correction is integrated into the site response, but it can be separated out in the linear range. The amplification at low rock ground-motion values provides the VS-kappa correction. A reference rock peak ground acceleration (PGA) value of 0.1 times the acceleration of gravity (g) is used for the linear range. The VS-kappa factors are computed for both the broadband analytical method and the IRVT method. The resulting VS-kappa factors are shown in Figure 2-6 for the nine combinations of target kappa (kappa_site) and target VS profile. The VS-kappa scaling is similar for the two approaches with the broad-band approach showing slightly less scaling at high frequencies even though the kappa is smaller for the broadband approach. SWUS-kHost =0.0341 1 0 t 1 o*1 .______.____.__,___.__.__._.._.__.__ _ _.___.__......._.__.L...L...L..'--L..--'---'---'---'-----'--'---'---'-' 10"1 10° 101 102 Frequency (Hz) -VsTarg C, kTarg = 0.04 -VsTarg C, kTarg = 0.05 -VsTarg C, kTarg = 0.03 ----* VsTarg U, kTarg = 0.04 ----* VsTarg U, kTarg = 0.05 ----* VsTarg U, kTarg = 0.03 ............. VsTarg L, kTarg = 0.04 ........... " VsTarg L, kTarg = 0.05 ............. VsTarg L, kTarg = 0.03 -wgtMean Figure 2-6-VS-Kappa Factors from the Best Kappa from IRVT (colored curves) and from Analytical Modeling (cyan curves). The mean for the analytical model is given by dashed black line. The mean for the IRVT method is shown by the solid black line. (From Reference 1 0)

3. ANALYTICAL SITE RESPONSE APPROACH 3.1 Site Response Approach PG&E Letter DCL-15-154 Enclosure Page 18 of 50 The site response approach does not provide amplification relative to the baserock site condition. Instead, the amplification is computed relative to the SWUS reference rock condition with VS30=760 m/s. The amplification is computed using ratios of the surface response spectra for the DCPP profile relative to the surface response spectra for the SWUS reference rock condition profile (Reference 8). For each profile, the surface response spectrum is computed using the point-source stochastic model. A magnitude 7 earthquake at a depth of 8 km is used for the input motion. A range of point source distances is used leading to a range of input motion levels. For each distance, the surface spectrum is computed for the velocity profile corresponding the SWUS reference rock site condition (called the host profile). Using the same distances, the surface spectrum is then computed for each of the alternative DCPP velocity profiles, kappa values, and nonlinear material properties (called the target profile). The amplification is defined as the ratio of surface spectrum for the DCPP site condition to the surface spectrum for the SWUS reference rock site condition and provides the combined effect of the linear kappa correction and nonlinear site effects. By using the ratio of the two surface spectra, this approach avoids the need for deconvolution. This process is illustrated in Figure 3-1. The logic tree for the analytical site response is shown in Figure 3-2. The alternative profiles were described in Section 2.2. The kappa values were described in Section 2.4. The nonlinear properties are described in Section 3.2 below. 3.2 Nonlinear Material Properties The material models (damping and modulus reduction) are modeled using three models: linear (M1); nonlinear rock (M2) per Electric Power Research Institute (EPRI) Report, "Guidelines for Determining Design Basis Ground Motions," (Reference 2); and nonlinear Peninsula Range (M3) per Silva et al's, "Description and Validation of the Stochastic Ground Motion Model," (Reference 11 ). For the linear model, the small strain damping is from the Peninsula Range model; however, the results are not sensitive to the selected small strain damping because additional small strain damping is added to the deeper part of the profile so that the total kappa matches the specified kappa value (Reference 8). The modulus and damping curves for the two nonlinear models are shown below in Figures 3-3 and 3-4. The nonlinear model is applied to the layers at depths up to 500 feet (152 meters). For layers at depths below 500 feet, a linear model is used. For the EPRI nonlinear model, there are 5 depth ranges from 0 to 500 feet as shown in Figure 3-3. For the Peninsula Range model, there are 2 depth ranges from 0 to PG&E Letter DCL-15-154 Enclosure Page 19 of 50 500 feet as shown in Figure 3-4. The numerical values for the 2 nonlinear models are listed in Table 3-1. Laboratory testing of the soft-rock material at DCPP was conducted in 1977 and 1978 (Reference 15). The strain dependence of the G/Gmax measurements and the damping are shown in Figures 3-5 and 3-6. These laboratory measurements can be compared with the three material models used in the analytical modeling. The range of the G/Gmax measurements are consistent with the range of the three models, with most of the data near the linear range. The lower end of the lab data is consistent with the EPRI model. Therefore, the linear and nonlinear approaches are given equal weight, and the two nonlinear models are also given equal weight. The logic tree weights are 0.5 for the linear model (M1) and 0.25 each for the two nonlinear models (M2 and M3). To avoid excessive nonlinear effects, the damping values in the site response calculation are limited to be less than 15 percent. The amplification depends on the linear amplification and the non-linear effects. The concept of limiting the amplification to be greater than or equal to 0.5 is intended to avoid large nonlinear effects that may not be reliable. Therefore, for the soil hazard calculation, the nonlinear part of the amplification is limited to be greater than or equal to 0.5, but the total amplification is not limited. For example, if the nonlinear amplification is 0.6 and the linear amplification is 0. 7, then the net amplification is 0.42 (i.e. 0.6 x 0. 7). This is allowed because the nonlinear amplification by itself is 0.6, which is above 0.5. The maximum strains at the 1 E-4 and 1 E-5 hazard levels for the two nonlinear models are given in the PEA report (Reference 8). 3.3 Profile Randomization For each of the three base profiles, 30 randomized profiles are developed based on the EPRI "footprint" model because the 3-D VS model provides local constraints on the VS profile. Because there is a gradient in the VS profile and there is not a clear depth to rock parameter, the depth to rock is not randomized. Only the VS values are randomized. 3.4 Example Results Examples of the results from the analytical approach for three ground motion levels are shown in Figures 3-7, 3-8, and 3-9. Figure 3-7 shows the amplification for a PGA of 0.2 g on the SWUS reference rock condition and reflects the linear site amplification (SA). Figures 3-8 and 3-9 show the amplification for a SWUS reference rock PGA values of 1.07 g and t.91 g which are close to the 1 E-4 and 1 E-5 hazard levels for the SWUS reference rock condition (Table B-1).

1.0 1.0 1.0 1.0 1.0 1.0 0.6 0.6 1.0 1.0 3.263 3.39 1.0 1.0 3.245 3.339 1.0 1.0 3.225 3.282 1.0 1.0 3.206 3.227 1.0 1.0 3.186 3.167 PG&E Letter DCL-15-154 Enclosure Page 20 of 50 Table 3-1. Modulus Reduction and Damping Curves* PR GENERIC SAND MODULUS REDUCTION CURVE; 0-50 FEET. 1.0 0.97 0.87 0.68 0.43 0.22 0.09 0.05 PR GENERIC SAND DAMPING CURVE; 0 -50 FEET. 1.2 1.64 2.8 5.49 10.2 15.0 15.0 15.0 PR GENERIC SAND MODULUS REDUCTION CURVE;51-500 FEET. 1.0 0.99 0.95 0.852 0.65 0.41 0.20 0.10 PR GENERIC SAND DAMPING CURVE;51-500 FEET. 0.6 0.81 1.2 2.5 5.3 10.27 15.0 15.0 EPRI GENERIC ROCK MODULUS REDUCTION CURVE; 0-20 FEET. 0.9716 0.8614 0.6294 0.383 0.1747 0.0714 0.0238 0.0084 EPRI GENERIC ROCK DAMPING CURVE; 0-20 FEET. 4.017 5.58 9.191 14.397 15.0 15.0 15.0 15.0 EPRI GENERIC ROCK MODULUS REDUCTION CURVE; 20-50 FEET. 0.9801 0.8844 0.6653 0.4177 0.1967 0.0821 0.0277 0.0098 EPRI GENERIC ROCK DAMPING CURVE; 20-50 FEET. 3.869 5.25 8.55 13.532 15.0 15.0 15.0 15.0 EPRI GENERIC ROCK MODULUS REDUCTION CURVE; 50-120 FEET. 0.9898 0.9121 0.7118 0.4655 0.229 0.0984 0.0338 0.012 EPRI GENERIC ROCK DAMPING CURVE; 50-120 FEET. 3.701 4.865 7.773 12.429 15.0 15.0 15.0 15.0 EPRI GENERIC ROCK MODULUS REDUCTION CURVE; 120-250 FEET. 0.9997 0.9417 0.7667 0.5264 0.2735 0.1224 0.0431 0.0154 EPRI GENERIC ROCK DAMPING CURVE; 120 -250 FEET. 3.534 4.463 6.926 11.14 15.0 15.0 15.0 15.0 EPRI GENERIC ROCK MODULUS REDUCTION CURVE; 250-500 FEET. 1.0 0.9668 0.8324 0.6119 0.3454 0.1649 0.0608 0.0222 EPRI GENERIC ROCK DAMPING CURVE; 250-500 FEET. 3.348 3.995 5.881 9.398 15.0 15.0 15.0 15.0

• The ten strain levels are (percent): 1.E-4.0, 1.E-3.5, 1.E-3.0, 1.E-2.5, 1.E-2.0, 1.E-1.5, 1.E-1.0, 1.E-0.5, 1.E-O.O, 1.E+0.5.

0.1 Host or Target VS P otile 10 requency Hz) 100 PG&E Letter DCL-15-154 Enclosure Page 21 of 50 c: .Q 0.8 .g 0.6 c. 0.4 0.2 0 -! 0.1 t 10 100 Frequency (Hz) 8km Point Source: M, DS, Rt Kbaserock,Q) Figure 3-1 -Cartoon of the Analytical Site Response Shallow VS model Deep Profile Kappa Gradient K=0.03 P1 (1.0) Upper model (0.2) Central model K=0.040 P1 (0.6) (0.6} (1.0) Lower mo<fel (0.2) P1 (0.20) (1.0) PG&E Letter DCL-15-154 Enclosure Page 22 of 50 Nonlinear rvlodel Linear (fv11 ) (0.5) Pen Range (M3) (0.25) EPRI Rock (fv12) (0.25) Unear (M1) (0.5) Pen Range (M3) (0.25) EPRI Rock (M2) (0.25) Linear (fv11) L:: ange (fv13) (0.25) EPRI Rock (M2) (0.25) Figure 3-2 -Logic Tree for Inputs to Analytical Site Response i;}.m CJ

• 10 w ...., ., Ul <:I * £o ., .. '"'0 Q E N II) * (JJ 0 PG&E Letter DCL-15-154 Enclosure Page 23 of 50 0 -4.0 -8.5 -8.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 m N QJ (.J &i I .-t Q ...., ttl

• cn.-l c 0... E II)

• Am LEGEND o-zo n zo-so Fr 50-120FT 120-250 FT 250-'500 FT -X '500-1000 Ff -+ 1000-2000 FT 2000-5000 FT 0 -4.0 -3.5 -3.0 '-2.5 -2.0 -1.5 -1.0 -0.5 0.0 Log (Shear Strain -Percent) MODULUS REDUCTION AND DAMPING CURVES FOR ROCK(EPRI) Figure 3-3 -Modulus and Damping curves for the EPRI Rock Model (M2) (From Reference 8)

PG&E Letter DCL-15-154 Enclosure Page 24 of 50 0 rl :;;.OJ 0 . I 0 w UJ """0 * "' 1! 'i U1 . " 0 -o 0 E '-N r(j

• U1 0 .. . . .. 0 -4.0 -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 N 't N .,...,o c N ClJ u '-ClJ Q_

• UJ I rl 0 ..,...., ro . a:::N .::n.-1 c a.. E Ill

• A CO LEG8'1D o-so n * *

• 51-500 FT * * * * .. * * * * * * * * * *

• a * * * * * * * * -4.0 -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 Log (Shear Strain -Percent) PENINSULAR RANGE MODULUS REDUCTION AND DAMPING CURVES Figure 3-4 -Modulus and Damping Curves for the Pen Range Model (M3) (From Reference 8)

N I Lr. ...... PR 0-50 ft PR 51-500 It EPRI 0-20 It EPRI20-50 It EPRI50-120 It EPRI120-250 It (j 1.4 EPRI 250-500 It ., :::1 1.1. ! 1. Qi e Tes.1 o Colunr T@S1 -Rock CUovft In SfiA.KE D I 0.0001 0.001 *HciJ\fll!.,:\)l:ta:,1'11ll' '.t*Jl.\Uro*nlh, Shear Strain (%j * *

• 0.01 Shear Strain (%) * **

• PG&E Letter DC L 154 Enclosure Page 25 of 50 ****----* 0.1 Figure 3-5 -1978 Lab Testing for DCPP Rock for G/Gmax

'" I o: 0> ,g "' 0:: 8 0.0001 PR 0-50 It PR 51-500 It EPRI 0-20 It EPRI20-50 II EPRI50-120 It EPRI 120-250 It EPRI 250-500 It 0.001 Sneer Slrain ('%) 10'2 0.01 Shear Strain (%) j EXPI.ANf\TION *

• PG&E Letter DCL-15-154 Enclosure Page 26 of 50 ( 0 Rason>nl Column Ta6l 1 -Rock Ct.We in SHAKE I OCf'P Rod\ 1 0.1 Figure 3-6 -1978 Lab Testing for Nonlinear Damping for DCPP Rock 2-0') *C\1 0 II < <.9 a.. X. 0 0 a: -Q,) a: (f) :::> 3 (/) (V) 0. E < (L Cl. 0 0 .1-0.'1 PG&E Letter DCL-15-154 Enclosure Page 27 of 50 I I I l I I ! I . ! I I ! I I l j I ) I I ' I i I l i I : l j I I I ! I I I ; ! l . ' -. I i I l ! ! ! ' * . . . .I I I . * * * * *

• I I ! *t .** I

• I , .J. .,l. 1 10 100 Frequency (Hz) Figure 3-7-Analytical Site Terms for a SWUS Reference Rock (760 m/s) PGA of 0.2 g (The green curves are for the lower VS profile; red curves are for the central VS profile; and the blue curves are for the upper VS profile. The short dashed lines are for the target kappa of 0.03 sec, the long dashed lines are for the target kappa of 0.05 sec, and the solid lines are for the target kappa of 0.04 sec. The black line is the mean.) (From Reference 9)

-*C) f'. 0 .,..... ll a.. <.,) 0 a: -Q) a: (j) s (f) -M a_ E < a.. a.. 0 0 2 1 Frequency (Hz) PG&E Letter DCL-15-154 Enclosure Page 28 of 50 Figure 3-8 -Analytical site terms for a SWUS reference rock (760 m/s) PGA of 1.07 g (The green curves are for the lower VS profile; red curves are for the central VS profile; and the blue curves are for the upper VS profile. The short dashed lines are for the target kappa of 0.03 sec, the long dashed lines are for the target kappa of 0.05 sec, and the solid lines are for the target kappa of 0.04 sec. The black line is the mean.) (From Reference 9)

-0) or-II <3 a.. c 0 :.;::; t1 0. E a.. a... u 0 1 PG&E Letter DCL-15-154 Enclosure Page 29 of 50 .1 1 10 100 Frequency (l-Iz) Figure 3-9 -Analytical site terms for a SWUS reference rock (760 m/s) PGA of 1.91 g, corresponding to the 1 E-5 hazard level. (The green curves are for the lower VS profile; red curves are for the central VS profile; and the blue curves are for the upper VS profile. The short dashed lines are for the target kappa of 0.03 sec, the long dashed lines are for the target kappa of 0.05 sec, and the solid lines are for the target kappa of 0.04 sec. The black line is the mean.) (From Reference 9) 4. EMPIRICAL SITE RESPONSE APPROACH 4.1 Residuals for ESTA27 and ESTA28 Station EST A27 recorded both the 2003 San Simeon and the 2004 Parkfield earthquakes. Station ESTA28 only recorded the 2004 Parkfield earthquake. The event-path corrected residuals are listed in Table 4-1. Following the methodology used in the DCPP Seismic Hazard and Screening Report (Reference 3), they are adjusted to account for the expected differences in the average SA due to the differences between the VS30 for the control point and the VS30 for the two free-field sites.

PG&E Letter DCL-15-154 Enclosure Page 30 of 50 The VS30 values for central models for ESTA27, EST A28, and the control point are listed in Table 4-2. The VS30 adjustment factors, based on the NGA-W2 GMPEs are listed in Table 4-3. The standard error of the DCPP site term, 8S2S(f), has three parts: (1) There is the standard error (SE) due to the number of observations at DCPP. PG&E use the phiO from Lin et al (2011 ), "Repeatable Source, Site, and Path Effects on the Standard Deviation for Empirical Ground-Motion Prediction Models," (Reference 13) as the estimate of the aleatory variability of the DCPP event-corrected residuals. This part of the SE is phiO I sqrt(N). Although there are 3 recordings, the data at ESTA27 and EST A28 for the Parkfield earthquake are correlated. So, N=2 is used as a conservative assumption. (2) The second part is the SE of the estimate of the event-path term, terms. For each event, this is theSE of the mean (sigma/sqrt(n)) for each earthquake. (3) The third part is SE of the VS30 adjustment (correcting the ESTA27 and ESTA28 residuals to the control point). The standard deviation of the VS30 at ESTA27 and ESTA28 is about 0.18 LN units and the standard deviation of the VS30 for the control point is about 0.23 LN units. These three sources of uncertainty are uncorrelated and can be combined by simple propagation of errors: The components of theSE are shown in Figure 4-1 and are listed in Table 4-4. The total SEs are smoothed. Using the standard deviations listed in Table 4-4, the VS values for the central profiles are scaled up and down by exp(1.6 SE). The resulting lower and upper profiles are listed in Table 4-5. 4.2 Changes from the Approach used in the DCPP Seismic Hazard and Screening Report There were three changes to the approach to empirical site terms used in the SHSR: 1) The control point was changed from the location of EST A28 at elevation 85 feet to being a hypothetical location that represents the center and range of profiles under the power-block and the turbine building.

PG&E Letter DCL-15-154 Enclosure Page 31 of 50 2) The epistemic uncertainty in the site was computed using the approach described in Section 4.1, rather than the simplified approach used in the SHSR based on the phiS2S from global data. 3) All three recordings at DCPP from the San Simeon and Parkfield earthquakes were used rather than just using the ESTA27 recording from San Simeon (adjusted to ESTA28) and the ESTA28 recording from Parkfield. All three are considered applicable to the average for the power-block and turbine building region. Table 4-1 -Event-Path Corrected Residuals(from Reference 9) Parkfield San Simeon Parkfield Period (sec) ESTA28 ESTA27 ESTA27 0.01 -0.296 -0.242 -0.028 0.02 -0.310 -0.259 -0.046 0.03 -0.330 -0.315 -0.140 0.05 -0.508 -0.427 -0.248 0.075 -0.537 -0.382 -0.310 ' 0.1 -0.726 -0.399 -0.480 0.15 -0.476 -0.315 -0.357 0.2 -0.628 -0.076 -0.283 0.25 -0.419 0.117 -0.285 0.3 -0.283 0.100 0.036 0.4 0.292 0.216 0.677 0.5 0.483 0.156 0.798 0.75 0.188 0.517 0.450 1 -0.231 0.560 0.071 1.5 -0.331 0.098 -0.064 2 -0.191 0.917 -0.049.

PG&E Letter DCL-15-154 Enclosure Page 32 of 50 Table 4-2-VS30 for Free-Field Sites and Hypothetical Control Point (From Reference 9) Location VS30 (m/s) ESTA27 856 ESTA28 777 Control Point (Power-Block and Turbine Building) 968 Table 4-3 -Linear VS30 Scaling from the Free-Field Sites to the Control Point (The scaling is computed using four NGA-West2 models for a M6.5 vertical strike-slip earthquake at a rupture distance of 50 km.) (From Reference 9) PSA (g) PSA (g) PSA (g) for for for VS30 Scale VS30 Scale Period VS30=856 V$30=777 VS30=968 Factor Factor for (sec.) (m/s) (m/s) (m/s) for 968/856 968/777 0.01 0.043 0.045 0.042 0.965 0.922 0.02 0.044 0.046 0.042 0.957 0.916 0.03 0.048 0.050 0.046 0.952 0.916 0;05 0.061 0.063 0.058 0.961 0.932 0.075 0.076 0.079 0.073 0.960 0.930 0.1 0.086 0.089 0.081 0.950 0.914 0.15 0.094 0.100 0.088 0.933 0.885 0.2 0.092 0.098 0.085 0.920 0.862 0.25 0.085 0.091 0.077 0.910 0.845 0.3 0.077 0.083 0.069 0.903 0.833 0.4 0.063 0.069 0.057 0.895. 0.821 0.5 0.053 0.058 0.047 0.890 0.812 0.75 0.035 0.039 0.031 0.885 0.804 1 0.025 0.028 0.022 0.882 0.798 1.5 0.015 0.017 0.013 0.879 0.793 2 0.010 0.012 0.009 0.890 0.805 3 0.006 0.007 0.006 0.919 0.844 4 0.004 0.004 0.004 0.939 0.868 5 0.003 0.003 0.003 0.943 0.889 7.5 0.002 0.002 0.001 0.949 0.904 10 0.001 0.001 0.001 0.956 0.916 PG&E Letter DCL-15-154 Enclosure Page 33 of 50 Table 4-4-Components of the Standard Error of DCPP Site Terms (From Reference 9) TotaiSE of DCPP Site Smoothed STD Dev of phiO Term Total SE VS30 (Reference SE of event (LN of DCPP 1.6*Smoothed Adjustment 13) -path term Units) Site Term TotaiSE Period (sec) (LN units) (LN units) (LN Units) (LN units) (LN units) 0.01 0.088 0.230 0.112 0.216 0.22 0.352 0.02 0.090 0.232 0.113 0.219 0.22 0.352 0.03 0.080 0.234 0.112 0.215 0.22 0.352 0.05 0.065 0.236 0.115 0.213 0.22 0.352 0.075 0.067 0.238 0.120 0.217 0.22 0.352 0.1 0.082 0.238 0.135 0.231 0.23 0.368 0.15 0.112 0.241 0.162 0.260 0.26 0.416 0.2 0.136 0.244 0.138 0.259 0.26 0.416 0.25 0.155 0.247 0.115 0.260 0.26 0.416 0.3 0.168 0.249 0.109 0.267 0.27 0.432 0.4 0.182 0.267 0.089 0.277 0.28 0.448 0.5 0.192 0.282 0.097 0.293 0.29 0.464 0.75 0.198 0.288 0.113 0.306 0.31 0.496 1 0.202 0.294 0.128 0.317 0.32 0.512 1.5 0.199 0.294 0.151 0.325 0.33 0.528 2 0.189 0.293 0.175 0.331 0.33 0.528 PG&E Letter DCL-15-154 Enclosure Page 34 of 50 Table 4-5 -DCPP Empirical Site Term with Epistemic Uncertainty (From Reference 9) Period Frequency Mean Upper Range Lower Range (sec) (Hz) (LN units) (LN units) (LN units) 0.01 100 -0.254 0.098 -0.606 0.02 50 -0.278 0.074 -0.630 0.03 33.3 -0.338 0.014 -0.690 0.05 20 -0.454 -0.102 -0.806 0.075 13.3 -0.456 -0.104 -0.808 0.1 10 -0.566 -0.198 -0.934 0.15 6.67 -0.455 -0.039 -0.871 0.2 5 -0.373 0.043 -0.789 0.25 4 -0.240 0.176 -0.656 0.3 3.33 -0.144 0.288 -0.576 0.4 2.5 0.207 0.655 -0.241 0.5 2 0.247 0.711 -0.217 0.75 1.33 0.260 0.756 -0.236 1 1 0.077 0.589 -0.435 1.5 0.667 0.000 0.560 -0.560 2 0.5 0.000 0.560 -0.560 3 0.333 0.000 0.560 -0.560 4 0.25 0.000 0.560 -0.560 5 0.2 0.000 0.560 -0.560 7.5 0.133 0.000 0.560 -0.560 10 0.1 0.000 0.560 -0.560

• * -Std Dev of VS30 Scale Factors PhiO PG&E Letter DCL-15-154 Enclosure Page 35 of 50 Std Err of Event/Path Terms (combined for two e{lk) Total SE of DCPP Site Term S oot ed Total SE *I I . 05 II O.'t 1 10 '100 Frequency (Hz) Figure 4-1 -Empirical Site Term for DCPP Relative to SWUS Reference Rock (From Reference 9)

-.!1 c ::J z _J E L-2 (/) 0.. E w a... 1. *t 0. ! il i! ll !!, II l I : I . l *

• Central --. Upper Lower PG&E Letter DCL-15-154 Enclosure Page 36 of 50

• Parkfield. ES A28

• Sa Sirneo ESTA27

• Pa field ESTA27 ! ! l ! l j j I : I I I :I I ) j f t l I I ! ' i I I ; i I ! ! I l I I I l ! I I 1 l ( i i 0 .'I 1 10 *too freq Figure 4-2-Empirical Site Term for DCPP Relative to SWUS Reference Rock (From Reference 9)