ML063030291

Draft Regulatory Guide DG-1146, a Performance-Based Approach to Define the Site-Specific Earthquake Ground Motion

ML063030291
Person / Time
Issue date:	10/30/2006
From:	Murphy A J NRC/RES/DFERR/DDERA
To:
Murphy A.J., 415-6982
References
DG-1146
Download: ML063030291 (64)

Text

This regulatory guide is being issued in draft form to involve the public in the early stages of the development of a regulatory positionin this area. It has not received staff review or approval and does not represent an official NRC staff position.Public comments are being solicited on this draft guide (including any implementation schedule) and its associated regulatoryanalysis or value/impact statement. Comments should be accompanied by appropriate supporting data. Written comments may be submitted to the Rules and Directives Branch, Office of Administration, U.S. Nuclear Regulatory Commission, Washington, DC 20555-0001. Comments may be submitted electronically through the NRC's interactive rulemaking Web page at http://www.nrc.gov/what-we-do/regulatory/rulemaking.html. Copies of comments received may be examined at the NRC'sPublic Document Room, 11555 Rockville Pike, Rockville, MD. Comments will be most helpful if received by December 14, 2006

.Requests for single copies of draft or active regulatory guides (which may be reproduced) or placement on an automatic distribution listfor single copies of future draft guides in specific divisions should be made to the U.S. Nuclear Regulatory Commission, Washington, DC 20555, Attention: Reproduction and Distribution Services Section, or by fax to (301)415-2289; or by email to Distribution@nrc.gov. Electronic copies of this draft regulatory guide are available through the NRC's interactive rulemakingWeb page (see above); the NRC's public Web site under Draft Regulatory Guides in the Regulatory Guides document collection of the NRC's Electronic Reading Room at http://www.nrc.gov/reading-rm/doc-collections/; and the NRC's Agencywide DocumentsAccess and Management System (ADAMS) at http://www.nrc.gov/reading-rm/adams.html, under Accession No. ML063030291. U.S. NUCLEAR REGULATORY COMMISSION October 2006OFFICE OF NUCLEAR REGULATORY RESEARCH Division 1 DRAFT REGULATORY GUIDEContact: A.J. Murphy (301) 415-6011A.M. Kammerer (301) 415-7964DRAFT REGULATORY GUIDE DG-1146A PERFORMANCE-BASED APPROACH TO DEFINETHE SITE-SPECIFIC EARTHQUAKE GROUND MOTIONA. INTRODUCTIONThe U.S. Nuclear Regulatory Commission (NRC) proposes this draft regulatory guideas an alternative to Regulatory Guide 1.165, "Identification and Characterization of Seismic Sourcesand Determination of Safe Shutdown Earthquake Ground Motion," (Ref. 1), for use in satisfyingthe requirements set forth in Section 100.23,"Geologic and Seismic Siting Criteria," of Title 10, Part 100, of the Code of Federal Regulations (10 CFR Part 100), "Reactor Site Criteria," as well as Appendix S,"Earthquake Engineering Criteria for Nuclear Power Plants," to 10 CFR Part 50, "Domestic Licensingof Production and Utilization Facilities." The NRC staff has recognized the need to provide alternativeguidance for incorporating recent developments in ground motion estimation models; updated models for earthquake sources; methods for determining site response; and new methods for defining a site-specific,performance-based safe shutdown earthquake ground motion (SSE).In 10 CFR 100.23, paragraph (c), "Geological, Seismological, and Engineering Characteristics,"and paragraph (d), "Geologic and Seismic Siting Factors," require that the geological, seismological, and engineering characteristics of a site and its environs be investigated in sufficient scope and detailto permit an adequate evaluation of the proposed site, provide sufficient information to support evaluationsperformed to arrive at estimates of the SSE, and permit adequate engineering solutions to actual or potentialgeologic and seismic effects at the proposed site. Data on the vibratory ground motion; tectonicand nontechtonic surface deformation; earthquake recurrence rates, fault geometry and slip rates,site foundation material; seismically induced floods and water waves; and other siting factors anddesign conditions should be obtained by reviewing pertinent literature and carrying out field investigations.

DG-1146, Page 210 CFR 100.23, paragraph (d)(1), "Determination of the Safe Shutdown Earthquake Ground Motion,"requires that uncertainty inherent in estimates of the SSE be addressed through an appropriate analysis,such as a probabilistic seismic hazard analysis (PSHA).In Appendix A, "General Design Criteria for Nuclear Power Plants," to 10 CFR Part 50,General Design Criterion (GDC) 2, "Design Bases for Protection Against Natural Phenomena,"

requires, in part, that nuclear power plant structures, systems, and components (SSCs) important to safetymust be designed to withstand the effects of natural phenomena (such as earthquakes) without lossof capability to perform their safety functions. Such SSCs must also be designed to accommodatethe effects of, and be compatible with, the environmental conditions associated with normal operation and postulated accidents.

Appendix S to 10 CFR Part 50 specifies, in part, requirements for implementing GDC 2with respect to earthquakes. Paragraph IV(a)(1)(i), of Appendix S to 10 CFR Pa rt 50, requires that the SSEmust be characterized by free-field ground motion response spectra at the free ground surface. In view ofthe limited data available on vibratory ground motions of strong earthquakes, it will usually be appropriatethat the design response spectra be smoothed spectra. In addition, the horizontal motions in the free fieldat the foundation level of the structures consistent with the SSE must be an appropriate response spectrumwith a peak ground acceleration of at least 0.1g.The two-step licensing practice, which required applicants to acquire a construction permit (CP),and then apply for an operating license (OL) during construction, has been modified to allow foran alternative procedure. The requirements and pro cedures applicable to the NRC's issuance of earlysite permits (ESPs) and combined licenses (COLs) for nuclear power facilities are set forth in Subparts Aand C of 10 CFR Part 52, respectively.This regulatory guide has been developed to provide general guidance on methods acceptable tothe NRC staff for (1) conducting geological, geophysical, seismological, and geotechnical investigations;(2) identifying and characterizing seismic sources; (3) conducting a PSHA; (4) determining seismic wavetransmission (soil amplification) characteristics of soil and rock sites; and (5) determining a site-specific,performance-based earthquake ground motion, satisfying the requirements of paragraphs (c), (d)(1),

and (d)(2) of 10 CFR 100.23, and leading to the establishment of an SSE to satisfy the requirementsof Appendix S to 10 CFR Part 50. NUREG-0800, "Standard Review Plan," Section 3.7.1, describesthe determination of the SSE, and a conceptual overview is provided in Regulatory Position 5.4 of this guide.This regulatory guide contains several appendices that address the objectives stated above.

Appendix A contains a list of defi nitions. Appendix B contains a list of abbreviations and acronyms. Appendix C discusses a procedure to characterize site geology, seismology, and geophysics. Appendix Ddescribes the procedure to determine the controlling earthquakes. Appendix E describes the procedurefor determining seismic wave transmission (soil amplification) characteristics of soil and rock sites. Appendix F provides criteria for developing and evaluating modified ground motions used for soilresponse analyses.The NRC issues regulatory guides to describe to the public methods that the staff considersacceptable for use in implementing specific parts of the agency's regulations, to explain techniquesthat the staff uses in evaluating specific problems or postulated accidents, and to provide guidanceto applicants. Regulatory guides are not substitutes for regulations, and compliance with regulatory guidesis not required. The NRC issues regulatory guides in draft form to solicit public comment and involvethe public in developing the agency's regulatory positions. Draft regulatory guides have not received complete staff review and, therefore, they do not represent official NRC staff positions.

1The CEUS and WUS Time History databases provided in NUREG/CR-6728 (Ref. 2) are being updated and will beavailable in the Summer of 2007.DG-1146, Page 3This regulatory guide may address information collections that are covered by the requirements of 10 CFR Part 50, 10 CFR Part 52, and 10 CFR Part 100, which the Office of Management and Budget(OMB) approved under OMB control numbers 3150-0011, 3150-0151, and 3150-0093, respectively.

The NRC may neither conduct nor sponsor, and a person is not required to respond to, an informationcollection request or requirement unless the requesting document displays a currently valid OMBcontrol number.

B. DISCUSSION BackgroundRegulatory Guide 1.165 (Ref. 1) provides general guidance on procedures acceptable to the NRCstaff to satisfy the requirements of 10 CFR 100.23. However, the NRC staff has recognized the need toprovide alternative guidance for incorporating recent developments in ground motion estimation models;updated models for earthquake sources; methods for determining site response; and new methods fordefining a site-specific, performance-based SSE. Therefore, the NRC staff is proposing this draftregulatory guide as an alternative to Regulatory Guide 1.165 (Ref. 1) to satisfy, in part, the requirements of 10 CFR 100.23 and Appendix S to 10 CFR Part 50.The general process to determine a site-specific, performance-based earthquake ground motionincludes the following:(1)site- and region-specific geological, seismological, geophysical, and geotechnical investigations(2)a probabilistic seismic hazard analysis (PSHA)

(3)a site response analysis to incorporate the effects of local geology and topography(4)and the selection of appropriate performance goals and methodologyGeological, seismological, and geophysical investigations are performed to develop an up-to-date, site-specific, earth science database that supports site characterization and a PSHA. The resultsof these investigations will also be used to assess whether new data and their interpretation are consistentwith the information used in accepted probabilistic seismic hazard studies.The PSHA is conducted with up-to-date interpretations of earthquake sources, earthquake recurrence,and strong ground motion estimation. The site seismic hazard characteristics are quantified by the seismichazard curves from a PSHA and the uniform hazard response spectra (UHRS) that cover a broad rangeof natural frequencies. The hazard curves are developed in part by identifying and characterizing each seismic source in terms of maximum magnitude, magnitude recurrence relationship, and source geometry. The rock-based ground motion at a site resulting from the combined effect of all sources can then bedetermined through the use of attenuation relationships.Seismic wave transmission (site amplification) procedures are necessary to obtain appropriateUHRS at the free-field ground surface if the shear wave velocity of the surficial material is less than thegeneric hard rock conditions appropriate for the rock-based attenuation relationships used in the PSHA. A database of earthquake time histories on rock fo r both the central and eastern United States (CEUS) and the western United States (WUS) has been developed to determine the dynamic site responsefor the soil or rock site conditions.

1 2The reference probability is computed as the median probability obtained from the distribution of median probabilitiesof exceeding the SSE at 29 sites in the CEUS. The sites selected were intended to represent relatively recent designs that used the Regulatory Guide 1.60 (Ref. 4), design response spectrum, or similar spectrum, as their design bases.

The use of the reference probability was to ensure an adequate level of conservatism in determining the SSE consistentwith recent licensing decisions.

3The P F , R P, and H D criteria corresponding to Seismic Design Category (SDC) 5 in Reference 3 (which are equivalent tonuclear power plant requirements) are used.DG-1146, Page 4A performance-based approach is described in Chapters 1 and 2 of ASCE/SEI Standard 43-05(Ref. 3), instead of the reference probability approach described in Appendix B to Regulatory Guide 1.165 (Ref. 1).2 The performance-based approach employs Target Performance Goal (P F), Probability Ratio (R P), and Hazard Exceedance Probability (H D) criteria 3 to ensure that nuclear power plants can withstandthe effects of earthquakes with desired performance, the desired performance being expressed as the targetvalue of 1E-05 for the mean annual probability of exceedance of onset of significant inelastic deformation(FOSID). Setting the performance goal to be equivalent to the FOSID of SSCs is conservative, since the seismic demand resulting in the onset of significant inelastic deformation is less than thatfor failure of the SSCs.One of the objectives in developing the performance-based SSE is to achieve approximateconsistent performance for SSCs, across a range of seismic environments, annual probabilities, and structural frequencies. The intent is to develop a site-specific SSE that achieves the P F and ensuresthat the performance of the SSCs related to safety and environmental protection is acceptable.Basis for Regulatory PositionsSite- and Region-Specific Investigations to Characterize Site Geology, Seismology and GeophysicsThe primary objective of geological, seismological, and geophysical investigations is to developan up-to-date, site-specific earth science database that supports site characterization and a PSHA. Theresults of these investigations will also be used to assess whether new data and their interpretation areconsistent with the information used in accepted probabilistic seismic hazard studies. If the new data, such as new seismic sources and new ground motion a ttenuation relationships, are consistent with theexisting earth science database, updating or modification of the hazard analysis is not required. It will benecessary to update seismic sources and ground motion attenuation relationships for sites where there is significant new information resulting from the site investigations.Features Discovered During ConstructionIt should be demonstrated that deformation features discovered during construction, particularlyfaults, do not have the potential to compromise the safety of the plant. Applying the combined licensing procedure to a site could result in the award of a license prior to the start of construction. During the construction of nuclear pow er plants licensed in the past, previously unknown faults were oftendiscovered in site excavations. Before issuance of the OL, it was necessary to demonstrate that faults inthe excavation posed no hazard to the facility. Under the combined license procedure, these kinds offeatures should be mapped and assessed as to their rupture and ground motion generating potential whilethe excavations' walls and bases are exposed, and the NRC staff should be notified when excavations are open for inspection.

DG-1146, Page 5Probabilistic Seismic Hazard Analysis MethodsA PSHA has been identified in 10 CFR 100.23 as a means to address the uncertainties in thedetermination of the SSE. Furthermore, the rule recognizes that the nature of uncertainty and the appropriate approach to account for uncertainties depend on the tectonic setting of the site and onproperly characterizing input parameters to the PSHA, such as the seismic source characteristics, therecurrence of earthquakes within a seismic source, the maximum magnitude of earthquakes within aseismic source, and engineering estimation of earthquake ground motion through attenuation relationships.Every site is unique; therefore, requirements for analyses and investigations vary. It is notpossible to provide procedures for addressing all situations. In cases that are not specifically addressedin this regulatory guide, prudent and sound engineering judgment should be exercised.Probabilistic procedures were developed specifically for nuclear power plant seismic hazardassessments in the CEUS, also referred to as the Stable Continent Region (SCR). Experience has beengained by applying this methodology at nuclear facility sites, both reactor and non-reactor sites,throughout the United States. The experience with these applications also served as the basis for theguidelines for conducting a PSHA in Reference 5. These procedures provide a structured approach fordecision making with respect to the SSE when performed with site-specific investigations. A PSHAprovides a framework to address the uncertainties associated with the identification and characterizationof seismic sources by incorporating multiple interpretations of seismological parameters. A PSHA also provides a means for evaluation of the likelihood of SSE occurrence during the design lifetime of a givenfacility, given the recurrence interval and recurrence pattern of earthquakes in pertinent seismic sources. Within the framework of a probabilistic analysis, uncertainties in the characterization of seismic sourcesand ground motions are identified and incorporated in the procedure at each step of the process forestimating the SSE. The role of geological, seismological, and geophysical investigations is to developgeoscience information about the site for use in the detailed design analysis of the facility, as well as toensure that the seismic hazard analysis is based on up-to-date information.Experience in performing seismic hazard evaluations in active plate-margin regions in the WUS(for example, the San Gregorio-Hosgri fault zone and the Cascadia Subduction Zone) has also identified uncertainties associated with the characterization of seismic sources (R efs. 6-8). Sources of uncertaintyinclude fault geometry, rupture segmentation, rupture extent, seismic-activity rate, ground motion, andearthquake occurrence modeling. As is the case for sites in the CEUS, alternative hypotheses andparameters must be considered to account for these uncertainties.Uncertainties associated with the identification and characterization of seismic sources intectonic environments in both the CEUS and the WUS should be evaluated. The same basic approachcan be applied to determine the SSE.

Central and Eastern United StatesThe CEUS is considered to be that part of th e United States east of the Rocky Mountain front, oreast of Longitude 105 West (Refs. 9, 10) A PSHA in the CEUS must account for credible alternativeseismic source models through the use of a decision tree with appropriate weighting factors that arebased on the most up to date information and relative confidence in alternative characterizations for eachseismic source. Seismic sources identified and characterized by the Lawrence Livermore NationalLaboratory (LLNL) (Refs 9-11) and the Electric Power Research Institute (EPRI) (Refs. 12, 13) wereused for studies in the CEUS in the past. These databases may still represent the latest informationavailable for some seismic sources. However, if more up to date information is available, it should be incorporated if significant.

DG-1146, Page 6In the CEUS, characterization of seismic sources is more problematic than in active plate-marginregions because there is generally no clear association between seismicity and known tectonic structuresor near-surface geology. In general, the observed geologic structures are a result of response to tectonicforces that may have little or no correlation with current tectonic forces. Therefore, it is important toaccount for this uncertainty by the use of multiple alternative seismotectonic models.The identification of seismic sources and reasonable alternatives in the CEUS considershypotheses currently advocated for the occurrence of earthquakes in the CEUS (e.g., the reactivation offavorably oriented zones of weakness or the local amplification and release of stresses concentratedaround a geologic structure). In tectonically active areas of the CEUS, such as the New Madrid SeismicZone, where geological, seismological, and geophysical evidence suggest the nature of the sources thatgenerate the earthquakes, it may be more appropriate to characterize those seismic sources by usingprocedures similar to those normally applied in the WUS.Western United StatesThe WUS is considered to be that part of the United States that lies west of the Rocky Mountainfront, or west of approximately 105 West Longitude. In some regions of the WUS, significant researchinto seismic sources is ongoing and the results of this research should be evaluated and addressed.The active plate-margin regions include, for example, coastal California, Oregon, Washington,and Alaska. For the active plate-margin regions, where earthquakes can often be correlated with knownfaults that have experienced repeated movements at or near the ground surface during the Quaternary, tectonic structures should be assessed for their earthquake and surface deformation potential. In theseregions, at least three types of sources may exist:(1)faults that are known to be at or near the surface(2)buried (blind) sources that may often be manifested as folds at the earth's surface(3)subduction zone sources, such as those in the Pacific Northwest and AlaskaThe nature of surface faults can be evaluated by conventional surface and near-surfaceinvestigation techniques to assess orientation, geometry, sense of displacements, length of rupture,Quaternary history, etc. Possibilities of multiple ruptures should be cons idered, as appropriate.

Buried (blind) faults are often associated with surficial deformation such as folding, uplift, orsubsidence. The surface expression of blind faulting can be detected by mapping the uplifted or down-dropped geomorphological features or stratigraphy, survey leveling, and geodetic methods. The nature ofthe structure at depth can often be evaluated by deep core borings and geophysical techniques.Continental U.S. subduction zones are located in the Pacific Northwest and Alaska. Seismic sources associated with subduction zones are sources within the overriding plate, on the interfacebetween the subducting and overriding lithospheric plates, and in the interior of the downgoing oceanicslab. The characterization of subduction zone seismic sources should consider the three-dimensionalgeometry of the subducting plate, rupture segmentation of subduction zones, geometry of historicalruptures, constraints on the up-dip and down-dip extent of rupture, and comparisons with othersubducting plates worldwide.The Basin and Range region of the WUS, and to a lesser extent the Pacific Northwest and thecentral United States, exhibit temporal clustering of earthquakes. Temporal clustering is bestexemplified by the rupture histories within the Wasatch fault zone in Utah, where several large, lateHolocene coseismic faulting events occurred at relatively close intervals (for hundreds to thousands ofyears) that were preceded by long periods of quiescence that lasted thousands to tens of thousands ofyears. Temporal clustering should be considered in these regions, or wherever paleoseismic evidence indicates that it has occurred.

DG-1146, Page 7 Lower-Bound Magnitude CutoffCurrent seismic hazard analysis methods generally utilize a lower-bound body wave magnitudecut-off value for earthquakes to be included in the PSHA. This lower-bound magnitude cut-off level is aconservatively defined value based on research studies whose objective was to estimate the damagepotential of smaller earthquakes. Reference 14 estab lishes an appropriate distribution of low-magnitudeearthquakes in the seismic hazard analysis through the use of the cumulative absolute velocity (CAV)model, in place of a lower-bound magnitude cutoff.Several ground motion measures such as peak ground acceleration, Arias intensity, root meansquare acceleration, and CAV were evaluated to determine the single ground motion measure that is bestcorrelated with the threshold of potential damage. The CAV was determined to be the best parametercorrelating damage with the Modified Mercalli Intensity Scale, and a CAV value of 0.16 g-sec was found to be a conservative characterization of the threshold between damaging and non-damaging earthquakemotions for buildings of good design and construction. An empirical model for estimating CAV in termsof magnitude, peak ground acceleration (PGA), and duration is needed because the PSHA calculationdoes not use time histories directly.Spectral Frequency Range Considered in the Probabilistic Seismic Hazard AnalysesThe frequency range considered in the PSHA should extend from a low frequency that can bereliably obtained from the current strong-motion data set from the WUS, principally California, to a high-frequency limit that permits the ratio of spectral acceleration to PGA to reach nearly 1 for hard rock siteconditions. For the CEUS this high-frequency limit is approximately 100 Hz. For soft rock conditions, the ratio of spectral acceleration to PGA will reach 1 at about 40-50 Hz. To obtain a smooth plot ofspectral response across the frequency range of interest, the hazard assessment should be conducted at a minimum of 30 frequencies, approximately equally spaced on a logarithmic frequency axis between 100 and 0.1 Hz.

Deaggregation of Mean Hazard High- and low-frequency controlling earthquakes are developed for the ground motion levelscorresponding to mean annual probabilities of 1 E-04, 1 E-05, and 1 E-06, by deaggregating the PSHA interms of earthquake magnitudes and distances. Multiple ground motion levels are used to obtain a morecomplete range of earthquake characteristics (i.e., mean magnitudes and distances) that contribute to thehigh-frequency (5 and 10 Hz) and low-frequency (1 and 2.5 Hz) hazard, than could be obtained from asingle ground motion level (e.g., 1 E-05/yr).Choice of Epsilon in Probabilistic Seismic Hazard AnalysesEpsilon, the number of standard deviations included in defining the distribution of groundmotions for each magnitude and distance scenario, can have a significant impact on the results of thePSHA. Care should be taken in choosing a value for epsilon large enough such that natural aleatory variability in ground motions is adequately addressed. A recent study (Ref. 15) found no technical basisfor truncating the ground motion distribution at a specified number of standard deviations (epsilons)below that implied by the strength of the geologic materials. Even though very large ground motionsmay have a low probability of occurrence, when the hazard is calculated for low annual frequencies ofexceedance, low probability events need to be considered.

4After the design of SSCs is completed by the performance-based method described above, the resulting SSCs should bedemonstrated to have HCLPF capacity 1.67 times the SSE response spectra for evolutionary and new reactors. HCLPF represents the seismic capacity corresponding to a 1-percent mean probability of failure and the representsthe composite uncertainty.DG-1146, Page 8 Site Response AnalysisFor rock or soil sites where the free-field ground surface shear wave velocity is less than thegeneric hard rock conditions used in the PSHA, seismic wave transmission (site amplification) procedures should be used to obtain appropriate UHRS at the free-field ground surface. Earthquake timehistories are used to determine the dynamic site response for the soil or rock site conditions. Reference 2contains a database of recorded time histories on rock for both the CEUS and WUS. The database isdivided into magnitude and distance bins, with each bin having a minimum of 15, three-component(two horizontal and one vertical) sets.Performance-Based MethodA performance-based approach is described in Chapters 1 and 2 of ASCE/SEI Standard 43-05(Standard) (Ref. 3). The Standard was developed by a number of contributors considered to beexperienced in the design of nuclear facilities, and reviewed by a number of outside agencies andpersonnel. It is a national consensus standard by the American Society of Civil Engineers (ASCE).The Standard was developed to provide performance-based criteria for the seismic design ofsafety-related SSCs for use in nuclear facilities. The Standard covers a broad range of facilities thatprocess, store, or handle radioactive materials. The goal of the Standard is to present seismic designcriteria using a graded approach commensurate with the relative importance to safety and safeguards,which will ensure that nuclear facilities can withstand the effects of earthquakes with desiredperformance, expressed as probabilistic target performance goals and hazard exceedance probabilities. The performance-based approach used for this regulatory guide is based on the premise that the seismicdemand and structural capacity evaluation criteria laid out by the NRC's Standard Review Plan (SRP)(NUREG-0800) are aimed at having sufficient conservatism to reasonably achieve both of the following:(1)less than about a 1% probability of unacceptable performance for the site-specific responsespectrum ground motion(2)less than about a 10% probability of unacceptable performance for a ground motion equal to150% of the site-specific response spectrum ground motion.The Standard, for the most demanding seismic design category (SDC 5), recommends using a P F = 1 E-05 for the FOSID, R P = 10, and H D = 1 E-04 for the minimum structural damage state, which isdescribed as essentially elastic behavior. Specifically, essentially elastic behavior means that localized inelasticity might occur at stress concentration points, but the overall seismic response will be essentiallyelastic. SDC 5 criteria are nearly identical to the NRC SRP (NUREG-0800), regulatory guides, andprofessional design codes and standards referenced therein. Also, setting the performance goal of1 E-05/yr to be equivalent to the FOSID of SSCs is conservative, since the seismic demand resulting in the onset of significant inelastic deformation is less than that for failure of the SSC.Thus, the performance-based approach combines a conservative characterization of groundmotion hazard with equipment/structure performance (fragility characteristics) to establish risk-consistent SSEs, rather than only hazard-consistent ground shaking, as occurs using the hazard referenceprobability approach in Appendix B to Regulatory Guide 1.165 (Ref. 1). The performance target (the mean annual probability of SSCs reaching the limit state of inelastic response) results from theintegration of the hazard function and the fragility function which is modeled by two parameters, the mean capacity of SSCs using the acceptance criteria provided in NUREG 0800, and the variability .4 DG-1146, Page 9The UHRS at the free-field ground surface is modified by a design factor to obtain theperformance-based site-specific response spectrum. The design factor achieves a relatively consistentannual probability of plant component failure across the range of plant locations and structuralfrequencies. It does this by determining a ground motion slope ratio (amplitude ratio), accounting for theslope of the seismic hazard curve, which changes with structural frequency and site location. The designfactor ensures that the site-specific response spectrum is equal to or greater than the mean 1 E-04 UHRS.The flow chart in Figure 1 depicts the procedures described in this regulatory guide.Geological, Geophysical, Seismological, andGeotechnical InvestigationsRegulatory Position 1Appendix CSeismic Sources Significant to theSite Seismic HazardRegulatory Position 2Appendix CProbabilistic Seismic Hazard Analysis(PSHA) ProceduresRegulatory Position 3Deaggregation of Mean HazardRegulatory Position 3.5Appendix D Seismic Wave Transmission(Soil Amplification) Characteristics of the SiteRegulatory Position 4Appendices E & FPerformance-Based Site-Specific EarthquakeGround Motion ProcedureRegulatory Position 5Figure 1. Flow Diagram DG-1146, Page 10C. REGULATORY POSITION1.Geological, Geophysical, Seismological, and Geotechni cal Investigations1.1Comprehensive Site Area and Region InvestigationsComprehensive geological, seismological, geophysical, and geotechnical engineeringinvestigations of the site area and region should be performed. For existing nuclear power plant siteswhere additional units are planned, the geosciences technical information originally used to validatethose sites may need to be updated. Depending on how much new or additional information has becomeavailable since the initial investigations and analyses were performed and the complexity of the site andregional geology and seismology, additional data may be required. This technical information should beused, along with all other available information, to plan and determine the scope of additional investigations. The investigations described in this regulatory guide are performed primarily to gatherinformation needed to confirm the suitability of the site and to gather data pertinent to the safe design and construction of the nuclear power plant. Appropriate geological, seismological, and geophysicalinvestigations are described in Appendix C to this regulatory guide. Geotechnical investigations aredescribed in Regulatory Guide 1.132 (Ref. 16). Another important purpose for the site-specificinvestigations is to determine whether there are any new data or interpretations that are not adequatelyincorporated into the existing PSHA databases. Appendix C also describes a method for assessing theimpact of new information, obtained during the site-specific investigations on the databases used for the PSHA.These investigations should be performed at four levels, with the degree of detail based on(a) distance from the site, (b) the nature of the Quaternary tectonic regime, (c) the geological complexityof the site and region, (d) the existence of potential seismic sources, and (e) the potential for surfacedeformations, etc. A more detailed discussion of the areas and levels of investigations and the bases forthem are presented in Appendix C to this regulatory guide. General guidelines for the levels ofinvestigation are characterized as follows.

1.1.1Within a Radius of 320 Kilometers (200 mi) of the Site (Site Region)Conduct regional geological and seismological investigations to identify seismic sources(seismogenic and capable tectonic sources). These investigations should include literature reviews, thestudy of maps and remote sensing data, and, if necessary, ground-truth reconnaissances.

1.1.2Within a Radius of 40 Kilometers (25 mi) of the Site (Site Vicinity)

Geological, seismological, and geophysical investigations should be carried out in greater detailthan the regional investigations, to identify and characterize the seismic and surface deformationpotential of any capable tectonic sources and the seismic potential of seismogenic sources, or todemonstrate that such structures are not present. Sites with capable tectonic or seismogenic sourceswithin a radius of 40 kilometers (km) (25 mi) may require more extensive geological and seismological investigations and analyses [similar in detail to investigations and analysis usually preferred within an8 km (5 mi) radius].

DG-1146, Page 11 1.1.3Within a Radius of 8 Kilometers (5 mi) of the Site (Site Area)Detailed geological, seismological, geophysical, and geotechnical engineering investigationsshould be conducted to evaluate the potential for tectonic deformation at or near the ground surface andto assess the transmission characteristics of soils and rocks in the site vicinity.

1.1.4 Within

a Radius of Approximately 1 Kilometer (0.6 mi) of the Site (Site Location)Very detailed geological, geophysical, and geotechnical engineering investigations should beconducted to assess specific soil and rock characteristics, as described in Reference 16.1.2Expanding the Areas of InvestigationThe areas of investigation may be expanded beyond those specified above in regions that includeseismogenic tectonic sources, relatively high seismicity, or complex geology, or in regions that haveexperienced a large, geologically recent earthquake identified in historical records or by paleoseismic data.1.3Features Discovered During ConstructionIt should be demonstrated that deformation features discovered during construction, particularlyfaults, do not have the potential to compromise the safety of the plant. The newly identified featuresshould be mapped and assessed as to their rupture and ground motion generating potential while theexcavations' walls and bases are exposed. A commitment should be made, in documents (SafetyAnalysis Reports) supporting the license application, to geologically map all excavations and to notify the NRC staff when excavations are open for inspection.1.4Justifying Assumpti ons and ConclusionsData sufficient to clearly justify all assumptions and conclusions should be presented. Becauseengineering solutions cannot always be satisfactorily demonstrated for the effects of permanent grounddisplacement, it is prudent to avoid a site that has a potential for surface or near-surface deformation. Such sites normally will require extensive additional investigations.1.5Characterize Lithologi c, Stratigraphic, Hydrologic, and Structural Ge ologic Conditions For the site and for the area surrounding the site, the lithologic, stratigraphic, hydrologic, andstructural geologic conditions should be characterized. The investigations should include themeasurement of the static and dynamic engineering properties of the materials underlying the site as wellas an evaluation of the physical evidence concerning the behavior during prior earthquakes of thesurficial materials and the substrata underlying the site. The properties needed to assess the behavior ofthe underlying material during earthquakes should be measured. These include the potential forliquefaction and the characteristics of the underlying material in transmitting earthquake ground motionsto the foundations of the plant (such as seismic wave velocities, density, water content, porosity, elastic moduli, and strength).

DG-1146, Page 122.Seismic Sources Significant to the Site Seismic Hazard2.1Evaluation on New Seismic SourcesFor sites in the CEUS, existing databases may be used to identify seismic sources to performPSHA. Previously unidentified seismic sources th at were not included in these databases should beappropriately characterized and sensitivity analyses performed to assess their significance to the seismichazard estimate. The results of investigation discussed in Regulatory Position 1 should be used, in accordance with Appendix C to this regulatory guide, to determine whether the seismic sources and theircharacterization should be updated. The guidance in Regulatory Positions 2.2 and 2.3 (below) and themethods in Appendix C to this regulatory guide may be used if additional seismic sources are to be developed as a result of investigations.2.2Use of Alternative Seismic SourcesWhen existing methods and databases are not used or are not applicable, the guidance inRegulatory Position 2.3 should be used for identification and characterization of seismic sources. Theuncertainties in the characterization of seismic sources should be addressed. "Seismic sources" is ageneral term referring to both seismogenic sources and capable tectonic sources. The main distinctionbetween these two types of seismic sources is that a seismogenic source would not cause surfacedisplacement, but a capable tectonic source causes surface or near-surface displacement.Identification and characterization of seismic sources should be based on regional and sitegeological and geophysical data, historical and instrumental seismicity data, the regional stress field, andgeological evidence of prehistoric earthquakes. Investigations to identify seismic sources are describedin Appendix C to this regulatory guide. The bases for the identification of seismic sources should bedescribed. A general list of characteristics to be evaluated for seismic sources is presented in

Appendix C.2.3Characterizing Seismic PotentialAs part of the seismic source characterization, the seismic potential for each source should beevaluated. Typically, characterization of the seismic potential consists of four equally importantelements:(1)selection of a model for the spatial distribution of earthquakes in a source(2)selection of a model for the temporal distribution of earthquakes in a source(3)selection of a model for the relative frequency of earthquakes of various magnitudes, includingan estimate for the largest earthquake that coul d occur in the source under the current tectonicregime(4)a complete description of the uncertaintyFor example, truncated exponential and characteristic earthquake models are often used for thedistribution of magnitudes. A stationary Poisson process is used to model the spatial and temporaloccurrences of earthquakes in a source. For a general discussion of evaluating the earthquake potentialand characterizing the uncertainty, refer to NUREG/CR- 6372 (Ref. 5).

DG-1146, Page 13 2.3.1Characterizing Seismic Potential When A lternative Methods and Databases Are UsedFor sites in the CEUS where previous source databases are not used, it is necessary to evaluatethe seismic potential for each source. The seismic sources and data accepted by the NRC in past licensing decisions may be used, along with the data gathered from the investigations carried out asdescribed in Regulatory Position 1.Generally, the seismic sources for the CEUS are area sources because there is uncertainty aboutthe underlying causes of earthquakes. This uncertainty is caused by a lack of active surface faulting, alow rate of seismic activity, or a short historical record. The assessment of earthquake recurrence forCEUS area sources commonly relies heavily on catalogs of historic earthquakes. Because these catalogsare incomplete and cover a relatively short period of time, the earthquake recurrence rate may not beestimated reliably. Considerable care must be taken to correct for incompleteness and to model theuncertainty in the rate of earthquake recurrence. To completely characterize the seismic potential for asource, it is also necessary to estimate the largest earthquake magnitude that a seismic source is capableof generating under the current tectonic regime. This estimated magnitude defines the upper bound ofthe earthquake recurrence relationship.Primary methods for assessing maximum earthquakes for area sources usually include a consideration of the historical seismicity record, the pattern and rate of seismic activity, the Quaternarygeologic record (2 million years and younger), characteristics of the source, the current stress regime(and how it aligns with known tectonic structures), paleoseismic data, and analogs to sources in otherregions considered tectonically similar to the CEUS. Because of the shortness of the historical catalogand low rate of seismic activity, considerable judgment is needed. It is important to characterize the large uncertainties in the assessment of the earthquake potential (Refs. 5 and 12).

2.3.2Characterizing Seismic Potential for Western United States Sites For sites located within the WUS, earthquakes can often be associated with known tectonicstructures with a high degree of certainty. For faults, the earthquake potential is related to thecharacteristics of the estimated future rupture, such as the total rupture area, the length, or the amount offault displacement. Empirical relations can be used to estimate the earthquake potential from faultbehavior data and also to estimate the amount of displacement that might be expected for a givenmagnitude. It is prudent to use several different rupture length or area-versus-magnitude relations toobtain an estimate of the earthquake magnitude. When such correlations are used, the earthquakepotential is often evaluated as the mean of the distribution. The difficult issue is the evaluation of theappropriate rupture dimension to be used. This is a judgmental process based on geological data for thefault in question and the behavior of other regional fault systems of the same type.In addition to maximum magnitude, the other elements of the recurrence model are generallyobtained using catalogs of seismicity, fault slip rate, and other data. All the sources of uncertainty mustbe appropriately modeled.

2.3.3Characterizing Seismic Potential for Sites Near Subduction Zones For sites near subduction zones, such as in the Pacific Northwest and Alaska, the maximummagnitude must be assessed for subduction zone seismic sources. Worldwide observations indicate thatthe largest known earthquakes are a ssociated with the plate interface, although intraslab earthquakes mayalso have large magnitudes. The assessment of plate interface earthquakes can be based on estimates ofthe expected dimensions of rupture or analogies to other subduction zones worldwide.

DG-1146, Page 143.Probabilistic Seismic Hazard Analysis ProceduresA PSHA should be performed to allow for the use of multiple source models to estimate thelikelihood of earthquake ground motions occurring at a site and to systematically take into accountuncertainties that exist in various parameters (such as seismic sources, maximum earthquakes, and ground motion attenuation). Alternative hypotheses are considered in a quantitative fashion in a PSHA.

They can be used to evaluate the sensitivity of the hazard to the uncertainties in the significantparameters and to identify the relative contribution of each seismic source to the hazard.The following steps describe a procedure acceptable to the NRC staff for performing a PSHA.3.1Perform Regional and Site InvestigationsPerform regional and site geological, seismological, and geophysical investigations inaccordance with Regulatory Position 1 and Appendix C to this regulatory guide.3.2Evaluation of New InformationFor CEUS sites, perform an evaluation of seismic sources, in accordance with Appendix C to thisregulatory guide, to determine whether they are consistent with the site-specific data gathered inRegulatory Position 3.1 or if they require updating. The PSHA should be updated if the new informationindicates that the current version of the seismic source model underestimates the hazard or if newattenuation relationships are available. In most cases, limited-scope sensitivity studies are sufficient todemonstrate that the existing database in the PSHA envelops the findings from site-specificinvestigations. Any significant update should follow the guidance of NUREG/CR- 6372 (Ref. 5).3.3Conduct a Probabilistic Seismic Hazard AnalysisPerform a site-specific PSHA with up-to-date interpretations of earthquake sources, earthquakerecurrence, and strong ground motion estimation using original or updated sources as determined inRegulatory Positions 3.1 and 3.2. Characterize epistemic uncertainties in a complete and defensible fashion. CAV filtering can be used in place of a lower-bound magnitude cutoff of 5.0 (Ref. 14). For theCEUS, the understanding of single- and double-corner ground-motion models is evolving, and the PSHAshould fully document the appropriateness of using either or both of these models. Aleatory variabilityof the ground motion as expressed in the choice of epsilon used in the PSHA should be consideredcarefully and described. The ground motion estimates should be made in the free field to develop theseismic hazard information base discussed in Appendix D to this regulatory guide.3.4Hazard AssessmentConduct the hazard assessment at a minimum of 30 frequencies, approximately equally spacedon a logarithmic frequency axis between 100 and 0.1Hz. Report fractile hazard curves at the following fractile levels (p) for each ground motion parameter: 0.05, 0.16, 0.50, 0.84, and 0.95, as well as the mean.

Report the fractile hazard curves in tabular as well as graphical format. Determine the mean UHRS forannual exceedance frequencies of 1 E-04, 1 E-05, and 1 E-06.

DG-1146, Page 153.5Deaggregate the Mean Probabilistic Hazard CharacterizationDeaggregate the mean probabilistic hazard characterization at ground motion levelscorresponding to the annual frequencies of 1 E-04, 1 E-05, and 1 E-06 in accordance with Appendix D to this regulatory guide to determine the controlling earthquakes (i.e., magnitudes and distances) anddocument the hazard information base as discussed in Appendix D.4.Seismic Wave Transmission Characteristics of the SiteThe hazard curves from the PSHA are defined for generic surficial hard rock conditions [i.e.,rocks with a shear wave velocity (V s) about 2.8 km/sec (9200 ft/sec)]. Because the surficial andsubsurface shear wave velocities at nuclear power plant sites are generally less than 2.8 km/sec (9200 ft/sec), the following proce dure should be used to define the UHRS at the free-field groundsurface. Additional discussion pertaining to this regulatory position is provided in Appendix E to thisregulatory guide.For both soil and rock sites, the subsurface mode l should extend to sufficient depth to reach thegeneric hard rock conditions as defined in the attenuation relationships used in the PSHA.4.1Site and Laboratory In vestigations and TestingPerform site and laboratory investigations and testing to obtain data defining the static anddynamic engineering properties of the site-specific soil and rock materials, and their spatial distribution. Methods acceptable to the NRC staff are described in Regulatory Guide 1.132 (Ref. 16), Regulatory Guide 1.138 (Ref. 17), and Subsection C.2.2.2 of Appendix C to this regulatory guide. Sufficient siteproperty data should be developed to adequately estimate the mean and variability of the properties in thesite response calculations. When soil models are deduced from laboratory studies, critically peer reviewthe sampling and testing programs to ensure that the generated soil dynamic nonlinear properties areappropriate to properly characterize site response.4.2Dynamic Site ResponseThe high- and low-frequency controlling earthquakes are derived in accordance with RegulatoryPosition 3.5 and Appendix D to this regulatory guide. Determine the dynamic site response by developing appropriate ground motion or earthquake time historie s for each of the controllingearthquakes. Reference 2 contains a database of recorded time histories on rock for both the CEUS andWUS. The database is divided into magnitude and distance bins, with each bin having a minimum of 15,three-component (two horizontal and one vertical) sets. Appendix F to this regulatory guide providescriteria acceptable to the NRC staff for developing and evaluating modified ground motions used for soilresponse and structural analyses. The time histories are scaled to match the response spectrum for thecorresponding controlling earthquake.Often vertically propagating shear waves are the dominant contributor to free-field groundmotions at a site. In these cases, a one-dimensional equivalent-linear analysis or nonlinear analysis thatassumes vertical propagation of shear waves may be appropriate. However, site characteristics (such as adipping bedrock surface, topographic effects or other impedance boundaries), regional characteristics(such as certain topographic effects), and source characteristics (such as nearby dipping seismic sources)may require that analyses are also able to account for inclined waves.

DG-1146, Page 16Use a Monte Carlo or equivalent procedure to accommodate the variability in soil depth, shearwave velocities, layer thicknesses, and strain-dependent dynamic nonlinear material properties at the site. Perform a sufficient number of convolution analyses using scaled earthquake time histories to adequatelycapture the effect of the site subsurface variability and the uncertainty in the soil properties. Generally,at least 60 convolution analyses should be performed to define the mean and the standard deviation of thesite response. The results of the site response analyses are used to develop high- and low-frequency siteamplification functions as detailed in Appendix E to this regulatory guide.4.3Site Amplification FunctionBased on the suite of site response analyses described in Regulatory Position 4.2, siteamplification functions are calculated. To determine the UHRS at the free-field ground surface for aspecific annual probability of exceedance, multiply the rock-based UHRS by the high-frequency and low-frequency site amplification functions separately, and envelop the two results. If the two controlling earthquake response spectral shapes cover a broad range of frequencies such that when scaled andenveloped they approximate the UHRS, then it is also acceptable to multiply the high- and low-frequencycontrolling earthquake spectra by the appropriate site amplification function and envelope the results. The surface 1 E-04 and 1 E-05 UHRS are used to determine the performance-based SSE.

5.Performance-Based Site-Specific Earthquake Ground Motion Procedure5.1Horizontal SpectrumThe performance-based site-specific earthquake ground motion is developed using a methodanalogous the development of the design response spectrum (DRS) that achieves the annual FOSID targetperformance goal (P F) of 1 E-05, and hazard exceedance probability (H D) of 1 E-04, described in Chapters 1 and 2 of Reference 3, and also discussed in Section B of this guide. The horizontal site-specific earthquake ground motion is obtained by scaling the site-specific mean UHRS by a design factor (DF), or Site-Specific Earthquake Ground Motion = UHRS x DFEquation (1)where UHRS is the mean 1 E-04 UHRS derived in accordance with Regulatory Position 4.4, andDF (from Reference 3) isDF = max { 1.0, 0.6(A R)0.8 }Equation (2)where A R is the ground motion slope ratio of spectral accelerations, frequency by frequency,from a seismic hazard curve corresponding to a 10-fold reduction in hazard exceedance frequency;

therefore, A R = mean 1 E-05 UHRS ÷ mean 1 E-04 UHRSEquation (3)Figure 2 provides a comparison of the performance-based site-specific horizontal designresponse spectrum and the mean 1 E-04 and 1 E-05 UHRS.The above discussion is based on the assumption that the hazard curves are approximated by apower law equation (i.e., linear on a log-log plot) in the range of 1 E-04 to 1 E-05. If A R is greater than4.2, then this assumption is not valid. In these cases, it is acceptable to use a value equal to 45 percent of the mean 1 E-05 UHRS. However, in no case should the site SSE be less than the minimum required byAppendix S to 10 CFR Part 50. (See Regulatory Position 5.4.)

DG-1146, Page 17Site-Specific Earthquake Ground MotionFigure 2. Comparison of the Performance-Base Site Specific Earthquake Ground Motion and the Mean 1E-04 and 1 E-05 UHRS DG-1146, Page 185.2Vertical SpectrumVertical response spectra are developed be combining the appropriate horizontal response spectraand the most recent V/H response spectral ratios appropriate for either CEUS or WUS sites. AppropriateV/H ratios for CEUS or WUS rock and soil sites should be determined from the most recent attenuationrelations. However, as there are currently no CEUS attenuation relations that predict vertical groundmotions, appropriate V/H ratios for CEUS rock sites provided in Reference 2 may be used. For CEUS soil sites, References 2 and 18 describe a procedure to determine a WUS-to-CEUS transfer function thatmay be used to modify the WUS V/H soil ratios. Other methods used to determine V/H ratios may also

be appropriate.5.3Location of the Site Safe Shutdown Earthquake Ground MotionThe horizontal and vertical SSE are determined in the free field on the ground surface. For siteswith one or more thin soil layers near the surface that will be excavated, the SSE is specified on anoutcrop or a hypothetical outcrop of competent material (V s 1000 fps) at a free surface.5.4Determination of Safe Shutdown EarthquakeThe SSE is the vibratory ground motion for which certain structures, systems, and componentsare designed to remain functional, pursuant to Appendix S to 10 CFR Part 50. The SSE for the site ischaracterized by both horizontal and vertical free-field ground motion response spectra at the free groundsurface. In addition, the horizontal component of motion in the free field at the foundation level of thestructures during the SSE must be represented by an appropriate response spectrum with a peak groundacceleration of at least 0.1g.Development of the SSE will be more fully described in NUREG 0800. However, a conceptualoverview of one method to meet the above criteria based on the site-specific earthquake ground motion isdiscussed and provided in Figure 3.Development of the SSE response spectrum begins with the site-specific earthquake groundmotion. This motion is developed at the free field ground surface, but must be supplemented by theregulatory requirements defined at the foundation level. The ground motion at the foundation levelsconsistent with the site-specific earthquake ground motion can be determined by either deconvolution orthrough a surface-to-foundation depth spectra amplification functions. Surface-to-foundation depthspectral amplification functions are developed by dividing the mean spectral values of the surface motionby the mean spectral value calculated from the output of the site response at the depth of the foundationfor each spectral period. This process is analogous to the development of the spectral amplificationfunctions used to compare rock outcrop motions to soil surface motions, as described in Appendix E.Once the motions at the foundation depth resulting from the site specific earthquake groundmotion have been determined, the motions at foundation level are compared against appropriate responsespectrum with a peak ground acceleration of at least 0.1g. If necessary, a composite spectrum thatenvelopes both spectra should be determined. This new composite spectrum is defined at the free fieldfoundation depth and should be used to determine a new free field surface spectrum through eitherappropriate site response analyses or through the use of the surface-to-foundation depth spectralamplification functions. This new free field surface spectrum is the SSE. In cases where the minimumspectra does not govern, this spectra should be very similar or the same as the site-specific earthquakeground motion determined using this guide.Once the SSE is developed, it is compared with the seismic design criteria in the designcertification documentation.

DG-1146, Page 19 Develop performance-based site specificearthquake ground motion procedure Use surface-to-foundation spectra amplification functions or deconvolu tion to develop soil motions at foundation level from site-specificearthquake ground motionCompare soil motions at foundation level against appropriate response spectrum with a peakground acceleration of at least 0.1g.

Envelop both spectra.Using enveloped spectra at foundation level,use site response analyses or spectra amplificationfunctions to determine spectra at the surface. The resulting surface spectra is the SSE.Figure 3. Possible Development Methodology of Safe Shutdown Earthquake DG-1146, Page 20 D. IMPLEMENTATIONThe purpose of this section is to provide information to applicants regarding the NRC staff'splans for using this draft regulatory guide. No backfitting is intended or approved in connection with its issuance.This draft regulatory guide identifies methods that the NRC staff considers acceptablefor (1) conducting geological, geophysical, seismological, and geotechnical investigations, (2) identifyingand characterizing seismic sources, (3) conducting probabilistic seismic hazard analyses, (4) determiningseismic wave transmission characteristics of soil and rock sites, and (5) determining a performance-basedsite-specific earthquake ground motion, for satisfying the requirements of 10 CFR 100.23. A brief overviewof techniques to develop the SSE in order to satisfy the requirements of A ppendix S to 10 CFR Part 50is also included in this guide and in NUREG-0800, "Standard Review Plan," Section 3.7.1.The NRC has issued this draft guide to encourage public participation in its development. Except in those cases in which an applicant or licensee proposes or has previously established an acceptablealternative method for complying w ith specified portions of the NRC's regulations, the methods to bedescribed in the active guide will reflect public comments and will be used in evaluating submittals in connection with applications for early site permits and combined licenses.

DG-1146, Page 21REGULATORY ANALYSIS1.Statement of the ProblemThe U.S. Nuclear Regulatory Commission (NRC) issued Regulatory Guide 1.165, in March1997. The regulatory guide provides general guidance on procedures acceptable to the NRC staff to satisfy the requirements of 10 CFR 100.23.The NRC staff recognized the need to provide alternative guidance for incorporating recentdevelopments in ground motion estimation models, updated models for earthquake sources, new methodsfor determining site response, and new methods for defining a site-specific, performance-based safe shutdown earthquake (SSE) ground motion. Therefore, this draft regulatory guide is being proposed bythe NRC staff as an alternative to Regulatory Guide 1.165 to satisfy the requirements of 10 CFR 100.23 and Appendix S to 10 CFR Part 50.2.ObjectiveThe objective of this regulatory action is to update the NRC's guidance in the area of siting anddefining a site-specific, performance-based SSE to give licensees and applicants an opportunity to usestate-of-the-art methods that are currently available.3.Alternatives and Consequences of the Pr oposed ActionThe NRC staff considered the following alternative approaches to the problem of outdatedguidance regarding the siting and seismic design:(1)Do not update Regulatory Guide 1.165.(2)Update Regulatory Guide 1.165.

(3)Develop a new Regulatory Guide.3.1Alternative 1: Do Not Update Regulatory Guide 1.165Under this alternative, the NRC would not update Regulatory Guide 1.165. Applicants andlicensees would continue to rely on the current version of the regulatory guide. This alternative isconsidered the baseline or "no-action" alternative.3.2Alternative 2: Update Regulatory Guide 1.165Under this alternative, the NRC would update Regulatory Guide 1.165. The update wouldincorporate improved methods for (1) conducting geological, geophysical, seismological, and geotechnicalinvestigations; (2) identifying and characterizing seismic sources; (3) conducting probabilistic seismichazard analyses (PSHA); (4) developing ground motion time histories for use in soil response(including a database of time histories for the CEUS and WUS); (5) determining seismic wave transmission(soil amplification) characteristics of soil and rock sites; and (6) defining a site-specific, performance-basedSSE. The staff has identified the following conse quences associated with adopting Alternative 2:(1)Applicants and licensees would have guidance on the use of the latest technology available,with consequent improvements in the siting and defining site-specific ground motion

for nuclear power plants.

DG-1146, Page 22(2)Regulatory efficiency would be improved by reducing uncertainty as to what is acceptable and byencouraging consistency in the siting of nuclear power plants. Benefits to the industry and theNRC will accrue to the extent this occurs. NRC reviews would be facilitated because licensee submittals would be more predictable and analytically consistent.(3)Both the NRC and the nuclear industry would realize cost savings. From the NRC's perspective,relative to the baseline, the NRC would incur one-time incremental costs to issue the revisedregulatory guide. However, the NRC should also realize cost savings associated with the reviewof licensee submittals. In the staff's view, the ongoing cost savings associated with these reviewsshould more than offset the one-time cost.

On balance, the NRC staff expects that industry would realize net savings, as the one-timeincremental cost to review and comment on the revised regulatory guide would be more thancompensated for by the efficiencies (e.g., reduced unnecessary conservatism, followup questions,and revisions) associated with each licensee submission.(4)Because the new approach is substantially different from the existing Regulatory Guide 1.165 approach, inclusion of both approaches in the same document could be confusing to stakeholdersand users of the document.3.3Alternative 3: Develop a New Regulatory GuideUnder this alternative, the NRC would develop a new regulatory guide reflecting theimprovements to Regulatory Guide 1.165 identified in Alternative 2. In addition to the consequencesassociated with Alternative 2, the staff has determined that the consequence associated with adoptingAlternative 3 is that Regulatory Guide 1.165 would remain as an alternative option for satisfying therequirements of 10 CFR 100.23.4.ConclusionBased on this regulatory analysis, the staff recommends that the NRC develop a new regulatoryguide. The staff concludes that the proposed action will reduce unnecessary burden on the part of boththe NRC and its licensees, while improving the process for siting of nuclear power plants. Furthermore,the staff sees no adverse effects with retaining Regulatory Guide 1.165, an acceptable alternative to thenew regulatory guide for satisfying the requirements of 10 CFR 100.23.BACKFIT ANALYSISThis regulatory guide gives applicants an opportunity to use currently available state-of-the-artmethods for satisfying the requirements of 10 CFR 100.23 and Appendix A to 10 CFR Part 50. As such,this draft regulatory guide does not require a backfit analysis as describe d in 10 CFR 50.109(c), becauseit does not impose a new or amended provision in the Commission's rules or a regulatory staff position interpreting the Commission's rules that is either new or different from a prev ious applicable staffposition. In addition, this regulatory guide does not require modification or addition to structures,systems, components, or design of a facility or the procedures or organization required to design,construct, or operate a facility. The application of this regulatory guide is voluntary. An applicant is freeto select a preferred method for achieving complianc e with a license or the rules or orders of theCommission as described in 10 CFR 50.109(a)(7). Applicants may continue to use Regulatory Guide 1.165, issued March 1997.

5All regulatory guides listed herein were published by the U.S. Nuclear Regulatory Commission. Where an accessionnumber is identified, the specified regulatory guide is available electronically through the NRC's AgencywideDocuments Access and Management System (ADAMS) at http://www.nrc.gov/reading-rm/adams.html. All otherregulatory guides are available electronically through the Public Electronic Reading Room on the NRC's public Website, at http://www.nrc.gov/reading-rm/doc-collections/reg-guides/. Single copies of regulatory guides may also beobtained free of charge by writing the Reproduction and Distribution Services Section, ADM, USNRC, Washington,DC 20555-0001, or by fax to (301) 415-2289, or by email to DISTRIBUTION@nrc.gov. Active guides may also bepurchased from the National Technical Information Service (NTIS) on a standing order basis. Details on this service may be obtained by contacting NTIS at 5285 Port Royal Road, Springfield, Virginia 22161, online at http://www.ntis.gov, or by telephone at (703) 487-4650. Copies are also available for inspection or copying for a feefrom the NRC's Public Document Room (PDR), which is located at 11555 Rockville Pike, Rockville, Maryland; the PDR's mailing address is USNRC PDR, Washington, DC 20555-0001. The PDR can also be reached by tele phoneat (301) 415-4737 or (800) 397-4205, by fax at (301) 415-3548, and by email to PDR@nrc.gov

.6Copies are available for inspection or copying for a fee from the NRC's Public Document Room at 11555 RockvillePike, Rockville, MD; the PDR's mailing address is USNRC PDR, Washington, DC 20555; telephone (301) 415-4737or (800) 397-4209; fax (301) 415-3548; email PDR@nrc.gov. In addition, copies are available at current ratesfrom the U.S. Government Printing Office, P.O. Box 37082, Washington, DC 20402-9328, telephone (202) 512-1800; or from the National Technical Information Service (NTIS), 5285 Port Royal Road, Springfield, VA 22161, http://www.ntis.gov, telephone (703) 487-4650.

7Copies may be purchased from the American Society for Civil Engineers (ASCE), 1801 Alexander Bell Drive, Reston, VA 20190 [phone: 800-548-ASCE (2723)]. Purchase information is available through the ASCE Web site at http://www.asce.org/bookstore/book.cfm?isbn=0784407622 8Copies are available for inspection or copying for a fee from the NRC's Public Document Room at 11555 RockvillePike, Rockville, MD; the PDR's mailing address is USNRC PDR, Washington, DC 20555; telephone (301) 415-4737or (800) 397-4209; fax (301) 415-3548; email PDR@nrc.gov

.DG-1146, Page 23 REFERENCES 1.Regulatory Guide 1.165, "Identification and Characterization of Seismic Sources andDetermination of Safe Shutdown Earthquake Ground Motion," U.S. Nuclear RegulatoryCommission, Washington, DC, March 1997, available in ADAMS under Accession

1. ML003740084.

52.McGuire, R.K., W.J. Silva, and C.J. Costantino, "Technical Basis for Revision of RegulatoryGuidance on Design Ground Motions: Hazard- and Risk-Consistent Ground Motion SpectraGuidelines," NUREG/CR-6728, U.S. Nuclear Regulatory Commission, Washington, DC, October 2001.

6 3.ASCE/SEI 43-05, "Seismic Design Criteria for Structures, Systems, and Components in NuclearFacilities," American Society of Civil Engineers/Structural Engineering Institute, 2005.

7 4.Regulatory Guide 1.60, "Design Response Spectra for Seismic Design of Nuclear PowerPlants," U.S. Nuclear Regulatory Commission, Revision 1, December 1973, available inADAMS under Accession #ML003740207.

45.Budnitz, R.J., et al., "Recommendations for Probabilistic Seismic Hazard Analysis: Guidance onUncertainty and Use of Experts," NUREG/CR-6372, Volumes 1 and 2, U.S. Nuclear RegulatoryCommission, Washington, DC, April 1997.

5 6.Pacific Gas and Electric Company, "Final Report of the Diablo Canyon Long-Term SeismicProgram: Diablo Canyon Power Plant," Docket Nos. 50-275 and 50-323, 1988.

8 DG-1146, Page 247.Rood, H., et al

., "Safety Evaluation Report Related to the Operation of Diablo Canyon NuclearPower Plant, Units 1 and 2," NUREG-0675, Supplement No. 34, U.S. Nuclear RegulatoryCommission, Washington, DC, June 1991.

58.Sorensen, G., Washington Public Power Supply System, Letter to Document Control Branch,U.S. Nuclear Regulatory Commission, Washington, DC,

Subject:

"Nuclear Project No. 3,Resolution of Key Licensing Issues, Response," February 29, 1988.

79.Bernreuter, D.L., et al., "Seismic Hazard Characterization of 69 Nuclear Plant Sites East of theRocky Mountains," NUREG/CR-5250, Volumes 1-8, U.S. Nuclear Regulatory Commission,Washington, DC, January 1989.

510.Sobel, P., "Revised Livermore Seismic Hazard Estimates for Sixty-Nine Nuclear Power PlantSites East of the Rocky Mountains," NUREG-1488, U.S. Nuclear Regulatory Commission,Washington, DC, April 1994.

511.Savy, J.B., et al., "Eastern Seismic Hazard Characterization Update," UCRL-ID-115111,Lawrence Livermore National Laboratory, June 1993, (Accession #9310190318 in NRC'sPublic Document Room).

712.Electric Power Research Institute (EPRI), "Probabilistic Seismic Hazard Evaluations at Nuclear Power Plant Sites in the Central and Eastern United States," NP-4726, All Volumes, 1989-1991.13.Electric Power Research Institute (EPRI), "The Earthquakes of Stable Continental Regions,"Volume 1: Assessment of Large Earthquake Potential, EPRI TR-102261-V1, 1994.14.Electric Power Research Institute (EPRI) and U.S. Department of Energy (DOE), "Program onTechnology Innovation: Use of Minimum CAV in Determining Effects of Small MagnitudeEarthquakes on Seismic Hazard Analyses," Report 1012965, December 2005.

15. Electric Power Research Institute (EPRI) and U.S. Department of Energy (DOE), "Program onTechnology Innovation: Truncation of the Lognormal Distribution and Value of the StandardDeviation for Ground Motion Models in the Ce ntral and Eastern Unite d States," Report 1013105,February 2006.

16.Regulatory Guide 1.132, "Site Investigations for Foundations of Nuclear Power Plants,"Revision 1, U.S. Nuclear Regulatory Commission, Washington, DC, October 2003, available inADAMS under Accession #ML032800710.

4 17.Regulatory Guide 1.138, "Laboratory Investigations of Soils and Rocks for EngineeringAnalysis and Design of Nuclear Power Plants," Revision 2, U.S. Nuclear RegulatoryCommission, Washington, DC, December 2003 available in ADAMS under Accession

1. ML003740184.

418.McGuire, R.K., W.J. Silva, and C.J. Costantino, "Technical Basis for Revision of RegulatoryGuidance on Design Ground Motions: Development of Hazard- and Risk-Consistent SeismicSpectra for Two Sites," NUREG/CR-6769, U.S. Nuclear Regulatory Commission, Washington, DC, April 2002.

5 Appendix A to DG-1146, Page A-1APPENDIX ADEFINITIONSCombined License. A combined construction permit and operating license with conditions for a nuclearpower facility issued pursuant to Subpart C of 10 CFR Part 52.Controlling Earthquakes. The earthquakes used to determine spectral shapes or to estimate groundmotions at the site. There may be several controlling earthquakes for a site. As a result of the probabilistic seismic hazard analysis (PSHA), controlling earthquakes are characterized as meanmagnitudes and distances derived from a deaggregation analysis of the mean estimate of the PSHA.Cumulative Absolute Velocity (CAV). For each component of the free-field ground motion, the CAV should be calculated as follows: (1) the absolute acceleration (g units) time-history is divided into1-second intervals, (2) each 1-second interval that has at least 1 exceedance of 0.025g is integrated overtime, and (3) all the integrated values are summed together to arrive at the CAV. The CAV is exceededif the calculation is greater than 0.16 g-second. The application of the CAV in siting requires thedevelopment of a CAV model (Ref. 14) because the PSHA calculation does not use time historiesdirectly.Deaggregation. The process for determining the fractional contribution of each magnitude-distance pairto the total seismic hazard. To accomplish this, a set of magnitude and distance bi ns are selected and theannual probability of exceeding selected ground acceleration parameters from each magnitude-distance pair is computed and divided by the total probability for earthquakes.Design Factor. The ratio between the site-specific earthquake ground motion and the UHRS. Thedesign factor is aimed at achieving the target annual probability of failure associated with the targetperformance goals.Early Site Permit. A Commission approval, issued pursuant to Subpart A of 10 CFR Part 52, for a siteor sites for one or more nuclear power facilities.

Earthquake Recurrence. The frequency of occurrence of earthquakes as a function of magnitude. Recurrence relationships or curves are developed for each seismic source, and they reflect the frequencyof occurrence (usually expressed on an annual basis) of magnitudes up to the maximum, includingmeasures of uncertainty.Frequency of Onset of Significant Inelastic Deformation (FOSID). The annual probability of theonset of significant inelastic deformation (OSID). OSID is just beyond the occurrence of insignificant(or localized) inelastic deformation, and in this way corresponds to "essentially elastic behavior." As such, OSID of a structure, system, or component (SSC) can be expected to occur well before seismicallyinduced core damage, resulting in much larger frequencies of OSID than seismic core damage frequency (SCDF) values. In fact, OSID occurs before SSC "failure," where the term failure refers to impairedfunctionality.Ground Motion Slope Ratio. Ratio of the spectral accelerations, frequency by frequency, from aseismic hazard curve corresponding to a 10-fold reduction in hazard exceedance frequency.

(See Equation 3 in Regulatory Position 5.1.)

Appendix A to DG-1146, Page A-2High Confidence in Low Probability of Failure (HCLPF)

Capacity. An earthquake acceleration levelfor which a given component, system, or plant is evaluated as having a 1% failure probability on the mean fragility curve.

Intensity. The intensity of an earthquake is a qualitative description of the effects of the earthquake at aparticular location, as evidenced by observed effects on humans, on human-built structures, and on the earth's surface at a particular location. Commonly used scales to specify intensity are the Rossi-Forel, Mercalli, and Modified Mercalli. The Modified Mercalli Intensity (MMI) scale describes intensities withvalues ranging from I to XII in the order of severity. MMI of I indicates an earthquake that was not feltexcept by a very few, whereas MMI of XII indicates total damage of all works of construction, eitherpartially or completely.

Magnitude. An earthquake's magnitude is a measure of the strength of the earthquake as determinedfrom seismographic observations and is an objective, quantitative measure of the size of an earthquake. The magnitude can be expressed in various ways based on seismographic records (e.g., Richter LocalMagnitude, Surface Wave Magnitude, Body Wave Magnitude, and Moment Magnitude). Currently, the most commonly used magnitude measurement is the Moment Magnitude, M w , which is based on theseismic moment computed as the rupture force along the fault multiplied by the average amount of slip,and thus is a direct measure of the energy released during an earthquake.Maximum Magnitude. The maximum magnitude is the upper bound to earthquake recurrence curves.Mean Site Amplification Function. The mean amplification function is obtained for each controllingearthquake, by dividing the response spectrum from the computed surface motion by the responsespectrum from the input hard rock motion, and computing the arithmetic mean of the individual response spectral ratios.Performance-Based Response Spectrum. A uniform hazard response spectrum (UHRS) that ismodified by design factors at various frequencies.Nontectonic Deformation. Nontectonic deformation is distortion of surface or near-surface soils orrocks that is not directly attributable to tectonic activity. Such deformation includes features associatedwith subsidence, karst terrain, glaciation or deglaciation, and growth faulting.Response Spectrum. A plot of the maximum responses (acceleration, velocity, or displacement) ofidealized single-degree-of-freedom oscillators as a function of the natural frequencies of the oscillatorsfor a given damping value. The response spectrum is calculated for a specified vibratory motion input at the oscillators' supports.Ring Area. Annular region bounded by radii associated with the distance rings used in hazarddeaggregation (Table D.1, "Recommended Magnitude and Distance Bins," in Appendix D to thisregulatory guide).Safe Shutdown Earthquake Ground Motion (SSE). The vibratory ground motion for which certainstructures, systems, and components are designed, pursuant to Appendix S to 10 CFR Part 50, to remainfunctional. The SSE for the site is characterized by both horizontal and vertical free-field ground motionresponse spectra at the free ground surface.Seismic Source. A general term referring to both seismogenic sources and capable tectonic sources.

Appendix A to DG-1146, Page A-3Capable Tectonic Source. A capable tectonic source is a tectonic structure that can generate bothvibratory ground motion and tectonic surface deformation such as faulting or folding at or near theearth's surface in the present seismotectonic regime. It is described by at least one of the following characteristics:a.presence of surface or near-surface deformation of landforms or geologic deposits of a recurringnature within the last approximately 500,000 years or at least once in the last approximately50,000 yearsb.a reasonable association with one or more moderate to large earthquakes or sustained earthquakeactivity that are usually accompanied by significant surface deformationc.a structural association with a capable tectonic source that has characteristics of either item a or b(above), such that movement on one could be reasonably expected to be accompanied bymovement on the otherIn some cases, the geological evidence of past activity at or near the ground surface along a potentialcapable tectonic source may be obscured at a particular site. This might occur, for example, at a sitehaving a deep overburden. For these cases, evidence may exist elsewhere along the structure from whichan evaluation of its characteristics in the vicinity of the site can be reasonably based. Such evidence is tobe used in determining whether the structure is a capable tectonic source within this definition.Notwithstanding the foregoing paragraphs, the association of a structure with geological structures thatare at least pre-Quaternary, such as many of those found in the central and eastern regions of the UnitedStates, in the absence of conflicting evidence, will dem onstrate that the structure is not a capable tectonicsource within this definition.Seismogenic Source. A seismogenic source is a portion of the earth that is assumed to have a uniformearthquake potential (same expected maximum earthquake and recurrence frequency), distinct from thatof surrounding sources

. A seismogenic source will generate vibratory ground motion but is assumed tonot cause surface displacement. Seismogenic sources cover a wide range of seismotectonic conditions,from a well-defined tectonic structure to simply a large region of diffuse seismicity (seismotectonicprovince).Seismic Wave Transmission (Site Amplification). The amplification (increase or decrease) ofearthquake ground motion by rock and soil near the earth's surface in the vicinity of the site of interest. Topographic effects, the effect of the water table, and basin edge wave-propagation effects are sometimes included under site response.Spectral Acceleration. Peak acceleration response of an osc illator as a function of period or frequencyand damping ratio when subjected to an acceleration time history. It is equal to the peak relative displacement of a linear oscillator of frequency, f, attached to the ground, times the quantity (2f)2. It isexpressed in units of gravity (g) or cm/second 2.Stable Continental Region (SCR). An SCR is composed of continental crust, including continentalshelves, slopes, and attenuated continental crust, and excludes active plate boundaries and zones ofcurrently active tectonics directly influenced by plate margin processes. It exhibits no significantdeformation associated with the major Mesozoic-to-Cenozoic (last 240 million years) orogenic belts. It excludes major zones of Neogene (last 25 million years) rifting, volcanism, or suturing.

Appendix A to DG-1146, Page A-4Stationary Poisson Process. A probabilistic model of the occurrence of an event over time (or space)that has the following characteristics: (1) the occurrence of the event in small intervals is constant overtime (or space), (2) the occurrence of two (or more) events in a small interval is negligible, and (3) theoccurrence of the event in non-overlapping intervals is independent.Target Performance Goal (P F). Target annual probability of exceeding the 1 E-05 frequency of onsetof significant inelastic deformation (FOSID) limit state.Tectonic Structure. A large-scale dislocation or distortion, usually within the earth's crust. Its extentmay be on the order of tens of meters (yards) to hundreds of kilometers (miles).Uniform Hazard Response Spectrum (UHRS). A plot of a ground response parameter (for example,spectral acceleration or spectral velocity) that has an equal likelihood of exceedance at different frequencies.

Within Motion. An earthquake record modified for use in a site response model. Within motions aredeveloped through deconvolution of a surface recording to account for the properties of the overburdenmaterial at the level at which the record is to be applied. The within motion can also be called the"bedrock motion" if it occurs at a high-impedance boundary where rock is first encountered.

Appendix B to DG-1146, Page B-1APPENDIX B ABBREVIATIONSApeak ground acceleration A Rground motion slope ratioASCEAmerican Society of Civil Engineers ASTMAmerican Society of Testing and MaterialsCAVcumulative absolute velocityCPconstruction permit CEUScentral and eastern United StatesDdistanceDpeak ground displacement DFdesign factorEPRIElectric Power Research Institute FOSIDfrequency of onset of significant inelastic deformation HCLPFhigh-confidence-low-probability-of-failure (1% mean probability of failure)

H Dmean annual hazard exceedance frequencyLLNLLawrence Livermore National Laboratory MmagnitudeMMIModified Mercalli IntensityNRCU.S. Nuclear Regulatory Commission OLoperating licenseOMBOffice of Management and Budget OSIDonset of significant inelastic deformation P Ftarget performance goalPGApeak ground acceleration PGVpeak ground velocity PSHAprobabilistic seismic hazard analysis R Pprobability ratioSCDFseismic core damage frequencySCRstable continental region SDCseismic design category SRPStandard Review Plan (NUREG-0800)

SSCsstructures, systems, and components SSEsafe shutdown earthquake ground motion StandardASCE/SEI Standard 43-05 Appendix B to DG-1146, Page B-2UHRSuniform hazard response spectrumUSNRCU.S. Nuclear Regulatory CommissionVpeak ground velocityV/Hratio of vertical to horizontal spectral accelerations V sshear wave velocityWUSwestern United States Appendix C to DG-1146, Page C-1APPENDIX CINVESTIGATIONS TO CHARACTERIZE SITE GEOLOGY, SEISMOLOGY AND GEOPHYSICSC.1IntroductionAs characterized for use in probabilistic seismic hazard analyses (PSHA), seismogenic sourcesare zones within which future earthquakes are likely to occur. Geological, seismological, andgeophysical investigations provide the information needed to identify and characterize source zoneparameters, such as size and geometry, and to estimate earthquake recurrence rates and maximum magnitudes. The amount of data available about earthquakes and their causative sources variessubstantially between the western United States (WUS) (west of the Rocky Mountain front) and the central and eastern United States (CEUS), or stable continental region (east of the Rocky Mountainfront). Furthermore, there are variations in the amount and quality of data within these regions.Seismogenic sources in active tectonic regions such as the WUS are generally readily identifiedbecause of their relatively high activity rate. In the CEUS, identifying seismogenic sources is more difficult because their activity rate is relatively low and because most seismic sources are covered bythick deposits. However, several significant tectonic structures exist and some of these have beeninterpreted as seismogenic sources (e.g., the New Madrid fault zone, Nemaha Ridge, and Meers fault).In the CEUS, it is most likely that the determination of the properties of the seismogenic source,whether it is a tectonic structure or a seismotectonic province, will be inferred rather than demonstratedby strong correlations with seismicity or geologic data. Moreover, it is not generally known whatrelationships exist between observed tectonic structures in a seismic source within the CEUS and thecurrent earthquake activity that may be associated with that source. The historical seismicity record, theresults of regional and site studies, and expert judgment play key roles in characterizing a source zone. If, on the other hand, strong correlations and data exist suggesting a relationship between seismicity andseismic sources, approaches used for more active tectonic regions can be applied. Reference C.1 may beused to assess potential for large earthquakes.The primary objective of geological, seismological, and geophysical investigations is to developan up-to-date, site-specific earth science database that supports site characterization and a PSHA. Theresults of these investigations will also be used to assess whether new data and their interpretation areconsistent with the information used in accepted probabilistic seismic hazard studies. If the new data, such as new seismic sources and new ground motion a ttenuation relationships, are consistent with theexisting earth science database, updating or modification of the hazard analysis is not required. It will benecessary to update seismic sources and ground motion attenuation relationships for sites where there is significant new information provided by the site investigation.The following are to be evaluated for seismic source characterization: Seismic source location and geometry (location and extent, both surface and subsurface). Thisevaluation will normally require interpretations of available geological, geophysical, andseismological data in the source region by multiple experts or a team of experts. The evaluation should include interpretations of the seismic potential of each source and relationships amongseismic sources in the region to express uncertainty in the evaluations. Seismic source evaluations generally develop four types of sources: (1) fault sources, (2) area sourcesrepresenting concentrated historic seismicity not associated with k nown tectonic structure, Appendix C to DG-1146, Page C-2(3) area sources representing geographic regions with similar tectonic histories, type of crust, and structural features, and (4) background sources. Background sources are generally used toexpress uncertainty in the overall seismic source configuration interpreted for the site region. Acceptable approaches for evaluating and characterizing uncertainties for input to a seismichazard calculation are contained in NUREG/CR-6372 (Ref. C.2).Evaluations of earthquake recurrence for each seismic source, including recurrence rate andrecurrence model. These evaluations normally draw most heavily on historical and instrumentalseismicity associated with each source and paleoearthquake information. Preferred methods andapproaches for evaluating and characterizing uncertainty in earthquake recurrence generally willdepend on the type of source. Acceptable methods are described in NUREG/CR-6372 (Ref. C.2). Evaluations of the maximum earthquake magnitude for each seismic source. These evaluationswill draw on a broad range of source-specific t ectonic characteristics, including tect onic historyand available seismicity data. Uncertainty in this evaluation should normally be expressed as amaximum magnitude distribution. Preferred methods and information for evaluating andcharacterizing maximum earthquakes for seismic sources vary with the type of source.

Acceptable methods are contained in NUREG/CR-6372 (Ref. C.2). Other evaluations, depending on the geologic setting of a site, such as local faults that have ahistory of Quaternary (last 2 million years) displacements, sense of slip on faults, fault length andwidth, area of faults, age of displacements, estimated displacement per event, estimatedearthquake magnitude per offset event, orientations of regional tectonic stresses with respect tofaults, and the possibility of seismogenic folds. Capable tectonic sources are not always exposedat the ground surface (such as blind thrust faults) in the WUS as demonstrated by the buriedreverse causative faults of the 1983 Coalinga, 1988 Whittier Narrows, l989 Loma Prieta, and 1994 Northridge earthquakes. These examples emphasize the need to conduct thoroughinvestigations not only at the ground surface but also in the subsurface to identify structures atseismogenic depths. Regional topographic featur es should also be closely studied. Whenever faults or other structures are encountered at a site (including s ites in the CEUS) in either outcropor excavations, it is necessary to perform adequately detailed specific investigations to determinewhether or not they are seismogenic or may cause surface deformation at the site. Acceptablemethods for performing these investigations are contained in NUREG/CR-5503 (Ref. C.3).C.2.Investigations To Evaluate Seismic SourcesC.2.1GeneralInvestigations of the site and region around the site are necessary to identify capable tectonicsources and to determine their potential for generating earthquakes and causing surface deformation. If itis determined that surface deformation need not be take n into account at the site, sufficient data to clearlyjustify the determination should be presented in the application for an early site permit or combinedlicense. Generally, any tectonic deformation at the earth's surface within the Site Area [8 km (5 mi)ofthe site] will require detailed examination to determine its significance. Potentially active tectonicdeformation within the seismogenic zone beneath a site will have to be assessed using geological, geophysical and seismological methods to determine its significance. Engineering solutions are generallyavailable to mitigate the potential vibratory effects of earthquakes through design. However, engineeringsolutions cannot always be demonstrated to be adequate for mitigation of the effects of permanent grounddisplacement phenomena such as surface faulting or folding, subsidence, or ground collapse. For thisreason, it is prudent to select an alternative site when the potential exists for permanent grounddisplacement at the proposed site (Ref. C.4).

Appendix C to DG-1146, Page C-3The level of detail for investigations should be governed by knowledge of the current and lateQuaternary tectonic regime and the geological complexity of the site and region. The investigations fordetermining seismic sources should be carried out in all the following levels, with areas described byradii of 320 km (200 mi) (Site Region), 40 km (25 mi) (Site Vicinity), and 8 km (5 mi) (Site Area) and 1 km (0.6 mi) (Site Location) from the site. The investigations should provide increasingly detailed information as they proceed from the regional level down to the site (e.g., from Site Region to SiteLocation). Whenever faults or othe r structures are encountered at a site in either outcrop or excavations,it is necessary to perform many of the investigations described below to determine whether or not they

are capable tectonic sources.

C.2.1.1 Site Region Investigations [within radii of 320 km (200 mi) - 40 km (25 mi) of the site]The Site Region investigations should be planned to identify seismic sources and describe theQuaternary tectonic regime. The data should be presented at a scale of 1:500,000 or larger. Theinvestigations are not expected to be extensive or in detail, but should include a comprehensive literaturereview supplemented by focused geological reconnaissances based on the results of the literature study (including topographic, geologic, aeromagnetic, and gravity maps, and airphotos). Some detailedinvestigations at specific locations within the region may be necessary if potential seismic sources thatmay be significant for determining the safe shutdown earthquake ground moti on (SSE), are identified.The large size of the area for the regional investigations is recommended because of thepossibility that all significant seismic sources, or alternative configurations, may not have been capturedor enveloped by previous source databases, such as the Lawrence Livermore National Laboratory(LLNL)/Electric Power Research Institute (EPRI) database. Thus, it will increase the chances of (1) identifying evidence for unknown seismic sources that might extend close enough for earthquake ground motions generated by that source to affect the site, and (2) confirming the PSHA's database. Furthermore, because of the relative rarity of seismic activity in the CEUS, the area should be largeenough to include as many historical and instrumentally recorded earthquakes for analysis as reasonablypossible. The specified area of study is expected to be large enough to incorporate any previouslyidentified sources that could be analogous to sources that may underlie or be relatively close to the site. In past licensing activities for sites in the CEUS, it has often been necessary, because of the absence ofdatable horizons overlying bedrock, to extend investigations out many tens or hundreds of kilometersfrom the site along a structure or to an outlying analogous structure to locate overlying datable strata orunconformities so that geochronological methods could be applied. This procedure has also been used toestimate the age of an undatable seismic source in the Site Vicinity by relating its time of last activity tothat of a similar, previously evaluated structure, or a known tectonic episode, the evidence of which maybe many tens or hundreds of kilometers away.It is necessary to evaluate the geological, seismological, and geophysical data obtained from thesite-specific investigations to demonstrate that th e data are sufficient and consistent with the PSHAmethodology. Specifically, the evaluation may need to include but not limited to seismic sourcecharacterization, ground motion attenua tion relationships and other PSHA-related contents, such as low-magnitude threshold and magnitude conversion relationships. If new information identified by the site-specific investigations were to result in a significant increase in the hazard estimate for a site, and if thisnew information were validated by a strong technical basis, the PSHA might have to be modified toincorporate the new technical information. Using sensitivity studies, it may also be possible to justify alower hazard estimate with an exceptionally strong technical basis. However, it is expected that largeuncertainties in estimating seismic hazard in the CEUS will continue to exist in the future, andsubstantial delays in the licensing process will result from trying to justify a lower value with respect to a specific site.

Appendix C to DG-1146, Page C-4 C.2.1.2 Site Vicinity Investigations [within radii of 40 km (25 mi) - 8 km (5 mi) of the site]Reconnaissance-level investigations, which may need to be supplemented at specific locations bymore detailed explorations such as geologic mapping, geophysical surveying, borings, and trenching,should be conducted for the Site Vicinity; the data should be presented at a scale of 1:50,000 or larger.

C.2.1.3 Site Area Investigations [within radii of 8 km (5 mi) - 1 km (0.6 mi) of the site]Detailed investigations should be carried out within the Site Area, and the resulting datapresented at a scale of 1:5,000 or larger. The level of investigations should be in sufficient detail todelineate the geology and the potential for tectonic deformation at or near the ground surface. Theinvestigations should use the methods described in the following subsections that are appropriate for thetectonic regime to characterize seismic sources.The areas of investigations may be asymmetrical and may cover larger areas than those describedabove in regions of late Quaternary activity, regions with high rates of historical seismic activity (felt orinstrumentally recorded data), or sites that are located near a capable tectonic source such as a fault zone.C.2.2Contents of Site Vicinity and Site Area InvestigationsThe following methods are suggested but they are not all-inclusive and investigations should notbe limited to them. Some procedures will not be applicable to every site, and situations will occur thatrequire investigations that are not included in the following discussion. It is anticipated that newtechnologies will be available in the future that will be applicable to these investigations.

C.2.2.1 Surface InvestigationsSurface exploration to assess the geology and geologic structure of the Site Area is dependent onthe Site Location and may be carried out with the use of any appropriate combination of the geological,geophysical, seismological, and geotechnical engineering techniques summarized in the followingparagraphs. However, not all of these methods must be carried out at a given site.

C.2.2.1.1Geological interpretations should be performed of aerial photographs and otherremote-sensing imagery, as appropriate for the particular site conditions, to assist in identifying rock outcrops, faults and other tectonic features, fracture traces, geologic contacts, lineaments, soil conditions,and evidence of landslides or soil liquefaction.

C.2.2.1.2Mapping topographic, geomorphic, and hydrologic features should be performedat scales and with contour intervals suitable for analysis and descriptions of stratigraphy (particularlyQuaternary), surface tectonic structures such as fault zones, and Quaternary geomorphic features. Forcoastal sites, or sites located near lakes or rivers, this includes topography, geomorphology (particularlymapping marine and fluvial terraces), bathymetry, submarine landslides, geophysics (such as seismicreflection), and hydrographic surveys to the extent needed to describe the site area features.

C.2.2.1.3Vertical crustal movements should be evaluated using (1) geodetic landsurveying and (2) geological analyses (such as analysis of regional dissection and degradation patterns),shorelines, fluvial adjustments (such as changes in stream longitudinal profile s or terraces), and otherlong-term changes (such as elevation changes across lava flows).

Appendix C to DG-1146, Page C-5 C.2.2.1.4Analysis should be performed to determine the tectonic significance of offset,displaced, or anomalous landforms such as displaced stream channels or changes in stream profiles or theupstream migration of knickpoints; abrupt changes in fluvial deposits or terraces; changes in paleo-channels across a fault; or uplifted, down-dropped, or laterally displaced marine terraces.

C.2.2.1.5Analysis should be performed to determine the tectonic significance ofQuaternary sedimentary deposits within or near tectonic zones, such as fault zones, including (1) fault-related or fault-controlled deposits such as sag ponds, graben fill deposits, and colluvial wedges formed by the erosion of a fault paleo-scarp and (2) non-fault-related, but offset, deposits such as alluvial fans,debris cones, fluvial terrace, and lake shoreline deposits.

C.2.2.1.6Identification and analysis should be performed of deformation features causedby vibratory ground motions, including seismically i nduced liquefaction features (sand boils, explosioncraters, lateral spreads, settlement, soil flows), mud volcanoes, landslides, rockfalls, deformed lakedeposits or soil horizons, shear zones, cracks, or fissures.

C.2.2.1.7Analysis should be performed of fault displacements, including the interpretionof the morphology of topographic fault scarps associated with or produced by surface rupture. Faultscarp morphology is useful in estimating the age of last displacement [in conjunction with the appropriategeochronological methods described in NUREG/CR-5562 (Ref. C.5)], approximate magnitude of theassociated earthquake, recurrence intervals, slip rate, and the nature of the causative fault at depth.

C.2.2.2 Subsurface Investigation ContentsSubsurface investigations at the site to identify and describe potential seismic sources or capable tectonic sources and to obtain required geotechnical information are described in Regulatory Guide 1.132(Ref. C.6). The investigations include, but may not be confined to, the following:

C.2.2.2.1Geophysical investigations useful in the past include magnetic and gravitysurveys, seismic reflection and seismic refraction surveys, bore-hole geophysics, electrical surveys, andground-penetrating radar surveys.

C.2.2.2.2Core borings to map subsurface geology and obtain samples for testing such asdetermining the properties of the subsurface soils and rocks and geochronological analysis.

C.2.2.2.3Excavating and logging of trenches across geological features to obtain samplesfor the geochronological analysis of those features.

C.2.2.2.4At some sites, deep unconsolidated mate rial/soil, bodies of water, or othermaterial may obscure geologic evidence of past activity along a tectonic structure. In such cases, theanalysis of evidence elsewhere along the structure can be used to evaluate its characteristics in thevicinity of the site.In the CEUS it may not be possible to reasonably demonstrate the age of youngest activity on atectonic structure with adequate deterministic certainty. In such cases, the uncertainty should bequantified; the NRC staff will accept evaluations using the methods described in NUREG/CR-5503(Ref. C.3). A demonstrated tectonic association of such structures with geolog ic structural features ortectonic processes that are geologically old (at least pre-Quaternary) should be acceptable as an ageindicator in the absence of conflicting evidence.

Appendix C to DG-1146, Page C-6C.2.3Site Location Investigations [within 1 km (0.6 mi)]Data from Site Location investigations should be presented at a scale of 1:500 or larger. Important aspects of the site investigations are the excavation and logging of exploratory trenches and themapping of the excavations for the plant structures, particularly those that are characterized as SeismicCategory I. In addition to geological, geophysical, and seismological investigations, detailedgeotechnical engineering investigations as described in Regulatory Guide 1.132 (Ref. C.6), should be conducted at the site.C.2.4Surface-Fault Rupture and Associated Deformation at the SiteAvoid a site that has a potential for fault rupture at or near the ground surface and associateddeformation. Where it is determined that surface deformation need not be taken into account, sufficientdata or detailed studies to reasonably support the determination should be presented. The presence orabsence of Quaternary faulting at the site needs to be evaluated to determine whether there is a potentialhazard that is caused by surface faulting. The potential for surface fault rupture should be characterizedby evaluating (1) the location and geometry of faults relative to the site, (2) the nature and amount ofdisplacement (sense of slip, cumulative slip, slip per event, and nature and extent of related foldingand/or secondary faulting), and (3) the likelihood of displacement during some future period of concern(recurrence interval, slip rate, and elapsed time since the most recent displacement). Acceptable methodsand approaches for conducting these evaluations are described in NUREG/CR-5503 (Ref. C.3);acceptable geochronology dating methods are described in NUREG/CR-5562 (Ref. C.5).For assessing the potential for fault displacement, the details of the spatial pattern of the faultzone (e.g., the complexity of fault traces, branches, and en echelon patterns) may be important as they may define the particular locations where fault displacement may be expected in the future. The amountof slip that might be expected to occur can be evaluated directly based on paleoseismic investigations orit can be estimated indirectly based on the magnitude of the earthquake that the fault can generate.Both non-tectonic and tectonic deformation can pose a substantial hazard to a nuclear powerplant, but there are likely to be differences in the approaches used to resolve the issues raised by the twotypes of phenomena. Therefore, non-tectonic deformation should be distinguished from tectonicdeformation at a site. In past nuclear power plant licensing activities, surface displacements caused byphenomena other than tectonic phenomena have been confused with tectonically induced faulting. Thesestructures, such as those found in karst terrain, and growth faulting, which occurs in the Gulf CoastalPlain or in other deep soil regions, cause extensive subsurface fluid withdrawal.Glacially induced faults generally do not represent a deep-seated seismic or fault displacementhazard because the conditions that created them are no longer present. However, residual stresses fromPleistocene glaciation may still be present in glaciated regions, although they are of less concern thanactive tectonically induced stresses. These features should be investigated with respect to their relationship to current in situ stresses.

The nature of faults related to collapse features can usually be defined through geotechnicalinvestigations and can either be avoided or, if feasible, adequate engineering fixes can be provided.

Appendix C to DG-1146, Page C-7Large, naturally occurring growth faults as those found in the coastal plain of Texas andLouisiana can pose a surface displacement hazard, even though offset most likely occurs at a much less rapid rate than that of tectonic faults. They are not regarded as having the capacity to generate damagingvibratory ground motion, can often be identified and avoided in siting, and their displacements can bemonitored. Some growth faults and antithetic faults related to growth faults and fault zones should beinvestigated in regions where growth faults are known to be present. Local human-induced growthfaulting can be monitored and controlled or avoided.If questionable features cannot be demonstrated to be of non-tectonic origin, they should betreated as tectonic deformation.C.2.5Site Geotechnical Invest igations and Evaluations C.2.5.1 Geotechnical InvestigationsThe geotechnical investigations should include, but not necessarily be limited to, (1) defining sitesoil and near-surface geologic strata properties as may be required for hazard evaluations, engineeringanalyses, and seismic design; (2) evaluating the effects of local soil and site geologic strata on groundmotion at the ground surface; (3) evaluating dynamic properties of the near-surface soils and geologicstrata; (4) conducting soil-structure interaction analyses; and (5) assessing the potential for soil failure ordeformation induced by ground shaking (liquefaction, differential compaction, and land sliding).Subsurface conditions should be investigated by means of borings, soundings, well logs,exploratory excavations, sampling, geophysical methods (e.g., cross-hole, down-hole, and geophysicallogging) that adequately assess soil and ground-water conditions and other methods described in NUREG/CR-5738 (Ref. C.7). Appropriate investigations should be made to determine the contributionof the subsurface soils and rocks to the loads imposed on the structures.A laboratory testing program should be carried out to identify and classify the subsurface soilsand rocks and to determine their physical and engineering properties. Laboratory tests for both static anddynamic properties (e.g., shear modulus, damping, liquefaction resistance, etc.) are generally required. The dynamic property tests should include cyclic triaxial tests, cyclic simple shear tests, cyclic torsionalshear tests, and resonant column tests, as appropriate. Both static and dynamic tests should be conductedas recommended in American Society for Testing and Materials (ASTM) standa rds or test proceduresacceptable to the staff. The ASTM specification numbers for static and dynamic laboratory tests can befound in the annual books of ASTM Standards, Volume 04.08. Sufficient laboratory test data should beobtained to allow for reasonable assessments of mean values of soil properties and their potentialvariability.

C.2.5.2 Ground Failure EvaluationsLiquefaction is a soil behavior phenomenon in which cohesionless soils (sand, silt, or gravel) under saturated conditions lose a substantial part or all of their strength because of high pore waterpressures generated in the soils by strong ground motions induced by earthquakes. Potential effects of liquefaction include reduction in foundation bearing capacity, settlements, land sliding and lateralmovements, flotation of lightweight structures (such as tanks) embedded in the liquefied soil, andincreased lateral pressures on walls retaining liquefied soil. Guidance in Regulatory Guide 1.198,"Procedures and Criteria for Assessing Seismic Soil Liquefaction at Nuclear Power Plant Sites"(Ref. C.8), should be used for evaluating the site for liquefaction potential.

Appendix C to DG-1146, Page C-8Investigations of liquefaction potential typically involve both geological and geotechnicalengineering assessments. The parameters controlling liquefaction phenomena are (1) the lithology of thesoil at the site, (2) the ground water conditions, (3) the behavior of the soil under dynamic loadings, and(4) the potential severity of the vibratory ground motion.C.3Evaluation of New Information Obtained from the Site-specific InvestigationsThe first step inreviewing the new information obtained from the site-specific investigationswithprevious interpretations is determining whether thefollowing existing parameters are consistent withthe new information: (1) the range of seismogenic sources as interpreted by the seismicity experts or teams involved in the study, (2) the range of seismicity rates for the region around the site as interpretedby the seismicity experts or teams involved in the studies, (3) the range of maximum magnitudes determined by the seismicity experts or teams, and (4) attenuation relations. The new information is considered not significant and no further evaluation is needed if it is consistent with the assumptionsused in the PSHA, no additional alternative seismic sources or seismic parameters are needed, or it supports maintaining the site mean seismic hazard.

1Copies are available for inspection or copying for a fee from the NRC's Public Document Room at 11555 RockvillePike, Rockville, MD; the PDR's mailing address is USNRC PDR, Washington, DC 20555; telephone (301) 415-4737or (800) 397-4209; fax (301) 415-3548; email PDR@nrc.gov. In addition, copies are available at current ratesfrom the U.S. Government Printing Office, P.O. Box 37082, Washington, DC 20402-9328, telephone (202) 512-1800; or from the National Technical Information Service (NTIS), 5285 Port Royal Road, Springfield, VA 22161, http://www.ntis.gov, telephone (703) 487-4650.

2All regulatory guides listed herein were published by the U.S. Nuclear Regulatory Commission. Where an ADAMSaccession number is identified, the specified regulatory guide is available electronically through the NRC's Agencywide Documents Access and Management System (ADAMS) at http://www.nrc.gov/reading-rm/adams.html. All other regulatory guides are available electronically through the Public Electronic Reading Room on the NRC'spublic Web site, at http://www.nrc.gov/reading-rm/doc-collections/reg-guides/. Single copies of regulatory guides mayalso be obtained free of charge by writing the Reproduction and Distribution Services Section, ADM, USNRC,Washington, DC 20555-0001, or by fax to (301) 415-2289, or by email to DISTRIBUTION@nrc.gov. Active guidesmay also be purchased from the National Technical Information Service (NTIS) on a standing order basis. Details on this service may be obtained by contacting NTIS at 5285 Port Royal Road, Springfield, Virginia 22161, online at http://www.ntis.gov, or by telephone at (703) 487-4650. Copies are also available for inspection or copying for a feefrom the NRC's Public Document Room (PDR), which is located at 11555 Rockville Pike, Rockville, Maryland; the PDR's mailing address is USNRC PDR, Washington, DC 20555-0001. The PDR can also be reached by tele phoneat (301) 415-4737 or (800) 397-4205, by fax at (301) 415-3548, and by email to PDR@nrc.gov

.Appendix C to DG-1146, Page C-9APPENDIX C REFERENCESC.1Electric Power Research Institute, "The Earthquakes of Stable Continental Regions," Volume 1: Assessment of Large Earthquake Potential, EPRI TR-102261-V1, 1994.C.2Budnitz, R.J., et al., "Recommendations for Probabilistic Seismic Hazard Analysis: Guidance onUncertainty and Use of Experts," NUREG/CR-6372, Volumes 1 and 2, U.S. Nuclear RegulatoryCommission, Washington, DC, April 1997.

1C.3Hanson, K.L., et al., "Techniques for Identifying Faults and Determining Their Origins,"NUREG/CR-5503, U.S. Nuclear Regulatory Commission, Washington, DC, 1999.

1C.4International Atomic Energy Agency (IAEA), "Earthquakes and Associated Topics in Relation toNuclear Power Plant Siting," Safety Series No. 50-SG-S1, Revision 1, 1991. C.5Sowers, J.M., et al., "Dating and Earthquakes: Review of Quaternary Geochronology and ItsApplication to Paleoseismology," NUREG/CR-5562, U.S. Nuclear Regulatory Commission,Washington, DC, 1998.

1 C.6Regulatory Guide 1.132, "Site Investigations for Foundations of Nuclear Power Plants,"Revision 1, U.S. Nuclear Regulatory Commission, Washington, DC, October 2003, available inADAMS under Accession #ML032800710.

2C.7Torres, N., et al.,"Field Investigations for Foundations of Nuclear Power Facilities,"NUREG/CR-5738, U.S. Nuclear Regulatory Commission, Washington, DC, 1999.

1 C.8Regulatory Guide 1.198 , "Procedures and Criteria for Assessing Seismic Soil Liquefaction at Nuclear Power Plant Sites," U.S. Nuclear Regulatory Commission, Washington, DC,November 2003, available in ADAMS under Accession #ML010650295.

2 Appendix D to DG-1146, Page D-1APPENDIX D DETERMINATION OF CONT ROLLING EARTHQUAKESAND DEVELOPMENT OF SEISMIC HAZARD INFORMATION BASED.1IntroductionThis appendix elaborates on the steps described in Regulatory Position 3 of this regulatory guidefor determining the controlling earthquakes used to develop an appropriate ground motion for the siteresponse analysis. The controlling earthquakes are developed from the deaggregation of the probabilisticseismic hazard analysis (PSHA) results at ground motion levels corresponding to the annual frequenciesof 1 E-04, 1 E-05, and 1 E-06. The deaggregation of the PSHA defines the contribution of individualmagnitude and distance ranges to the overall seismic hazard and results in the magnitude and distancevalues of the controlling earthquakes at the average of 1 and 2.5 Hz and the average of 5 and 10 Hz.Using the controlling earthquakes, Regulatory Position 4 and Appendix E to this regulatory guidedescribe a procedure for determining the appropriate ground motion for the site response analysis to determine the performance-based safe shutdown earthquake ground motion (SSE).D.2Procedure To Determine Controlling EarthquakesThe following approach is acceptable to the Nuclear Regulatory Commission (NRC) staff fordetermining the controlling earthquakes. This procedure is based on a deaggregation of the probabilistic seismic hazard in terms of earthquake magnitudes and distances. When the controlling earthquakes havebeen obtained, the site specific earthquake ground motion can be developed according to the proceduresdescribed in Appendix E. The SSE response spectrum can then be determined according to the procedures described in Regulatory Position 5.Step 1Perform a site-specific PSHA. The PSHA methodology attributes and results described inRegulatory Positions 3.3 and 3.4 of this regulatory guide are summarized as follows:Perform the PSHA for actual or assumed rock conditions with up-to-date interpretations ofearthquake sources, earthquake recurrence, and strong ground motion estimation using originalor updated sources as determined in Regulatory Positions 3.1 and 3.2.CAV filtering can be used in place of a lower-bound magnitude cutoff (Ref. D.1).

Conduct the hazard assessment at a minimum of 30 frequencies, approximately equally spacedon a logarithmic frequency axis between 100 and 0.1Hz.Report fractile hazard curves at the following fractile levels (p) for each ground motionparameter: 0.05, 0.16, 0.50, 0.84, and 0.95, as well as the mean. Report the fractile hazard curves in tabular as well as graphical format.Determine the mean uniform hazard response spectra (UHRS) for annual exceedance frequenciesof 1 E-04, 1 E-05, and 1 E-06.Steps 2 through 7 are repeated at annual probability levels corresponding to 1 E-04, 1 E-05, and1 E-06. The resultant hazard curves for the annual probability level of 1 E-05 are shown in Figure D.1.

1The recommended magnitude and distance bins and procedure used to establish controlling earthquakes weredeveloped for application in the CEUS, where the nearby earthquakes generally control the response in the 5 to 10 Hz frequency range, and larger but distant events can control the lower frequency range. For other situations, alternative binning schemes as well as a study of contributions from various bins will be necessary to identify controllingearthquakes, consistent with the distribution of the seismicity.Appendix D to DG-1146, Page D-2Step 2 (a)Using the annual probability levels of 1 E-04, 1 E-05, and 1 E-06, determine the ground motionlevels for the spectral accelerations at 1, 2.5, 5, and 10 Hz from the total mean hazard obtained in

Step 1.(b)Calculate the average of the ground motion level for the 1 and 2.5 Hz and the 5 and 10 Hz spectral acceleration pairs.Step 3Perform a deaggregation of the PSHA for each of the spectral accelerations and annualprobability levels as enumerated in Step 2, and provide a table (such as illustrated in Table D.1

1) for eachof the 12 cases. The deaggregation is performed to obtain the relative contribution to the hazard for eachof the magnitude-distance bins for each case.Figure D.1. Total Mean Hazard Curves Appendix D to DG-1146, Page D-3Table D.1Recommended Magnitude and Distance BinsMoment Magnitude Range of BinsDistanceRange of Bin (km)5 - 5.55.5 6.56.5 - 7>7 0 - 15 15 - 25 25 - 50 50 - 100 100 - 200 200 - 300>300Step 4If the deaggregated results for each of the 12 cases is not in terms of the fractional contribution tothe total hazard then perform Step 4 as described below. Otherwise, average the low-frequency (1 and2.5 Hz) and high-frequency (5 and 10 Hz) deaggregation results.From the deaggregated results of Step 3, the mean annual probability of exceeding the groundmotion levels of Step 2(a) (spectral accelerations at 1, 2.5, 5, and 10 Hz) are determined for eachmagnitude-distance bin. These values are denoted by H mdf.Using Hmdf values, the fractional contribution of each magnitude and distance bin to the totalhazard for the average of 1 and 2.5 Hz, P(m,d) 1, is computed according to the following:P(m,d)1 = [( Hmdf ) / 2 ] ÷ [ ( Hmdf ) / 2 ](Equation 1) f = 1,2 m d f = 1,2where f = 1 and f = 2 represent the ground motion measure at 1 and 2.5 Hz, respectively.The fractional contribution of each magnitude and distance bin to the total hazard for the averageof 5 and 10 Hz, P(m,d) 2, is computed according to the following:P(m,d)2 = [( Hmdf ) / 2 ] ÷ [ ( Hmdf ) / 2 ](Equation 2) f = 1,2

m d f = 1,2where f = 1 and f = 2 represent the ground motion measure at 5 and 10 Hz, respectively.

Appendix D to DG-1146, Page D-4Step 5Review the magnitude-distance distribution for the average of 1 and 2.5 Hz to determine whetherthe contribution to the hazard for distances of 100 kilometer (km) (63 mi) or greater is substantial (on theorder of 5 percent or greater).If the contribution to the hazard for distances of 100 km (63 mi) or greater exceeds 5 percent,additional calculations are needed to determine the controlling earthquakes using the magnitude-distancedistribution for distances greater than 100 km (63 mi). This distribution, P > 100 (m,d) 1 , is defined bythe following:P > 100 (m,d) 1 = [ P(m,d) 1 ] ÷ [ P(m,d) 1 ](Equation 3)

m d > 100The purpose of this calculation is to identify a distant, larger event that may controllow-frequency content of a response spectrum.The distance of 100 km (63 mi) is chosen for CEUS sites. However, for all sites the results offull magnitude-distance distribution should be carefully examined to ensure that proper controllingearthquakes are clearly identified.Step 6Calculate the mean magnitude and distance of the controlling earthquake associated with theground motions determined in Step 2 for the average of 5 and 10 Hz. The following relation is used tocalculate the mean magnitude using results of the entire magnitude-distance bins matrix:

M c (5 - 10 Hz) = m P(m,d)2 (Equation 4) m dwhere m is the central magnitude value for each magnitude bin.The mean distance of the controlling earthquake is determined using results of the entiremagnitude-distance bins matrix:

Ln {D c (5 - 10 Hz)} = Ln(d) P(m,d)2 (Equation 5)

d mwhere d is the centroid distance value for each distance bin.Step 7If the contribution to the hazard calculated in Step 5 for distances of 100 km (63 mi) or greaterexceeds 5 percent for the average of 1 and 2.5 Hz, calculate the mean magnitude and distance of thecontrolling earthquakes associated with the ground motions determined in Step 2 for the average of 1 and2.5 Hz. The following relation is used to calculate the mean magnitude using calculations based onmagnitude-distance bins greater than distances of 100 km (63 mi) as discussed in Step 5:

M c (1 - 2.5 Hz) = m P > 100 (m,d) 1 (Equation 6) m d > 100where m is the central magnitude value for each magnitude bin.

Appendix D to DG-1146, Page D-5The mean distance of the controlling earthquake is based on magnitude-distance bins greater thandistances of 100 km (63 mi) as discussed in Step 5 and determined according to the following:

Ln {D c (1 - 2.5 Hz)} = Ln(d) P > 100 (m,d) 2 (Equation 7) d > 100 mwhere d is the centroid distance value for each distance bin.When more than one earthquake magnitude-distance pair contributes significantly to the spectralaccelerations at a given frequency, it may be necessary to use more than one controlling earthquake for determining the spectral response at the frequency.Step 8Document the high- and low-frequency controlling earthquakes for each target annual probabilityof exceedance in the tabular form illustrated in Table D.2.Table D.2High- and Low-Frequency Controlling EarthquakesHazardMagnitude (m b)DistanceMean 1 E-04High-Frequency (5 and 10 Hz)Mean 1 E-04Low-Frequency (1 and 2.5 Hz)Mean 1 E-05High-Frequency (5 and 10 Hz)Mean 1 E-05Low-Frequency (1 and 2.5 Hz)Mean 1 E-06High-Frequency (5 and 10 Hz)Mean 1 E-06Low-Frequency (1 and 2.5 Hz)Step 9Develop response spectra using the controlling earthquake (high- and low-frequency) magnitudesand distances from Table D.2, and appropriate ground motion models. Scale these spectra to match thesite rock spectral accelerations at 5 and 10 Hz (high-frequency) and 1 and 2.5 Hz (low-frequency).

Appendix D to DG-1146, Page D-6D.3Example for a Central and Eastern United States SiteTo illustrate the procedure in Section D.2, calculations are shown here for a CEUS site. In thisexample, only the annual probability level of 1 E-05 is used.It must be emphasized that the recommended magn itude and distance bins and procedure used toestablish controlling earthquakes were developed for application in the CEUS, where the nearbyearthquakes generally control the response in the 5 to 10 Hz frequency range, and larger but distantevents can control the lower frequency range. For other situations, alternative binning schemes as well as a study of contributions from various bins will be necessary to identify controlling earthquakesconsistent with the distribution of the seismicity.Step 1The PSHA used methods described in the 1993 LLNL seismic hazard methodology (Refs. D.2and D.3). For this example, the databases and seismic sources identified in the LLNL or EPRImethodologies for CEUS sites (Refs. D.2-D.6) were evaluated to ensure that they contain up-to-dateinterpretations of earthquake sources, earthquake recurrence, and strong ground motion estimation. Theanalyses were performed for spectral acceleration at 1, 2.5, 5, and 10 Hz. The resultant hazard curves areplotted in Figure D.1.Step 2The hazard curves at 1, 2.5, 5, and 10 Hz obtained in Step 1 are assessed at the annual probabilitylevels of 1 E-04, 1 E-05, and 1 E-06. As an example, the corresponding ground motion level values forthe annual probability level of 1 E-05 are given in Table D.3. See Figure D.1.The average of the ground motion levels at the 1 and 2.5 Hz, Sa1-2.5, and 5 and 10 Hz, Sa5-10 , aregiven in Table D.4.Table D.3Ground Motion LevelsFrequency (Hz)12.5510Spectral Acc. (cm/s/s)88258351551Table D.4Average Ground Motion Values Sa 1-2.5 (cm/s/s)173 Sa 5-10 (cm/s/s)451 Appendix D to DG-1146, Page D-7Step 3The mean seismic hazard is deaggregated for the matrix of magnitude and distance bins as givenin Table D.1.The hazard values corresponding to the ground motion levels found in Step 2, and listed inTable D.3, are then determined from the hazard curv e for each bin for spectral accelerations at 1, 2.5, 5,and 10 Hz. This process is illustrated in Figure D.2. The vertical line corresponds to the value 88 centimeter/second/second (cm/s/s) listed in Table D.3 for the 1 Hz hazard curve and intersects the hazardcurve for the 25 - 50 bin, 6 - 6.5 bin, at a hazard value (probability of exceedance) of 2.14E-08/yr. Tables D.5 to D.8 list the appropriate hazard value for each bin for 1, 2.5, 5, and 10 Hz, respectively.It should be noted that if the mean hazard in each of the 35 bins is added up, it should equal1 E-05.Figure D.2. 1 Hz Mean Hazard Curve forDistance Bin 25 - 50 km & Magnitude Bin 6 - 6.5 Appendix D to DG-1146, Page D-8Table D.5Mean Exceeding Probability Values for Spectral Accelerationsat 1 Hz (88 cm/s/s)Distance Rangeof Bin (km)Magnitude Range of Bin5 - 5.55.5 6.56.5 - 7>7 0 - 151.90E-079.04E-071.09E-0700 15 - 253.86E-082.47E-072.30E-0800 25 - 501.65E-082.90E-072.05E-0700 50 - 1002.25E-091.47E-077.14E-072.40E-070 100 - 2009.58E-112.26E-088.17E-075.84E-060 200 - 30001.82E-101.53E-081.76E-070> 300008.61E-119.87E-111.62E-09Table D6Mean Exceeding Probability Values for Spectral Accelerationsat 2.5 Hz (258 cm/s/s)Distance Rangeof Bin (km)Magnitude Range of Bin5 - 5.55.5 6.56.5 - 7>7 0 - 151.18E-061.76E-062.18E-0700 15 - 252.85E-076.35E-075.71E-0800 25 - 501.38E-078.89E-073.38E-0700 50 - 1002.07E-083.32E-077.72E-072.16E-070 100 - 2007.94E-104.13E-085.66E-072.51E-060 200 - 3003.79E-131.10E-103.95E-092.66E-080> 30008.04E-142.61E-124.79E-141.25E-14 Appendix D to DG-1146, Page D-9Table D.7Mean Exceeding Probability Values for Spectral Accelerationsat 5 Hz (351 cm/s/s)Distance Rangeof Bin (km)Magnitude Range of Bin5 - 5.55.5 6.56.5 - 7>7 0 - 152.38E-062.81E-062.48E-0700 15 - 254.51E-079.70E-076.53E-0800 25 - 501.32E-078.83E-073.63E-0700 50 - 1005.90E-081.60E-074.79E-071.37E-070 100 - 2003.87E-115.47E-091.22E-077.44E-070 200 - 30001.15E-121.31E-101.93E-090> 300 00000Table D.8Mean Exceeding Probability Values for Spectral Accelerationsat 10 Hz (551 cm/s/s)Distance Rangeof Bin (km)Magnitude Range of Bin5 - 5.55.5 6.56.5 - 7>7 0 - 153.19E-063.21E-062.38E-0700 15 - 255.94E-071.08E-068.95E-0800 25 - 501.18E-076.74E-072.96E-0700 50 - 1001.70E-096.60E-081.98E-077.78E-080 100 - 2003.62E-124.85E-101.91E-081.56E-070 200 - 30001.12E-141.77E-126.72E-110> 300 00000Step 4Using deaggregated mean hazard results, the fractional contribution of each magnitude-distancepair to the total hazard is determined.Tables D.9 and D.10 show P (m,d) 1 and P (m,d) 2 for the average of 1 and 2.5 Hz and 5 and10 Hz, respectively, using Equations 1 and 2.

Appendix D to DG-1146, Page D-10Step 5 Because the contribution of the distance bins greater than 100 km in Table D.9 contains morethan 5 percent of the total hazard for the average of 1 and 2.5 Hz, the controlling earthquake for thespectral average of 1 and 2.5 Hz is calculated using magnitude-distance bins for distance greater than100 km. Table D.11 shows P > 100 (m,d) 1 for the average of 1 to 2.5 Hz, using Equation 3.Table D.9P(m,d)1 for Average Spectral Accelerations 1 and 2.5 HzCorresponding to the 1 E-05 Annual ProbabilityDistance Rangeof Bin (km)Magnitude Range of Bin5 - 5.55.5 6.56.5 - 7>7 0 - 150.0690.1330.0160.0000.000 15 - 250.0160.0440.0040.0000.000 25 - 500.0080.0590.0270.0000.000 50 - 1000.0010.0240.0740.0230.000 100 - 2000.0000.0030.0690.4180.000 200 - 3000.0000.0000.0010.0100.000> 3000.0000.0000.0000.0000.000 Table D.10P(m,d)2 for Average Spectral Accelerations 5 and 10 HzCorresponding to the 1 E-05 Annual ProbabilityDistance Rangeof Bin (km)Magnitude Range of Bin5 - 5.55.5 6.56.5 - 7>7 0 - 150.2780.3010.024 0.0000.000 15 - 250.0520.1020.0080.0000.000 25 - 500.0130.0780.0330.0000.000 50 - 1000.0030.0110.0340.0110.000 100 - 2000.0000.0000.0070.0450.000 200 - 3000.0000.0000.0000.0000.000> 3000.0000.0000.0000.0000.000 Appendix D to DG-1146, Page D-11 Table D.11P>100 (m,d) 1 for Average Spectral Accelerations 1 and 2.5 HzCorresponding to the 1 E-05 Annual ProbabilityDistance Rangeof Bin (km)Magnitude Range of Bin5 - 5.55.5 6.56.5 - 7>7 100 - 2000.0000.0060.1380.8330.000 200 - 3000.0000.0000.0020.0200.000> 3000.0000.0000.0000.0000.000Figures D.3 to D.5 show the above information in terms of the relative percentage contribution.Steps 6 and 7To compute the controlling magnitudes and distances at 1 to 2.5 Hz and 5 to 10 Hz for theexample site, the values of P > 100 (m,d) 1 and P (m,d) 2 are used with m and d values corresponding to themid-point of the magnitude of the bin (5.25, 5.75, 6.25, 6.75, 7.3) and centroid of the ring area (10, 20.4,38.9, 77.8, 155.6, 253.3, and somewhat arbitrarily 350 km). Note that the mid-point of the last magnitudebin may change because this value is dependent on the maximum magnitudes used in the hazard analysis.Step 8The high- and low-frequency controlling earthquakes for all frequency levels are shown inTable D.12. Provide a description of the source of each high- and low-frequency controlling earthquake.

Appendix D to DG-1146, Page D-12Figure D.3. Full Distribution for Average of 5 and 10 Hz 0-1515-25 25-50 50-100 100-200 200-300>3005-5.55.5-66-6.56.5-7>70.000.050.100.150.200.250.300.350.40 M a g n i t u d e b i n s D i s t a n c e b i n sContribution to HazardFull Distribution for Average of 5 and 10 Hz Appendix D to DG-1146, Page D-13Figure D.4. Full Distribution for Average of 1 and 2.5 Hz0-1515-2525-5050-100100-200200-300>3005-5.55.5-66-6.56.5-7>70.000.050.100.150.200.250.300.350.400.45 M a g n i t u d e b i n s D i s t a n c e b i n s Contribution to HazardFull Distribution for Average of 1 and 2.5 Hz Appendix D to DG-1146, Page D-14Figure D.5. Renormalized Hazard Distribution for Distances > 100 km for Average of 1 and 2.5 Hz5-5.55.5-66-6.56.5-7>7100-200200-300>3000.00.10.20.30.40.50.60.70.80.9 D i s t a n c e b i n s M a g n i t u d e b i n sContribution to HazardRe-Normalized Hazard Distribution for Distances > 100 km for Average of 1 and 2.5 Hz Appendix D to DG-1146, Page D-15 Table D.12 Magnitude and Distancesof High- and Low-Frequency Controlling Earthquakesfrom the LLNL Probabilistic AnalysisHazardMagnitude (m b)DistanceMean 1 E-04High-Frequency (5 and 10 Hz)From Steps 2 - 7Mean 1 E-04Low-Frequency (1 and 2.5 Hz)Mean 1 E-05High-Frequency (5 and 10 Hz)5.717 km (10 mi)Mean 1 E-05Low-Frequency (1 and 2.5 Hz)6.7157 km (97 mi)Mean 1 E-06High-Frequency (5 and 10 Hz)From steps 2 - 7Mean 1 E-06Low-Frequency (1 and 2.5 Hz)Step 9The response spectra used for the site response analysis is obtained using the controllingearthquake (high- and low-frequency) magnitudes and distances in Table D.12, using appropriate groundmotion models (e.g., the EPRI ground motions models in Reference D.7 for the CEUS), and then scalingthese spectra to match the site rock spectral accelerations at 5 and 10 Hz (high-frequency) and 1 and2.5 Hz (low-frequency).D.4Sites Not in the Central and Eastern United StatesThe determination of the controlling earthquakes and the seismic hazard information base forsites not in the CEUS is also carried out using the procedure described in Sec tion D.2 of this appendix. However, because of differences in seismicity rates and ground motion attenuation at these sites, alternative magnitude-distance bins may have to be used.

2Copies are available for inspection or copying for a fee from the NRC's Public Document Room at 11555 RockvillePike, Rockville, MD; the PDR's mailing address is USNRC PDR, Washington, DC 20555; telephone (301) 415-4737or (800) 397-4209; fax (301) 415-3548; email PDR@nrc.gov. In addition, copies are available at current ratesfrom the U.S. Government Printing Office, P.O. Box 37082, Washington, DC 20402-9328, telephone (202) 512-1800; or from the National Technical Information Service (NTIS), 5285 Port Royal Road, Springfield, VA 22161, http://www.ntis.gov, telephone (703) 487-4650.Appendix D to DG-1146, Page D-16APPENDIX D REFERENCESD.1Electric Power Research Institute (EPRI), and U.S. Department of Energy (DOE), "Program onTechnology Innovation: Use of Minimum CAV in Determining Effects of Small MagnitudeEarthquakes on Seismic Hazard Analyses," Report 1012965, December 2005.D.2 Sobel.P., "Revised Livermore Seismic Hazard Estimates for Sixty-Nine Nuclear Power PlantSites East of the Rocky Mountains," NUREG-1488, U.S. Nuclear Regulatory Commission,Washington, DC, April 1994.

2D.3 Savy, J.B., et al., "Eastern Seismic Hazard Characterization Update," UCRL-ID-115111,Lawrence Livermore National Laboratory, June 1993, available in the NRC's Public DocumentRoom under Accession #9310190318) 2D.4Bernreuter, D.L., et al., "Seismic Hazard Characterization of 69 Nuclear Plant Sites East of theRocky Mountains," NUREG/CR-5250, Volumes 1-8, U.S. Nuclear Regulatory Commission,Washington, DC, January 1989.

2D.5 Electric Power Research Institute (EPRI), "Probabilistic Seismic Hazard Evaluations at Nuclear Power Plant Sites in the Central and Eastern United States," NP-4726, All Volumes, 1989-1991.D.6Electric Power Research Institute (EPRI), "The Earthquakes of Stable Continental Regions,"Volume 1: Assessment of Large Earthquake Potential, EPRI TR-102261-V1, 1994.D.7Electric Power Research Institute (EPRI), et al., "CEUS Ground Motion Project Final Report, Report 1009684, 2004.

Appendix E to DG-1146, Page E-1APPENDIX ESEISMIC WAVE TRAN SMISSION ANALYSISE.1IntroductionThis appendix elaborates on the procedure described in Regulatory Position 4 of this regulatoryguide to develop the uniform hazard response spectrum (UHRS) at the free-field ground surface. Thehazard curves from the probabilistic seismic hazard analysis (PSHA) are defined for generic hard rockconditions usually with a shear wave velocity (V s) about 2.8 km/sec (9200 ft/sec). To properly addressamplification or deamplification effects of the soils between the generic hard rock and the ground surfaceand other interfaces in between, the following procedure should be used:(1)Develop the site-specific soil profile.

(2)Develop appropriate modified earthquake time histories to be used in site response analyses.(3)Perform a suite of site response analyses to determine mean site amplification functions for arange of frequencies.(4)Develop the UHRS at the free ground surface based on the generic rock-based PSHA and themean site amplification functions.The horizontal and vertical safe shutdown earthquake ground motion (SSE) should bedetermined in the free field on the ground surface. For sites with V s less than the assumed hard-rockshear wave velocity, as defined in the PSHA, a site response analysis should be performed. The inputground motion used in the analyses are specified based on the PSHA, which assumes that an outcrop or ahypothetical outcrop exists at the free-field surface. If the rock layer (as defined by shear wave velocity) first occurs at depth, the input ground motions are deconvolved to develop "within" motions and imposedat the highest layer of the generic hard rock [greater than 2.8 km/sec (9200 ft/sec)].E.2Site-Specific Soil ProfileSite and laboratory investigations and testing are performed to obtain data defining the static anddynamic engineering properties of soil and rock materials, and their spatial distribution. The proceduresidentified in Regulatory Guide 1.132 (Ref. E.1), Regulatory Guide 1.138 (Ref. E.2), and subsectionC.2.2.2 of Appendix C to this regulatory guide are acceptable to the NRC staff.To be acceptable, the seismic wave transmission characteristics (spectral amplification ordeamplification) of the materials overlying bedrock at the site are described over a range of frequenciesthat include the significant structural frequencies. The wave transmission characteristics should bedeveloped for levels of seismic motion ranging from those causing very small strains (i.e., no soil degradation) to those consistent with strain levels anticipated in the SSE. The following materialproperties should be determined for each stratum under the site: (1) thickness, (2) compressionalvelocity, (3) shear wave velocity, (4) bulk densities, (5) soil index properties and classification, (6) shearmodulus and damping variations with strain level, and (7) the water table elevation and its variationthroughout the site. Site specific soil profiles should be deep enough to fully capture the amplification tothe lowest structural frequency of interest and should extend to the depth at which the input groundmotion is to be imposed [typically the depth at which V s = 2.8 km/sec (9200 ft/sec)].

Appendix E to DG-1146, Page E-2Often vertically propagating shear waves are the dominant contributor to free-field groundmotions at a site. In these cases, a one-dimensional equivalent-linear analysis or nonlinear analysis thatassumes vertical propagation of shear waves may be appropriate. However, site characteristics (such as adipping bedrock surface or other impedance boundaries), regional characteristics (such as topographic effects), and source characteristics (such as nearby dipping seismic sources) may require that analyses beable to also account for inclined waves.E.3Site Response AnalysisTo estimate site response, input ground motions are needed for the analyses performed. Thechoice of input ground motions have significant impact on the amplification of motion observed in thesoil column. A range of key earthquake scenarios, known as the controlling earthquakes, are developedto focus the analyses on the types of scenarios most likely to impact the site.To develop these controlling earthquakes and to better understand how different sources within aseismic source model contribute to the overall seismic hazard, the seismic hazard developed by the PSHA is deaggregated into a series of magnitude-distance bins for a variety of spectral frequencies andannual exceedance frequencies. Based on the results of the deaggregation, the earthquake scenarios that most contribute to the high-frequency (5 and 10Hz) and low-frequency (1 and 2.5 Hz) spectral motionswith annual exceedance frequencies of 1 E-04, 1 E-05 and 1 E-06 are determined. The distance-magnitude bins from the 5 and 10 Hz frequencies are averaged, and the 1 and 2.5 Hz frequencies areaveraged, and the controlling earthquakes (dominant distance and magnitude bin) is determined for thehigh- and low-frequency motions for each of the 3 annual frequencies.The response spectra for each of the six individual controlling earthquakes determined above aredeveloped using appropriate attenuation relationships. The resulting response spectra are then scaled tomatch the site rock spectral accelerations at either 5 and 10 Hz (high-frequency) or at 1 and 2.5 Hz (low-frequency).Once the spectra for each controlling earthquake are determined and scaled, the spectra areplotted and compared against the natural frequency of the soil to assure that the spectra from at least onekey earthquake scenario have sufficient energy content at the natural frequency of the site as determinedby the full rock-based UHRS. A visual comparison of the spectra from the controlling earthquakes, theUHRS spectra, and the natural frequency of the soil column should be provided. For sites with numerouscontributing earthquakes, or if the natural frequency of the site is not adequately covered by the use ofsix controlling earthquakes, more complex methodologies should be applied.After determining the scaled response spectra from each of the characteristic earthquakes,corresponding acceleration time histories are identified and their spectra compared to the targetcharacteristic earthquakes spectra. The acceleration time histories are modified to match the targetspectra. The scaled time histories are imposed in the developed subsurface model (as equivalent"within" motions) at the bedrock level. The imposed motions propagate through the analytical model ofthe site soils to determine the free-field surface ground motions (Regulatory Position 4.2). Onlyacceleration time histories modified from recorded earthquakes are used for this purpose.The required soil parameters for the site response analyses include the depth, soil type, density,shear wave velocity, shear modulus and damping, and their variations with strain levels for each of thesoil layers. Internal friction angle, cohesive strength, and over-consolidation ratio for clay are alsoneeded for non-linear analyses. The strain-dependent shear modulus and damping curves are developedbased on site-specific testing results and supplemented as appropriate by published data for similar soils.

Appendix E to DG-1146, Page E-3The strain-dependent shear modulus and damping curves incorporated into the analysis should bejustified against observed data for each soil at the site. When site-specific laboratory data is used, theresult should be compared to earthquake recordings on similar soils. A maximum critical damping valueof 15 percent is allowed in site response analyses. The effects of confining pressures (that reflect thedepths of the soil) on these strain-dependent soil dynamic characteristics are assessed and considered insite response analyses. The variability in these properties is accounted for in the site response analyses. Steeply dipping soil layers or other contrasting impedance interfaces, such as fault zones or valley walls, should also be considered because the existence of those interfaces will influence the wave propagationfrom the reference hard rock to the ground surface. Multi dimensional soil models may be needed ifcomplex geologic and geotechnical conditions exist.As a minimum, the results of the site response analysis should show the input motion (rockresponse spectra), output motion (surface response spectra), and spectral amplification function (siteground motion transfer function) at the surface. It is common practice to also provide a plot showing thepeak accelerations in the soil as a function of depth. In addition to developing the spectral amplificationsfunctions at the surface, it is also useful to simultaneously develop the spectral amplification functionsfor the foundation level for use in development of the SSE. However, because the location of thefoundation is not always known at the time at which the analyses are undertaken, the functions may bedetermined at several depths including the deepest depth from the Plant Parameter Envelope. In addition,determining and providing the spectral amplification functions at a variety of depths may support reviewof the results used in determining the soil UHRS and the SSE.To determine the UHRS at the free-field ground surface, the site amplification functions (spectralratios) for each characteristic earthquake scenario are computed. To capture the variation of the soilproperties at the site, it usually requires that 60 randomized shear velocity profiles are paired with 60 setsof randomized shear modulus and damping curves (i.e., one shear velocity profile with one set of modulus reduction and damping curves). The mean site amplification function is obtained for eachcharacteristic earthquake scenario by dividing the response spectrum from the computed surface motionby the response spectrum from the input hard-rock surface motion, and computing the arithmetic mean ofthese 60 individual response spectral ratios.Figure E.1 compares the 1 E-05 rock UHRS, the target spectra for the high- and low-frequencycontrolling earthquakes, and the natural frequency of the soil column at the site.Figure E.2 shows the computed high- and low-frequency average site amplification functions forthe mean 1 E-05 hazard level characteristic earthquake scenarios. As shown in Figure E.2, the soilcolumn amplifies the input hard rock motion over the fairly wide frequency range. The lines shown inFigure E.2 are the final average site amplification function for each controlling earthquake based on asuite of site response analyses that take into account uncertainty in the soil model and use a variety ofappropriate time histories.Once the soil amplification functions are developed, they are applied to the free-field rock UHRSto develop two free-field soil spectra. To determine the soil UHRS at the free-field ground surface, foreach of the annual exceedance frequencies (1 E-04, 1 E-05 and 1 E-06), multiply the rock-based UHRS at all 25 points and the natural frequency of the site soil column by the site amplification functions, andenvelop the results. These two curves are enveloped to determine the final free-field soil UHRS. If thetwo controlling earthquake response spectral shapes cover a broad range of frequencies such that when scaled and enveloped they approximate the UHRS, then it is also acceptable to multiply the high- andlow-frequency controlling earthquake spectra by the appropriate site amplification function and envelopethe results. Figure E.3 shows a comparison of the results when each of the soil amplification functionsare applied to the free-field rock UHRS and the final free-field soil UHRS.

Appendix E to DG-1146, Page E-4 0 1 20.1110100Frequency (Hz)Spectral Acceleration (g)Mean Rock UHRSLow-Frequency Controlling Earthquake Target SpectraHigh-Frequency Controlling Earthquake Target SpectraNatural Frequency of the SoilFigure E.1. Comparison of the 1 E-05 Rock UHRS, the Target Spectrafor the High- and Low-Frequency Controlling Earthquakes, and the Natural Frequency of the Soil Column Appendix E to DG-1146, Page E-50.1 1 100.1110100Frequency (Hz)

Response Spectral RatioLow-Frequency Site Amplification FunctionHigh-Frequency Site Amplification FunctionFigure E.2. Mean Site Amplification Functions for Low- and High-Frequency Characteristic Earthquakes for 1 E-05 Annual Probability Level Appendix E to DG-1146, Page E-6 0 1 2

30.1110100Frequency (Hz)Spectral Acceleration (g)1 E-5 Mean Soil UHRSMean Rock UHRS Scaled By Low-Frequency Amplification FunctionMean Rock UHRS Scaled By High-Frequency Amplification FunctionFigure E.3. Free-Surface Soil Uniform Hazard Response Spectra Compared to Spectra Resulting from Scaling of Mean Rock UHRSby Low- and High-Frequency Amplification FunctionsE.4Free-Field Ground Surface Uniform Hazard Re sponse SpectraIn accordance with Regulatory Position 4.3 of this regulatory guide, the UHRS at the free-fieldground surface is determined for the site by first scaling the appropriate UHRS at the rock surface by themean site amplification functions developed for both the low- and high-frequency ranges and thenenveloping both sets of results by a smooth curve. These smooth UHRS are used in Regulatory Position 5of this regulatory guide to develop the performance-based horizontal and vertical response spectra.

1All regulatory guides listed herein were published by the U.S. Nuclear Regulatory Commission. Where an ADAMSaccession number is identified, the specified regulatory guide is available electronically through the NRC's Agencywide Documents Access and Management System (ADAMS) at http://www.nrc.gov/reading-rm/adams.html. All other regulatory guides are available electronically through the Public Electronic Reading Room on the NRC'spublic Web site, at http://www.nrc.gov/reading-rm/doc-collections/reg-guides/. Single copies of regulatory guides mayalso be obtained free of charge by writing the Reproduction and Distribution Services Section, ADM, USNRC,Washington, DC 20555-0001, or by fax to (301) 415-2289, or by email to DISTRIBUTION@nrc.gov. Active guidesmay also be purchased from the National Technical Information Service (NTIS) on a standing order basis. Details on this service may be obtained by contacting NTIS at 5285 Port Royal Road, Springfield, Virginia 22161, online at http://www.ntis.gov, or by telephone at (703) 487-4650. Copies are also available for inspection or copying for a feefrom the NRC's Public Document Room (PDR), which is located at 11555 Rockville Pike, Rockville, Maryland; the PDR's mailing address is USNRC PDR, Washington, DC 20555-0001. The PDR can also be reached by tele phoneat (301) 415-4737 or (800) 397-4205, by fax at (301) 415-3548, and by email to PDR@nrc.gov

.Appendix E to DG-1146, Page E-7APPENDIX E REFERENCES E.1Regulatory Guide 1.132, "Site Investigations for Foundations of Nuclear Power Plants,"Revision 1, U.S. Nuclear Regulatory Commission, Washington, DC, October 2003, available inADAMS under Accession #ML032800710.

1 E.2Regulatory Guide 1.138, "Laboratory Investigations of Soils and Rocks for EngineeringAnalysis and Design of Nuclear Power Plants," Revision 2, U.S. Nuclear RegulatoryCommission, Washington, DC, December 2003 available in ADAMS under Accession

1. ML003740184.

1 Appendix F to DG-1146, Page F-1APPENDIX FCRITERIA FOR DEVELOPING TIME HISTORIESThis appendix provides criteria for developing and evaluating modified ground motions used forsoil response and structural analyses. Recorded or modified recorded earthquake ground motion timehistories may be used for linear seismic analysis. Actual recorded earthquake ground motion or modifiedrecorded ground motions should be used for nonlinear seismic analysis.The general objective is to generate a modified recorded accelerogram that achievesapproximately a mean-based fit to the target spectrum. That is, the average ratio of the spectralacceleration calculated from the accelerogram to the target, where the ratio is calculated frequency byfrequency, is only slightly greater than one. The aim is to achieve an accelerogram that does not have significant gaps in the Fourier amplitude spectrum, but which is not biased high with respect to the target. Time histories biased high with respect to a spectral target may overdrive (overestimate damping andstiffness reduction) a site soil column or structure when nonlinear effects are important. Ground motionsthat are generated to "match" or "envelop" given design response spectral shapes should comply withthe following six steps:(1)The time history should have a sufficiently small time increment and sufficiently long durations. Time histories should have a Nyquist frequency of at least 50 Hz (e.g., a time increment of atmost 0.010 seconds) and a total duration of 20 seconds. If frequencies higher than 50 Hz are ofinterest, the time increment of the record must be suitably reduced to provide a Nyquistfrequency (N y = 1/(2 t), where t = time increment) above the maximum frequency of interest. The total duration of the record can be increased by zero packing to satisfy these frequency

criteria.(2)Spectral accelerations at 5 percent damping are computed at a minimum of 100 points perfrequency decade, uniformly spaced over the log frequency scale from 0.1 Hz to 50 Hz or the Nyquist frequency. If the target response spectrum is defined in the frequency range from 0.2 Hzto 25 Hz, the comparison of the modified motion response spectrum with the target spectrum ismade at each frequency computed in this frequency range.(3)The computed 5 percent damped response spectrum of the average of all accelerograms shouldnot fall more than 10 percent below the target spectrum at any one frequency. To prevent spectra in large frequency windows from falling below the target spectrum, the spectra within afrequency window of no larger than +/- 10 percent centered on the frequency should be allowed to fall below the target spectrum. This corresponds to spectra at no more than nine adjacent frequency points defined in (2) above from falling below the target spectrum.(4)The mean of the 5 percent damped response spectra should not exceed the target spectrum at anyfrequency by more than 30 percent (a factor of 1.3) in the frequency range between 0.2 Hz and25 Hz. If the spectrum for the accelerogram exceeds the target spectrum by more than 30 percentat any frequency in this frequency range, the power spectral density of the accelerogram needs to be computed and shown to not have significant gaps in energy at any frequency over thefrequency range.

1The directional correlation coefficient is a measure of the degree of linear relationship between two earthquakeaccelerograms. For accelerograms X and Y, the directional correlation coefficient is given by the following equation:

n _ _ x y = { (1 / n) ( [ (X i - x ) (Y i - y) ] ) } ÷ { x y } i = 1 _ _where n is the number of discrete acceleration-time data points, x and y are the mean values, and x and yare the standard deviations of X and Y, respectively.Appendix F to DG-1146, Page F-2(5)Because of the high variability in time domain characteristics of recorded earthquakes of similarmagnitudes and at similar distances, strict time domain criteria are not recommended. However,modified motions defined as described above should typically have durations (defined by the 5percent to 75 percent Arias intensity), and ratios V/A and AD/V 2 (A, V, and D are peak groundacceleration, ground velocity, and ground displacement, respectively), which are generally consistent with characteristic values for the magnitude and distance of the appropriate controllingevents defining the uniform hazard response spectra. However, in some tectonic settings thesecriteria may need to be altered to accommodate phenomena, such as, directivity, basin effects, and hanging wall effects.(6)To be considered statistically independent, the directional correlation coefficients between pairsof time histories should not exceed a value of 0.30

1. Simply shifting the starting time of a givenaccelerogram does not constitute the establishment of a different accelerogram.