ML042430314

Draft Regulatory Guide 1138, Draft Appendix C, NRC Staff Regulatory Position on ANS External Hazards PRA Standard to Regulatory Guide 1.200 for Trial Use, an Approach for Determining the Technical Adequacy of PRA Results for Risk Informed A

ML042430314
Person / Time
Issue date:	08/31/2004
From:	Mary Drouin NRC/RES/DRAA/PRAB
To:
Drouin M (301)415-6675
References
DG-1138 RG-1.200
Download: ML042430314 (30)

Text

U.S. NUCLEAR REGULATORY COMMISSION August 2004 OFFICE OF NUCLEAR REGULATORY RESEARCH Division 1 Draft DG-1138 DRAFT REGULATORY GUIDE

Contact:

M.T. Drouin FOR COMMENT Draft Appendix C NRC Staff Regulatory Position on ANS External Hazards PRA Standard to Regulatory Guide 1.200 For Trial Use An Approach for Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk Informed Activities The NRC has issued for comment draft Regulatory Guide DG-1138 which is a preliminary draft of the staffs regulatory position on ANSI/ANS 58.21-2003, External Events PRA Methodology Standard.

The staffs position is documented in Appendix C to Regulatory Guide (RG) 1.200, An Approach for Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk-Informed Activities. RG 1.200 was issued for trial use in February 2004 and did not contain Appendix C. The NRC staff is only soliciting comments on Appendix C to RG 1.200; Appendix C has not been issued for use. It is the staffs intent to issue a draft Revision 1 to RG 1.200 with Appendix C for public review and comment before issuing a final Revision 1 to RG 1.200 for use in mid-2005.

This regulatory guide is being issued in draft form to involve the public in the early stages of the development of a regulatory position in 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 regulatory analysis 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 or downloaded through the NRCs interactive web site at

<WWW.NRC.GOV> through Rulemaking. Copies of comments received may be examined at the NRC Public Document Room, 11555 Rockville Pike, Rockville, MD. Comments will be most helpful if received by October 29, 2004.

Requests for single copies of draft or active regulatory guides (which may be reproduced) or for placement on an automatic distribution list for 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 NRCs interactive web site (see above); the NRCs web site <WWW.NRC.GOV> in the Electronic Reading Room under Document Collections, Regulatory Guides; and in the NRCs ADAMS Documents at the same web site, under accession number ML042430314

DG-1138 DRAFT APPENDIX C NRC STAFF DRAFT REGULATORY POSITION ON ANS EXTERNAL HAZARDS PRA STANDARD Introduction The American Nuclear Society has published ANSI/ANS-58.21-2003, External Events PRA Methodology Standard. The standard states that it sets forth requirements for external-event probabilistic risk assessments (PRAs) used to support risk informed decisions for commercial nuclear power plants, and prescribes a method for applying these requirements for specific applications. The NRC staff has reviewed ANSI/ANS 58.21-2003 against the characteristics and attributes for a technically acceptable PRA as discussed in Chapter 3 of Regulatory Guide 1.200.

The staffs draft position on each requirement (referred to in the standard as a requirement, a high-level requirement, or a supporting requirement) in ANSI/ANS 58.21-2003 is categorized as no objection, no objection with clarification, or no objection subject to the following qualification, and defined as follows:

• No objection: the staff has no objection to the requirement.

• No objection with clarification: the staff has no objection to the requirement.

However, certain requirements, as written, are either unclear or ambiguous and therefore, the staff has provided its understanding of these requirements.

• No objection subject to the following qualification: the staff has a technical concern with the requirement and has provided a qualification to resolve the concern.

Table C-1 provides the staff draft position on each requirement in ANSI/ANS 58.21-2003. A discussion of the staff concern (issue) and the staff proposed resolution is provided. In the proposed staff resolution, the staff clarification or qualification to the requirement is indicated either in bolded text (i.e., bold) or strikeout text (i.e., strikeout); that is, the necessary additions or deletions to the requirement (as written in ANSI/ANS 58.21-2003) for the staff to have no objection are provided.

DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution SECTION 1 1.1 The standard is only for Clarification The objectives of this standard are to set current generation LWRs, forth requirements for external-event the requirements may not be probabilistic risk assessments (PRAs) used sufficient or adequate for to support risk-informed decisions for other types of reactors current commercial light water reactor nuclear power plants, and to prescribe a method for applying these requirements for specific applications (additional or revised requirements may be needed for other reactor designs).

1.2 ---------------------- No objection ---------------------

1

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 1.3 Scope Second Paragraph can Qualification Delete the 2nd para.

potentially lead to confusion and misinterpretation, Add a para:

concerning when the term Although both seismic PRA and SMA are PRA is inclusive of SMA intended to support risk-informed and when it is not. Further, the applications, the distinction between them distinction between the seismic regarding their applicability to develop risk PRA and SMA methodologies insights needs to be clearly understood.

needs to be clearly stated. The SMA is a deterministic risk methodology, and in this context, a well executed SMA analysis can provide qualitative, and limited quantitative risk insights that could be used to support an intended application. However, for situations where detailed quantitative risk insights are necessitated, a seismic PRA is needed to obtain the required insights.

1.3.2 The term full-scope PRA is Clarification ...that use aspects of PRA methodology but are misleading in the context of RG not full-scope complete PRAs themselves (see 1.200. 3.4, for example).

1.3.2 The demonstrably Clarification ...(Herein, the phrases bounding analysis and conservative and bounding demonstrably conservative analysis are used analyses are performed using interchangeably.)

different approaches, and should not be used interchangeably.

1.3.3 ---------------------- No objection ------------------------

1.3.4 The effects of the external event Clarification The analysis of the LERF endpoint proceeds in (e.g., earthquake) on the the same way as the analysis of the CDF integrity of the containment endpoint, with one major exception, as follows:

boundary should be discussed. There are some accident sequences, leading to A potential LERF may be core damage but not to large early releases in mitigated by containment for an the internal-events PRA model, that need to be internal event initiator. However, elevated to potential LERF sequences when the effective containment may be initiator is an external event. One set of compromised by physical sequences are those where the effects of the damage/weakening of the external initiators might compromise containment boundary due to containment integrity and thereby possibly the external event. contribute to LERF. The other set These are sequences in which offsite protective action (specifically, the evacuation of nearby populations) is impeded due to the external event. The same sequence that might not be a LERF sequence due to any internal initiator may perhaps affect nearby populations who cannot evacuate as effectively.

1.3.5-1.3.6 ---------------------- No objection -----------------------

2

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 1.4, The interpretation of supporting Qualification Furthermore ... , it is understood that the 5th para. requirements (SR) that use the interpretation is somewhat graded, with more same word under more than one detail, or more specificity, or more realism, or a capability category is different combination thereof, required for the higher from that currently adopted in Capability Category than for the lower one. , it RG 1.200. applies equally to each Capability Category without any need to identify a corresponding Capability Category. The differentiation between capability categories is made in other SRs.

1.4, It is inappropriate to make Qualification Concerning the requirement ... from the EPRI 2nd to the last statements regarding the quality guidance report. Essentially every SMA that has para. and uniformity of past SMA been completed using the EPRI SMA method analyses for IPEEE in the followed the EPRI guidance closely, with only standard. minor deviations. Thus there exists little gradation among the SMAs accomplished to date, and it is anticipated that if another SMA were to be done it too would exhibit very little difference from those already completed.

Therefore, it has been judged ...

1.4, The last para needs greater Clarification the last para. clarity of intent. A choice of The SMA covered in Section 3.6 and the words such as As a matter of Seismic PRA covered in Section 3.7 may be philosophy could lead an used together. As a matter of philosophy, an analyst to do things outside the analyst can augment an SMA with issue-focused requirements of this standard. specific PRA evaluations and seismic-PRA evaluations to support an application. The analyst would need justify the adequacy of the blended or enhanced treatment, and peer review is to be relied upon to verify the treatment. This standard permits the use of issue-focused specific PRA evaluations to augment an SMA.

The analyst needs to document the technical basis for the adequacy of the methodology, and a peer review needs to verity it.

1.4, Table 1 The table does match the Table Qualification Replace with the table 1.3-1 of Addenda to of Addenda to ASME RA-Sa- ASME RA-Sa-2003.

2003 3

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 1.5 To be consistent with the ASME Qualification Shall, Should, and May: The high-level Standard, the word SHALL, requirements contained herein are phrased in should only appear in a high the usual language of standards, namely the level requirement. The words, language of "shall," "should," or "may." These should and may are three terms are defined in Section 2. These permissives and do not provide definitions are repeated here:

a minimum requirement. Action verbs should be used in all SRs. shall - used to state a mandatory requirement should - used to state a recommendation may - used to state an option to be implemented at the user's discretion.

SHALL is used to state a high-level requirement.

Action Verbs: Some of the Supporting requirements are phrased in "action verb" form, to conform to the format in the ASME standard (ASME, 2002). Whenever an action verb is used, the requirement is to be understood as if the "shall" form were used. As an example, the requirement REQ. EXT-B4 reads in part, "REVIEW any significant changes since the NRC operating license was issued." This is to be understood as equivalent to "Any significant changes since the NRC operating license was issued SHALL BE REVIEWED."

1.5, 3rd para The word "consider" is Qualification The Word Consider: ... pay particular permissive and inappropriate for attention to this topic.

SRs. Action verbs should be used.

1.6-1.10 ---------------------- No objection ----------------------

SECTION 2 2.1 Acronysms and Initialisms Clarification HLR - High-Level Requirement SR - Supporting Requirements 2.2 Definition of the bounding Clarification analysis should be provided bounding analysis: Analysis that uses assumptions such that the assessed outcome will meet or exceed the maximum severity of all creditable outcomes.

Definition of the demonstrably Clarification demonstrably conservative analysis: Analysis conservative analysis should be that uses assumptions such that the assessed provided outcome will be conservative relative to the expected outcome.

4

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution Composite The term uncertainty should be Clarification Composite variability, the composite variability variability used consistent with the aleatory includes the randomness variability aleatory and epistemic uncertainty (randomness) uncertainty (R) and the definitions epistemic (modeling) uncertainty (U). The logarithmic standard deviation of composite variability, c, is expressed as ( R2 + U2)1/2 Core See issue discussed in RG Clarification core damage: ....enough of the core, if Damage 1.200 Table A-1, Chapter 2, 2.2, released, to result in offsite public health Core Damage. effects to cause a significant release.

Dependency The definition should be Clarification dependency: Requirement external to an item consistent with the ASME RA- and upon which its function depends and is Sa-2003. associated with dependent events that are determined by, influence by, or correlated to other events or occurrences.

5

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution Dominant See Significant contributor in Clarification dominant contributor: A component, a system, contributor Table A-1, Reg. Guide 1.200, and an accident class, or as accident sequence Appendix A. that has a major impact on the CDF or on the LERF. significant contributor: (a) in the context of an accident sequence, a significant basic event or an initiating event that contributes to a significant sequence; (b) in the context of an accident progression sequence, a contributor which is an essential characteristic (e.g., containment failure mode, physical phenomena) of a significant accident progression sequence, and if not modeled would lead to the omission of the sequence.

significant basic event: those basic events that have a Fussell-Vesely importance greater than 0.005 OR a risk-achievement worth greater than 2.

significant cutset (relative to sequence):

those cutsets, when rank ordered by decreasing frequency, comprise 95% of the sequence CDF OR that individually contribute more than 1% to the sequence CDF.

significant cutset (relative to CDF): those cutsets, when rank ordered by decreasing frequency, comprise 95% of the CDF OR that individually contribute more than 1% to CDF.

significant accident sequence: a significant sequence is one of the set of sequences, defined at the functional or systemic level that, when rank ordered by decreasing frequency, comprise 95% of the core damage frequency (CDF)), OR that individually contribute more than ~1% to the CDF.

Significant accident progression sequence:

one of a set of containment event tree sequences that, when rank ordered by decreasing frequency, comprise 95% of the large early release frequency (LERF), OR that individually contribute more than ~1% to the LERF.

Failure mode This is an incorrect definition. Clarification failure mode: A condition ... or a system. a Use ASME definition. specific functional manifestation of a failure (i.e., the means by which an observer can determine that a failure has occurred) by precluding the successful operation of a piece of equipment, a component, or a system (e.g.,

fails to start, fails to run, leak).

6

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution Fractile Definition of terms lacks clarity. Clarification fractile hazard curves - A set of hazard curves hazard curves used to reflect the uncertainties associated with estimating seismic hazard. A common family of hazard curves used in describing the results of a PSHA is consists of curves of fractiles of the probability distributions of estimated seismic hazard as a function of the level of ground motion parameter.

Fragility The use of uncertainty should Clarification ...Fragility of an SSC is the conditional be consistent with the aleatory probability of its failure at a given hazard input and epistemic uncertainty level. The input could be earthquake motion, definitions wind speed, or flood level. The fragility model used in seismic PRA is known as a double lognormal model with three parameters, Am, R and U which are respectively, the median acceleration capacity, logarithmic standard deviation of aleatory (randomness) uncertainty in capacity and logarithmic standard deviation of the epistemic (modeling) uncertainty in the median capacity.

Large early Inconsistent with ASME Clarification ...protective actions such that there is a release definition potential for early health effects.

Screening Inconsistent with ASME Clarification ... An analysis that eliminates items from further analysis definition consideration based on their negligible contribution to the probability of a significant an accident or its consequences.

Success path Success path is usually defined Clarification ...A set of systems and associated components at the system level rather than that can be used to bring the plant to a stable components. hot or cold condition and maintain this condition for at least 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />.

SECTION 3 3.1-3.2 --------------------- No objection -----------------------

3.3 1st para. --------------------- No objection ----------------------

2nd para. To ensure the quality of the Clarification The high-level requirements ... and the peer outcome of the application of review team (see Section 5). Further, the this standard, the minimum analysis team needs to be experienced in qualifications of the analyst performing activities associated with all need to be clearly stated. elements of the PRA. As a minimum, the analysis team must show capability by direct experience from previous PRA studies of the methodology, and by training in the use of computer codes used in the analyses.

7

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3rd para. The Supporting Requirements Qualification The High Level Requirements and the depend on the Capability Supporting Requirements, taken together, are Category selected for the PRA. formulated in a way that is intended to support The category may be different the applications being considered. Specifically, for different systems or a PRA can meet the High Level Requirements elements included in the PRA. and Supporting Requirements at various The analyst should specify levels-of-detail and various scopes, that need which SRs are being used and not extend beyond what is adequate to support justify their use for the intended the intended application. The analysis team application. needs to identify the SRs used in the PRA and justify the selection of Capability Category from which they have been selected.

3.4 Title The title lacks clarity. Clarification Probabilistic Risk Assessment for Other External Events: Requirements for Identification and Screening and Conservative Analysis 3.4.1 ---------------------- No objection ----------------------

3.4.2, The demonstrably Clarification ...(Herein, the phrases bounding analysis and 1at para., conservative and bounding demonstrably conservative analysis are used item (3) analyses are performed using interchangeably.)

different approaches, and should not be used interchangeably.

3.4.2, Since this section pertains to Clarification ...subjected to the requirements in 3.5, 3.6, 3.7, last para., external events screening other 3.8, or ...

3.4.3, than seismic event, references 2nd para. to 3.6 and 3.7 requirements should be removed.,

3.4.3 The section is entitled Qualification HLR-EXT-A: All pPotential external events (i.e.,

HLR-EXT-A Requirements for Screening and all natural...... SHALL be identified considered Conservative Analysis. and conservative analysis), or detailed analysis.

However, the HLR has a SHALL be subjected to either screening requirement to perform a bounding analysis (demonstrably conservative screening, bounding, or detailed analysis), or detailed analysis.

analysis. The latter is inconsistent with the intent.

Furthermore, the supporting requirements only address identification of external hazards. The screening is performed in HLR-EXT-B and HLR-EXT-C.

3.4.3 -------------- No objection ----------------

HLR-EXT-B 3.4.3 The demonstrably Clarification HLR-EXT-C: A bounding or (demonstrably HLR-EXT-C conservative and bounding conservative) analysis, if used ...

analyses are performed using different approaches, and should not be used interchangeably.

8

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.4.3 -------------- No objection ----------------

HLR-EXT-D to HLR-EXT-E 3.4.4 The section is entitled Qualification HLR-EXT-A: All pPotential external events (i.e.,

HLR-EXT-A Requirements for Screening and all natural...... SHALL be identified considered Conservative Analysis. and conservative analysis), or detailed analysis.

However, the HLR has a and SHALL be subjected to either screening requirement to perform a bounding analysis (demonstrably conservative screening, bounding, or detailed analysis), or detailed analysis.

analysis. The latter is inconsistent with the intent.

Furthermore, the supporting requirements only address identification of external hazards. The assessment is performed in HLR-EXT-B and HLR-EXT-C.

3.4.4, Permissive MAY is Qualification ... and this list MAY be used as PROVIDES one REQ. EXT-A1 inappropriate for a SR acceptable way to meet this requirement.

requirement.

3.4.4, -------------------- No objection ----------------------

REQ. EXT-A2 and Note EXT-A2 3.4.4, Permissives should not be used Qualification ...the following screening criteria MAYbe used as REQ. EXT-B1 in SRs. PROVIDE an acceptable basis:

3.4.4, Permissives should not be used Qualification ..., the following screening criterion MAY be REQ. EXT-B2 in SRs. used as PROVIDES an acceptable basis...

3.4.4, ---------------------- No objection ----------------------

REQ. EXT-B3 3.4.4, The word "consider" is Qualification REVIEW... In particular, CONSIDER in the REQ. EXT-B4 permissive and inappropriate for review REVIEW all of the following:

SRs. Action verbs should be used.

3.4.4 The demonstrably Clarification HLR-EXT-C: A bounding or (demonstrably HLR-EXT-C conservative and bounding conservative) analysis, if used ...

and NOTE analyses are performed using HLR-EXT-C different approaches, and NOTE HLR-EXT-C: Herein, the phrases should not be used bounding analysis and demonstrably interchangeably. conservative analysis are used interchangeably.

9

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.4.4, Permissive MAY should not be Qualification For screening out an external event, the analytst REQ. EXT-C1 used in SRs. ... screening criteria is met: any one of the following three screening criteria PROVIDE an acceptable basis for bounding analysis or demonstrably conservative analysis:

3.4.4, The demonstrably Clarification NOTE EXT-C1: The bounding or (demonstrably NOTE EXT- conservative and bounding conservative) analysis ...

C1 analyses are performed using different approaches, and should not be used interchangeably.

3.4.4, This SR addresses the hazard Qualification BASE the estimation of the mean frequency and REQ. EXT-C2 analysis. The SR contains two opther parameters of the design-basis hazard on alternatives. The first is a state- state-of-the-art modeling and recent data ( .......

of-the-art hazard analysis, the ), or BOUND the estimation for the purposes of a second is a bounding analysis. demonstrably conservative analysis>

The SR should reflect the CONSIDER the uncertainties in modeling and minimum requirement which is data in this hazard evaluation.

that for a bounding analysis. In ESTIMATE the frequency and other the ASME Standard, the term parameters of the hazard using a bounding state-of-the-art is used to analysis or a demonstrably conservative correspond to a capability analysis.

category III. To conform to that meaning, the term should not be used here. Furthermore, the last sentence is appropriate for a detailed analysis but not for a bounding analysis.

3.4.4, The demonstrably Clarification NOTE EXT-C2: The spirit of a bounding or NOTE conservative and bounding (demonstrably conservative) analysis ...

EXT-C2 analyses are performed using different approaches, and should not be used interchangeably.

3.4.4, The requirement in the standard Qualification In estimating the mean conditional core damage REQ. EXT-C3 should represent the minimum, probability (CCDP), USE a systems model of the which is a demonstrably plant that meets the systems-modeling conservative analysis. requirements in ASME-RA-S-2002 insofar as they apply [1]. For the purposes of this screening analysis, a demonstrably conservative approach to the analysis is acceptable.

Calculate the CCDP using a bounding analysis or a demonstrably conservative analysis.

3.4.4, There is no requirement that Qualification NEW SR: Identify those SSCs required to REQ. EXT- identifies the impact of the maintain the plant in operation or that are C3a hazard on the plant SSCs. required to respond to an initiating event to prevent core damage, that are vulnerable to the hazard, and determine their failure modes.

10

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.4.4, There is no requirement that Qualification NEW SR: ESTIMATE the CCDP taking into REQ. EXT- addresses the incorporation of account the initiating events caused by the C3b the impact of the hazard into the hazard, and the systems of functions estimation of the CCDP rendered unavailable. Modifying the internal events PRA model as appropriate, using conservative assessments of the impact of the hazard (fragility analysis), is an acceptable approach.

3.4.4, Permissives should not be used Qualification BASE...This includes not only the hazard REQ. EXT-C4 in SRs. analysis but also any fragility analysis that may be necessary is applicable.

3.4.4, Since section 3.4 provides Clarification ...(See 3.5, 3.6, 3.7, 3.8, and 3.9.)

REQ. EXT-C5 requirements for external event hazards other than seismic, reference to sections dealing with SMA and seismic PRA should be removed.

3.4.4, --------------------- No objection ----------------------

REQ. EXT-D1, D2 3.4.4, ----------------- No objection ----------------------

NOTE EXT-D1 3.4.4, ------------------ No objection -----------------------

REQ. EXT-E1- E3 3.5 3.5.1 As currently written, the scope of Qualification Scope: ...The term other external events refers this section allows analyses of to external events other than earthquakes, high wind hazards and external winds, and external floods.

flooding hazards to be performed using the Applicability: ... external event. Alternatively, the requirements of this section. requirements in 3.8...then all of the requirements However, requirements for therein apply.

analyses of wind and external flooding hazards are explicitly provided in sections 3.8 and 3.9.

Therefore, the scope of section 3.5 should be narrowed.

3.5.3, The last sentence in the Clarification The analysis ... a mixture of the two. The models HLR-ANA-A statement of the high level used for ..... short term trends in the requirement contains the phrase frequencies.

"SHOULD NOT be unduly influenced by ... ", but there is no supporting requirement that relates to this. It is not, in fact clear what this last sentence means. If there is a real trend in frequencies this should in fact be included in the assessment.

11

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.5.3, --------------------- No objection ---------------------

HLR-ANA-B thru. HLR-ANA-D 3.5.4, The last sentence in the Clarification The analysis ... a mixture of the two. The models HLR-ANA-A statement of the high level used for ..... short term trends in the requirement contains the phrase frequencies.

"SHOULD NOT be unduly influenced by ... ", but there is no supporting requirement that relates to this. It is not, in fact clear what this last sentence means. If there is a real trend in frequencies this should in fact be included in the assessment.

3.5.4, -------------------- No objection -----------------------

REQ.ANA-A1 3.5.4, The word properly in the Clarification ... ACCOUNT properly for and ...

REQ.ANA-A2 statement ACCOUNT properly for and ... is superfluous.

3.5.4, The note contains a discussion Qualification NEW SR: To develop the PRA model, define NOTE ANA- on the parameterization of the the hazard curve in terms of the parameter A2 hazard curve(s). This does not that best represents a measure of the intensity clarify the requirement, but of the hazard.

suggests that another requirement be added.

3.5.4, ------------------- No objection -------------------------

REQ.ANA-A3 thru. B1 3.5.4, The word "consider" is Qualification ... CONSIDER INCORPORATE the findings of a REQ.ANA-B2 permissive and inappropriate for plant walkdown in this evaluation.

SRs. Action verbs should be used.

3.5.4, The note contains discussions Qualification NEW SR: Define the fragility curve for each NOTE ANA- that should be requirements. failure mode as a function of the same B3 parameter used to represent the intensity of the hazard.

12

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.5.4, There is no requirement to Qualification NEW SR: Identify those SSCs required to REQ.ANA-C1 identify the SSCs affected by maintain the plant in operation or that are the hazard, nor the initiating required to respond to an initiating event to events caused by the hazard. prevent core damage, that are vulnerable to The supporting requirements do the hazard, and determine their failure modes.

not support the HLR as stated.

NEW SR: ESTIMATE the CCDP taking into There is no requirement that account the initiating events caused by the addresses the incorporation of hazard, and the systems of functions the impact of the hazard into the rendered unavailable. Modifying the internal estimation of the CCDP events PRA model as appropriate, using conservative assessments of the impact of the hazard (fragility analysis), is an acceptable approach.

3.5.4, The word "consider" is Qualification ASSESS the accident sequences initiated by the REQ.ANA-C1 permissive and inappropriate for external event to estimate CDF and LERF SRs. Action verbs should be contribution. In the analysis, USE as used. appropriate the appropriate applicable hazard curves and the fragilities of structures and equipment.

3.5.4, ------------------- No objection -------------------------

REQ.ANA-D1 thru. D7 3.6 7th para, 2nd These sentences need greater Clarification As discussed in 1.4, the SMA covered in Section and 3rd clarity of intent. A choice of 3.6 and the Seismic PRA covered in Section 3.7 sentences words such as As a matter of may be used together. As a matter of philosophy could lead an philosophy, an analyst can augment an SMA analyst to do things outside the with issue-focused specific PRA evaluations and requirements of this standard. seismic-PRA evaluations to support an application. The analyst would need justify the adequacy of the blended or enhanced treatment, and peer review is to be relied upon to verify the treatment. this standard permits the use of issue-focused specific PRA evaluations to augment an SMA. The analyst needs to document the technical basis for the adequacy of the methodology, and a peer review needs to verity it. ...

3.6.1, ------------------- No objection ----------------------

HLR-SM-A 3.6.1, The last phrase, ..."...following Clarification "...following an earthquake equal to or larger HLR-SM-B an earthquake larger than the than the RLE".

RLE" , could be misinterpreted.

3.6.1, --------------------- No objection ----------------------

HLR-SM-C 3.6.1, --------------------- No objection ----------------------

HLR-SM-D 13

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.6.1, Plant walkdown is a major part Clarification ..., through the review of design documents, HLR-SM-E of the margin assessment including plant-specific analysis and test reports process (not a supplemental , and the results of a plant walkdown part) for identifications of SSC supplemented by earthquake experience data, failure modes. fragility test data, and generic qualification test data. , and by a walkdown 3.6.1, ------------------ No objection ---------------------

HLR-SM-F 3.6.1, ------------------ No objection ---------------------

HLR-SM-G 3.6.1, The wording ..applying the PRA Clarification ...applying the PRA and updating it its HLR-SM-H and updating it, needs to application and update ...

changed. The term PRA should not be used in an HLR for an SMA.

3.6.2, -------------------- No objection -----------------------

REQ SM-A1 to REQ SM-C1 3.6.2, Permissives should not be used Qualification ..., realistic seismic responses MAY be are REQ SM-C2 in SRs. obtained...

3.6.2, ---------------------- No objection -----------------------

REQ SM-C3 to SM-D4 3.6.2, The word FOCUS does not Clarification FOCUS the walkdown on During the walkdown, REQ SM-D5 provide a direction regarding IDENTIFY the potential for ...

and NOTE what actions should be taken.

SM-D5 3.6.2, NOTE SM-D6, related to II/I Qualification NOTE SM-D6: For SMA, A a II/I issue refers to NOTE SM-D6 issue is misleading in the the condition ... safety equipment. any object context of SMA. Any object (whether seismically qualified to the plant (whether seismically qualified to design basis or not) that can fall on and the plant design basis or not) damage any item on the SSEL. The HCLPF that can fall on and damage any capacity of the falling object may control the item on the SSEL is a II/I issue HCLPF capacity of the success path and for SMA. The HCLPF capacity of potentially the plant HCLPF capacity if it is the falling object may control the less than the HCLPF capacity of the weakest HCLPF capacity of the success path and potentially the plant item on the SSEL .

HCLPF capacity if it is less than the HCLPF capacity of the weakest item on the SSEL .

3.6.2, ---------------------- No objection -----------------------

REQ. SM-E1 3.6.2, The word "consider" is Qualification CONSIDER EXAMINE all relevant failure REQ. SM-E2 permissive and inappropriate for modes...

SRs. Action verbs should be used.

3.6.2, REQ. SM-F1 duplicates HLR- Clarification (REQ. SM-F1) BASE ...test data.

REQ. SM-F1 SM-F, and is less prescriptive.

14

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.6.2, ---------------------- No objection -----------------------

REQ. SM-F2 and NOTE SM-F2 3.6.2, ---------------------- No objection -----------------------

REQ. SM-F3 and NOTE SM-F3 3.6.2, -------------------- No objection --------------------

REQ. SM-G1 and NOTE SM-G1 3.6.2, Seismic upgrade is interpreted Clarification (REQ. SM-G2) REPORT ... have been done.

REQ. SM-G2 to mean a physical plant and NOTE modification to increase the SM-G2 seismic capacity of a weak SSC. Note SM-G2: If the plant ... would have been This is not part of the SMA done.

methodology just as performing seismic upgrade as a result of a seismic PRA is not part of the PRA methodology.

3.6.2, ---------------------- No objection -----------------------

REQ. SM-H1 thru H5 and NOTE SM-H5 3.7 3.7, ----------------------- No objection ----------------------

3.7.1.1 3.7.1.2, --------------------- No objection ---------------------

HLR-HA-A to HLR-HA-B 3.7.1.2, The word "consider" is Qualification ... SHALL consider all examine ... SHALL be HLR-HA-C permissive and inappropriate for considered addressed in characterizing the SRs. Action verbs should be ground motion propagation.

used.

3.7.1.2, The word "consider" is Qualification ... SHALL account for all examine credible ...

HLR-HA-D permissive and inappropriate for Both the aleatory ... be considered addressed ...

SRs. Action verbs should be used.

3.7.1.2, --------------------- No objection ----------------------

HLR-HA-E 3.7.1.2, --------------------- No objection ---------------------

HLR-HA-F 15

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.7.1.2, The reference to NUREG/CR- Qualification For further use in the SPRA, the spectral shape HLR-HA-G 0098 broad band spectrum SHALL be based on a site-specific evaluation shape should be made in a taking into account the contributions of supporting requirement. Further, deaggregated magnitude-distance results of the NURGE/CR-0098 spectral PSHA. Broad-band, smooth spectral shapes, shapes are not always such as those presented in NUREG/CR-0098 [6]

appropriate, particularly for (for lower-seismicity sites such as most of those CEUS sites. east of the U.S. Rocky Mountains) may also be used taking into account the site conditions. The use of uniform hazard response spectra may also be appropriate is acceptable if it reflects the site-specific shape.

NEW SR HA-G1a: Broad-band, smooth spectral shapes, such as those presented in NUREG/CR-0098 [6] are acceptable if they are shown to be appropriate for the site.

NEW NOTE HA-G1a: Recent developments

[42] indicate that these spectral shapes are not appropriate for CEUS sites where high frequency content is dominant at hard rock sites.

3.7.1.2, --------------------- No objection ---------------------

HLR-HA-H to HLR-HA-J 3.7.1.3, ---------------------- No objection ---------------------

HLR-HA-A 3.7.1.3, -------------------- No objection ----------------------

HA-A1 3.7.1.3, This requirement contains two Qualification As the parameter to characterize both hazard HA-A2, separate requirements. and fragilities, USE the spectral accelerations, or Cat. I and II the average spectral acceleration over a There is a requirement to selected band of frequencies, or peak ground capture the frequencies of SCCs acceleration. In the selection of frequencies to that are dominant to the PRA determine spectral accelerations or average results and insights. This can spectral acceleration, CAPTURE the frequencies not be a priori. of those SSCs that are of interest and are dominant contributors to the PRA results and insights.

NEW SR HA-A2a: In the selection of frequencies to determine spectral accelerations or average spectral acceleration, CAPTURE the frequencies of those SSCs that are of interest and dominant contributors to significant in the PRA quantification results ans insights.

16

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.7.1.3, This requirement contains two Qualification As the parameter to characterize both hazard HA-A2, separate requirements. and fragilities, USE the spectral accelerations, or Cat. III the average spectral acceleration over a There is a requirement to selected band of frequencies. In the selection of capture the frequencies of SCCs frequencies to determine spectral accelerations that are dominant to the PRA or average spectral acceleration, CAPTURE the results and insights. This can frequencies of those SSCs that are of interest not be a priori. and are dominant contributors to the PRA results and insights.

NEW SR HA-A2b: In the selection of frequencies to determine spectral accelerations or average spectral acceleration, CAPTURE the frequencies of those SSCs that are of interest and dominant contributors to significant in the PRA quantification results ans insights.

3.7.1.3, As stated, the requirement is Clarification In developing the PSHA results, whether they HA-A3 difficult to meet. are characterized by spectral accelerations, peak ground accelerations or both, EXTEND them to large enough values (consistent with the physical data and interpretations) so that the truncation does not significantly impact the numerical results. final numerical results, such as core damage frequency, reflect accurate estimates of risk, and the delineation and ranking of seismic-initiated sequences are not affected.

3.7.1.3, ---------------------- No objection ---------------------

HLR-HA-B 3.7.1.3, For Capability Category III Clarification In performing the PSHA, BASE it on available HA-B1 applications, the available data and developed comprehensive geological, base must be able characterize seismological, and geophysical and local effects on site response. geotechnical data bases that reflect the current state-of-the-knowledge, and that are used by experts/analysts to develop interpretations and inputs to the PSHA. For Category III applications, INCLUDE site specific laboratory data for site soils including their potential uncertainty to characterize local site response effects .

3.7.1.3, The use of term the amount of Qualification ... The difference between Capability Category II NOTE HA-B1 resources and sophistication... and III is ... the databases.

as the reason for the distinction between Capability Categories II and III is inconsistent with the bases for PRA capability categories.

3.7.1.3, --------------- No objection -----------------------

HA-B2 and HA-B3 17

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.7.1.3, The use of term the amount of Qualification ... The difference between Capability Category II NOTE HA-B2 resources and sophistication... and III is ... the databases.

and as the reason for the distinction NOTE HA-B3 between Capability Categories II and III is inconsistent with the bases for PRA capability categories.

3.7.1.3, The word "consider" is Qualification ... SHALL consider all examine ... SHALL be HLR-HA-C permissive and inappropriate for considered addressed in characterizing the SRs. Action verbs should be ground motion propagation.

used.

3.7.1.3, --------------- No objection -----------------------

HA-C1 - C4 3.7.1.3, The word "consider" is Qualification ... SHALL account for all examine credible ...

HLR-HA-D permissive and inappropriate for Both the aleatory ... be considered addressed ...

SRs. Action verbs should be used.

3.7.1.3, Since attenuation relationships Qualification ACCOUNT in ... Seismicity data (including HA-D1 for characterizing the ground strong motion data), and c) Current attenuation motion propagation are models in the ground motion estimates.

developed based on empirical data and subjective inputs, several attenuation models may exist.

3.7.1.3, -------------- No objection ---------------------

HA-D2 -D4 3.7.1.3, ---------------------- No objection ---------------------

HLR-HA-E 3.7.1.3, ----------------. No objection ---------------------

HA-E1, 3.7.1.3, The site-specific transfer Clarification The purpose of a local site response analysis...

Note HA-E1 functions that are used to modify for the site characteristic [41]. Probabilistic the rock ground motions should estimates of site properties should be used in computed using probabilistic determining the site-specific functions.

estimates of site properties.

3.7.1.3, --------------- No objection -----------------------

HA-E2 and Note HA-E2 3.7.1.3, ---------------------- No objection ---------------------

HLR-HA-F 3.7.1.3, --------------- No objection -----------------------

HA-F1 to HA-F3 18

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.7.1.3, The reference to NUREG/CR- Qualification For further use in the SPRA, the spectral shape HLR-HA-G 0098 broad band spectrum SHALL be based on a site-specific evaluation shape should be made in a taking into account the contributions of supporting requirement. Further, deaggregated magnitude-distance results of the NURGE/CR-0098 spectral PSHA. Broad-band, smooth spectral shapes, shapes are not always such as those presented in NUREG/CR-0098 [6]

appropriate, particularly for (for lower-seismicity sites such as most of those CEUS sites. east of the U.S. Rocky Mountains) may also be used taking into account the site conditions. The Also, the last sentence is use of existing uniform hazard response spectra inconsistent with that stated by (UHSs) is acceptable unless evidence comes to 3.7.1.2 HLR-HA-G light that would challenge these UHS spectral shapes if it reflects the site-specific shape.

NEW SR HA-G1a: Broad-band, smooth spectral shapes, such as those presented in NUREG/CR-0098 [6] are acceptable if they are shown to be appropriate for the site.

NEW NOTE HA-G1a: Recent developments

[42] indicate that these spectral shapes are not appropriate for CEUS sites where high frequency content is dominant at hard rock sites.

3.7.1.3, --------------- No objection -----------------------

HA-G1 3.7.1.3, Spectral shapes used to Clarification NOTE HA-G1: The issue of which spectral Note HA-G1 evaluate in-structure SSCs shape should be used in the screening of must include the effects of structures, systems, and components (SSCs) amplification from both local site and in quantification of SPRA results requires conditions and SSI. careful consideration. For screening purposes, the spectral shape used should have Based on IPEEE reviews, amplification factors, including effects from certain UHS shapes used for both local site conditions as well as soil-CEUS were not appropriate for structure interaction, such that the demand the screening purpose. resulting from the use of this shape is higher than that based on the design spectra. This will preclude premature screening of components and will avoid anomalies such as the screened components (e.g., surrogate elements) being the dominant risk contributing components.

Additional discussion on this issue can be found in Ref. 22. In the quantification of fragilities and of final risk results, it is important to use as realistic a shape as possible. Semi-site specific shapes, such as those given in NUREG-0098, have been used in the past and are considered may be adequate for this purpose, provided that they are shown to be reasonably appropriate for the site [42]. The UHS is acceptable for this purpose if it can be shown that the UHS shape is appropriate for the site.

unless evidence comes to light (e.g., within the technical literature) that these UHS do not reflect the spectral shape of the site-specific events.

3.7.1.3, ---------------------- No objection ---------------------

HLR-HA-H 19

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.7.1.3, ------------------- No objection ---------------------

Note HA-H 3.7.2 3.7.2.1 ----------------------- No objection --------------------

3.7.2.2, Words: important, significant Clarification The seismic-PRA systems models SHALL HLR-SA-A used to characterize the include all important seismic-caused initiating contribution to CDF should be events and that can lead to core damage or clearly stated in quantitative large early release, and SHALL include all other manner. important failures that can contribute significantly to CDF or LERF, including seismic-induced SSC failures, non-seismic-induced unavailabilities, and human errors. , that give rise to significant accident sequences and/or significant accident progression sequences.

3.7.2.2, ----------------------- No objection --------------------------

HLR-SA-B to HLR-SA-F 3.7.2.3, Words: important, significant Clarification The seismic-PRA systems models SHALL HLR-SA-A used to characterize the include all important seismic-caused initiating contribution to CDF should be events and that can lead to core damage or clearly stated in quantitative large early release, and SHALL include all other manner. important failures that can contribute significantly to CDF or LERF, including seismic-induced SSC failures, non-seismic-induced unavailabilities, and human errors. , that give rise to significant accident sequences and/or significant accident progression sequences.

3.7.2.3, To more closely follow the Clarification ENSURE that significant earthquake-caused SA-A 1 ASME Standard, this SLR initiating events that give rise to significant should conclude with the accident sequences and/or significant statement using a systematic accident progression sequences are included process, and there needs to be in the seismic-PRA system model using a a definition of significant. systematic process.

3.7.2.3, The note does not identify Clarification NOTE SA-A1: It is ...br thoroughly investigated.

NOTE SA-A 1 systematic process. One approach that has been used successfully is to perform an FMEA of the seismic failures identified by the fragility analysis...

3.7.2.3, The requirement is unclear. Clarification To be resolved.

SA-A2 20

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution SA-A3 1st paragraph: The SR Qualification ENSURE that the PRA system model reflect all contains the word all, which is significant earthquake-caused failures and all inappropriate in a Standard. significant nonseismically induced There needs to be a definition of unavailabilities and human errors that give rise significant. to significant accident sequences and/or significant accident progression sequences 2nd paragraph: Permissives The analysis MAY It is acceptable to group should not be used in Srs. earthquake-caused failures in the analysis if the Move to new SA-A3b below. leading failure in the group is modeled.

NOTE SA-A3:

NEW SA-A3a: USE the event trees and fault The note contains two issues trees from the internal-events full-power PRA that should be requirements.

model as the basis for the seismic event trees.

Last sentence of th 1st para refers to the use of NOTE SA-A3a: The event trees and fault trees supercomponent. Although from the internal-events full-power PRA model supercomponent could greatly are generally used as the basis for the seismic simplify system modeling, it could also lead to a situation event trees. This is done both to capture the where the supercomponent thinking that has gone into their development, becomes a dominant contributor and to assist in allowing comparisons between and the risk insights of SSCs the internal-events PRA and the seismic PRA to within the supercomponent be made on a common basis ... The lumping of could be masked, if it is not certain groups of individual components into applied properly. so-called "supercomponents" in the systems model is also a valid approximation in many situations. However, it is cautioned that supercomponents should be used in a manner that they will not become significant contributors to the seismic CDF.]

In special circumstances ... Further, it is then especially important that a peer review be undertaken that concentrates on these aspects.

NEW SA-A3b: INCLUDE in the PRA system models, the consequences of those earthquake caused failures of structures and components that are not included in the internal event models. The analysis MAY It is acceptable to group earthquake-caused failures in the analysis if the leading failure in the group is modeled.

Note for SA-A3b: Earthquakes can cause failures that are not explicitly represented in the internal-events models, primarily (but not exclusively) due to damage to structures and other passive items ... This means that initiating events and SSC failures that could lead to LERF-type consequences need to be included in the systems model even if the CDF frequency is quite low. (See FR-F4 and NOTE FR-F4.)

3.7.2.3, ----------------------- No objection --------------------------

HLR-SA-B 21

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.7.2.3, -------------------- No objection ----------------------

SA-B1 3.7.2.3, The word "consider" is Qualification In the human reliability analysis (HRA) aspect, SA-B2 permissive and inappropriate for CONSIDER EXAMINE that whether...

SRs. Action verbs should be used.

3.7.2.3, Permissives should not be used Qualification The analysis MAY It is acceptable to use SA-B3, in SRs. generic dependency and correlation values in 2nd para, the analysis and PROVIDE bases if justified.

cat. I and II 3.7.2.3, ------------------ No objection ------------------

SA-B4 3.7.2.3, The word "consider" is Qualification CONSIDER EXAMINE the effects ...

SA-B5 permissive and inappropriate for SRs. Action verbs should be used.

3.7.2.3, The word "consider" is Qualification CONSIDER EXAMINE the possibility ...

SA-B7 permissive and inappropriate for SRs. Action verbs should be used.

3.7.2.3, The word "consider" is Qualification CONSIDER EXAMINE the likelihood ...

SA-B8 permissive and inappropriate for SRs. Action verbs should be used.

3.7.2.3, Permissives should not be used Qualification It is acceptable to use conservative recovery SA-B8, in SRs. values MAY be used.

2nd para, Cat. I.

3.7.2.3, The word "consider" is Qualification CONSIDER EXAMINE the effect of including ...

SA-B9 permissive and inappropriate for SRs. Action verbs should be used.

3.7.2.3, ------------------- No objection -----------------------

SA-B10 3.7.2.3, ----------------------- No objection --------------------------

HLR-SA-C 3.7.2.3, The phrase demonstrating ... Clarification To ensure that the systems-analysis models SA-C1 significantly alter... is reflect the as-built, as-operated plant, JUSTIFY permissive and inappropriate any important conservatisms or other distortions for the requirement. introduced by demonstrating that they do not significantly alter the seismic-PRA's validity for applications is maintained.

3.7.2.3, ---------------------- No objection -------------------------

SA-D1 to SA-E1 3.7.2.3, Permissives should not be used Qualification It is acceptable to use broad groupings MAY be SA-E2, in SRs. used.

2nd para, cat I 22

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.7.2.3, Permissives should not be used Qualification The analysis MAY It is acceptable to use SA-E4, in SRs. generic dependency and correlation values in 2nd para, cat I the analysis and PROVIDE the basis for such and II application if justified.

3.7.2.3, ----------------------- No objection -------------------------

HLR-SA-F 3.7.2.3, The tem dominant risk Qualification NOTE SA-F1: The major outputs of a seismic NOTE SA-F1 contributors is not defined. PRA, such as mean CDF, mean LERF, uncertainty distributions on CDF and LERF, results of sensitivity studies, significant dominant risk contributors, and so on are examples of the PRA results that are generally documented.

3.7.2.3, ----------------------- No Objection -------------------------

SA-F2 to SA-F3 3.7.3 3.7.3.1, ---------------------- No objection -------------------------

HLR-FR-A to HLR-FR-B 3.7.3.1, Permissives should not be used Qualification The seismic fragility evaluation SHALL be based HLR-FR-C in SRs. on realistic seismic response that the SSCs experience at their failure levels. Depending on the site conditions and response analysis methods used in the plant design, DEVELOP realistic seismic response MAY be obtained by an appropriate combination of scaling, new analysis and new structural models.

3.7.3.1, --------------------- No objection ---------------------

HLR-FR-D thru.

HLR-FR-G 3.7.3.2, ---------------------- No objection -------------------------

HLR-FR-A 3.7.3.2, ---------------------- No objection -------------------------

FR-A1 and FR-A2 3.7.3.2, ---------------------- No objection -------------------------

HLR-FR-B 3.7.3.2, Permissives should not be used Qualification ...For example, it is acceptable to apply FR-B1 in Srs. guidance given in EPRI NP-6041 and NUREG/CR-4334 MAY be used to screen out components...

3.7.3.2, ---------------------- No objection -----------------------

FR-B2 23

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.7.3.2, Permissives should not be used Qualification The seismic fragility evaluation SHALL be based HLR-FR-C in SRs. on realistic seismic response that the SSCs experience at their failure levels. Depending on the site conditions and response analysis methods used in the plant design, DEVELOP realistic seismic response MAY be obtained by an appropriate combination of scaling, new analysis and new structural models.

3.7.3.2, Spectral shape issues for Clarification ESTIMATE the seismic responses that the FR-C1 Capability Category I and II components experience at their failure levels on a realistic basis using site-specific earthquake response spectra in three orthogonal directions, anchored to a ground motion parameter such as peak ground acceleration or average spectral acceleration over a given frequency band, or .

ENSURE that the spectral shape used reflects or bounds the site-specific considerations conditions.

3.7.3.2, Spectral shape issues for Clarification ESTIMATE the seismic responses that the FR-C1 Capability Category III components experience at their failure levels on a realistic basis using site-specific earthquake response spectra in three orthogonal directions, anchored to a ground motion parameter such as peak ground acceleration or average spectral acceleration over a given frequency band.

3.7.3.2, Probabilistic parameters for Clarification If probabilistic response analysis is performed to FR-C2 Capability Category I obtain realistic structural loads and floor response spectra, ENSURE that the number of simulations done (e.g., Monte Carlo simulation and Latin Hypercube Sampling) is large enough to obtain stable median and 85%

non-exceedance responses for free-field site response. In the response analysis, appropriately ACCOUNT for the entire spectrum of input ground motion levels displayed in the seismic hazard curves.

3.7.3.2, Probabilistic parameters for Clarification If probabilistic response analysis is performed to FR-C2 Capability Category II obtain realistic structural loads and floor response spectra, ENSURE that the number of simulations done (e.g., Monte Carlo simulation and Latin Hypercube Sampling) is large enough to obtain stable median and 85%

non-exceedance responses for free-field site response. In the response analysis, appropriately ACCOUNT for the entire spectrum of input ground motion levels displayed in the seismic hazard curves.

24

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.7.3.2, Probabilistic parameters for Clarification PERFORM probabilistic seismic response FR-C2 Capability Category III analysis taking into account the uncertainties in the input ground motion and structural and, site soil properties and structural parameters.

CALCULATE joint probability distributions of the responses of different components in the building.

3.7.3.2, Update reference Clarification NOTE FR-C2: For a description of the NOTE FR-C2 probabilistic seismic response analysis, the reader is referred to Ref. 49 and Ref. 42 3.7.3.2, ----------------------- No objection --------------------

FR-C3 to FR-C5 3.7.3.2, The word "consider" is Qualification ... dominate the seismically induced core FR-C6, permissive and inappropriate for damage frequency. CONSIDER ACCOUNT for Cat I and II SRs. Action verbs should be the uncertainties in the SSI analysis ... The used. minimum value of Cv SHALL be is 0.5. ...

3.7.3.2, ---------------------- No objection -------------------------

HLR-FR-D 3.7.3.2, The word "consider" is Qualification CONSIDER EXAMINE all relevant failure modes FR-D2 permissive and inappropriate for of structures ...

SRs. Action verbs should be used.

3.7.3.2, ---------------------- No objection -------------------------

HLR-FR-E 3.7.3.2, Fragility calculations should Clarification CONDUCT a detailed walkdown of the plant, FR-E1 incorporate effects of potential focusing on equipment anchorage, lateral seismic interaction including seismic support, spatial interactions and both structural and functional potential systems interactions (both structural interactions. and functional interactions).

3.7.3.2, Walkdown team qualifications Clarification DOCUMENT the walkdown procedures, FR-E2 should be documented. walkdown team composition and its members qualifications, walkdown observations and conclusions.

3.7.3.2, If a component is screened out Clarification If components are screened out during or FR-E3 by the walkdown team, the basis following the walkdown, DOCUMENT anchorage for the screening should be calculations or some other and PROVIDE the provided. basis justifying for such screening.

3.7.3.2, --------------------- No objection --------------------

FR-E4 3.7.3.2, Masonry wall failures and Clarification During the walkdown, EXAMINE potential FR-E5 potential sources for seismic-fire sources of interaction (e.g., II/I issues, impact interactions should also be between cabinets, masonry walls, flammable examined. and combustion sources, flooding and spray) and consequences of such interactions on equipment contained in the systems model.

25

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution 3.7.3.2, The II/I issues should also Qualification A "II/I issue" refers to situations where a NOTE FR-E5 include situations where a low non-seismically qualified object could fall on and seismic capacity object falls on damage a seismically qualified item of safety and damages an SSC item with equipment, and also situations where a low higher seismic capacity. In such seismic capacity object falls on and damages case, the fragility of the higher an SSC item with higher seismic capacity. In capacity SSC item may be such case, the fragility of the higher capacity controlled by the low capacity SSC item may be controlled by the low object. capacity object.

3.7.3.2, ---------------------- No objection -------------------------

HLR-FR-F 3.7.3.2, ------------------ No objection --------------------

FR-F1 to FR-G4 3.8 3.8.1 The organization of high level Clarification Insert: 3.8.2 High Level requirements and list requirements is inconsistent with all high level requirements consistent with other other sections of the Standard parts of the Standard.

3.8.2 Section number should be Clarification Change the section number to 3.8.3. See changed to 3.8.3. See comment Resolution for 3.8.1.

for 3.8.1 26

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution WIND-A1, The six elements described in Qualification In the tornado wind hazard analysis, ...

Cat. II and III NOTE WIND-A1 provide the ACCOUNT properly for and ...a mean hazard and NOTE details required for the tornado curve can be derived.

WIND-A1 wind hazard analysis and should be included in WIND-A1. INCLUDE the following elements in the tornado wind hazard analysis:

The word properly is superfluous. (1) Variation of tornado intensity with occurrence frequency (The frequency of tornado occurrence decreases rapidly with increased Intensity);

(2) Correlation of tornado width and length of damage area; longer tornadoes are usually wider; (3) Correlation of tornado area and intensity; stronger tornadoes are usually larger than weaker tornadoes; (4) Variation in tornado intensity along the damage path length; tornado intensity varies throughout its life cycle; (5) Variation of tornado intensity across the tornado path width.

(6) Variation of tornado differential pressure across the tornado path width.

NOTE WIND-A1: State-of-the-art methodologies are given ... can be found in Refs. 13, 56, and 57.

Tornado wind hazard analysis SHOULD include the following elements:

(1) Variation of tornado intensity with occurrence frequency (The frequency of tornado occurrence decreases rapidly with increased Intensity);

(2) Correlation of tornado width and length of damage area; longer tornadoes are usually wider; (3) Correlation of tornado area and intensity; stronger tornadoes are usually larger than weaker tornadoes; (4) Variation in tornado intensity along the damage path length; tornado intensity varies throughout its life cycle; (5) Variation of tornado intensity across the tornado path width.

(6) Variation of tornado differential pressure across the tornado path width.

27

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution WIND-A4, The word "consider" is Qualification ... CONSIDER EXAMINE specific features Cat. II and III permissive and inappropriate for ...large early release.

SRs. Action verbs should be used.

WIND-A4, There is no requirement for Qualification NEW SR WIND-A4a: SURVEY the plant Cat. II calculating the population of building and surroundings to assess the missiles. number, types, and locations of potential missiles.

WIND-A4, There is no requirement for Qualification NEW SR WIND-A4a: SURVEY the plant Cat. III calculating the population of building and surroundings and to catalog the missiles. number, types, and locations of potential missiles.

HLR-WIND-B Permissive may should not be Qualification (HLR-WIND-B): ... whose failure may contribute used in HLR. to core damage or large early release.

A requirement missing for NEW SR WIND-B1a: IDENTIFY plant identifying those plant structures, systems and components that are structures, systems and vulnerable to the wind hazards. ACCOUNT for components which are both wind effect and wind-borne missiles vulnerable to the wind hazards. effect.

WIND-B1 The word "consider" is Qualification ... In this evaluation, CONSIDER INCLUDE the permissive and inappropriate for findings of a plant walkdown.

SRs. Action verbs should be used.

NOTE WIND- In the 5th para., the phrase Clarification ...for nonseismic Category I II structures...

B1 ...nonseismic Category I structures should be Category II.

HLR-WIND-C Use of words All and Qualification The wind-PRA systems model SHALL include all important is improper. important significant wind-caused initiating events and other important significant failures that can lead to core damage or large early release.

WIND-C1 The word "consider" is Qualification ASSESS accident sequences initiated by high permissive and inappropriate for winds to estimate CDF and LERF contribution.

SRs. Action verbs should be In the analysis, CONSIDER USE the site-used. specific wind hazard curves and the fragilities of structures and equipment.

28

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution WIND C-2 ---------------- No objection ----------------------

through D-7 3.9 3.9.1 The organization of high level Clarification Insert: 3.9.2 High Level requirements and list requirements is inconsistent with all high level requirements consistent with other other sections of the Standard parts of the Standard.

3.9.2 Section number should be Clarification Change the section number to 3.9.3. See changed to 3.8.3. See comment Resolution for 3.9.1.

for 3.9.1 FLOOD-A1 Permissives should not be used Qualification In the hazard analysis for extreme local in SR. precipitation, USE up-to-date data for the relevant phenomena, it is acceptable to utilize both site-specific and regional data MAY be utilized.

FLOOD-A2 Permissives should not be used Qualification In the hazard analysis for extreme river flooding, in SR. including floods due to single or cascading dam failures, USE up-to-date data for the relevant phenomena. It is acceptable to utilize both site-specific and regional data MAY be used.

NOTE ------------------ No objection -------------------

FLOOD-A2 FLOOD-A3 Permissives should not be used Qualification In the hazard analysis for extreme ocean in SR. (coastal and estuary) flooding, USE up-to-date data for the relevant phenomena, it is acceptable to use both site-specific and regional data MAY be used.

FLOOD-A4 The word "consider" is Qualification ... CONSIDER ACCOUNT for high water levels, permissive and inappropriate for ...

SRs. Action verbs should be used.

FLOOD-A5 Permissives should not be used Qualification In the hazard analysis for extreme tsunami in SR. flooding, USE up-to-date data for the relevant phenomena, it is acceptable to use both site-specific and regional data MAY be used.

HLR-FLOOD- Permissives should not be used Qualification (HLR-FLOOD-B): ... whose failure may B in HLR. contribute to core damage or large early release, or both.

FLOOD-B1 The words "consider" and Qualification In the evaluation of flood fragilities of structures MAYare permissives and and exposed equipment (low-lying equipment on inappropriate for SRs. Action the site, intake and ultimate-heat-sink verbs should be used. equipment, etc.), USE plant-specific data. In this evaluation, CONSIDER INCLUDE the findings of A requirement missing for a plant walkdown. It is acceptable in the identifying those plant fragility analysis for both capacity and demand structures, systems and MAY be based on to apply the standard components which are methodology used for seismic events, with vulnerable to the wind hazards. appropriate modifications unique to the flooding event being studied.

NEW SR FLOOD-B1a: IDENTIFY plant structures, systems and components that are vulnerable to the flood hazards.

29

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution HLR-FLOOD- Use of words All and Qualification The external-flooding-PRA systems model C important is improper. SHALL include all important significant flood-caused initiating events and other important significant failures that can lead to core damage or large early release...

FLOOD-C1 The word "consider" is Qualification To estimate CDF and LERF contributions, permissive and inappropriate for ASSESS accident sequences initiated by SRs. Action verbs should be external flooding. In the analysis, CONSIDER used. USE where applicable the appropriate flooding hazard curves and the fragilities of structures and equipment.

FLOOD-C2 to ------------------ No objection --------------------

FLOOD- D7 SECTION 4: Table A1of APPENDIX A, Chapter 5 applies.

SECTION 5: Table A1 of APPENDIX A, Chapter 6 applies.

5.1 Regarding reference to ASME Clarification See comments for R.G. 1.200, APPENDIX A, PRA Standard, see issues for Chapter 6 of Table A-1.

R.G. 1.200, APPENDIX A, Chapter 6 of Table A-1.

5.1, 3rd papa. The purpose stated lacks clarity. Clarification The purpose of the peer review is fundamentally to provide an independent review of the PRA or SMA, to ensure concurrence with This means reviewing the analysis vis--vis the applicable Requirements in the Standard. The composition and qualifications of the peer review team are important, as is its independence; these aspects are covered in the ASME Standard's requirements (ASME, 2002) that are incorporated here by reference. Other process issues, including the need for a team leader and the need for a methodology for the review, are also covered in the ASME Standard.

5.2-5.4 ----------------------- No objection --------------------------

SECTION 6 6.1 Regarding reference to ASME Clarification See comments for R.G. 1.200, APPENDIX A, PRA Standard, see issues for Chapter 3 of Table A-1.

R.G. 1.200, APPENDIX A, Chapter 3 of Table A-1.

6.2 See Appendix D, general Quantification Delete 2nd para.

comment 1 SECTION 7 -------------------- No objection ------------------------

APPENDIX A ------------------- No objection -------------------------

APPENDIX B ------------------- No Objection -------------------------

Equation (B2) This example does not contains Qualification Select an example of a cutset which will contain non-seismic failures. both seismic and non-seismic failures.

APPENDIX C 30

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution C.1 Incorrect reference to Section Clarification Change 3.5.1.1" to 3.6.1".

Introduction, 3.5.1.1 2nd para.

C.2 The word stylized is not Clarification Delete the word stylized.

Seven Steps appropriate.

(Step 1)

C.2 ----------------------- No Objection -----------------------

Seven Steps (Steps 2 -7)

C.3 ----------------------- No Objection -----------------------

Enhancement s

C.4 Mitigating small LOCA accidents Clarification (2) select a primary success path and an Seven Steps - should be an objective of at alternate success path for the SMA, eliminating Detailed least one of the success paths those elements or paths that cannot be Discussion evaluated for seismic adequacy economically.

(C.4.4) Ensure that one of these two paths is capable of mitigating a small loss-of-coolant accident.

It is important....

C.4 The last sentence under Step 6 Clarification HCLPF capacities are documented for all Seven Steps - is not correct if only one elements in the primary and alternate success Detailed success path can mitigate a paths which have capacities less than the Discussion SLOCA and that success path specified RLE. The element with the lowest (C.4.6) has a lower HCLPF. In this HCLPF capacity in a success path establishes scenario, the plant HCLPF is the seismic HCLPF capacity for the path. The governed by the SLOCA higher seismic HCLPF capacity of the primary success path HCLPF. and alternative success paths is the seismic HCLPF capacity of the plant-as-a-whole if both paths can mitigate an SLOCA or only one path mitigate an SLOCA but the SLOCA path has a higher HCLPF than the other path. However, in the case where only one success path can mitigate an SLOCA and that path also has a lower HCLPF than the other path, then the plant HCLPF is governed by the SLOCA success path HCLPF.

C.4 There is no C.4.7. Looks like Clarification Change subsection number to C.4.7.

Seven Steps - C.4.8 should be C.4.7.

Detailed Discussion (C.4.8 ?)

C.5 ----------------------- No Objection -----------------------

Four En-hancements

- Detailed Discussion (C.5.1 thru 3)

APPENDIX D 31

DG-1138 DRAFT Table C-1 Draft Staff Position on ANSI/ANS 58.21-2003 Index No Issue Position Resolution General Appendix D attempts to expand Clarification Delete Appendix D.

Comment 1 the range of applicability of SMA considerably beyond its stated objectives, in order to support risk-informed applications for regulatory relief. The staff cannot accept a priori the possible enhancements described in the appendix. At the same time, the staff has no basis to reject these enhancements. The staff will need to conduct a case-by-case evaluation of (1) the implementation of a specific enhancement, and (2) the specific results and conclusions obtained. The standard would be vastly improved from a regulatory perspective if Appendix D is deleted from the standard.

General Assuming that ANS does NOT Clarification Revise Appendix D to focus on the applicability Comment 2 delete Appendix D from the of SMA and its limitations in developing risk standard, Appendix D should be insights. If desired, clearly and concisely list and rewritten to focus strictly on the describe possible enhancements in one section risk insights directly derivable of the appendix, with an introduction clearly from a SMA and present stating that implementation of any of these examples of its applicability and enhancements requires specific peer review, limitations. Implementation of and is subject to regulatory review on a case-by-any enhancements will require case basis.

specific staff review.

General Throughout Appendix D, ANS Clarification Revise the statements and examples in Comment 3 takes the position that the plant Appendix D to consider the case where the only HCLPF capacity is defined by success path capable of mitigating a SLOCA the HCLPF capacity of the more has the lower HCLPF capacity.

seismically rugged success path. The staff takes exception to this position. This is only true if both success paths can mitigate a SLOCA or the SLOCA path has higher HCLPF.

The SMA requirement is that only one success path has to be capable of mitigating a SLOCA.

This was previously identified under Index No. C.4 (C.4.6).

32