ML19345G009
| ML19345G009 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 02/17/1981 |
| From: | Rolonda Jackson Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML17150A134 | List: |
| References | |
| NUDOCS 8102190719 | |
| Download: ML19345G009 (26) | |
Text
{{#Wiki_filter:_ __ 02/17/81 0 UNITED STATES OF AMERICA NUCLEAR REGULt. TORY COMMISSION i BEFORE THE ATOMIC SAFETY AND LICENSING APPEAL BOARD In the Matter of ) ) PUBLIC SERVICE COMPANY OF ) Docket Nos. 50-443 NEll HAMPSHIRE, ET _AL. ) 50-444 (Seabrook Station, Units 1 ) and 2) ) TESTIMONY OF ROBERT E. JACKSON, Ph.D. Q.l. Please state your name and present position. A.l. My name is Robert E. Jackson. I am employed as Branch Chief, 'ieosciences Branch, Division of Engineering, Office of Nuclear Reactor Regulation, U.S. Nuclear Regulatory Commission, Washington, D.C. 20555. Q.2. Please describe your educational background and previous positions held. A.2. I received a B.S. degree in Geology from the University of Rhode Island, and a Ph.D. degree in Geology from the University of North Carolina with a specialty in structural geology - rock mechanics. I have been employed by NRC since August 1974 'in the areas of Structural Geology and Fault Geology as applied to the evaluation of the suitability of nuclear power plant sites. My area of expertise includes structural geology of igneous, metamorphic and sedimentary rocks, rock mechanics, teconophysics, petrology, and fault identification and behavior. I am experienced in field analysis. From 1973 to 1974 I was employed by Martin Marietta Laboratories in Baltimore, Maryland as a research scientist. My work for this corporate 8102190b
2-research and development laboratory consisted of a variety of problem-solving programs in rock mechanics relating to quarrying and blastinc. I also contributed to procrars in rapid tunneling and excavation technology. I also was in.olved in evaluating and developing new technologies for the crushed stone quarrying industry. I directed a prccran of investigation of sliding friction as it relates to earthquake source mechanisms and frictional behavior of fault zones. While at Martin Marietta Laboratories, I was an author or co-author or 12 professional papers in these various fields. Fron 1969 to 1973, I was a research assistant and teaching assistant at the University of North Carolina. My activity as a research assistant was in the development of a triaxial rock mechanics laboratory. One project I conducted in this lab was a study af experimental rock dilatancy as an earthauake nechanism. My dissertation was a study of sliding friction in foliated rocks and fault nylonites, including the behavior of fault gouge. Teaching experience consisted of teaching undergraduate and advanced structural geology laboratories as well as teaching field nappinc. I have presented papers at national meetings of professional societies to which I belong, including the American Geophysical Union, Geological Society of America, and the International Society for Rock Mechanics. From September 1978 to May 1979, I served as Leader, Geology and Seismology Section, Geosciences Branch. Since May, 1979 I have been Chief, Geosciences Branch. 0.3. Please briefly describe your responsibilities as Chief, Geosciences Bra nch.
l A.3. In this capacity I am responsibic for the review and evaluation of the neological and seismological aspects of power plant sites and I supervise the Geology and Seismology Sections within the Branch. In this capacity I am under the supervision of the Assistarit Director for Structures and Components. i The Geosciences Branch provides technical evaluation relating to ceology and seismology issues to the Division of Licensing. 1 0.4 Please address the question of whether Dr. Mi:hael Chinnery's hypothesis that there is an empirical relationship between earthquake intensity and earthquake recurrence time has factual validity. A.4 Since Dr. Reiter will be discussing detailed seismological and comparative aspects of Dr. Chinnery's work, I will discuss a part of this question from a more generalized geosciences and regulatory standpoint and provide insight into the proposition that it is geologically and seismologially reasonable to conclude that there may be a limit to the size of earthquakes in a given area and this fact bears directly on the validity of extrapolation of smaller earthquake events to an assumed occurrence of a larger event. I state this even though scientists probably cannot yet provide absolute proof of this. Dr. Chinnery is a well recognized seismologist and he has approached l the question of the possibility of maximum possible earthquakes vigorously. l The quality of his work and the importance of gaining as cuch insight as l possible into this question resulted in NRC Research contractural funding for Dr. Chinnery (NRC Contract: NRC-04-77-019) resulting in two publications (Chinnery,1979a, Chinnery 1979b). An abstract of Chinnery,1979a I (NUREG CR-0563) indicates: 1 -e
This report consists of a re /iew and assessment of available literature on the determination of maximum possible earthquakes. An attempt was made to find evi-dence of earthquake upper bounds based on data garnered from earthquake catalogs, such as the International Seisnological Center. It was found that determination of upper bound earthquakes was not possible because of severe limitation imposed by incomplete reporting and magnitude scale saturations. 1 Upper Limit to Earthquake Size Based on my understanding of Dr. Chinnery's work, I don't believe he is culing out either the concept or the fact that there may indeed be an uppsr limit to earthquake size in a given area. Instead, he indicates that he has not been able to develop a pure scientific basis for establishing such a level. Indeed he carefully notes that " maximum" is not always used with the same meaning. The first definition of " maximum" is the largest possible event that can occur given physical conditions of the source. The second definition includes the concept of probability or the largest event likely during some finite time period with due consideration of engineering design standards. As scientists working in a regulatory role, the Staff must provide due consideration of both possible definitions as they relate J to the final decision on the adequacy of the site and design of a nuclear t power plant. Such an evaluation must include sound scientific and professional judgment in cognizance of applicable regulations and guidance. But in the end, a decision must be made. I don't believe Dr. Chinnery's work has lead. him or other scientists and engineers to conclude that all homes, hospitals, dams, or other critical facilities located everywhere should be designed and built to withstand the maximum earthquake possible anywhere in the world.
Indeed, as a society, we clearly do not do this.
- t is completely necessary that we make such judgments.
Geologic observations must also be included in any corolete evaluation of earthquake hazard for a nuclear facility. Dr. Chinnery notes (Chinnery, 1979a) that the historical record of earthquakes is relatively short in most areas of the world and therefore it becomes very difficult to establish empirical data for maximum possible earthquakes in certain regions. In my view this is one of the primary reasons that the original evaluators of nuclear plant sites (USGS, NOAA) and, eventually, the authors of Appendix A to 10 C.F.R.100 gave such extensive weight to geologic factors. In fact, the USGS continues to maintain this approach to date. This is further supported by noting that future modifications to the Algermissen and Perkins (1976) risk map will include detailed consideration of local and regional geologic considerations as provided by regional experts. Based on both their licensing and scientific experience and expertisc, j the authors of Appendix A indicated that the maximum historical earthquake that had occurred in a given tectonic province 1/ should be used in the t evaluation and determination of the Safe Shutdown Earthquake. I believe that they clearly recognized that a greater earthquake was possible than that derived by the proposed methodology (App. A), albeit the likelihood dL/ Such provinces were envisioned as the classical large tectonic provinces and the authors also clearly favored that specific earthquakes that could reasonably be correlated with structure should not be assumed to occur elsewhere in the tectonic province (App. A, 1 IV, (a), (6).)
l 4
- of its occurrence sufficiantly small not to be considered for evaluation
- i j
of the site. This seems consistent with Dr. Chinnery's second definition of " maximum," a definition which includes the conceot of probability or i largest event likely during some finite time period with due consideration of engineering design standards. In many of the earlier site evaluations, the controlling earthquake and the accelerations assigned were provided i by reviewers who also assisted in the development of Appendix A are, generally, the maximum historical earthquakes. Occasionally, the SSE was 6pparently increased in size when, in their expert technical judgment, geological and seismologic conditions warranted. It is extrenely important to note that Dr. Chinnery's three recommendations for the most promising avenues for future investigations into maximum possible earthauakes are all basically geolooical in nature. First, he notes that j the need for more inf'rmation on the nature of strain and stress fields in seismic zones. Seconc.. he states that there is a need to improve our under-standing of the strength of crustal materials because it seems likely that the true uoper bound is controlled by the size of the region accumulating stress. Thirdly, he notes that information from geological and geomorphical j data on long term fault slip may place some constraintt on largest possible i earthquakes. Although, geologic arguments of which I am aware do not conclusively demonstrate that an upper bound exists in a given region, many observations I have been made that would indicate that it is indeed reasonable to reach such a conclusion. Slemmons (1977) as shown in figure 1 and 2 (attached), based on a comprehensive compilation of many earthquakes occurrences where surface i 4 f
i l rupture occurred, a relationship between earthquake magnitude and length of surface rupture and earthquake magnitude to recurrences interval and strain rate across faults, respectively. Dr. Slemmons (1977) notes, Modern engineering practice includes design of structures to accommodate strong earthquake motion. Most active faults have, by man's time framework, very long intervals between the shallow focus faulting events which are associated with most damaging earthquakes. The recurrence intervals or time gaps for these events are measured in hundreds, thousands, ten of thousands, or even hundreds of thousands of years, These long time intervals, in combination with the short and often non-representative nature of the historic and seismologic records, re-quire the geologic methods be used by alternatives or supplements to historic methods of establishing design events." (P. 10) Dr. Slemmons references numerous other studies which compare earth-quake magnitude to geologic observation. Wyss,1979 in his paper, "Esti-mating Expectable Magnitudes of Earthquake from Fault Dimensions," notes that the evaluation of seismic risk at locations where sensitive man-made structures are planned depends critically on a correct estimate of maximum expectable earthquake magnitude and recommends a relationship between fault area and magnitude be used (Figure 3, attached, Wyss, Figure 2). Recent studies performed by Woodward-Clyde consultants for the San l Onofre site review (Woodward-Clyde, 1979; Southern California Edison, i l 1980) provides a relationship establishing a Maximum Earthquake Limit as a result of a relationship between average geologic slip rate and l earthquake magnitude (figure 4, attached, Figure 361.45-4 SONGS FSAR). l l 1 x ....y.. -, _. -..,,-_. ~._ ,__.7 _v ,y r,,,
Y . 1 Obviously, all the references above relate to establishing magnitude estimates in areas of active faulting and require reliance to some extent on surface rupture observations. These techniques do, however, indicate that reasonable magnitude estimates can be made using proper consideration of geologic parameters. It should be noted that the eastern U.S. does present a different problem since surface faulting is not common in the east as compared with i the western United States. In the east, however, we do know that the average rate of tectonism is relatively slow. The absence of surface faults, the subdued topography and relative slow rate of uplift and denudation in this region (with due recognition of the anomalous areas of Charleston, South Carolina and New Madrid, Missouri areas) indicate that strain rates are relatively low, therefore, leading inherently to longer return periods for earthqt".kes. As Dr. Chinnery also notes, crustal properties also affect maximum earthquake potential. I would expand that to indicate that otner strength param-eters and nature of fault gouge material or mineralization, depth of burial with equivalent consideration of temperatures, pressures, mineral conposition (plutonic bodies), strength of asperities along faults, are all parameters that relate to the ability of a fault to sustain strain (stress) accumulation to achieve the ability to generate a larger earthquakes. Although we may not be able to resolve conclusively that the greatest earthquake will not occur, I believe that geologic observations indicate that upper bounds to earthquakes over long periods of time are certainly geologically reasonable. Dr. Chinnery's hypothesis that there is some em-w, --n--,- -cer.-., ,,,m,.,v-, e -- +w, e v nw -,.,,..w-awe,, ng ,y. n,- n .,e,.. -,m -y---- e., -, -- -e n.,
chance that an earthquake larger ti.3n the maximum historical can occur is one working hypothesis which should not, and has not been disregarded. Based on the above discussion, I believe that the question of the validity of Dr. Chinnery's method must be viewed in the total seismic design context. In that context, the answer to the question of its validity remains 4 equivocal, however, because strong consideration must be given to the realistic possibility of a geologic basis for limitation to earthquake size in a region. Q.5. Assuming only for the purposes of ergumer.t that Dr. Chinnery's methodo~ logy has factual validity, what is the impact, if any, of the application of Dr. Chinnery's methodology to the present seismic design of Seabrook as intensity VIII,.25g. Reg. Guide 1.60? A.5 First, I believe that it is incorrect to characterize I Dr. Chinnery's papers (Chinnery,1979a,1979b) as a total methodology. His work on this issue is better represented as linear extrapolation of frequency intensity data to a determination of expected probabilities for the occurrence of larger earthquakes. He indicates that the assumptions outlined in the paper seem to be a useful working hypothesis. This approach l is far different from a development of a total methodology for making a complete seismic design review recommendation. A common error in the treatment of a site review work by expert techrical professionals in a specific discipline is that the expert necessarily is unwilling or unable to consider the total methodology. As a regulatory revie.v staff, it is necessary to evaluate each parameter with a good knowledge of how that parameter will be utilized in the total finding. As an example, many
. expert seismologists focus on consideration of peak ground acceleration values and, indeed, based on the requirements of Appendix A, the best available judgment of such a value is important. Many experts are not i aware that.he response spectrum is far more important to total design consideration than the so-called "g" value (peak ground acceleration). Even with the same "g" value used as an anchor point, the actual e design could vary drastically dependinq upon the response spectrum used. For example, prior to the development of Reg. Guide 1.60, spectra which are somewhat less conservative than Reg. Guide 1.60 were comraonly utilized. Indeed, even under current requirements, response spectra less than Reg. Guide 1.60 may be utilized in certain instances where significant justification for use of such spectra exists. My point is that by being aware of the total view, a better judgment can be made. For example, if it is known ahead of time that a particularly low spectra is to be utilized then it probably is wise to require a higher anchor point and vice versa. As a review staff and under the guidance of Appendix A, we view each review step as an inf.erface, i.e., the geologist considers what the seis-mologist is trying ta achieve and the seismologist is in turn cognizant of the structural cagineer's needs. The most important aspect is whether the final answer is reasonably correct and conservative. Appendix A was written by geologists, seismologists, and engineers and,I believe, was intended to be used with true recognition of the above i described interface process. i y e -s.e -,..,, + -, ~ ,e---- -+,+,4 .-m.. w
J. Dr. Chinnery's extrapolation is used by NECNP to estimate that the maximum intensity that should be used for the Seabrook seismic design 1 determination should be a Modified Mercalli Intensity IX. I do not see how this finding (of a IX) can be made using the Chinnery extrapolation without due consideration by NECNP of what is considered to be an acceptable quantita' ave level of conservatism for the SSE in the region (seismo-tectonic province or zone). The only apparent reason for indicating that an MM IX is the correct intensity value as compared to an VIII must be basically due to a perception of what constitutes an acceptable level of hazard to an NPP by NECNP due to an earthquake. Appendix A generally specifies the use of the maximum historical earthquake i a given tectonic 7 province be used unless geological and seismological conditions warrant j something greater. j This guidance contained in Appendix A to 10 C.F.R.100 therefore implicitly specifies an acceptable level (use of maximum historical earthquake) of earthquake hazard for an NPP, albeit not in quantitative terms. Although the NRC Staff follows Appendix A to 10 C.F.R. Part 100, neither Dr. Chinnery nor Dr. Trifunac (as will be discussed in the next l questian) feels that they are obliged to do so as evidenced by their 2 response to ir' rrogatories or deposi'. ion answers. f 2/ Dr. Chinnery (NECNP responses to Applicants Interrogatories) in response to question 77 indicates that there are several parts to i Appendix A which are phrased very loosely from a scientific point of view. He further indicates that Appendix-A should specify an i annual probability level which constitutes an acceptable level of risk. In addition, he further indicates as an example, that the SSE shall be that earthquake which shall be established as leading 4 to actual failure of critical plant components with an annual probability not exceeding 10-7 and that such a computation should include both earthquake and plant component considerations. In h9 suggestion, where Dr. Chinnery clearly recognizes the need and potential trade-offs the necessary seist. ology and engineering inter-action required in evaluating earthquake hazard as part of the total seismic risk. ~._
I 1 l Over the last several years the NRC Staff, both in licensing and research has pursued the possible use of probabilistic methods. The NRC Office of Research has implemented the Seismic Safety Margins Research Program (SSMRP) for the purpose of assessing conservatism in design and i construction. As part of the ongoing review of a number of older operating reactors, the ONRR has contracted for the development of a Seismic Hazard Analysis Methodology for the Eastern United States (TERA, l 1980 - NUREG/CR-1582, Vols. 2 and 3). Af ter review of this information, the NRC has utilized this work to provide an initial review and recom-mendation for site specific spectra at Systematic Evaluation Program I (SEP) sites (Jackson,1980). l These recommended spectra will be used in conjunction with expert engineering judgment to assess the total seismic adequacy of the plants j under review. Initial comparison of these site specific spectra for the sites evaluated indicate that they are generally less than would result from a literal application of Apperdix A 10 C.F.R.100 and the current standard review plan throughout the frequency range of interest for nuclear power plants (Jackson,1980). The SEP methodology incorporates into its development a solicitation of expert opinion and includes in it a distri-bution of earthquakes, including earthquakes greater than the historical maximum. It also included a variety of seismic source zones provided by the solicitatio of expert opinion. Based on the discussion above, I believe that Dr. Chinnery's approach, if considered in the total seismic siting and design context, would have no impact on the adequacy of the present seismic design basis for the Seabrook site. I
k I 13 - J Q.6. Please address the ouestion of whether the Staff's methodology for correlating vibratory around motion (acceleration) is consistent with Appendix A to 10 C.F.R. Part 100. i A.6. As with the previcus subissues, some parts of my response to those l ? issues are also applicable to this isnie. Stepp and Coplan indicate ] (Prefiled Testimony,1975, p. 7) that "in keeping with our practice, we f chose a value of 0.25 g as being representative of the mean of the data I for an intensity VIII." The staff practice for reviewing a proposed seismic f design acceleration ancnor point ("g" value) ar.d response spectrum consists of a number of steps. The final product of these steps, the response i spectrum, is the most importa7t aspect since that is what the engineers use in their analysis and design of a nuclear power plant. For an eastern U.S. site, the initial step ccosists of choosing the controlling earthquake, I which is the maximum historical earthquake in a given province. Its size { is usually expressed in Modified Mercalli Intensity. Using that intensity, for the Seabrook site it is MMI VIII; it has bh$n' the staff practice to l l use an intensity-acceleration relationship to obtain an acceleration value l which is used to anchor a design response spectrum. Various such relationships l existed in the literature. Coulter, Waldron and Devine, 1973; Neumann, 1954; Trifunac and Brady,1975 and Murphy and O'Brien,1978 tre a few such P relationships. For the Seabrook site, an acceleration value of.25g is used to anchor a Regulatory Guide 1.60 spectrum. The NRC staff has always been cognizant of the spread of the acceierations values that have been recorded and plotted for a given intensity. In earlier siting investigations and reviews, a variety of relationships and response sepctrum had been utilized to. determine the seismic design for NPP. With 4 9 ,w.wm~,,, ew.,,,-._,,,,,,r .--.sw ,, w ~,..,_ y-,w,,,y,-.,,, .-,mu,,,,,.wwyp,y ,.~.,.,_,.,,.ms,,- ,,.,w,--y.-,m,~,,.,.e. f m m-y,,o ww.,_w.,
i the development of the Standard Review Plan (SRP) (1975) section 2.5.1-5, the staff standardized the procedures used to facilitate the review effort by providing one method that is considered acceptable to the staff. I might also note that the Standard Review Plan is not intended to supplant either 4 regulation or regulatory guides, but is intended to relate an acceptable interpretation, although exceptions are clearly allowed as long as supporting i bases are provided. Since issuance of the SRP, the staff has favored the i 3/ use of trend of the mean of the Trifunace and Brady (1975) relationship.~ 4 la an effort to further improve information relating to the intensity acceleration relationship, the NRC contracted for further development by encouraging expansion of the available data base. This work resulted in the publication of f1urphy and O'Brien,19/8, NUREG-0402. This more recent l analysis of intensity-acceleration using a large data set results in a mean pesk acceleration of 0.159 for an intensity VIII earthquake. This relationship has r.ot been adopted by the staff because it does have shortcomings in terms of the data gathering =nd the consistency of worldwide data as noted by Dr. Trifunac in his deposition (Tr. p.10,11) c:id primarily because, when values from this relationship are used to anchor R.G.1.60, the staff's scientific judgment is that it may not describe a sufficiently i conservative seismic design. The acceleration value utilized must be chosen in the total context of the how it is to be used, that is with a R.G.1.60 spectra or some lower spectra; the total methodology should result in a conservative design response spectrum. This methodology, when applied to jV Dr. Reiter in his testimony has addressed this question in detail. i e i --=ec ,--.,r-r-.v.,,,,- --m -.,x n- ,,,-1.w,,- w, yv--,.,,-,,-v w-,.,, ,w-ww<, yr,,--w,,,w--...-~w,.w,,- ,,-#s.w-y--..,-e,- y,-w.- ,-, -- r r,< -yy-
- - - - - _ ) 't T The Seabrook provides a siesmic design specification which is conservative. seismic design of.25g and Reg. Guide 1.60 is as high as any seismic design a specified for a nuclear power plant in the U.S. east of the Rocky Mountains. Th.'e designs have been developed over the course of many years by many experts and have been reviewed by a variety of NRC and AEC staff members as well as U.S. Geological Survey and NOAA prior to the presence of an NRC staff and the Advisory Committee on Reactor Safeguards. A comparison of the.25g, R. G.1.60 spectrum with spectrum developed for specific earthquakes is essentially at the upper limit of those di.ta available in the western U.S. for Intensity VIII (Figure 5). It is also between the mean plus one standard deviation and maximum for earthquakes of about the 3 ] same magnitude, distance, and site conditions (Figure 6). There is virtually no disagreement between the experts involved in the Seabrook proceedings that I am aware o, (Dr. Stepp, Dr. Newmark and r Dr. Trifunac) that the acceleration value determined and assigned for purposes of developing a seismic design should be something less than the absolute peak instrumental ground acceleration measured for a given intensity. Some of these experts believe that a " reference" or "effe:tive acceleration" should be applied when used in conjunction with a proper design response spectrum (i.e., R. G.1.60 or some other acceptable spectrum). There is, however, disagreement as to how one quanititatively derives this value. This decrease is accomplished differently depending on the background, skill, and expertise of the individual or group making the determination. " Effective" peak acceleration is defined in Geological Survey-Professional Paper 1114, Procedures for Estimating Earthquake Grou,d liotions, (1980), as the peak ground acceleration after the ground-motion record has been filtered to remove the very high frequencies that have little influence
_ upon structural response. Such a definition is by no means widely accepted yet and due to this fact, NRC ORES has recently initiated a large research project with Woodward-Clyde consultants entitled, Effective Peak Acceleration for Nuclear Power Plant Design. In their technical proposal (Woodward-Clyde,1980), they note (p.1-1) that the development of design or " effective" accelerations from free field instrumental peaks has been d the subject of considerable study in the past several years. It is noted that the characteristics of ground motion are approached from several perspectives including consideration of aspects significant to structural performance, cyclic loading, and variations of motions over the width of i the foundation. They indicate: "Although the available research results and empirical evidence I have provided considerable insight into ground motions and structural response and performance, no procedures or approaches for developing effective peak accelerations and response spectra from instrumental values has won wide acceptance at present." (Woodward-Clyde, 1980, p. 1-1). Since no precise formula exists, such a finding still remains an expert judgment. Appendix A to 10 C.F.R. Part 100 clearly contemplates such judgmeits. As wr.h any regulaticn, modifications to Appendix A are permitted and contemplated to allow for better quanitification and clarification of the intent of the regulation. The strmigth of any such regulation like Appendix A lies in its being used with sound judgments by technical professionals who are cognizant of the total seismic design consideration. As previously discussed, neither Dr. Chinnery nor Dr. Trifunac feel bound by Appendix A. Dr. Chinnerv suggests how he would rewrite the regulation (NECNP Responses to Applicant's Interrogatories, No.17). l l
17 - i Dr. Trifurac (Deposition, p. 5) indicates that Appendix A sets out general guidelines by which the acceleration level is chosen for a given SSE earthquake. However, it does not tell by what procedure or procedures one should pick accelerations for ground motion. In response to the question as to his opinion as to how Appendix A should be used to establish an acceleration anchor point for a standardized spectra, Dr. Trifunac stated (Deposition, p. 6), "That opini7n is that Appendix A states that you are to select the largest vibratory grou id motion that can be
- perienced at the site, and my interpretation of Appendix A is that, as an engineer who does this type of calculation, I have some degree of aedom to select what seems the most rational and most reasonable present state of art of accomplishing that, and there are several alternatives that can be selected that depend on the local conditions and the local information that is available at the particular site." Tne scope of 10 C.F.R. Part 100 (100.2) states:
"In the application of these criteria which are deliberately flexible, tne safeguards provided - either site isolation or engineered features - should reflect the lack of certainty that only experience can provide." The purpose of Appendix A to 10 C.F.R. Part 100 indicates: "These criteria are based on the limited geophysical and geological information available to date concerning faults and earthquake occurrence and effect. They will be revised as necessary when more complete infcrmation becomes available." (App. A I. Purpose). There is a need to consider modifications to Appendix A to both clarify the intent and to include a discussion of uncertainties, the role of judgment, and to encourage more rapid aesimilation of changes by placing specific procedures in Regulatory Guides rather than in the regulation which takes an extreme effort to modify. Until such revisions are made to reflect the
. i experience gained both in making siting decisions and in the vigorous applicati)n of the regulation, Appendix A must be utilized with flexibility and judgment in light of the total methodology for establishing a tenservative seismic design. The staff's methodology for establishing the total seismic design including correlating vibratory ground motion with intensity is consistent with Appendix A to 10 C.F.R. Part 100 because this regulation must be applied in a conservative manner with sound professional judgment. This has been the case with respect to Seabrook Station, Units 1 and 2. 4 I t
References, Algermissen, S. T., and Perkins, D. M.,1976, A Probabilistic Estimate of Maximum Acceleration in Rock in the Contiguous United States, U. S. Geological Survey Open File Report 76-416, 45 p. Chinnery, M. A.,1979a, Investigations of the Seismological Input to the Safety Design of Nuclear Power Reactors in New England, NUREG/CR-0563, O. 'S. Nuclear Regulatory Commission, p. 72. Chinnery, M. A.,1979b, "A Comparison of the Seismicity of Three Regions of the Eas tern U. S.", Bull. Seism, Soc. Amer., Vol. 69, No. 3, pp. 757-772. Coulter, H. W., H. H. Waldron, and J. F. Devine,1973, Seismic and Geologic Siting Considerations for Nuclear Facilities, Proc. World Conf. Earthquake Engr., 5th, Rome, Italy. Hays, W. W.,1980, Procedures for Estimating Earthquake Ground Motions, U. S. Geological Survey Professional Paper 1114, 77 p. Jackson, R. E.,1980, Memorandum to D. Crutchfield, Initial Review and Recomn.endations for Site Specific Spectra at SEP sites, U. S. Nuclear Regulatory Commission. Murphy, J. R., and O'Brien, L. J.,1976, Analysis of a Worldwide Strong l Motion Data Sample to Develop an Improved Correlation Between Peak j Acceleration, Seismic Intensity and Other Physical Parameters, Computer Sciences Corporation, NUREG-0402, U. S. Nuclear Regulatory Commission. Neumann, F.,1954, Earthquake Intensity and Related Ground Motion, Univ. Press, Seattle, Washington, 77 pp. O'Brien, L. J.,1980, The Correlation of Response Spectral Amplitudes with Seismic Intensity, Computer Sciences Corporation, U. S. Nuclear Regulatory Commission, NUREG/CR-12d9. 4 Slemmons, D. B.,1977, Faults and Earthquake Magnitude, Misc. Paper S-73-1, U. S. Arr.y Engineer Waterways Experiment Station, Vicksburg, Miss., Report 6, p. 166. Southern California Edison Company and San Diego Gas & Electric Company, 1980a San Onofre Generating Station Units 2 & 3, Response to NRC Questions 361.37 through 361.62. Tera Corporation,1980, Seismic Hazard Analysis, Vol. 2, A Methodology for l the Eastern United States, U. S. Nuclear Regulatory Commission, NUREG/CR-1582. l Tera Corporation,1980, Seismic Hazard Analysis, Vol. 3, Solicitation of - Expert Opinion, U. 5. Nuclear Regulatory Commission, NUREG/CR-1582. i Trifunac, M. D. and Brady, A. G.,1975, 'On the Correlation of Seismic Intensity Scales with Peaks of Recorded Strong Ground Motion," Bull. Seism. Soc. Amer., Vol. 65, p. 139. Woodward-Clyde Consultants,1979, Report of the Evaluation of Maximum Earthquake and Site Ground Motion Parameters Associated with the Offshore Zone of Deformation San Onofre Nuclear Generating Station, Report prepared for Southern California Edison Co., Rosemead, Califtrnia. Woodward-Clyde Consultants,1980, Effective Peak Acceleration for Nuclear Power Plant Design, Technical Proposal, Response to RFP No. RS-RE3-80-192. Wyss, M.,1979, Estimating Maximum Expectable Magnitude of Farthquakes from Fault Dimensions, Geology, Vol. 7, p. 336-340, July. i
Figure 1 DO:: t i j i f- [~ ,./- soo r-l- 300 1 1 o P ( t k ,c: l a, /. _~ ~ 3 5 y ios g o~ so 3 t .s if ? 33: l# z 1 = 5 g g w e i E E /. ? t 5 [' :/ [ gi io a s ? e / '/ a f [ e [ 5 3 ~ [ s ~ r 8 "" ^
- 8"C""'
III .cno 'cr. eu rvers - _.rss 4. > ease"% J~ i ~ ~ ?: 12 '.?. ': E ' ~ .. e ( l,2 U, y: .w s /u F/ J f g i $:/ si y n i = x; g g - see.4 a iroes / j I I /l I / li i a s a 3 s e s tamTHWant wag #estvoE Figure 1. Belation of carthquake magnitu:le to length of the ene of surface faulting wN c fu, UJL 4
@igure 2 23.00C.003 F AUL T114* ats'LL C' e' tee L*:'"' Wh thVh_t 94(&AVMfA R W h = = "' Y /4
- n n u rn,a,a = =
E v1Cf 40t OF A T a vi t, 5,... ag y,, A < N,_ _Dk f," WW.CCS E WDf G ~ 3 373 m.'s 13WMT. M/Wh%9W'/M f. 8
- A f
_} . _ [- d '0-l 'I C V ' w ^j 'n* M2 / i
- /
y 4 ~ f [ [ LLE CE 50 E _~ ( E ttt%;t at;%,v %ce pg4;g 2 Tf 0TC%<C 80s%;aa ts ,3, 5.000 3 = g .i-:0 ew v 4 g g g I 5' 7.....,,,,, e,, :..,, e b c%Y...'ak'EaYS$d!Drr. f t E. a.%:a a m: m., s,, c --y [_ ~ 70%t i A%L C;g ag est., ; l z I i iststw u f I / 100 / / / q. = .f SC d o 4 Y so / / l I I/ 2 E a m T ** Q u a et C. w a 0. t u C E Figure 2. Relation between tino or recurrence interval (in years), strain rates across fault zor.as (ir cr/yr), and earthquake magni-tude, using data free Table 11 in<l Figure 25 and using methods based en Wallace.30 Matsuda5I classified the rates shcvn by letters on the right side of the diagra
r.: n ~ 1 i e i i i 4 0 10 T Ms = log A + 4.15 } A g C 0 E 00 a 4 10 A 0 4 A 3 Ao.o e o A Q og A 4 A U A o O m D 3 @ 10 ja a AA A 8 A w x 4* co.A 4 ( 3 ,Ag' 6 A 2 A 48 r I A i 2 _A E10 g W e I 60 70 80 90 MAGNITUDE Fepure 2. Earthqualt rupture area as a function of magnitude (M.J"3.M
- 1. Data from Tables t
I and 2. For facilitating identifkation, data from Kanamori and Anderson (1975L Ovn and Winar (1977). and Ambrase)s and Zatorck i1%8) are differentiated by 9. lid dots open circles, and open triangles revectivel). Large car:1. quakes for which Mw mas estimated from rupture area mere not used. I i f f li
P1gure 4 ___l__ L i ~,C; _..c.__ , - __ o i + ~ J l ~ m t v' w a _a i; o r---.,--,i--, lQl 'E,l l _j 1 ii i ~~) s g j i,-{., 10 is E _, _, _ _ i ,,, i i _o_ _, _] .a a. i,. _2, v ie 9 13: iic#s]}Y,l J 3 L.m i ~ 9.10.14 y [i liiifo_T.,i l 114Q J J: I E ~ 18 .I i 1 F-u f{lTt i ____s ] ' ' ": ~ , -l,ele i e i C C,- l3~l u. ~ E au - l C~l c i e -] E 1.0 4 m h I I e l t b \\ Line Bounding ] [ ~----7'~ 3 E xtremes of l Bracketed Ranges m 9 } of Data (MEL) o l o J O wa
- 0.1
. - _ -- n O- ] _J ew 4 l r m i j d For Fault Names and Data Base See Table 361.45 - 3 i i 1 0.01 5 6 7 8 9 EARTHOUAKE MAGNITUDE, M 3 E XPL AN ATION Figure 361.45 - 4 Maximum Earthquake Limit O Maumum ;- .,entai record'"9 Geologic Slip Rate VS Historica' O Manmu-cinstrumente estimate Magnitude for Strike-Slip F aults - Ra.cr cover which smaller earthquakes occur Bo= sipetsents most !ik elv range of geologic 0 st p rate data and possib'e error ranse of 20.2 in M.egnatude calculation. Dashed box represents g= uise itsinty of pre instrumenta! estimates. I g % .- O Us 3s W
Figure 5 100.0 R.G.160 0.25g o w w ~ 5 10.0 l'\\ t-i o l
- .i:
o f a ~.:. UPPER I / 'j w BOUND p 1.0 _a w cc MEAN ~ 6 a LOWER ' 3 BOUND ~ ww 1 0.1 _- I i DATA SET - 5 WESTERN U.S. RELORDINGS l AT INTENSITY Vill (MAXIMUM HORIZONTAL COMPONENTS) l l 0.01 i i i i iiiil i i i n i i iil i i i v i ii 0.01 0.1 1.0 10.0 t PERIOD, SEC l FlGURE 3.4: Average PSRV Spectrum Compared with the Upper and Lower Bounds of the Observed Data, Modified Mercalli Intensity Level Vill (NUR EG/ CR-1259)
Figure 6 i.1EAN. MEAN PLUS ONE STANDARD DEVIATION. MAXIMUM AND MINIMUM RESPONSE SPECTRA FOR THIRTEEN US AND ITALY EARTHQUAKES - 7% DAMPING 100 R. G. 1.60 0.25g / / 0 ,/ MAXIMUM / / MEAN+ / 1 STVD. DEV / MEAN / 1 / / / I MINIMUM l 0.1 / l l 1 i O.1 .1 1 10 PERIOD (SECS) Response Sepctra from nearby (Rs27), Magnitude 5.8 ( 0.5), Earti quakes Recorded at Rock Sites (Justification of the Seismic Design Criteria used for the Sequouyah, Watts Bar and Bellefonte Nuclear Power Plants, Phase 11, TVA, August 1978)}}