ML20079M023

Submits Results of Review of Geologic & Seismic Data in FSAR Sections 2.5.1 Through 2.5.3 & Other Related Studies.Site Geologically Acceptable & Meets NRC Requirements

ML20079M023
Person / Time
Site:	Catawba
Issue date:	11/17/1983
From:	Newton C, Planner H LOS ALAMOS NATIONAL LABORATORY
To:	Rolonda Jackson Office of Nuclear Reactor Regulation
References
ESS-3-1411-83, NUDOCS 8401270123
Download: ML20079M023 (4)

Text

-

.. y.

,, - d. : , . , .

h hhh DATE; IN REPLY RsFsR TO-November 17, 1983 ESS-3-1411-83

,LosAlamos NatiomiLaboratory un sro" C335

' Los Alamos.New Mexico 87545 retsPuous:

(505) 667-8464 Dr.-Robert E. Jackson Chief, Geoscience ' Branch Division of Engineering U.S. Nuclear Regulatory Commission /

MS P-514 '

D~ N '

Washington, DC 20555

SUBJECT:

GEOSCIENCE CASE REVIEWS II, CATAWBA NUCLEAR STATION, SOUTH CAROLINA The Los Alamos staff working on this case has reviewed the geologic and seismic data compiled for the Catawba Nuclear Station, South Carolina, which is presented in FSAR (Final Safety Analysis Report) sections 2.5.1 through 2.5.3, together with other related studies developed since the Construction

. Permit Safety Evaluation Report. This review and evaluation has been con-ducted in accordance ~1th NRC Standard Review Plans sections 2.5.1, 2.5.2, and 2.5.3 and Appendix A to 10 CFR Part 100.

The. result of the geology review can be sunanarized as follows:

1. Detailed field investi;;ations of faulted and brecciated bedrock exposed during site excavation indicate that faults at the site do not- represent a safety hazard to . the plant. Last movement on these faults has been demonstrated by radiometric dating to have occurred prior to 86 +/- 30 mybp. No capable faults, therefore, are known to exist at the plant site that would cause earthquakes to be cente- J there.

2. No other evidence exists to suspect differential dis-placement at or near the ground surface in the plant site area.

3. Due primarily to the uncertain relationship between Charleston, SC, seismicity and regionni tectonic structure, there is presently no reason to believe that an 1886 Charleston-type earthquake could occur outside the Charleston Epicentral Area seismotectonic region.

4. With the plant located over adamellite (quartz monzonite) crystalline bedrock, the potential for detrimental subsidence affecting the plant is considered non-existent.

The Los Alamos staff has concluded, based on the geologic data compiled in tha' FSAR and related studies, that the site is geologically acceptable and meets the requirements established by the NRC.

8401270123 831117 PDR ADOCK 05000413 PDR

[fJ fg[ ,

A An Equal Opportunity Employer / operated by University of Cahfornia i

1

- , , , , - , , - - , - - , --.,iy- ,,-----,--my n--,r-n-+

. f.

Dr. R.~E.lJ:ckcon, US NRCE November 17, 1983 ESS-3-1411-83, page 2 s

The Los Alamos staff wishes to express its concern, however, for the general ~ 1ack of compiled structural data for regions surrounding the site and

. for lack of'an interpretation of available in situ stress measurements in relation to regional structural. orientationE This concern is based on exist-ing provisional tectonic interpretations for .the Coastal Plain and Piedmont

- that" frequently invoke deep ' structural features'. FSAR Appendix 2A, which

. contains pertinent; geclogic and seismologic information developed since first submittal of the FSAR, is a reminder of the rapidly changing tectonic inter-pretation for, the Southern Appalachians. Certainly the many hypotheses sum-marized in this Appendix will change or be modified within the lifetime of the reactor facility. Therefore, there is need to stress the regional data, not 7 only to meet the basic acceptance criteria but-to anticipate the consequences

. from a number of tectonic scenarios.

- A number of measurements' for stress field magnitude and direction have been made for the Southern Appalachians and Coastal Plain. Some of these data are summarized in FSAR Table 2.5.1 through 2.5.3. Additional stress field measurements are cited by Hainson (1977; 1981), Tarr and Rhea (1982),

Wentworth and Mergner-Keefer (1982), and Zoback et al. (1978). The data are-sparse but valuable for initiating a regional evaluation. In a few cases there is ' sufficient information, at least in theory, to indicate strike and range in dip of possible fault orientations favorable to reactivation. There-fore, the PSAR could report on the orientation of tectonic structures located

'iri the Charlotte, King's Mountain, and Carolina Slate Belts, which could constitute a potential hazard to the integrity of the plant based on such an

. -analysis. Separate summaries on fault, dike, and joint orientations for each of the three belts (data that are not available ' from the FSAR) would be resourceful to making this analysis now or as new data become available.

The seismology review was restricted to the issue of reservoir-induced seismicity .(RIS), and the summary opinion is that the maximum RIS potential falls below the design levels for Catawba. The supporting observations and interpretations are as_follows.

The impoundment of artificial lakes is now known to induce seismic

~ activity under certain. causative conditions that are not fully understood.

. The first observed case was the filling of Lake Mead behind Boulder (aka Hoover) Dam in Arizona. Boulder Dam is 221 m high and can store 38 km3 of water. In 1936, with' a 100-m-high water column, the first shocks were felt in

! this area mistakenly thought to be aseismic. Dean Carder was among the first, in 1945, to associate the water impoundment with the seismic activity there.

l Therefore, when the Catawba River was dammed in 1904 and, again, when Lake Wylie was filled to its present level in 1925, there was no recognized basis o :for monitoring local seismicity.

1 During the almost 80 years of the pond's existence at Lake Wylie, there have been no reports of local scismicity, at least of a magnitude to warrant an entry into earthquake catalogs of the southeastern United States. Given ss a

_ _ _ _ _ -~

?9 Dr. R. E.-J ckton, US'NRC Navsmbar 17, 1983

+

' ESS-3-1411-83, page 3 the population of Charlotte, NC, and the reporting of an intensity V earth-quake near there more than 100 years ago, the lack of catalogued seismicity is a credible indication of no induced earthquakes of magnitude, ML =. 3 or greater at Lake Wylie.

.There have been several recent occurrances of RIS in the Piedmont physio-

' graphic province. Probable to definite RIS has occurred at Lake Sinclair, CA, ,

and Clark Hill' Reservoir, Lake Keowee, Lake Jocassee, and Mouticello Reservoir '

in South Carolina. Lake Wylie ranges from about 25 to 75% in volume and maximum water depth compared to these other Piedmont reservoirs. The most con-

~

L vincing cases.of RIS are at Monticello, where the maximum magnitude observed is ML = 2.9 in more than 5 years t f monitoring, and at Jocassee, where the largest earthquake (Mt = 3.9) occurred about 6 years after full pond was reached. The sleeper was at Clark Hill, where an ML = 4.3 earthquake was observed 22 years af ter impoundment; this is an ambiguous case of RIS along an old fault. Most of the Piedmont RIS can be characterized by shallow, diffuse foci associated with jointing in heterogeneous metamorphic rocks within hetero-geneous stress fields, the same kind of geologic environment as Lake Wylie.

On a worldwide basis, RIS has occurred at magnitudes suf ficient to damage engineered structures. These have occurred in tectonic environments where these earthquakes, although in some cases not anticipated, were shown by site studies to be normal with respect to fault capacity and regional stresses.

These environments differ in many ways from Lake Wylie: the dans are built across much deeper canyons, the plate deviatoric stresses are much greater, and the fault (s) moving is(are) classified aa active.

Since the causes of RIS involve a complex interaction of reservoir loading, pore pressure increases and/or fluctuations, the ambient stress field and geologic structure, we are in no better position to predict RIS than we have predicted natural earthquakes. Two possibilities are obvious for RIS differing from natural seismicity. One is that the RIS is brought forward in time, but at the same magnitude distribution as would have naturally occurred over a longer period. The other is that the RIS upper magnitude will be greater or lesser, depending on whether strains are brought to a sudden total

,raliaf or are more slowly cycled downward; that is, the frequency distribution of earthquake magnitudes described by the seismologist's b-value may increase or decrease. The change in b-value probably depends upon whether the focal mechanisam are thrust type, strike-slip, or. normal (extensional) movement.

In spite of the variety of possibilit[es for reservoirs inducing ear'thquakes, a comprehensive study of the potential for naturally occurring

. earthquakes at a given site should be sufficient. for providing bounds on potential RIS. Such bounds incorpo~ rate RIS s'ob' ervations, interpretations of site conditions, and a margin for error in judgment. Our analysis of the RIS potential at Lake Wylie is dominated by the lack of observed RIS with an estimated threshold of M t;= 2.5. Furthermore, the site conditions and dam-1ake parameters resemble mo'st nearly those at Lake Monticello, where we have

~

t J '

h io - - ..

W Dr.,R. E.~ J:ck on, US NRC Novsabsr 17, 1983

>+

ESS-3-1411-83, ~ page 4 previously estimated the RIS magnitude' limit at ML = 4.5, one and one-half magnitude units above the maxioma RIS event to date. Although we infer that the Lake Wylie~ RIS potential is less _than at Lake Monticello, we think that an

'ML = 4.5 upper limit provides a comfortably large margin for error. There-fore, the design earthquake is a naturally occurring regional tectonic earth-quake, not one induced by Lake Wylie.

kt Harry N. Planner Carl A. Newton Geology Reviewer Seismology - Reviewer

References:

Hainson, B. C. (1977), A stress measurement in West Virginia and the state of stress in the southern Appalachians, EOS, Trans. Amer. Geophys. Union,

v. 58, p. 493.

Hainson, B. C. (1981), State of stress in the central and eastern United

. States, Geol. Soc. Amer. Abstr. with Programs, v.13(7), p. 466.

Tarr,- A. C., and S. Rhea (1982), Seismicity near Charleston, South Carolina, March 1973 to December 1979, ~in ." Studies Related to the Charleston, South Carolina, Earthquake of 1886: ~Tactonics and Seismicity (Collected Abstracts)," G. S. Gohn (ed. ), 'JS Geol. Survey Open File Report 82-134,

-pp. 32-33.

' Wentworth, C. M. , and M. . Mergner-Keefer (1982), Regenerate faults of small Cenozoic offset--probable earthquake sources in the southeastern United States, in " Studies Related to the Charleston, South Carolina Earthquake i of 1886: Tectonics and Seismicity (Collected Abstracts)," G. S. Gohn (ed.),

.US Gani. Survey Open-File Report 82-134, pp. 34-35.

( Zoback,'M. D..-J. H. Healy, J. C. Roller, G. S. Gohn, and B. B. Higgins (1978), Normal faulting and in, situ stress in the South Carolina coastal plain near Charleston, Geology, v. 6, pp. 147-152.

t i-xc: Jay Boudreau. ES/NP, MS F671 John T.; Whetten, ESS-DO, MS D446 i; Wayne A. Morris, ESS-2, MS D462 C. L. Edwards, ESS-3, MS C335

, T. A. Weaver, ESS-3, MS C335 L Dan J. Cash, ESS-3, MS C335 L CRM-4, MS A150' (2)

! ESS-3 File L- H. Sullivan, Q-D0, K552 i

!