ML20211G722

Proposed Earth Science Seismotectonic Research Program for Siting High Level Radwaste Repository

ML20211G722
Person / Time
Issue date:	05/05/1986
From:	NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
To:
Shared Package
ML20211G719	List: ML20211G715 ML20211G722
References
8606200109, PROC-860505, NUDOCS 8606200114
Download: ML20211G722 (46)

Text

_ m u.. a .m - , - - . - - - - - - - - _

4 _

se I'

.' EARTH SCIENCE SEISM 0 TECTONIC RESEARCH PROGRAM FOR THE

SITING OF A HIGH LEVEL RADI0 ACTIVE WASTE REPOSITORY ,

1 l

Purpose and Scope: l l

The purpose of this document is to define the earth science research issues necessary to support licensing positions and actions in the siting and design of geologic repositories for the disposal of high-level radioactive waste. The focus of the document is on tectonics and related sub-fields of seismology, neotectonics, and engineering geology. It addresses issues about which there is considerable uncertainty and for which the NRC requires guidance in estab-lishing investigation needs and design criteria. The issues addressed here are l

l basically generic iss.ues or of regional significance which are considered appro-

priate areas of responsibility of the Office of Nuclear Regulatory Research.

j- Regulatory Research Role:

i The licensing role of NRC in the siting, construction and operation of a high .

level radioactive waste disposal repository is to make informed judgment on the

+

adequacy of the license application and to verify that the repository will meet  ;

the performance requirements in 10 CFR Part 60.and the EPA standard in 40 CFR

^

Part 191. Some of the criteria of 10 CFR Part 60 are essentially the same as

i. ' for conventional civil structures. In those cases informed licensing judgment can be based on the present engineering and scientific state of practice. ,

2. 3 However, other criteria and performance objectives are unique to geologic repositories because of (1) the goal of complete isolation of radionuclides

from the accessible environment, (2) the 10,000 year requirement for the isola-tion within the underground storage facility, and (3) the expected high thermal stress in the in-situ rock. In those cases informed licensing judgment requires decisions based on considerably greater uncertainties in predicting long term g stability than have been heretofore required. It is the role of research.to investigate methods and obtain information not currently available that will

reduce uncertainty in the repository licensing effort.

05/05/86 1 EARTH SCIENCE RESEARCH 8606200114 860604

~ >lC *

+ PDR RES * -

86062001O9 _

PDR -

Tie following discussion proVides suggestions for research topics considered likely to reduce uncertainty and to enhance NRC's efforts to provide guidance and informed judgement in the licensing of a high-level radioactive waste

, repository. ,

I. TECTONICS The plate tectonics model has provided an understanding of linear and arcuate earthquake belts as plate boundaries where stored strain is released because of the movement of plate's with respect to each other. It has, however, been less

~

instructive with respect to the causes of intraplate seismicity. It is generally assumed that stresses at the plate margins are transnitted to plate interiors and are at least in part responsible for the elastic strains that result in sudden releases as earthquakes. It is also assumed that because the transmis-sion of stresses over great distances cust be slow, the rate of strain build-up is also very slow. Futhermore, unlike the narrow zones of seismicity. associated with plate boundaries, intraplate seismicity is more diffuse and less subject to prediction. Strain releases in continental plate interiors may occur at great depths on unknown structures, or near the surface on pre-existing struc-tures fortuitously oriented with respect to stress directions, or on structures formed within the " contemporary stress regime. Intraplate seismicity may be of such low magnitudes that it is only detectable by sensitive instruments, as "microseismicity," or violently destructive like the New Madrid 1811-12 events or the Charleston, SC,1886 event.

Despite the diffuse character of continental seismicity, and the low rates of

- deformation, concentrated research efforts in geophysics, geology, and neotec-tonics, much of it sponsored by the NRC, have delineated several active seismic zones, determined the limiting seismic substructures and/or the probable return periods for events such as the New Madrid and Charleston earthquakes, using deterministic evidence developed during those investigations.

In evaluating candidate sites for a high level radioactive waste repository in bedrock, regional tectonic strains, surface and subsurface structures, and the likely resulting seismic occurrences in the region close to the sites are necessary considerations for determining seismogenic structures and the design 2 EARTH SCIENCE RESEARCH 05/05/86 i ..

basis maximum magnitude events. Repository performance, bedrock integrity, modifications of hydraulic conductivity and potential resultant radionuclide releases depend in large part on the contemporary tectonic setting. The more that is known or understood of the tectonic environment", stress orientations, strain rates, and active structures, the fewer uncertainties are introduced into the scenarios that will be used to demonstrate the suitability of candidate sites in meeting the technical criteria specified in 10 CFR Part 60. As much of the data will be subject to interpretation, independent assessment of the seismotectonic information will reduce uncertainty and provide confidence in licensing positions relative to repository performance assessments.

To address these critical issues and to develop an independent data base for evaluating the information and interpretations to be included in the site char-acterization reports and the final safety analysis reports, the following re-search programs in the related subfields of Seismology, Neotectonics and Engineering are recommended.

II. SEISMOLOGY A. SEISMOLOGY RESEARCH ISSUES The siting, design and construction of a geologic repository will require that several new seismological judgments be made, because (1) The length of time.that the underground structure is required to perform is so long that the site will likely experience ground shak-

~~ ing in excess of that experienced in the region in historical times.

~

(2) Underground structures have never been designed to meet as many per-formance objectives as are required of a radioactive waste repository.

These objectives include not only structural stability but also the o.

integrity of the canisters, the integrity of borehole and shaft seals, i and the integrity of the rock mass as a part of the isolation system.

(3) The long-term isolation of the repository depends critically on

' control of groundwater flow. Modification of flow paths due to a 3 EARTH SCIEi4CE RESEARCH 05/05/86 l '

l - l J 7- - - -

\

_ a

major seismic event is of concern in evaluating performance objec-tives and potential release rates. ,

One of the areas with significant levels of uncertainty,is seismic wave propa- ~

gation at depth, because of the lack of data and related analytical techniques.

Included under that broad topic are: ground motion variations arising from ,

changes in propagation across geological discontinuities; frequency and strain level dependence of seismic wave propagation; and the relationship between I

seismic / tectonic forces and ground water flow.

1. Uncertainty in Ground Motion There are few data on seismic ground motion at repository depths. Even for surface facilities the data base on strong ground motion is far from complete.

During the last several decades there has been considerable information gathered and models developed on the relationship of surface ground motion to magnitude and distance. However, very limited data is available with respect to subsur-face facilities. Much of the data base on subsurface ground motion is summarized in a Lawrence Livermore National Laboratory document, UCID-20505, " Effects of

' Earthquakes on Underground Facilities; Literature Review and Discussion" by D. W. Carpente and D. H. Chung.

In general, ground motinn at depths of a few thousand feet, both accelerations and displacements, are thought to be less than those at the surface. How much at.tenuation occurs is variable, however, and seems to depend strongly on site conditions. The work of Ralph J. Archuleta, conducted in part for NRC, shows that a particular source parameter, the corner frequencies, of subsurface and surface recorded ground motions is nearly the same for a particular experimental set-up over a depth interval of about 160 m. He noted that recorded motions at '

the surface are more complex than those at depth and much of that complexity is due to surficial conditions. These observations, however, involve relatively shallow depths and the transmission of ground motions through soil, not bedrock.

There is little data on attenuation relations at depth since deep underground structures are rarely instrumented to measure vibratory ground motion associated with earthquakes.

4 EARTH SCIENCE RESEARCH 05/05/86

i t

In summary, little is known about subsurface accelerations in absolute terms, and as compared with surface accelerations. Data that exist indicate subsurface motions in general are less than surface motions but the amount depends on the surface conditions. This is not surprising in that individual site conditions have been known for some time to be a major factor in determining the character of ground motion at a site. Although there is no direct test of the problem, all studies suggest that the difference between subsurface and surface strong ground motions at a "hard rock site" should be considerably less than for a

" soft soil site". A significant part of the major problem is a lack of data.

2. Uncertainty in Response of Underground Structures to Earthquakes The response of underground structures to a seismic event is a concern overlap-ping the disciplines of seismology and engineering. The seismologist is interested in predicting the size of the earthquake likely to affect the site, the resultant level of ground motion at depth to which the repository may be .

subject, and the wave properties such as frequencies. The engineer, however, is concerned with predicting the response of the underground structure to the possible accelerations and frequencies, and to provide state of the art tech-niques to mitigate adverse effects. Although engineering issues are addressed under another heading, it is important to recognize that uncertainties in seis-mology in the areas of subsurface ground motions, accelerations and frequencies, lead to further uncertainties in engineering predictions of seismic response of the structure. More seismological information is needed to prevent such com-pounded uncertainties in the licensing process.

"- B. RESEARC4 TO ADDRESS SEISMIC UNCERTAINTIES

1. The lack of data concerning ground motions at depth contributes a large part of the uncertainty related to seismic issues in repos.itory performance.

Therefore, a program should be instituted to increase the underground strong motion data base by placing seismometers in deep (1000'-3000') boreholes or mines with known geometries in active seismic areas. An array of instruments would be preferable so that attenuation relations could be established. ' Where f j

possible, the seismometers should be installed in regions where HLW repositories

' are proposed within rocks that are similar to the rocks selected for HLW repost-tories and at the same depths.

5 EARTH SCIENCE RESEARCH

05/05/86 i

I l j, j ,.- . - I

i l

i 2. Data concerning the differences and similarities of seismic response .

at depth and at the surface may be obtained with a comparative study of surface and subsurface seismic effects and wave characteristics in (1) bedrock surface vs. bedrock subsurface, and (2) deep soil surface vs. b6drock subsurface. Com-parisons should be made of:

a. Variation of ground motion amplit'udes

b. Wave form and frequency content

c. Attenuation relationships: can high frequency attenuation relation- ,

ships and effects on wave form for high attenuation rates be extra-polated to regions of low attenuation--or are they entirely different?

d. Contribution of peak particle velocity vs. peak acceleration to near-field damage.

Where possible, bedrock types should be those of the candidate sites: welded tuff, basalt, granite bodies, salt domes and beds.

a

3. An LLNL report on the effects of earthquakes on underground facil-ities observed that small near-field earthquakes required further investiga-tion because experimental, analytical and blast data indicate that 4

(1) Small near-field earthquakes generated high frequency waves that damaged

'model underground openings which were otherwise stable when larger more distance event records were used; (2) For the same level of ground motion, nearby small earthquakes are more likely to cause damage than larger, more distant earthquakes because the ground motion from small events is more impulsive, inducing higher stress.

(3) With respect to small earthquakes, there' appears to be a closer relationship between peak velocity and underground damage than between acceleration and damage.

With these observations and the small amount of data available concerning small earthquakes, it is recommended that a study of the relationship between peak velocities and source parameters be undertaken to determine if it is true that, 6 EARTH SCIENCE RESEARCH 05/05/86 q--

according to the LLNL report, a small fault in the near-field to a nuclear waste repository could have seismic significance disproportionate to its size.

Such a study should attempt to: $

a. Relate such parameters to regional stresses and type of faulting;

b. Determine near-field effects on subsurface rock openings such as tunnels and shafts;

c. Compare with data for large earthquakes; and

d. Evaluate the significance of results of the investigation with respect to repository performance objectives, and to preclosure structures, systems and components important to safety.

4. The relationship of altered ground-water flow and earthquakes is poorly understood. Hydrologic changes in response to a seismic event were recorded in the San Francisco earthquake of 1906, and studied in more detail in connection with the 1985 Borah Peak, Idaho, earthquake. Significant' increases in discharge, ,

out to 90 Km beyond the rupture surface, in streams, wells, and springs, and elevation of water tables by up to 4 meters within the epicentral region suggest a need to learn more about the mechanism and geo-hydrologic relationships. As ground-water adjustments to earthquakes have serious implications for the iso-lation of subsurface geologic repositories, it is advisable that an investigative program of the relationship between earthquakes and ground-water flow be under-taken. Sucyaprogramshould:

a. Develop a world-wide data base on ground-water flow path modifications and adjustments in response to fault movement and/or a seismic event.

Information should include geologic and tectonic setting, seismic history and details of seismic events, and/or surface fault movements involving ground-water response, pre-event ground-water data (aquifer character and depth, flow rates and conduits), and changes of any of these, duration of flow path and rate modifications, variations in response for different events in same locality, any other relevant information.

7 EARTH SCIENCE RESEARCH 05/05/86 7

I

.I -

b. Develop predictive models for ground-water response to seismic events

- with respect to seismic character and geologic / tectonic settings based on evaluation of data base. Emphasis should be on characteristics of candidate repository geologic / seismic settingi. }

III. NE0 TECTONICS .

A. NE0 TECTONICS RESEARCH COMMON TO ALL SITES.

Because the tectonic framework is specific for each geologic setting, neotec-tonic investigations are necessarily site-related. The types of investigations and kinds of information sought are similar, although the nature of the geologic setting and rock types, proximity to other regions, the age, and the amount of information already available will dictate the variations in the investigative program and the specific approach.

Although site investigations are the responsibility of 00E, independent studies by the NRC are important to form the basis by which NRC reviewers can assess the investigations and analyses performed by 00E.

In general, the types of information to be included in a tectonic investigation for all candidate sites are -

~

1. Mapping of all regional and local structures within the site region, determination of age of latest movement, and sense of

~

displacement;

2. Determination of the relationship of macro- and microseismicity to mapped structures, gravity and magnetic anomalies in the site region;

3. Direction and magnitude of principal regional stresses; rela-tionship to mapped structures and displacements; 4

4. Strain rates and relationship of these to seismic recurrence intervals;

' f l

a l 8 EARTH SCIENCE RESEARCH

,j 05/05/86 i

v .

, ; . .} r - ,

k

5. Changes in strain rates based on historical and instrumental seismic data, geodetic changes, and structural history;

6. Investigation of individual surface structures that indicate ~

recency of movement and a potential for generating earthquakes or surface displacement. Data requirements include -

a. Sense, amNnt, and age of last displacement;

b. History of displacements;

c. Length of fault and of segment of recent movement (s); and

d. Strike, dip and area of fault surface;

7. Relationship of surface lineaments and geophysical anomalies to tectonic structures;

8. Establishment of an updated earthquake catalogue of the region; and

9. Development of a tectonic model or models to explain all the observations and relationships of 1 through 8.

Investigations outlining details of more site related studies will be found in the following sections. ,

B. TECTONICS RESEARCH RELATED TO REPOSITORIES LOCATED IN BEDDED BASA

~~ The Basalt Waste Isolation Project (BWIP) is assessing the feasibility of a geologic repository in the Pasco Basin, which is founded on basalt, a dense and massive rock. Basalt is, however, a brittle rock of low ductility under a wide range of pressure and temperature conditioris. The basalt, hundreds of meters thick, has been deformed by folding and faulting by north-south compres-sion during the Tertiary Period (65-2 million years before the present (MYBP)).

Jointing and shear fracturing are common in exposures of the basalt around the Pasco Basin, resulting from the deformation.

The reference repository site in the Pasco Basin is in the core of a syncline, the Cold Creek syncline, in the basalt. The syncline is buried beneath 9 EARTH SCIENCE RESEARCH 05/05/86 l

,!-L .

sedimentary deposits, so much of its character is not observed. Inasmuch as the isolation of the repository is dependent upon the integrity of the bedrock barrier in large part, a study of the subsurface characteristics of a syncline in basalt may increase confidence in the likelihood of a repository in such a structure to meet the performance objectives of 10 CFR 60.

In addition, other uncertainties concerning structural characteristics of basalt are addressed in the following suggestions:

1. Cold Creek Syncline, Hanford, Washington.

a. Study the relationship'between fracturing and folding of syn-clines by studying surface analogues to the Cold Creek Syncline, such as the Vantage Syncline, several miles north of the Hanford site, to clarify the following issues -

(1) is fracturing more intense along fold axes than within the 1

flanks;

,(2) what are relative contributions of stress directions and fold geometry to fracture development and orientation; (3) is residual stress more intense in synclinal hinges than on limbs; and (4) do residual stress orientations depend on position in the syncline or on regional stresses?

b. Using worldwide data, determine whether or not there is a relationship bet' ween synclinical axes and seismicity.

2. Ambiguities relating to the structure make it uncertain whether the many faults in the basalt in the Hanford Region project downward to the base of the basalt, flatten off at some depth, or die out at shallow depths. Theories vary as to the significance and/or extent of faulting. One view interprets the faults as shallow, discontinuous secondary effects of the primary folding of brittle rock, that die out within the folds.,_Another interpretation views the 10 EARTH SCIENCE RESEARCH 05/05/86 ie p.- ~ _

I

. . 1 folds as secondary effects of large scale, low angle thrust faulting, which j flattren off at depth. The uncertainties concerning the possible steep projec-tion of faults to depths where earthquakes occur, or the flattening of faults

• that may then pass through or under the reference repository may be reduced by ,

investigations to determine the character of the faults in the fold belt.

An investigative program therefore is recordmended to determine the relationship between folds and faults in the Yakima Fold Belt, just west of the Pasco Basin, the mechanisms and tectonic causes of the deformation. Are faults primary, deep seated fractures, or do they flatten with depth and may therefore pass through'or under the reference repository? Is there evidence for the decoupl-ing of the basalt and underlying bedrock during deformation?

3. The nature and structure of the rocks below the basalt have not been determined from data, but by extrapolation, because the dense basalt has not allowed any geophysical penetration to the rocks below. There is a very large gap in our knowledge between the surface and sub-basalt geologic character.

Since earthquakes originate in the deeper crust, it would be important to know if any structures capable of localizing earthquakes are present, and if these have propagated upward into the basalt. Up to now only oil company drilling programs have provided direct information, but the data have not yet been made available to the public. This information will reduce uncertainties concerning the sub-basalt potential for seismic activity and for surface structures to be upward propagations of sub-basalt features, in the vicinity of the BWIP repository. l

"- A research p ygram, therefore, is suggested to define the nature of the tectonic structure in rocks below the Columbia River Basalts. Determine whether or not Source structures within the basalt reflect structures in the rocks below it.

of data includes oil company drilling and magnotelluric exploratory results.

C. TECTONICS RESEARCH RELATED TO THE WELDED TUFF REPOSITORY AT NEVADA TE SITE

1. Geological, geophysical, and geodetic investigations indicate that the Sierra Nevada Mts. west of the Nevada site are rising and rotating about an 11 EARTH SCIENCE RESEARCH

. . 05/05/86 j1-~ .

approximately N-S horizontal axis. The effects of this tectonic activity on the tectonics and seismicity of the Basin and Range Province in general, and on the Yucca Flat region in particular, should be assessed. This should be done through analysis of the published information and any other seismic, tectonic and/or geodetic data.

2. Establish a data base from world-wide data to increase the accuracy of predicting fault attitudes at depth in welded tuff from surface and near-surface deformation. Information should include:

a. Regional geologic setting

b. Contemporary stress orientations

c. Fault and joint orientations, characteristics, and variations

d. Thickness and shape of the tuff body

e. Boundary conditions of the tuff body

f. Nature of bounding rock bodies, structures, and orientation Compare NTS regional and local data for the same features to determine if there is a consistent pattern that reduces uncertainty in prediction of structural characteristics.

3. Recent investigations suggest the possibility that many of the Basin and Range faults flatten with depth and that the bedrock is currently sliding and rotating along some of the faults. Several faults in the Yucca Flat vicinity are extensions of Basin and Range faults and may have some of the same potential for flattening and sliding. Such faults may pass through or under the Yucca Flat Repository, and therefore present uncertainty concerning the stability of the site. An investigative program is suggested to determine the nature, atti-tude and recency of faulting in the NTS area, using already available inforniation and investigative techniques necessary to determine subsurface orientations and potential for sliding.

4. Determine the potential for renewed volcanic activity in the Nevada Test Site region. Evaluate the potential effects on a repository in tuff of possible volcanic hazards such as magma rise, renewed hydrothermal activity, or high heat flow from depth.

12 EARTH SCIENCE RESEARCH 05/05/86

~

(, e-

D. TECTONICS RESEARCH RELATED TO THE BEDDED SALT REPOSITORIES ,

1. Many of the northeast trending features in the Paradox Basin, includ-l ingtheColoradoRiverbelowMoab, Utah,maybestructuhallycontrolledbybase- }

a ment shear zones or wrench faults. These northeast striking features may be a l small part of a more extensive system called the Colorado lineament. The l Colorado lineament is a system of Precc. brian faults that extends southwest f l

from Lake Superior through Minnesota, South Dakota, Nebraska, across the Colo-rado Mountains and the Colorado Plateau. On the Colorado Plateau, the Colorado f

• I River follows the faults along much of their trend. There is a spatial coin-cidence between the Colorado lineament and seismicity. There is an apparent ,

i alignment of epicenters along the Colorado River in the west-central part of I the Paradox Basin between Moab and the confluence of the Green and Colorado Rivers. The hypocenters of these earthquakes extend from the top of Precambrian basement to depths of 10 kms. Fault plane solutions exhibit strike slip fault- ,

ing along northeast strikes.

.t It is suggested, therefore, that a study be undertaken of the section of the I

Colorado lineament between the confluence of the Colorado and Green Rivers and Moab, Utah to determine its seismic potential and possible effect on a candi- ,f' date salt site in the Paradox Basin. This program should include micro- l earthquake monitoring, geological mapping, determination of the characteris- l tics of faults in the lineament including ages, seismicity studies, geo- 3 physics, state of stress, trenching and a synthesis of the results of all of these investigations.

2. At least one west-northwest striking fault of the Wichita Frontal j t

Fault system of southern Oklahoma, the Meers Fault, has experienced substantial displacement within the past several thousand years, but is not characterized ,

by seismicity. Although this system of faults apparently passes north of the Palo Duro Basin, a similar west-northwest structural grain (basement faults and g joints) is predominant within the basin. Numerous Landsat lineaments have been identified across the basin with a west-northwest trend. Many of these linea-ments are caused by linear scarps, alignments of playa lake depressions, and h h

linear stream segments. ,

9 I

13 EARTH SCIENCE RESEARCH

[ 05/05/86 f

n.

• 4 g

a. Investigate the west-northwest striking linear features in the Palo Ouro Basin and adjacent areas to determine if (1) they are controlled by tectonic structures, (2) there is a relationship between these and the Wichita Fault System, and (3) there is any evidence that they may have experienced recent movement. The significance of the results for a proposed depository in nearby salt beds should be assessed.

b. Study the seismicity and tectonics of the Amarillo-Wichita Uplift, and associated faults adjacent to the Palo Ouro Basin. Evaluate the potential of this tectonic region to affect the integrity of the pro-posed salt site in west Texas.

3. Specific research is needed with respect to bedded salt repositories to improve the extent to which the characteristics of geological features such as structures or stratigraphy, as determined by mapping at the surface, can be extrapolated to the depths of h proposed HLW repository. Geologic structures are mapped at the surface in these regions and it is not clear what the charac-teristics of'those structures are as they penetrate downward into the layered evaporites and sediments (i.e., do they flatten into bedding, ramp down, or con-tinue at the same attitude?). Conversely, there are structures identified by geophysical techniques in basement rock beneath the basins. It is not known

- whether or not they propagate upward through the strata in the basin or what their character is if they do. Research is needed in methods to identify and define such structures in bedded salt and to determine their significance to the repositories.

4. Determine if there is a relationship between structural features such l as the Grabens and Needles Fault Zones in the Paradox Basin and dissolution j

activity or features in the salt beds below. Do such structures extend down through the salt beds to provide hydrologic conduits or do they die out above the salt? l

~

05/05/86 14 EARTH SCIENCE RESEARCH j

1

, b.r ~

IV. ENGINEERING GEOLOGY / ROCK MECHANICS A. ENGINEERING / ROCK MECHANICS RESEARCH ISSUES -

Engineering issues related to a deep geological repository are concerned with two major considerations:

(1) The overall bulk rock properties, such as density, porosity and per-meability, fracture patterns, fracture density, etc.

(2) The response of the repository to seismic ground motion, which depends not only on the mechanical properties of the rock, but on the geometry of the repository such as the cross-sectional shape and interconnected

" tunnels, and on the seismic wave characteristics, such as frequency and peak velocity.

4 i

The role of bulk rock properties and seismic parameters needs to be understood in assessing the behavior of different rock materials, backfills and seals under This infor-varying static (in situ), dynamic (seismic), and thermal stresses.

' mation is needed in predicting the likelihood, nature and extent of damage to Such damage can affect underground openings in response to seismic stresses.

the hydraulic conductivity of the repository rock medium, the stability of the openings, rock integrity, and pre-existing planar discontinuities such as faults

! and joints.

- Large uncertainties exist with respect to these issues because the state of the

' - - art has not yet provided the necessary technologies for analyzing the three-dimensional response of subsurface openings to seismic ground motion in complex or inhomogeneous rock bodies. Nor have engineering studies developed numerical methods for assessing, in three-dimensional space, the behavior of openings in I rock in response to dynamic stresses.

J As licensing judgments, therefore, must be made r' elating to the long period of performance and the extreme environmental conditions, research to identify methods of dealing with and reducing uncertainty in estimating the site bulk 15 EARTH SCIENCE RESEARCH 05/05/86

. l .

.,,w ,,.. ,,_,/. g .y

_.- >--_yy._,

. _, w y - _.y_.,,

rock properties and seismic response of underground openings are high priority items because they affect those judgments.

B. RESEARCH TO ADDRESS ROCK MASS UNCERTAINTIES .' .

The engineering properties of a rock mass are to a large extent determined by the distribution of inhomogeneities, which.may be grain size, composition, porosity, fractures and other discontinuities within the rock mass. The distri ~

bution and orientation of fractures is extremely important in determining the strength and hydraulic properties of a rock mass.

Even after the repository is completely excavated, there will be an unreducible uncertainty about the engineering properties in the unexcavated rock around and '

within the repository. That unexcavated rock will be important for the perfor-mance of the repository as its behavior will affect the response of the under-ground structure. Also, it is important for the isolation of the waste as it is part of the pathway to the accessible environment.

l It is desirable, therefore, to establish the uncertainties in estimating the engineering properties or variability of bulk rock mass in the unexcavated part t

of the repository rock body. These uncertainties can be estimated by a critical l statistical evaluation of case histories of underground construction projects.

There are two well developed data bases for this study, the civil engineering community and the mining industry. In both of these, where underground struc-tures are proposed, pre-excavation deterministic and geostatistical site char-acterization provides for prediction of underground site conditions and the information base for the design of the structure. Subsequent excavation may uncover new information about the site that will require design modification.

Mining and civil engineering records kept of "as predicted" and "as discovered" data can provide a data base for statistical evaluation of the success or varia- f tions between the prediction of site conditions based on the pre-excavation site characterization and the observed conditions following excavation. This could provide the empirical data base on which to evaluate uncertainties in 4 subsurface site characterization.

i 6

16 EARTH SCIENCE RESEARCH .

05/05/86 l.-

A strong effort should be made to tap the data base within the regions in which HLW repositories are proposed; and a similar effort should be made, worldwide if necessary, to obtain data from large volumes of rock similar to those within .

which an HLW repository is planned. ,

C. RESEARCH TO ADDRESS SEISMIC STRESSES AND THE EFFECTS UPON UNDERGROUND EXCAVATIONS In general, underground structures apparently experience less damage in an earthquake than surface facilities. This is in part due to the expected reduc-tion in ground motion with depth. However, it is in part due to what appears to be an inherent robustness of underground structures with respect to ground motions. Data concerning this issue is also summarized in UCID-20505. There is a reasonably good qualitative data base on underground structure response to earthquakes. Dowding's work on the subject is reviewed in the above document.

From his review of 71 tunnels subjected to earthquakes he concluded the follow-ing: (1) for surface acceleration less than 0.19g, no damage has been noted to underground structures; (2) for acceleration of 0.19g to 0.4g, minor damage is reported; (3) major damage can occur at levels of acceleration in excess of 0.4g; (4) collapse is observed only at levels of acceleration above 0.5g.

There are a number of complicating factors, however. Damage to underground structures is greater in areas where shafts and portals intersect because of

. high stress concentrations. In addition, the frequencies of motion that cause damage to underground structures are different from those that cause damage to surface structures. Also, there are indications that the damage that occurs

- . for a given level of ground shaking is a strong function of the quality of the rock, i.e. , the nature and density of joints, fractures or other discontinuities.

The consequence of this uncertainty is that the overall response of the waste facility to ground shaking is not well known. Thus, there is significant uncertainty related to the response of the repository to, a design basis event.

The character of the response is of concern primarily in the preclosure period because of surface operations and the requirement for retrievability, both of which may be compromised if damage results from a seismic event.

17 EARTH SCIENCE RESEARCH l . 05/05/86 I .

(

~

f ' , _O e, , w , e w

The deformation of underground structures such as tunnels in response to seismic ,

motion is of three types: axial, curvature, and hoop. Axial and curvature f deformations result when seismic waves propagate parallel with or oblique to

' the tunnel. Axial deformation involves alternating regions of compressive and

• tensile strain in the rock along the tunnel axis. Curvature deformation creates a sinusoidal curvature of the tunnel. Hoop deformation refers to the change of cross-sectional shape of the tunnel when th'e wave travels normal (perpendicular) or near normal to the tunnel axis. For these types of deformation, there are two-dimensional analytical methods to analyze static and seismic stresses and stress concentrations in homogeneous rock. However, analytical methods in in-homogeneous rock and in three dimensions are extremely difficult so that there are only some simplified techniques for dealing with these.

In order to determine the strength of the rock, the level of stress concentra-tion, and the level of ground motion the tunnel can withstand, it is necessary to evaluate the applicability of current analytical aad numerical modeling methods to the real rock situation.

To identify the uncertainties inherent in the present state of the art tech-4 niques the following studies are recommended:

1. Identify and test methods for the 3-dimensional analytical, and numer-ical modeling of'the seismic response of an opening in rock such as a tunnel or shaft.

(a) Evaluate the applicability of current simplified 2-dimensional models in homogenous media to 3-dimensional structures in J

inhomogeneous media. Determine the limits and uncertainties inherent in the simplifications of the real rock conditions.

(b) Identify and test 3-dimensional analytical solutions for seismic design and support requirements for openings in rock, such as tunnels and shafts using properties of a variety of rock types such as those proposed for repositories.

18 EARTH SCIENCE RESEARCH 05/05/86 i ,

17 --

- _ ,_ -. -- ._. . . _ . . ._ ,,-...a

e (c) Identify methods using 3-dimensional models, for the numerical

~

evaluation of seismic stresses on tunnel-shaped openings in rock due to axial and cross-sectional (curvature) deformations in a

• variety of rock types proposed for repositories. Computerized .

numerical modeling of static and dynamic behavior of underground structures are used to determine the stability of underground structures in response to in situ and earthquake stresses. Such models make possible the study of potential seismic damage such as new cracks or the enhancement of preexisting ones. Up to now, primarily 2-dimensional models have been studied, although computer technology can permit the more complex three-dimensional analysis.

(d) Determine the effect of the complex geometry of underground facilities on their response to seismic loading. The number of interconnections and the total number of tunnels are likely to affect response. The effects of the direction of dynamic load-ing, whether parallel, perpendicular, or oblique, with respect to tunnel shape, orientation, and depth need to be .eisidered.

2. To enhance the experimental and theoretical data bases, an analysis is proposed of the existing data bases of (a) the effects of strong ground motion on underground structures that were developed by LLL during the nuclear devices testing program, and (b) of the seismic response of underground struc-tures in high temperatures, high stress regimes such as exist in the deep South African and Indian gold mines. .

t

3. Identify and test enoineering measures to overcome high thermal stress in combination with seismic ground motion, with particular attention to (a) shaft stability, (b) retrievability.

4. Determine the potential for induced seismicity, rock burst, and frac-turing due to stress release within the site rock as a' result of excavation for a repository. Determine the time-dependence of these hazards for various repository rock types in varying stress and thermal regimes. Test various measures to mitigate the hazards studied.

19 EARTH SCIENCE RESEARCH 05/05/86 J .. .

5. Identify and test preventive or remedial engineering measures to seal the potential damaged rock zone (DRZ) that may result from excavation activity.

The requirements to seal exploratory penetrations (e.g., shafts, boreholes) to

• keep them from becoming preferential pathways for water [is stated specifically ~

in 10 CFR 60 Section 134. A migration path closely associated with excavation If procedures is the damaged or stress-relieved rock around such excavations.

the DRZ develops during or after construction of a repository, it could become the most direct flow path or high conductivity zone from the repository to the Furthermore, the existence of the ORZ accessible environment or vice-versa.

The has major implications with regard to retrieval, canister loading, etc.

most common mechanisms for the development of the DRZ around excavations such as shafts, tunnels, rooms, etc., include -

a. slip of rock blocks along pre-existing discontinuities such as joints;

b. rotation of rock blocks;

c. development of new fractures as a result of the excavation process; and

d. rock fracturing as a consequence of excess stress concentration.

Since the development of the DRZ during and after construction of the repository is likely, the impact of the DRZ on waste isolation performance and preventive or remedial actions need to be addressed.

D. ENGINEERING / ROCK MECHANICS RESEARCH IN SALT SITES

1. Mechanical Behavior of Homogeneous Salt Rock: Major uncertainties

- continue to exist with regard to predicting the mechanical behavior of homo-Evidence to date suggests that geneous salt, especially at high temperatures.

there are substantial differences between predicted and measured borehole, shaft and room closures in salt. Salt creep uncertainties have serious implications with regard to HLW disposal and retrieval in this medium. Included in the impacts are -

a. canister loading;

b. retrieval;

c. closures of emplacement rooms, access drifts, etc.; and

d. backfill compression.  ;

20 EARTH SCIENCE RESEARCH 05/05/86 g lc. -

2. Effects of Impurities and Inhomogeneities in Salt. While the discus-sion above deals with pure salt (halite itself), major rock mechanics problems remain concerning the influence of inhomogeneities and nonuniformities in and

• around salt formations, e.g. faults, breccias, brine and gas pockets, mud, ,

shale, anhydrites, etc. Most tests to date have been performed on relatively pure salt. Research taking into account the impurities could provide more Other impor-realistic insights into the behavior of in-situ salt formations.

tant topics that need to be looked into include -

a. long-term performance (up to closure) of salt reinforcement (bolts) especially due to effects of heat and brine migration;

b. brine migration considering the coupled thermal, mechanical, hydrologic and chemical effects;

c. compaction of salt backfill; and

d. damaged zone and healing thereof around salt shafts, drifts, rooms, j

etc.

3. Salt Dissolution Salt Salt is one of the most water-soluble of natural geologic materials.

domes are commonly characterized by peripheral and internal weak zones that are created by the processes of solutioning and erosion of salt. The presence or absence of these weak zones is extremely imgortant to the long-term integrity I of the salt repository. The intrusion of groundwater into the dome .is permitted by discontinuities such as faults, joints, karst or other permeable structures.

Research is recommended, Bedded salt shares this potential for such features.

~ ~ therefore, to improve the state of knowledge in identifying and defining geo-logic and geomorphic features that have the potential of promoting dissolution, and in determining the relationship between those structures, their potential J

as hydrologic pathways, and salt dissolution zones. .,

F

a. Conduct research to establish whether or not there is a relationship L

between deep interior dissolution features,. thinning of beds, base-h ment faults, surface faults, lineaments, playas, regional joint systems or drainage features. Identify measures to mitigate the f effects of the presence of these features if a relationship exists. [

. 'i1 EARTH SCIENCE RESEARCH g 21 u I 05/05/86

' 5 I i .I

, e

~

b ~

,.e.

k

, . . l l

b. Institute a program to determine whether or not there is a relation- 1 ship between dissolution of salt and karst features in carbonate rocks above and below potential repository host salt layers. Test f

measurestomitigatetheeffectsofsuchrelaf.fonships.

c. Conduct research to evaluate the rates of dissolution of salt at the surface and at depth in the presence of faults and other pathways.

3. Identify and test methods of determining the heterogeneity of salt domes. Anomalous zones contain pure salt, clay, brine, gas pockets or brec-ciated zones. It would be helpful to be able to locate these zones and ident-ify the material within them, in order to identify and evaluate potential migra-tion of these anomalous zones.

4. In predicting the performance of a geologic repository, it is impor-tant to take into account the scale relationships between the structure and I

distribution of heterogeneities such as impurities, changes in composition, fractures, etc. For testing to be relevant, the scale of the sample must be large enough, or sample selection random enough, to include a representative number of heterogeneities in the rock. Therefore, test the validity of apply--

ing laboratory test results of core taken from salt domes or salt beds to the repository design in the rock mass, considering heterogeneities in the rock mass and other uncertainties. Assess how such laboratory tests reduce uncer-tainty, and characterize the remaining uncertainty.

5. Establish a data base of the characteristics cf faults located between salt domes and surrounding sedimentary rocks.

6. Examine worldwide data on storage of equipment in mines in domes and salt beds to develop a data base for rock characteristics and impurities, room closures and retrievability, and to determine the viability of a geologic repository in salt.

7. Perform research to improve the reliability of geophysical explora-tory techniques that are used to identify inhomogeneities in bedded salt. The geophysical methods can be verified by other investigative techniques with 22' EARTH SCIENCE RESEARCH 05/05/86 I -

i Ja n

4 greater resolution, but which cannot be used, or must be limited in the reposi-tory area, such as borings, test shafts, tunnels, etc. The features that con-tribute to the inhomogeneity of the salt rock mass include: thickening and .

thinning of beds, clay interbeds or seams, muddy salt beds, joints and frac-tures, brine / gas pockets, and inclusions. A data base of facts about the occur-rence of such features in salt strata should be established from this research to provide a basis for more accurate predictions regarding the heterogeneity of salt repositories. Such heterogeneities impact on the predictability of the mechanical behavior of the salt.

S

- . g 4

9 e

J 23 EARTH SCIENCE RESEARCH 05/05/86

~

.m ..

EARTH SCIENCE SEISM 0 TECTONIC RESEARCH PROGRAM FOR THE SITING OF A HIGH LEVEL RADI0 ACTIVE WASTE REPOSITORY ,

Purpose and Scope:

The purpose of this document is to define the earth science research issues necessary to support licensing positions and acticns in the siting and design of geologic repositories for the disposal of high-level radioactive waste. The focus of the document is on tectonics.and related sub-fields of seismology, neotectonics, and engineering geology. It addresses issues about which there is considerable uncertainty and for which the NRC requires guidance in estab-lishing investigation needs and design criteria. The issues addressed here are basically generic issues or of regional significance which are considered appro-priate areas of responsibility of the Office of Nuclear Regulatory Research. -

Regulatory Research Role: -

The licensing role of NRC in the siting, construction and operation of a high .

level radioactive waste disposal repository is to make informed judgment on the adequacy of the license application and to verify that the repository will meet the performance requirements in 10 CFR Part 60 and the EPA standard in 40 CFR Part 191. Some of the criteria of 10 CFR Part 60 are essentially the same as for conventional civil structures. In those cases informed licensing judgment can be based on the present engineering and scientific state of practice. ,

1 However, other criteria and performance objectives are unique to geologic repositories because of (1) the goal of complete isolation of radionuclides from the accessible environment, (2) the 10,000 year requirement for the isola-tion within the underground storage facility, and (3) the expected high thermal stress in the in-situ rock. In those cases informed licensing judgment requires decisions based on considerably greater uncertainties in predicting long term stability than have been heretofore required. It is the role of research-to investigate methods and obtain information not currently available that will reduce uncertainty in the repository licensing ef fort.

05/05/86 1 EARTH SCIENCE RESEARCH

+, - -

.o .

The following discussion provides suggestions for research topics considered likely to reduce uncertainty and to enhance NRC's efforts to provide guidance and informed judgement in the licensing of a high-level radioactive waste

~

repository. -

I. TECTONICS The plate tectonics model has provided an understanding of linear and arcuate earthquake belts as plate boundaries where stored strain is released because of the movement of plates with respect to each other. It has, however, been less instructive with. respect to the causes of intraplate seismicity. It is generally

) I assumed that stresses at the plate margins are transmitted to plate interiors and are at least in part responsible for the elastic strains that result in sudden releases as earthquakes. It is also assumed that because the transmis-sion of stresses over great distances must be slow, the rate of strain build-up is also very slow. Futhermore, unlike the narrow zones of seismicity associated with plate boundaries, intraplate seismicity is more diffuse and less subject to prediction. Strain releases in continental plate interiors may occur at great depths on unknown structures, or near the surface on pre-existing struc-tures fortuitously oriented with respect to stress directions, or on structures formed within the contemporary stress regime. Intraplate seismicity may be of such low magnitudes that it is only detectable by sensitive instruments, as '

"microseismicity," or violently destructive like the New Madrid 1811-12 events or the Charleston, SC, 1886 event.

Despite the diffuse character of continental seismicity, and the low rates of

- deformation, concentrated research efforts in geophysics, geology, and neotec-tonics, much of it sponsored by the NRC, have delineated several active seismic zones, determined the limiting seismic substructures and/or the probable return periods for events such as the New Madrid and Charleston earthquakes, using deterministic evidence developed during those investigations.

. In eval- candidate sites for a high level radioactive waste repository in bedrock, ,sgional tectonic strains, surface and subsurface structures, and the likely resulting seismic occurrences in the regicn close to the sites are necessary considerations for determining seismogenic structures and the design 2 EARTH SCIENCE RESEARCH 05/05/86 s .

l,_, - . __ _ _ _

I basis maximum magnitude events. Repository performance, bedrock integrity, modifications of hydraulic conductivity and potential resultant radionuclide releases depend in large part on the conttmporary tectonic setting. The more that is known or understood of the tectoaic environment, stress orientations, strain rates, and active structures, the fewer uncertainties are introduced into the scenarios that will be usea to demonstrate the suitability of candidate sites in meeting the technical criteria specified in 10 CFR Part 60. As much of the data will be subject to interprethlion, independent assessment of the seismotectonic information will reduce uncertainty and provide confidence in licensing positions relative to repository performance assessments.

To address these critical issues and to develop an independent data base for evaluating the information and interpretations to be included in the site char-acterization reports and the final safety analysis reports, the following re-search programs in the related subfields of Seismology, Neotectonics and Engineering are recommended.

II. SEISMOLOGY A. SEISMOLOGY RESEARCH ISSUES -

The siting, design and construction of a geologic repository will require that several new seismological judgments be made, because (1) The length of time.that_the underground structure is required to perform is so long that the site will likely experience ground shak-

~ ~

ing in excess of that experienced in the region in historical times.

(2) Underground structures have never been designed to meet as many per-formance objectives as are required of a radioactive waste repository.

These objectives include not only structural stability but also the integrity of the canisters, the integrity of borehole and shaft seals, and the N.egrity of the rock mass as a part of the isolation system.

(3) The long-term isolation of the repository depends critically on control of groundwater flow. Modification of flow paths due to a 3 EARTH SCIENCE RESEARCH i 05/05/86 t

i

, _ g. .

,i

major seismic event is of concern in evaluating performance objec-tives and potential release rates.

l One of the areas with significant levels of uncertainty,'is seismic wave propa-gation at depth, because of the lack of data and related analytical techniques.

Included under that broad topic are: ground motion variations arising from changes in propagation across geological discontinuities; frequency and strain level dependence of seismic wave propagation; and the relationship between seismic / tectonic forces and ground water flow.

1. Uncertainty in Ground Motion There are few data on seismic ground motion at repository depths. Even for surface facilities the data base on strong ground motion is far from complete.

During the last several decades there has been considerable information gathered and models developed on the relaticnship of surface ground motion to magnitude and distance. However, very limited data is available with respect to subsur-face facilities. Much of the data base on subsurface ground motion is summarized in a Lawrence Livermore National Laboratory document, UCID-20505, " Effects of ,

' Earthquakes on Underground Facilities; Literature Review and Discussion" by D. W. Carpenter and D. H. Chung.

In general, ground motion at depths of a few thousand feet, both accelerations and displacements, are thought to be less than those at the surface. How much attenuation occurs is variable, however, and seems to depend strongly on site conditions. The work of Ralph J. Archuleta, conducted in part for NRC, shows that a particular source parameter, the corner frequencies, of subsurface and surface recorded ground motions is nearly the same for a particular experimental set-up over a depth interval of about 160 m. He noted that recorded motions at the surface are more complex than those at depth and much of that complexity is due to surficial conditions. These observations, however, involve relatively shallow depths and the transmission of ground motions through soil, not bedrock.

1 There is little data on attenuation relations at depth since deep underground l structures are rarely instrumented to measure vibratory ground motion associated ll.

with earthquakes.

4 EARTH SCIENCE RESEARCH 05/05/86

- r

?

i i  !

In summary, little is known about subsurface accelerations in absolute terms, and as compared with surface accelerations. Data that exist indicate subsurface motions in general are less than surface motions but the amount depends on the '

surface conditions. This is not surprising in that individual site conditions have been known for some time to be a major factor in determining the character of ground motion at a site. Although there is no direct test of the problem, all studies suggest that the difference between subsurface and surface strong ground motions at a "hard rock site" should be considerably less than for a

" soft soil site". A significant part of the major problem is a lack of data.

2. Uncertainty in Response of Underground Structures to Earthouakes ,

The response of underground structures to a seismic event is a concern overlap- l ping the disciplines of seismology and engineering. The seismologist is interested in predicting the size of the earthquake likely to affect the site,  ;

the resultant level of ground motion at depth to which the repository may be subject, and the wave properties such as frequencies. The engineer, however, is concerned with predicting the response of the underground structure to the i possible accelerations and frequencies, and to provide state of the art tech- .!

niques to mitigate adverse effects. Although engineering issues are addressed i under another heading, it is important to recognize that uncertainties in seis-mology in the areas of. subsurface ground motions, accelerations and frequencies, lead to further uncertainties in engineering predictions of seismic response of the structure. More seismological information is needed to prevent such com-pounded uncertainties in the licensing process.

i

"- 8. RESEARCK TO ADDRESS SEISMIC UNCERTAINTIES

1. The lack of data concerning ground motions at depth contributes a large l part of the uncertainty related to seismic issues in repository performance.

Therefore, a program should be instituted to increase the underground strong motion data base by placing seismometers in deep (1000'-3000') boreholes or mines with known geometries in active seismic areas. An array of instruments would be preferable so that attentation relations could be established. Where possible, the seismometers should be installed in regions where HLW repositories are proposed within rocks that are similar to the rocks selected for HLW reposi-tories and at the same depths.

5 EARTH SCIENCE RESEARCH

. 05/05/86 i

jj.. .

2. Data concerning the differences and similarities of seismic response at depth and at the surface may be obtained with a comparative study of surface and subsurface seismic effects and wave characteristics in (1) bedrock surface vs. bedrock subsurface, and (2) deep soil surface vs. b6drock subsurface. Com-parisons should be made of:

f

a. Variation of ground motion amplit'udes

b. Wave form and frequency content

c. Attenuation relationships: can high frequency attenuation relation-ships and effects on wave form for high attenuation rates be extra-polated to regions of low attenuation--or are they entirely different?

d. Contribution of peak particle velocity vs. peak a;celeration to near-field damage.

Where possible, bedrock types should be those of the candidate sites: welded tuff, basalt, granite bodies, salt domes and beds.

3. An LLNL report on the effects of earthquakes on underground facil-ities observed that small near-field earthquakes required further investiga-tion because experimental, analytical and blast data indicate that (1) Small near-field earthquakes generated high frequency waves that damaged model underground openings which were otherwise stable when larget more distance event records were used; (2) For the same level of ground mction, nearby small earthquakes are more likely to cause damage than larger, more distant earthquakes because the ground motion from small events is more impulsive, inducing higher stress.

(3) With respect to small earthquakes, there appears to be a closer relationship between peak velocity and underground damage than between acceleration and damage.

With these observations and the small amount of data available concerning small earthquakes, it is recommended that a study of the relationship between peak velocities and source parameters be undertaken to determine if it is true that, 6 EARTH SCIENCE RESEARCH 05/05/86 1

p.;-.

2 .

according to the LLNL report, a small fault in the near-field to a nuclear waste '

repository could have seismic significance disproportionate to its size.

Such a study should attempt to: [

a. Relate such parameters to regional stresses and type of faulting;

b. Determine near-field effects on subsurface rock openings such as tunnels and shafts;

c. Compare with data for large earthquakes; and

d. Evaluate the significance of results of the investigation with respect I to repository performance objectives, and to preclosure structures, systems and components important to safety.

4. The relationship of altered ground-water flow and earthquakes is poorly understood. Hydrologic changes in response to a seismic event were recorded in the San Francisco earthquake of 1906, and studied in more detail in connection

- with the 1985 Borah Peak, Idaho, earthquake. Significant increases in discharge, .

out to 90 Km beyond the rupture surface, in streams, wells, and springs, and elevation of water tables by up to 4 meters within the epicentral region suggest j a need to learn more about the mechanism and geo-hydralogic relationships. As ground-water adjustments to earthquakes have serious implications for the iso-lation of subsurface geologic repositories, it is advisable that an investigative program of the relationship between earthquakes and ground-water flow be under- ,

taken. Sucyaprogramshould:

a. Develop a world-wide data base on ground-water flow path modifications and adjustments in response to fault movement and/or a seismic event.

Information should include geologic and tectonic setting, seismic history and details of seismic events, and/or surface fault movements involving ground-water response, pre-event ground-water data (aquifer character and depth, flow rates and conduits), and changes of any of 1

these, duration of flow path and rate modifications, variations in response for different events in same locality, any other relevant information.

7 EARTH SCIENCE RESEARCH

05/05/86 i

t r 4

~ ,

-T- g a - ,-t wy v ---7 w +- mr , a-.-r r ys- es<e a -e

i

b. Develop predict.ive models for ground-water response to seismic events with respect to seismic character and geologic / tectonic settings based on evaluation of data base. Emphasis should be on characteristics of

• candidate repository geologic / seismic settingi.

III. NE0 TECTONICS A. NE0 TECTONICS RESEARCH COMMON TO ALL SITES.

Because the tectonic framework is specific for each geologic setting, neotec-tonic investigations are necessarily site-related. The types of investigations and kinds of information sought are similar, although the nature of the geologic setting and rock types, proximity to other regions, the age, and the amount of information already available will dictate the variations in the' investigative program and the specific approach.

Although site investigations are the responsibility of DOE, independent studies by the NRC are important to form the basis by which NRC reviewers can assess the investigations and analyses performed by DOE.

In general, the types of information to be included in a tectonic investigation for all candidate sites are -

1. Mapping of all regional and local structures within the site region, determination of age of latest movement, and sense of displacement;  ;

2. Determination of the relationship of macro- and microseismicity to mapped structures, gravity and magnetic anomalies in the site region;

3. Direction and magnitude of principal regional stresses; rela-tionship to mapped structures and displacements;

4. Strain rates and relationship of these to seismic recurrence intervals; 4

8 EARTH SCIENCE RESEARCH

, 05/05/86 I, - -

(.-..-

5. Changes in strain rates based on historical and instrumental seismic data, geodetic changes, and structural history; i

6. Investigationofindividualsurfacestruhturesthatindicate '

recency of movement and a potential for generating earthquakes or surface displacement. Data requirements include -

a. Sense, amount, and age of last displacement; ,

b. History of displacements;

c. Length of fault and of segment of recent movement (s); and

d. Strike, dip and area of fault surface;

7. Relationship of surface lineaments and geophysical anomalies to tectonic structures;

8. Establishment of an updated earthquake catalogue of the region; and

9. Development of a tectonic model or models to explain all the observations and relationships of 1 through 8.

Investigations outlining details of more site related studies will be found in the following sections.

B. TECTONICS RESEARCH RELATED TO REPOSITORIES LOCATED IN BEDDED BASAL

~~ The Basalt Waste Isolation Project (BWIP) is assessing the feasibility of a geologic repository in the Pasco Basin, which is founded on basalt, a dense and massive rock. Basalt is, however, a brittle rock of low ductility under a wide range of pressure and temperature conditioris. The basalt, hundreds of meters thick, has been deformed by folding and faulting by north-south compres-

• sion during the Tertiary Period (65-2 million years'befef% *.he present (MYBP)).

Jointing and shear fracturing are common in exposures of the basalt around the Pasco Basin, resulting from the deformation.

i

' The reference repository site in the Pasco Basin is in the core of a syncline, the Cold Creek syncline, in the basalt.' The syncline is buried beneath 4' 9 EARTH SCIENCE RESEARCH 05/05/86  :

!I s  !

sedimentary deposits, so much of its character is not observed. Inasmuch as the isolation of the repository is dependent upon the integrity of the bedrock barrier in large part, a study of the subsurface characteristics of a syncline in basalt may increase confidence in the likelihood of a repository in such a structure to meet the performance objectives of 10 CFR 60.

~

In addition, other uncertainties concerning structural characteristics of basalt are addressed in the following suggestions:

1. Cold Creek Syncline, Hanford, Washington.

a. Study the relationship between fracturing and folding of syn-clines by studying surface analogues to the Cold Creek Syncline, such as the Vantage Syncline, several miles north of the Hanford site, to clarify the following issues -

(1) is fracturing more intense along fold axes than within the flanks; (2) what are relative contributions of stress directions and fold geometry to fracture development and orientation; (3) is residual stress more intense in synclinal hinges than on limbs; and (4) do residual stress orientations depend on position in the syncline or on regional stresses? j i

b. Using worldwide data, determine whether or not there is a l relationship between synclinical axes and seismicity.

2. Ambiguities relating to the structure make it uncertain whether the l many faults in the basalt in the Hanford Region project downward to the base of  ;

the basalt, flatten off at some depth, or die out at shallow depths. Theories vary as to the significance and/or extent of faulting. One view interprets the faults as shallow, discontinuous secondary effects of the primary folding of brittle rock, that die out within the folds. Another interpretati5n views the 05/05/86 10 EARTH SCIENCE RESEARCH

.'*T

- folds as secondary effects.of large scale, low angle thrust faulting, which flatten off at depth. The uncertainties concerning the possible steep projec-tion of faults to depths where earthquakes occur, or the flattening of faults

• that may then pass through or under the reference repository may be reduced by investigations to determine the character of the faults in the fold belt.

An investigative program therefore is recoEmended to determine the relationship between folds and faults in the Yakima Fold Belt, just west of the Pasco Basin, the mechanisms and tectonic causes of the deformation. Are faults primary, deep seated fractures, or do they flatten with depth and may therefore pass through or under the reference repository? Is there evidence for the decoupi-ing of the basalt and underlying bedrock during deformation?

3. The nature and structure of the rocks below the basalt have not been determined from data, but by extrapolation, because the dense basalt has not allowed any geophysical penetration to the rocks below. There is a very large gap in our knowledge between the surface and sub-basalt geologic character.

Since earthquakes originate in the deeper crust, it would be important to know if any structures capable of localizing earthquakes are present, and if these have propagated upward into the basalt. Up to now only oil company drilling programs have provided direct information, but the data have not yet been made available to the public. This information will reduce uncertainties concerning the sub-basalt potential for seismic activity and for surface structures to be upward propagations of sub-basalt features, in the vicinity of the BWIP repository.

~"- A research papgram, therefore, is suggested to define the nature of the tectonic structure in rocks below the Columbia River Basalts. Determine whether or not structures within the basalt reflect struct,utes in the rocks below it. Source of data includes oil company drilling and magnotelluric exploratory results.

C. TECTONICS RESEARCH RELATED TO THE WELDED TUFF REPOSITORY AT NEVADA TES SITE

1. Geological, geophysical, and geodetic investigations indicate that the Sierra Nevada Mts. west of the Nevada site are rising and rotating about an ,

11 EARTH SCIENCE RESEARCH

05/05/86

. i ---

approximately N-S horizontal axis. The effects of this tectonic activity on the tectonics and seismicity of the Basin and Range Province in general, and on the Yucca Flat region in particular, should be assessed. This should be done throegh analysis of the published information and any other seismic, tectonic ar.d/or geodetic data.

2. Establish a data base from world-wide data to increase the accuracy of predicting fault attitudes at depth in welded tuff from surface and near-surface deformation. Information should include:

i

a. Regional geologic setting

b. Contemporary stress orientations

c. Fault and joint orientations, characteristics, and variations

d. Thickness and shape of the tuff body

e. Boundary conditions of the tuff body

f. Nature of bounding rock bodies, structures, and orientation Compare NTS regional and local data for the same features to determine if there is a consistent pattern that reduces uncertainty in prediction of structural characteristics.

3. Recent investigations suggest the possibility that many of the Basin and Range faults fistten with depth and that the bedrock is currently sliding and rotating along some of the faults. Several faults in'the Yucca Flat vicinity are extensions of Basin and Range faults and may have some of the same potential for flattening and sliding. Such faults may pass through or under the Vecca Flat Repository, and therefore present uncertainty concerning the stability of 1 the site. An investigative program is suggested to determine the nature, atti-tude and recency of faulting in the NTS area, using already available information and investigative techniques necessary to determine subsurface orientations and potential for sliding.

4. Determine the potential for renewed volcanic activity in the Nevada Test Site region. Evaluate the potential effects on a repository in tuff of possible volcanic hazards such as magma rise, renewed hydrothermal activity, or high heat flow from depth.

12 EARTH SCIENCE RESEARCH 05/05/86 6

. _. . -- l

. ~ _ . - - -= - - - .. -- . . . - . - __ _

D. TECTONICS RESEARCH RELATED TO THE BEDDED SALT REPOSITORIES i

i

1. Many of the northeast trending features in the Paradox Basin, includ-

• ing the Colorado River below Moab, Utah, may be structurally controlled by base- '

ment shear zones or wrench faults. These northeast striking features may be a

{

small part of a more extensive system called the Colorado lineament. The Colorado lineament is a system of Precambrian faults that extends southwest from Lake Superior through Minnesota, South Dakota, Nebraska, across the Colo-rado Mountains and the Colorado Plateau. On the. Colorado Plateau, the Colorado l

River follows the faults along much of their trend. There is a spatial coin-cidence between the Colorado lineament and seismicity. There is an apparent alignment of epicenters along the Colorado River in the west-central part of the Paradox Basin between Moab and the confluence of the Green and Colorado Rivers. The hypocenters of these earthquakes extend from the top of Precambrian basement to depths of 10 kms. Fault plane solutions exhibit strike slip fault-

. ing along northeast strikes.

O It is suggested, therefore, that a study be undertaken of the section of the Colorado lineament between the confluence of the Colcrado and Green Rivers and Moab, Utah to determine its seismic potential and possible effect on a candi- ,

date salt site in the Paradox Basin. This program should include micro-earthquake monitoring, geological mapping, determination of the characteris-tic.s of faults in the lineament including ages, seismicity studies, geo-physics, state of stress, trenching and a synthesis of the results of all of these investigations.

2. At least one west-northwest striking fault of the Wichita Frontal 1

Fault system of southern Oklahoma, the Meers Fault, has experienced substantial displacement within the past several thousand years, but is not characterized by seismicity. Although this system of faults apparently passes north of the Palo Duro Basin, a similar west-northwest structural grain (basement faults and joints) is predominant within the basin. Numerous Landsat lineaments have been identified across the basin with a west-northwest trend. Many of these linea-ments are caused by linear scarps, alignments of playa lake depressions, and .

linear stream segments.

13 EARTH SCIENCE RESEARCH

05/05/86

.. e g e a t .- -

^

_ . _ _ _.~ .1..__...._, '

i

a. Investigate the west-northwest striking linear features in the Palo Ouro Basin and adjacent areas to determine if (1) they are controlled by tectonic structures, (2) there is a relationship between these and the Wichita Fault System, and (3) there is any evidence that they may have experienced recent covement. The significance of the results for a proposed depository in nearby salt beds should be assessed.

b. Study the seismicity and tectonics of the Amarillo-Wichita Uplift, and associated faults adjacent to the Palo Ouro Basin. Eval. ate the potential of this tectonic region to affect the integrity of the pro-posed salt site in west Texas. '

3. Specific research is needed with respect to bedded salt repositories to improve the extent to which the characteristics of geological features such as structures or stratigraphy, as determined by mapping at the surface, can be extrapolated to the depths of a proposed HLW repository. Geologic structures are mapped at the surface in these regions and it is not clear what the charac-teristics of those structures are as they penetrate downward into the layered l

! evaporites and s&diments (i.e., do they flatten into bedding, ramp down, or con- l tinue at the same attitude?). Conversely, there are structures identified by geophysical techniques in basement rock beneath the basins. It is not known

- whether or not they propagate upward through the strata in the basin or what their character is if they do. Research is needed in methods to identify and define such structures in bedded salt and to determine their significance to tha repositories.

4. Determine if there is a relationship between structural features such f as the Grabens and Needles Fault Zones in the Paradox Basin and dissolution activity or features in the salt beds below. Do such structures extend down through the salt beds to provide hydrologic conduits or do they die out above .

the salt?

e j 05/05/86 14 EARTH SCIENCE RESEARCH

, *g hA

IV. ENGINEERING GEOLOGY / ROCK MECHANICS A. ENGINEERING / ROCK MECHANICS RESEARCH ISSUES -

Engineering issues related to a deep geological repository are concerned with two major considerations: .

(1) The overall bulk rock properties, such as density, porosity and per-meability, fracture patterns, fracture density, etc.

(2) The response of the repository to seismic ground motion, which depends not only on the mechanica? properties of the rock, but on the geometry of the repository such as the cross-sectional shape and interconnected 1 tunnels, and on the seismic wave characteristics, such as frequency and peak velocity.

i The role of bulk rock properties and seismic parameters needs to be understood in assessing the behavior of different rock materials, backfills and seals under '

This infor-4 varying static (in situ), dynamic (seismic), and thermal stresses.

mation is needed in predicting the likelihood, nature and extent of damage to underground openings in response to seismic stresses. Such dtmage can affect the hydraulic conductivity of the repository rock medium, the stability of the openings, rock integrity, and pre-existing planar discontinuities such as faults and joints.

Large uncertainties exist with respect to these issues because the state of the

- art has not yet provided the necessary technologies for analyzing the three-dimensional response of subsurface openings to seismic ground motion in complex or inhomogeneous rock bodies. Nor have engineering studies developed numerical methods for assessing, in three-dimensional space, the behavior of openings in rock in response to dynamic stresses.

i As licensing judgments, therefore, must be made relating to the long period of performanca and the extreme environmental conditions, research to identify methods of dealing with and reducing uncertsinty in estimating the site bulk 15 EARTH SCIENCE RESEARCH

) 05/05/86 l

i

L.

i rock properties and seismic response of underground openings are high priority items because they affect those judgments.

i l B. RESEARCH TO ADDRESS ROCK MASS UNCERTAINTIES .' .

• The engineering properties of a rock mass are to a large extent determined by the distribution of inhomogeneities, which.'may be grain size, composition, porosity, fractures and other discontinuities within the rock mass. The distri-bution and orientation of fractures is extrr..nely important in determining the strength and hydraulic properties of a rock mass.

'- Even after the repository is completely excavated, there will be an unreducible uncertainty about the engineering properties in the unexcavated rock around and within the repository. That unexcavated rock will be important for the perfor-mance of the repository as its behavior will affect the response of the under-ground structura. Also, it is important for the isolation of the waste as it is part of the pathway to the accessible environment.

It is desirable, therefore, to establish the uncertainties in estimating the engineering properties or variability of bulk rock mass in the unexcavated part of the repository rock body. These uncertainties can be estimated by a critical j

statistical evaluation of case histories of underground construction projects.

f There are two well developed data bases for this study, the civil engineering community and the raining industry. In both of these, where underground struc-J tures are proposed, pre-excavation deterministic and geostatistical site char-acterization provides for prediction of underground site conditions and the information base for the design of the structure. Subsequent excavation may uncover new information about the site that will require design modification.

Mining and civil engineering records kept of "as predicted" and "as discovered" data can provide a data base for statistical evaluation of the success or varia- ,

tions between the prediction of site conditions based on the pre-excavation f site characterization and the observed conditions following excavation. This l 8

could provide the empirical data base on which to evaluate uncertainties in subsurface site characterization.

4 Y

16 EARTH SCIENCE RESCARCH I 05/05/86 f

.I

. g _ ..

P t

I A strong effort should be made to tap the data base within the regions in which j

HLW repositories are proposed; and a similar effort should be made, worldwide if necessary, to obtain data from large volumes of rock similar to those within which an HLW repository is planned.

• C. RESEARCH TO ADDRESS SEISMIC STRESSES AND THE EFFECTS UPON UNDERGROUND EXCAVATIONS I

i In general, underground structures apparently experience less damage in an earthquake than surface facilities. This is in part due to the expected reduc- l

)

tion in ground motion with depth. However, it is in part due to what appears j to be an inherent robustness of underground structures with respect to ground motions. Data concerning this issue is also summarized in UCID-20505. There i t

is a reasonably good qualitative data base on underground structure response to earthquakes. Dowding's work on i.he subject is reviewed in the above document. f From his review of 71 tunnels subjected to earthquakes he concluded the follow- ,

ing: (1) for surface acceleration less than 0.19g, no damage has been noted to l underground structures; (2) for acceleration of 0.19g to 0.4g, minor damage is l

)

reported; (3) major damage can occur at levels of acceleration in excess of 0.4g; (4) collapse is observed only at levels of acceleration above 0.5g.

There are a number of complicating factors, however. Damage to underground structures is greater in areas where shafts and portals intersect because of high stress concentrations. In addition, the frequencies of motion that cause damage to underground structures are different from those that cause damage to st:rface structures. Also, there are indications that the damage that occurs

i. for a given hevel of ground shaking is a strong function of the quality of the rock, i.e., the nature and density of joints, fractures or other discontinuities.

The consequence of this uncertainty is that the overall response of the waste facility to ground shaking is not well known. Thus, there is significant uncertainty related to the response of the repository to, a design basis event.

The character of the response is of concern primarily in the preclosure period because of surface operations and the requirement for retrievability, both of which may be compromised if damage results from a seismic event.

17 EARTH SCIENCE RESEARCH

. 05/05/86

[,7 .

E

The deformation of underground structures such as tunnels in response to seismic motion is of three types: axial, curvature, and hoop. Axial and curvature deformations result when seismic waves propagate parallel with or oblique to the tunnel. Axial deformation involves alternating regions of compressive and tensile strain in the rock along the tunnel axis. Curvature deformation creates a sinusoidal curvature of the tunnel. Hoop deformation refers to the change of cross-sectional shape of the tunnel when th'e wave travels normal (perpendicular) or near normal to the tunnel axis. For these types of deformation, there are two-dimensional analytical methods to analyze static and seismic stresses and stress concentrations in homogeneous rock. However, analytical methods in in-homogeneous rock and in three dimensions are extremely difficult so that there are only some simplified techniques for dealing with these.

In order to determine the strength of the rock, the level of stress concentra-tion, and the level of ground motion the tunnel can withstand, it is necessary to evaluate the applicability of current analytical and numerical modeling methods to the real rock situation.

To identify the uncertainties inherent in the present state of the art tech-niques the following studies are recommended:

Identify and test methods for the 3-dimensional analytical, and numer-i 1.

ical modeling of the seismic response of an opening in rock such as a tunnel or i

shaft.

(a) Evaluate the applicability of current simplified 2-dimensional models in homogenous media to 3-dimensional structures in inhomogeneous media. Determine the limits and uncertainties inherent in the simplifications of the real rock conditions.

(b) Identify and test 3-dimensional analytical solutions for seismic '

design and support requirements for openings in rock, such as tunnels and shafts using properties of a variety of rock types a

such as those proposed for repositories.

18 EARTH SCIENCE RESEARCH

+ 05/05/86 .

. l 3- --

i

-- ~__ .- -, . . .L

(c) Identify methods using 3-dimensional models, for the numerical

~

evaluation of seismic stresses on tunnel-shaped openings in rock due to axial and cross-sectional (curvature) deformations in a variety of rock types proposed for repositories. Computerized .'

numerical modeling of static and dynamic behavior of underground structures are used to determine the stability of underground structures in response to in situ and earthquake stresses. Such models make possible the study of potential seismic damage such as new cracks or the enhancement of preexisting ones. Up to now, primarily 2-dimensional models have been studied, although computer technology can permit the more complex three-dimensional analysis.

(d) Determine the effect of the complex geom'etry of underground facilities on their response to seismic loading. The number of interconnections and the total number of tunnels are likely to affect response. The effects of the direction of dynamic load-ing, whether parallel, perpendicular, or oblique, with respect to tunnel shape, orientation, and depth need to be considered.

2. To enhance the experimental and theoretical data bases, an analysis is proposed of the existing data bases of (a) the effects of strong ground motion on underground structures that were developed by LLL during the nuclear devices testing program, and (b) of the seismic response of underground struc-tures in high temperatures, high stress regimes such as exist in the deep South African and Indian gold mines. .

t

3. Identify and test engineering measures to overcome high thermal stress in combination with seismic ground motion, with particular attention to (a) shaft stability, (b) retrievability.

4. Determine the potential for induced seismicity, rock burst, and frac-turing due to stress release within the site rock as a' result of excavation for a repository. Determine the time-dependence of these hazards for various repository rock types in varying stress and thermal regimes. Test various measures to mitigate the hazards studied.

19 EARTH SCIENCE RESEARCH 05/05/86

, l .. ~ .

5. Identify and test preventive or remedial engineering measures to seal the potential damaged rock zone (DRZ) that may result from excavation activity.

The requirements to seal exploratory penetrations (e.g. , shafts, boreholes) to

• keep them from becoming preferential pathways for water',is stated specifically '

in 10 CFR 60 Section 134. A migration path closely associated with excavation If procedures is the damaged or stress-relieved rock around such excavations.

the DRZ develops during or after construction of a repository, it could become the most direct flow path or high conductivity zone from the repository to the Furthermore, the existence of the ORZ accessible environment or vice-versa.

The has major implications with regard to retrieval, canister loading, etc.

most common mechanisms for the development of the DRZ around excavations such as shafts, tunnels, rooms, etc., include -

1 j a. slip of rock blocks along pre-existing discontinuities such as joints;.

7

b. rotation of rock blocks; j

• c. development of new fractures as a result of the excavation process;

and

! d. rock fracturing as a consequence of excess stress concentration.

Since the development of the ORZ during and after construction of the repository l is likely, the impact of the DRI on waste isolation performance and preventive or remedial actions need to be addressed.

'. D. ENGINEERING / ROCK MECHANICS RESEARCH IN SALT SITES

1. Mechanical Behavior of Homogeneous Salt Rock: Major uncertainties continue to exist with regard to predicting the mechanical behavior of homo- 1 Evidence to date suggests that j geneous salt, especially at high temperatures.

there are substantial differences between predicted and measured borehole, shaft and room closures in salt. Salt creep uncertainties have serious implications l with regard to HLW disposal and retrieval in this medium. Included in the I

i impacts are -

a. canister loading;

b. retrieval;

c. closures of emplacement rooms, access drifts, etc.; and i '

d. backfill compression.

20 EARTH SCIENCE RESEARCH l 05/05/86 3

i 7 .

• I

b . .

i

2. Effects of Impurities and Inhomogeneities in Salt. While the discus-sion above deals with pure salt (halite itself), major rock mechanics problems j

remain concerning the influence of inhomogeneities and nonuniformities in and l ',

around salt formations, e.g. faults, breccias, brine and gas pockets, muo, shale, anhydrites, etc. Most tests to date have been performed on relatively -

pure salt. Research taking into account the impurities could provide more Other impor-e realistic insights into the behavior of in-situ salt formations.

tant topics that need to be looked into include -

a. long-term performance (up to closure) of salt reinforcement (bolts) i especially due to effects of heat and brine migration;  !

! b. brine migration considering the coupled thermal, mechanical, I hydrologic and chemical effects; i

c. compaction of salt backfill; and

d. damaged zone and healing thereof around salt shafts, drifts, rooms, etc.

3. Salt Dissolution i,

i Salt is one of the most water-soluble of natural geologic materials. Salt l

domes are commonly characterized by peripheral and internal weak zones that are l

i created by the processes of solutioning and erosion of salt. The presence or

[

absence of these weak zones is extremely important to the long-term integrity

' of the salt repository. The intrusion of groundwater into the dome is* permitted by discontinuities such as faults, joints, karst or other permeable structures.

f Bedded salt shares this potential for such features. Research is recommended, l

~~ therefore, to improve the state oh knowledge in identifying and defining geo-logic and geomorphic features that have the potential of promoting dissolution,

~

and in determining the relationship between those structures, their potential f

as hydrologic pathways, and salt dissolution zones.

i i a. Conduct research to establish whether or not there is a relationship between deep interior dissolution features, thinning of beds, base-ment faults, surface faults, lineaments, playas, regional joint systems or drainage features. Identify measures to mitigate the effects of the presence of these features if a relationship exists.

i 21 EARTH SCIENCE RESEARCH 05/05/86 ff

^

'l L , -

b. Institute a program to determine whether or not there is a relation-I ship between dissolution of salt and karst features in carbonate rocks above and below potential repository host salt layers. Test measurestomitigatetheeffectsofsuchrela?ionships.

c. Conduct research to evaluate the rates of dissolution of salt at the surface and at depth in the presence of faults and other pathways.

3. Identify and test methods of determining the heterogeneity of salt domes. Anomalous zones contain pure salt, clay, brine, gas pockets or brec-ciated zones. It would be helpful to be able to locate these zones and ident-ify the material within them, in order to identify and evaluate potential migra-tion of these anomalous zones.

4. In predicting the performance of a geologic repository, it is impor-tant to take into account the scale relationships between the structure and distribution of heterogeneities such as impurities, changes in composition, fractures, etc. For testing to be relevant, the scale of the sample must be large enough, or sample selection random enough, to include a representative

~

number of heterogeneities in the rock. Therefore, test the validity of apply-ing laboratory test results of core taken from salt domes or salt beds to the l

repository design in the rock mass, considering heterogeneities in the rock mass and other uncertainties. Assess how such laboratory tests reduce uncer-tainty, and characterize the remaining uncertainty *.

i 5. Establish a data base of the characteristics of faults located between salt domes and surrounding sedimentary rocks.

6. Examine worldwide data on storage of equipment in mines in domes and salt beds to develop a data base for rock characteristics and impurities, room I closures and retrievability, and to determine the viability of a geologic -

repository in salt.

7. Perform research to improve the reliability of geophysical explora-tory techniques that are used to identify inhomogeneities in bedded salt. The

~

geophysical methods can be verified by other investigative techniques with 22 EARTH SCIENCE RESEARCH 05/05/86

+ l l8

. i j'r- .

t greater resolution, but which cannot be used, or must be limited in the reposi-tory area, such as borings, test shafts, tunnels, etc. The features that con-tribute to the inhomogeneity of the salt rock mass include: thickening and .

thinning of beds, clay interbeds or seams, muddy salt beds, joints and frac- ,

tures, brine / gas pockets, and inclusions. A data base of facts about the occur-rence of such features in salt strata should be established from this research to provide a basis for more accurate predf.ctions regarding the heterogeneity of salt repositories. Such heterogeneities impact on the predictability of the mechanical behavior of the salt.

S

.. 3 i

23 EARTH SCIENCE RESEARCH l 05/05/86 l l

!l