- Email: [email protected]
Geophysical investigations in sweden for the characterization of a site for radioactive waste disposal — An overview
Geoexplorotion, 22 (1984) 187-201 Elsevier Science Publishers B.V., Amsterdam -Printed
in The Netherlands
187
GEOPHYSICAL INVESTIGATIONS IN SWEDEN FOR THE CHARACTERIZATION OF A SITE FOR RADIOACTIVE WASTE DISPOSAL - AN OVERVIEW
0. OLSSON,
0. DURAN, A. JAMTLID
Sueriges Geologiska AB (SGAB), (Received December
and L. STENBERG
Uppsala (Sweden)
2, 1983; accepted January 11, 1984)
ABSTRACT Olsson, O., Duran, O., Jiimtlid, A. and Stenberg, L., 1984. Geophysical investigations in Sweden for the characterization of a site for radioactive waste disposal - an overview. Geoexploration, 22: 187-201. The geophysical measurements carried out for the investigations of prospective sites for radioactive waste disposal in Sweden are described. The investigations comprise detailed surface measurements with magnetic, electromagnetic, electric and seismic techniques. Measurements are made with a line spacing of 40 m and a station spacing of 20 m in an area of 4-6 km2. At each site 7-10 deep boreholes are drilled. A number of geophysical logs are run in the boreholes, including several electric logs, providing data on temperature, salinity, and gamma ray. These geophysical data are evaluated together with the results from geological and hydrological investigations to make a three-dimensional geological and tectonical model of the site. This model defines the geometry used in the numerical modelling of the groundwater flow. In the case of thin, resistive overburden the surface investigations proved to be effective in the detection of dipping fracture zones. A line spacing as low as 40 m is needed to get adequate mapping of the fracture zones. The methods used cannot be considered efficient for mapping sub-horizontal fracture zones. The borehole logs provide data for the identification of fracture zones as well as different lithological units and give qualitative information of the hydraulic conditions. The temperature and salinity logs have provided the best hydraulic information, whereas, after testing, the induction log (borehole slingram), the differential resistance log and the VLF-log were found to be not effective for our investigations. The borehole techniques used, normally give information only about a relatively small volume of rocks around the boreholes. To obtain a detailed description of the bedrock between the boreholes, development of cross-hole geophysical, high-resolution techniques of considerable range, is needed.
INTRODUCTION
The high-level radioactive waste from nuclear reactors is to be stored in a repository at considerable depth below the earth surface. To obtain a final storage which will protect the environment against the radioactive wastes for several hundredthousands of years, several barriers are constructed 0016-7142/84/$03.00
o 1984 Elsevier Science Publishers B.V.
188
to minimize the transport of radioactive nuclides. The barriers consist of the fuel material itself, the canister, the buffer material and the bedrock surrounding the repository (SKBF/KBS, 1983). In this context the transport of radionuclides through the bedrock will depend on the groundwater flow and the sorption of radionuclides to fracture minerals. The properties of the bedrock are important also because they may influence the performance of other barriers, e.g. the corrosion of copper canisters is dependent on the chemical properties of the groundwater. Comprehensive investigations have been carried out in Sweden at several different sites, mainly in gneissic or granitic rock. The objective has been to characterize the geological and hydrological conditions at each site and to produce the site-specific data necessary for a safety analysis of a repository situated at a depth of about 500 m. The investigations at a site are carried out as a closely integrated program in geology, geophysics and hydrology. The main objective of these investigations is to describe the factors essential for the safety of a final repository. Emphasis is thus put on describing the groundwater flow. In crystalline rock, a large part of the flow takes place in the fracture zones. Thus a large effort is made to determine the location and the hydraulic properties of the fracture zones. The detailed investigations on the surface cover an area of 4-6 km2 and the bedrock is investigated to a depth of about 700 m. The data are used to produce a descriptive geological and tectonical model of the site and a numerical model of the groundwater flow. The investigations in Sweden started in the mid-1970’s and since then several sites have been investigated (Fig. 1). The four sites, which have been investigated after 1979, have been investigated essentially according to the general site investigation program described by Ahlbom et al. (1983a). However, modifications in this general program have been made as called for by changes in the geological and hydrological conditions. In the present paper special attention will be paid to the geophysical part of the program and to some of the experiences gained. The four sites investigated according to this program are: (1) Site Fjallveden, 80 km southwest of Stockholm. This site is characterized by flat topography with minor fracture valleys in a northwesterly direction. The bedrock consists of veined gneiss with steeply dipping strata of granite gneiss (Ahlbom et al., 1983b). (2) Site Gide%, 30 km northeast of &nsktildsvik. This site is situated within a plateau-area of 100 km2. The topography of this site is flat. The bedrock consists of veined gneiss with strata of granite gneiss with a small dip (Ahlbom et al., 1983c). (3) Site Kamlunge, 65 km northeast of LuleP. This site is situated on a plateau with an area of 16 km2, about 100 m above the surrounding valleys. The bedrock consists of gneisses and red granite. Within the site a horizontal fracture zone has been found at a depth of 500 m. (Ahlbom et al., 1983d).
189
(4) Site Svartboberget, 12 km west of Edsbyn, is situated on a 2.5 km wide and 5 km long ridge. The surrounding valleys, which have a northwesterly direction, are at a level 75--85 m below the top of the ridge. The bedrock consists of a strongly metamorphosed gneiss, migmatite. Characteristic for this site is the presence of minor fracture zones with a mutual separation of 30-80 m. (Ahlbom et al., 1983e). The major part of the geological, geophysical and hydrological investigations have been performed by Sveriges Geologiska AB on commission from the Swedish Nuclear Fuel Supply Co.
SVARTBOBERGET
l
Fig. 1. The location of studied sites. The sites investigated after 1979 by filled dots, previously investigated sites by unfilled dots.
are indicated
190 CHOICE OF SITE
Reconnaissance studies are carried out continuously to find new sites which are to be studied in detail. Detailed studies are normally carried out at one or two sites each year. The following factors are considered in the selection of a site. ~0~0~~~~~ : a flat topography gives a small hydraulic gradient. Major fracture zones: the site should not be intersected by any major fracture zones. Minor fracture zones: the frequency of minor fracture zones should be small within the site. Rock type!: sites with different types of bedrock are to be studied within the long term program - until now the investigations have been concentrated to gneissic and granitic rocks. Ore mineralizations: a site likely to contain potential orebodies is avoided. The selection of a site for detailed investigations is made initially from available maps of geology, topography, and geophysical me~urements. The topo~aphic~ maps are studied to get a prelimin~ indication of the hydraulic gradients and the location of groundwater recharge and discharge areas. Data on well capacity is used to get a preliminary estimate of the hydraulic conductivity of the region and of the type of bedrock. Data from air photographs, satellite images, etc., are used to obtain info~ation about geologic st~c~res and the presence of important lineaments. Geologic field studies are made at prospective sites to gather additional data on outcrops, fracturing and rock-type distribution. If the site survives also this screening a preliminary geophysical study is made. Geophysical me~uremen~ are made as reconn~s~ce profiles on the ground surface. Measurements are made with proton magnetometer, horizontal-loop EM (slingram) and VLF. This combination of methods is normally sufficient to give a preliminary indication of the presence of larger fracture zones and the bedrock structure at the site. Interpretation is made of the dip, strike and thickness of fracture zones. The station spacing used is 20 m, except for the magnetic measurements where it is 5 m. If the data from the reconnaissance studies are still considered favorable, a reconnaissance borehole is drilled to get information about the bedrock properties at depth. The borehole is normally vertical with a depth of 800-1000 m. Geological mapping is made of the core and geophysical and hydrological measurements are made in the hole. The geophysical measurements will be described below. Following the evaluation of these studies a decision is taken whether the investigations are to be continued or not.
191 GEOPHYSICAL
SURFACE
INVESTIGATIONS
AT A SITE
The surface investigations at a site begin with a regional geologic and tectonic study covering an area of about 100 km’. After this, detailed geological and geophysical mapping is made of the rock type distribution and the fracture zones. The detailed investigations cover an area of 4-6 km2 in which a coordinate system is set up by means of stake lines to serve as a reference for all investigations to be made at the site. Geophysical measurements are made both in the coordinate system and in profiles outside of the coordinate system. The methods used in the coordinate system are: (a) proton magnetometer measurements; (b) horizontal loop EM (slingram) measurements, frequency 18 kHz, intercoil spacing 60 m; and (c) resistivity and time domain IP, measured using gradient array. The equipment used for the investigations is described by Almen et al., 1983. A line spacing of 40 m and a station spacing of 20 m is used. It has proven advantageous to decrease the station spacing of the magnetic measurements to 5 m as this greatly facilitates accurate dip determinations of thin dikes. In the investigations made before 1978 a line spacing of 80 m was used, but with this line spacing the interpretation of the strike of fracture zones often became ambiguous. The magnetic measurements are mostly used as an indicator of bedrock structure and normally indicate lithological changes and the presence of, e.g., dolerite dykes. In many cases the magnetic measurements also indicate the presence of faults and fracture zones. The water in the fracture zones may oxidize magnetite to hematite which implies that a fracture zone may be associated with a magnetic minimum (Henkel and Guzman, 1977). In Fig. 2 the magnetic measurements and the corresponding interpretation from site Gide% are shown. Several magnetic indications are to be seen which correspond to dolerite dykes (Ahlbom et al., 1983c). The high-frequency horizontal loop EM (slingram) is used to indicate the presence of fracture zones. The frequency is made high and the intercoil spacing is made large to obtain a high induction number making it possible to detect fracture zones, which can be considered to be poor sheetlike conductors. The large intercoil spacing also makes the depth of investigation acceptable. Especially in areas with a thin and/or highly resistive overburden, the high frequency EM method is efficient. However, the results obtained in areas covered by conductive clay, have not been satisfactory. In these cases the clay causes anomalies much larger than those which can be expected from fracture zones. An example of the results obtained and the corresponding interpretation at the test site Gide% is given in Fig. 3. In this area the bedrock resistivity is around 10,000 ohm m and overburden resistivity is about 200-1,000 ohm m, the overburden thickness varies from zero to about 5 m. In most
DISTINCT INDICATION WEAK INDICATION FAULT
INTERPRETATION
X80’ Gil @e-
STRIKE AND DIP DIAMOND DRILLED BOREHOLE PERCUSSION DRILLED BOREHOLE
_--
400 m
_
SWEDISH GEoLoGKAl
0
GIDEA
Fig. 2. Magnetic measurements and the corresponding interpretation from site Gide%.
---
-
LEGEND
MAGNETIC
IMAGINARY
LNTERPRETATION
Fig.
3.
Horizontal
loop
EM
(sling-ram)
LEGEND STRONG INDtCATlON X30’ DISTINCT INDICATION Gil@--WEAK INDICATION a-‘- WEAK RESLSTIVITY INDICATION
SLINGRAM
measurement,
imaginary
component
ST RlKE AND DIP DIAMOND DRILLED BOREHOLE PERCUSSION DRILLED BOREHOLE
.
,
,
LpOm
(quadrature)
and
I f
hi’
from site Gide%.
5.00 3.03 I.011 -1.00 -3. w -5.W -1O.W -15.w
IQ.03
15.w
NiJRNALF~LXT
ANPNALI
interpretation
SWEDISH GEOIBGICAL
q
GIDEA
194
cases the anomaly from a fracture zone will be about 3-5% in the imaginary (quadrature) component. Most of the “strong ” indications have been found after drilling to correspond to fracture zones with a width of 5-25 m (Ahlbom et al., 1983c). The resistivity method will also show the presence of good conductors such as fracture zones and mineralizations. The gradient array is used with a large-current electrode separation, thus the depth penetration should be considerable. The distance between the current electrodes is about 1.5 km and the distance between the potential electrodes is 20 m. Indications of conducting minerals have been obtained at depths of about 100 m but no definite indication of deeply situated fracture zones have yet been obtained. When the fracture zones are perpendicular to the direction of the electric field, the resistivity measurements seem to give a better indication of the width of the fracture zones compared to the high-frequency EM measurements. However, in most cases a better resolution of fracture zones is obtained with the high-frequency EM measurements. One advantage with the resistivity method is that it gives a measure of the bedrock resistivity between the fracture zones. Generally, high resistivity corresponds to bedrock with few fractures, The resistivity measurements may thus be used to identify blocks of rock with low fracturing. The resistivity measurements will also be affected by the presence of clay but not to the same extent as the high-frequency EM me~urements. The IP measurements, which are carried out simultaneously and with the same equipment as for the resistivity measurements, are used to discriminate between conductors which are due to mineralizations or fracture zones, The presence of a mineralization will cause both high IP-values and low resistivities while only a low r~sistivity is expected from fracture zones. In some cases the IP measurements give information about the distribution of different rock types and may thus indirectly indicate the presence of fracture zones. The gneissic sites often have disseminations of sulphides and graphite and thus have fairly large background IP values. Seismic refraction me~urements are made in profiles within the area of detailed investigations. The total length of the seismic profiles has normally been 4-6 km. The results are interpreted in terms of overburden thickness, substratum velocity and presence of fracture zones. The seismic method has normally been a good complement to the other methods and, in the case of conductive overburden, a necessary tool for inte~~tation of dubious anomalies which might be due to either fracture zones or mineralizations. To obtain information about the regional geologic structure in the vicinity of the site, profile measurements are made with proton magnetometer, slingram (horizontal loop EM), and VLF. Extensive profile measurements are made where complete airborne geophysical measurements are not available.
195
The objective of the geophysical measurements at the ground surface is to find the location and the orientation of fracture zones within the site. These data are used in the construction of a preliminary geologic and tectonic model of the site. The surface measurements are also used in combination with the geological information for positioning the boreholes that are made to investigate the bedrock and the fracture zones at large depth. GEOPHYSICAL
BOREHOLE
INVESTIGATIONS
AT A SITE
Percussion boreholes are made to investigate the shallow parts of fracture zones. These boreholes have a length of 100-150 m and dip 50-90”. The diameter is 110 mm. The boreholes are used to some extent for hydraulic interference tests. The following geophysical logs are run in these boreholes: borehole deviation; point resistance; natural gamma. The geophysical logs are interpreted together with data on well capacity and drilling speed to determine the intersection of fracture zones in the boreholes. The results from the deviation me~urements indicate that there is a significant curvature of the percussion boreholes. In some areas almost all percussion boreholes deviate significantly downward compared to the expected direction and in some areas upward. An example of the deviation of the boreholes in the horizontal plane is given in Fig. 2 where the boreholes are plotted in their actual position. At each site 7-10 deep diamond boreholes are drilled to investigate the properties of fracture zones at depth. The holes have a diameter of 56 mm, and a length of about 700 m. Most boreholes are drilled with a dip of about 60” and at least one borehole is vertical. A comprehensive investigation program is carried out in the holes including geologic mapping of the cores, geophysical logging, hydraulic water injection tests, piezometric measurements and chemical water analyses. Samples are also taken from the drill cores for determination of several physical parameters and detailed fracture mineral studies. The geophysical data are used to describe the properties of the bedrock in the vicinity of the borehole, such as lithology, hydraulic properties, and water chemistry. The data are also used to determine sections in the boreholes for detailed measurements of hydraulic conductivity and chemical water sampling. The geophysical investigations are an important part of the integrated borehole investigation program to describe the properties of bedrock and to identify and ch~acterize fracture zones at depth. In all core boreholes a standard well logging program is conducted comprising the following methods: borehole deviation; natural gamma; point resistance; resistivity, normal and lateral, array length 1.6 m; self-potential; temperature; salinity (borehole fluid resistivity); induced polarization (made only in a limited number of holes at each site). To investigate the relation between the physical properties of the bedrock and the results obtained with the logs’ petrophysical measurements
196
are made on core samples. The following parameters are measured: density, porosity, resistivity, IP, susceptibility, and remanent magnetization. Since most of the logs used are standard geophysical logs they require little further explanation. Some comments will be made nevertheless on the experience gained as these might be of general interest and on the purpose of each log in the investigation program. The natural gamma log gives information on changes in the mineral composition of the bedrock, specifically the contents of radioactive minerals. The point-resistance log shows the presence of fractures and electrically conductive minerals in the borehole wall. The electrode of the point-resistance log is made very short (5 cm) and the probe is made so as to almost completely fill the borehole. The diameter of the probe is 53 mm which is 3 mm less than the diameter of the borehole. This design makes it possible to obtain a good resolution, and in some cases even single fractures are resolved. The effect of the borehole fluid is reduced by the large diameter of the probe, but in case of saline water (borehole fluid resistivity less than 10 ohm m), the resolution will deteriorate. The resistivity logs (normal and lateral) give information on the presence of fracture zones and conducting minerals. The resistivity probes also have a diameter of 53 mm. This makes it unnecessary to perform borehole fluid corrections in most cases. The electric logs are used together with the results from the core logging to define the width of the fracture zones. The normal resistivity may also be used to determine the porosity (Archie, 1942; Dahknov, 1962). The results obtained from the electric logs in a borehole in Gide”a are shown in Fig. 1. The resistivity of the unfractured rock at Gidei is in the range 50,000-100,000 ohm m. A fracture zone containing clay gives a large resistivity anomaly at a depth of about 380 m. Two minor fracture zones are also seen. Note the high resolution obtained with the pointresistance log and its correlation with the fracture frequency. The self-potential log indicates the presence of electrically conductive minerals (sulphides, graphite) and water flow in or out of the borehole. The self-potential measurements are made with a Cu-CuSO, electrode in the borehole. With these electrodes improved reproducibility is obtained compared to lead electrodes. The temperature log will of course give the temperature of the rock and the temperature gradient. As the temperature is actually measured in the borehole fluid, information is also obtained on water permeable fracture zones and water flow along the borehole. The borehole fluid resistivity log will give the salinity of the groundwater. Salinity variations indirectly indicate the presence of permeable fractures. The salinity is measured with a four-electrode system to minimize errors caused by oxidation of the electrodes In Fig. 5, the borehole logs from the borehole Km 12 in Kamlunge are shown. The upper part of the borehole intersects several fracture zones.
197
Fig. 5. Results obtained from temperature, salinity, single-point resistance and core logging and hydraulic conductivity from the borehole Km 12 at site Kamlunge.
i
4
1
4
3
2
1
LOGGlNG Km 12
BOREHOLE BOREHOLE
KAMLLJNGE
199
The induced polarization log will indicate the presence of electrically conductive minerals (e.g. sulphides, graphite) in the vicinity of the borehole. The zones are clearly indicated on the single-point resistance and the fracture frequency logs. Some of the fracture zones also have high hydraulic conductivity. The permeable zones cause temperature anomalies which are very easily seen on the temperature gradient plot. The temperature log is made only a few days after the drilling is completed, so that stable temperature conditions have not yet been obtained. One effect of the drilling is that warm water from deeper levels is injected into the permeable zones. The warm water will cause a temperature anomaly at the fracture which will decay with time (Drury, 1982). This type of anomaly can be seen at the fracture zones 2 and 3. The permeable fracture zone 4 is associated with a change in salinity. An interesting example of the lithologic information obtained from the geophysical logs is the borehole Fj5 (Fig. 6) in Fj’Zllveden. The dominating rock type in the borehole is a veined gneiss or gneiss migmatite. There are also strata of granite gneiss, which are important because they have a higher hydraulic conductivity than the veined gneiss (Ahlbom et al., 198313). The granite gneiss has higher contents of radioactive mineral, i.e., a higher gamma radiation level, and lower contents of sulphides, i.e., lower IP-effect and higher resistivity, compared to the veined gneiss. The
Fig. 6. Results obtained Fj 5 at site Fjiillveden.
from
resistivity,
gamma,
IP and core
logging
of the borehole
200
granite gneiss at a depth of about 400 m stands out clearly. The logs have been a substantial aid in the core logging and for the identification of the granite gneiss strata in the percussion boreholes. Some of the basic rocks are also identified due to their lower gamma radiation levels. Especially in rock with low content of electrically conductive minerals this combination of logging techniques has been efficient in the localization and characterization of fracture zones. However, at sites where the sulphide content is relatively high, there will be difficulties in discriminating between waterbearing fractures and the presence of conducting minerals. In the evaluation of the borehole data the results from the geological, geophysical and hydrological investigations in adjacent holes and from the surface measurements are correlated to build a three-dimensional model of the site. The orientation, extent and width of the fracture zones are inferred from the data. In some cases the data are ambiguous or too scanty to make possible a reliable model of the site. In such a case complementary investigations are made, e.g. extension of the area measured at the surface, additional boreholes and measurements therein. The mise-‘a-la-massemethod has also been used with some success to find the extent and orientation of fracture zones (Jamtlid et al., 1982). During the execution of the site investigation program there has been careful consideration of what logs to include in the standard investigations. The most obvious complement to the present program is the sonic log. The basic reason why the sonic log has not been used so far, is the lack of readily available equipment for 56-mm holes. However, the introduction of a sonic log in the program will soon be made. Over the last few years of investigations the methods used have changed. The following logs have been considered not to be effective and thus abandoned: induction log (borehole slingram), differential resistance, and VLF (Brotzen et al., 1980). CONCLUSIONS
The geophysical program described has been employed at several sites in granitic and gneissic rocks in Sweden. In the case of thin, resistive overburden the surface investigation program has proven to be effective in the detection of the dipping fracture zones. The methods used are not likely to detect sub-horizontal fracture zones, at least not those at some depth. The geophysical investigations give valuable data for the geological and hydrological characterization of a site. The geophysical logging techniques are normally useful in defining the fracture zones and different lithological units. In a detailed study by Magnusson and Duran (1983), a good correlation between fracture frequency and resistivity has been found. However, there was hardly any correlation between the resistivity and the hydraulic conductivity. The logs which give the best hydraulic infor-
201
mation, are actually the temperature and the salinity fogs, There is a need for new. geophysical logs which can give quantitative hydrologic~ information at a cost essentially lower than the hydraulic-injection tests. In this type of investigations there is a need for obtaining data on the properties of the bedrock and the fracture zones at considerable distance from the borehole. The borehole techniques used in this program normally give info~ation only about a relatively small volume around the borehole (in the order of 1 m). The development of cross-hole geophysical techniques which have high resolution and considerable range is needed. ACKNOWLEDGEMENTS
The financial contribution by the Swedish Nuclear Fuel Supply Co. for the preparation of this paper and the permission to publish these results is appreciated. REFERENCES Ahlbom, K., Carlsson, L. and Olsson, O., 1983a. Final disposal of spent nuclear fuel - geological, hydrogeological and geophysical methods for site characterization. KBS TR, 83-43, Stockholm. Ahlbom, K., Carlsson, L., Carlsten, S-E., Duran, O., Larsson, N-A. and Olsson, O., 1983b. Evaluation of the geological, geophysical and hydrogeological conditions at Fjzlveden. KBS TR, 83-52, Stockholm. Ahlbom, K., Albino, B., Carlsson, L., Nilsson, G., Olsson, O., Stenberg, L. and Timje, H., 1983~. Evaluation of the geological, geophysical and hydrogeological conditions at Gide%. KBS TR, 83-53, Stockholm. Ahlbom, K., Albino, B., Carlsson, L., Danielson, J., Nilsson, G., Olsson, O., Sehlstedt, S., Stejskal, V. and Stenberg, L., 1983d. Evaluation of the geological, geophysical and hydrogeologic~ conditions at Kamlunge. KBS TR, 83-54, Stockholm. Ahlbom, K., Carlsson, L., Gentzschein, B., JZmtlid, A., Olsson, 0. and Tiren, S., 1983e. Evaluation of the geological, geophysical and hydrogeological conditions at Svartboberget. KBS TR, 83-55, Stockholm. Almen, K., Hansson, K., Johansson, B. -E., Nilsson, G., Andersson, O., Wikberg, P. and ahagen, H., 1983. Final disposal of spent nuclear fuel - equipment for site ch~acter~ation. KBS TR, 83-44, Stockholm. Archie, G.E., 1942, The electrical resistivity log as an aid in determining some reservoir characteristics. Trans. A.I.M.E., 146: 54-62. Brotzen, O., Magnusson, K-B. and Duran, O., 1980. Evaluation of geophysical borehole studies. Prav. Rep., 4.14., Stockholm. Dahknov, V.N., 1962. Geophysical we11logging. Color. Sch. Mines, Golden, Cola. Drury, M.J., 1982. Borehole temperature logging for the detection of water flow. In: Geophysical Investigation in Connection with Geological Disposal of Radioactive Waste. OECD/NEA, Ottawa. (Also in: Geoexploration, 22: 231-243.) Henkel, H. and Guzman, M., 1977. Magnetic features of fracture zones. Geoexploration, 15: 173-181. Jgmtlid, A., Magnusson, K-A, and Olsson, O., 1982. Electrical borehole measurements for the mapping of fracture zones in crystalline rock. In: Geophysical Investigation in Connection with Geological Disposal of Radioactive Waste. OECD, Ottawa. Magnusson, K. -A. and Duran, O., 1982. Comparison between core log and hydraulic and geophysical measurements in boreholes. In: Geophysical .Investigation in Connection with Geological Disposal of Radioactive Waste. OECD, Ottawa, 1982. SKBF/KBS, 1983. Final Storage of Spent Nuclear Fuel. SKBF/KBS, Stockholm.
in The Netherlands
187
GEOPHYSICAL INVESTIGATIONS IN SWEDEN FOR THE CHARACTERIZATION OF A SITE FOR RADIOACTIVE WASTE DISPOSAL - AN OVERVIEW
0. OLSSON,
0. DURAN, A. JAMTLID
Sueriges Geologiska AB (SGAB), (Received December
and L. STENBERG
Uppsala (Sweden)
2, 1983; accepted January 11, 1984)
ABSTRACT Olsson, O., Duran, O., Jiimtlid, A. and Stenberg, L., 1984. Geophysical investigations in Sweden for the characterization of a site for radioactive waste disposal - an overview. Geoexploration, 22: 187-201. The geophysical measurements carried out for the investigations of prospective sites for radioactive waste disposal in Sweden are described. The investigations comprise detailed surface measurements with magnetic, electromagnetic, electric and seismic techniques. Measurements are made with a line spacing of 40 m and a station spacing of 20 m in an area of 4-6 km2. At each site 7-10 deep boreholes are drilled. A number of geophysical logs are run in the boreholes, including several electric logs, providing data on temperature, salinity, and gamma ray. These geophysical data are evaluated together with the results from geological and hydrological investigations to make a three-dimensional geological and tectonical model of the site. This model defines the geometry used in the numerical modelling of the groundwater flow. In the case of thin, resistive overburden the surface investigations proved to be effective in the detection of dipping fracture zones. A line spacing as low as 40 m is needed to get adequate mapping of the fracture zones. The methods used cannot be considered efficient for mapping sub-horizontal fracture zones. The borehole logs provide data for the identification of fracture zones as well as different lithological units and give qualitative information of the hydraulic conditions. The temperature and salinity logs have provided the best hydraulic information, whereas, after testing, the induction log (borehole slingram), the differential resistance log and the VLF-log were found to be not effective for our investigations. The borehole techniques used, normally give information only about a relatively small volume of rocks around the boreholes. To obtain a detailed description of the bedrock between the boreholes, development of cross-hole geophysical, high-resolution techniques of considerable range, is needed.
INTRODUCTION
The high-level radioactive waste from nuclear reactors is to be stored in a repository at considerable depth below the earth surface. To obtain a final storage which will protect the environment against the radioactive wastes for several hundredthousands of years, several barriers are constructed 0016-7142/84/$03.00
o 1984 Elsevier Science Publishers B.V.
188
to minimize the transport of radioactive nuclides. The barriers consist of the fuel material itself, the canister, the buffer material and the bedrock surrounding the repository (SKBF/KBS, 1983). In this context the transport of radionuclides through the bedrock will depend on the groundwater flow and the sorption of radionuclides to fracture minerals. The properties of the bedrock are important also because they may influence the performance of other barriers, e.g. the corrosion of copper canisters is dependent on the chemical properties of the groundwater. Comprehensive investigations have been carried out in Sweden at several different sites, mainly in gneissic or granitic rock. The objective has been to characterize the geological and hydrological conditions at each site and to produce the site-specific data necessary for a safety analysis of a repository situated at a depth of about 500 m. The investigations at a site are carried out as a closely integrated program in geology, geophysics and hydrology. The main objective of these investigations is to describe the factors essential for the safety of a final repository. Emphasis is thus put on describing the groundwater flow. In crystalline rock, a large part of the flow takes place in the fracture zones. Thus a large effort is made to determine the location and the hydraulic properties of the fracture zones. The detailed investigations on the surface cover an area of 4-6 km2 and the bedrock is investigated to a depth of about 700 m. The data are used to produce a descriptive geological and tectonical model of the site and a numerical model of the groundwater flow. The investigations in Sweden started in the mid-1970’s and since then several sites have been investigated (Fig. 1). The four sites, which have been investigated after 1979, have been investigated essentially according to the general site investigation program described by Ahlbom et al. (1983a). However, modifications in this general program have been made as called for by changes in the geological and hydrological conditions. In the present paper special attention will be paid to the geophysical part of the program and to some of the experiences gained. The four sites investigated according to this program are: (1) Site Fjallveden, 80 km southwest of Stockholm. This site is characterized by flat topography with minor fracture valleys in a northwesterly direction. The bedrock consists of veined gneiss with steeply dipping strata of granite gneiss (Ahlbom et al., 1983b). (2) Site Gide%, 30 km northeast of &nsktildsvik. This site is situated within a plateau-area of 100 km2. The topography of this site is flat. The bedrock consists of veined gneiss with strata of granite gneiss with a small dip (Ahlbom et al., 1983c). (3) Site Kamlunge, 65 km northeast of LuleP. This site is situated on a plateau with an area of 16 km2, about 100 m above the surrounding valleys. The bedrock consists of gneisses and red granite. Within the site a horizontal fracture zone has been found at a depth of 500 m. (Ahlbom et al., 1983d).
189
(4) Site Svartboberget, 12 km west of Edsbyn, is situated on a 2.5 km wide and 5 km long ridge. The surrounding valleys, which have a northwesterly direction, are at a level 75--85 m below the top of the ridge. The bedrock consists of a strongly metamorphosed gneiss, migmatite. Characteristic for this site is the presence of minor fracture zones with a mutual separation of 30-80 m. (Ahlbom et al., 1983e). The major part of the geological, geophysical and hydrological investigations have been performed by Sveriges Geologiska AB on commission from the Swedish Nuclear Fuel Supply Co.
SVARTBOBERGET
l
Fig. 1. The location of studied sites. The sites investigated after 1979 by filled dots, previously investigated sites by unfilled dots.
are indicated
190 CHOICE OF SITE
Reconnaissance studies are carried out continuously to find new sites which are to be studied in detail. Detailed studies are normally carried out at one or two sites each year. The following factors are considered in the selection of a site. ~0~0~~~~~ : a flat topography gives a small hydraulic gradient. Major fracture zones: the site should not be intersected by any major fracture zones. Minor fracture zones: the frequency of minor fracture zones should be small within the site. Rock type!: sites with different types of bedrock are to be studied within the long term program - until now the investigations have been concentrated to gneissic and granitic rocks. Ore mineralizations: a site likely to contain potential orebodies is avoided. The selection of a site for detailed investigations is made initially from available maps of geology, topography, and geophysical me~urements. The topo~aphic~ maps are studied to get a prelimin~ indication of the hydraulic gradients and the location of groundwater recharge and discharge areas. Data on well capacity is used to get a preliminary estimate of the hydraulic conductivity of the region and of the type of bedrock. Data from air photographs, satellite images, etc., are used to obtain info~ation about geologic st~c~res and the presence of important lineaments. Geologic field studies are made at prospective sites to gather additional data on outcrops, fracturing and rock-type distribution. If the site survives also this screening a preliminary geophysical study is made. Geophysical me~uremen~ are made as reconn~s~ce profiles on the ground surface. Measurements are made with proton magnetometer, horizontal-loop EM (slingram) and VLF. This combination of methods is normally sufficient to give a preliminary indication of the presence of larger fracture zones and the bedrock structure at the site. Interpretation is made of the dip, strike and thickness of fracture zones. The station spacing used is 20 m, except for the magnetic measurements where it is 5 m. If the data from the reconnaissance studies are still considered favorable, a reconnaissance borehole is drilled to get information about the bedrock properties at depth. The borehole is normally vertical with a depth of 800-1000 m. Geological mapping is made of the core and geophysical and hydrological measurements are made in the hole. The geophysical measurements will be described below. Following the evaluation of these studies a decision is taken whether the investigations are to be continued or not.
191 GEOPHYSICAL
SURFACE
INVESTIGATIONS
AT A SITE
The surface investigations at a site begin with a regional geologic and tectonic study covering an area of about 100 km’. After this, detailed geological and geophysical mapping is made of the rock type distribution and the fracture zones. The detailed investigations cover an area of 4-6 km2 in which a coordinate system is set up by means of stake lines to serve as a reference for all investigations to be made at the site. Geophysical measurements are made both in the coordinate system and in profiles outside of the coordinate system. The methods used in the coordinate system are: (a) proton magnetometer measurements; (b) horizontal loop EM (slingram) measurements, frequency 18 kHz, intercoil spacing 60 m; and (c) resistivity and time domain IP, measured using gradient array. The equipment used for the investigations is described by Almen et al., 1983. A line spacing of 40 m and a station spacing of 20 m is used. It has proven advantageous to decrease the station spacing of the magnetic measurements to 5 m as this greatly facilitates accurate dip determinations of thin dikes. In the investigations made before 1978 a line spacing of 80 m was used, but with this line spacing the interpretation of the strike of fracture zones often became ambiguous. The magnetic measurements are mostly used as an indicator of bedrock structure and normally indicate lithological changes and the presence of, e.g., dolerite dykes. In many cases the magnetic measurements also indicate the presence of faults and fracture zones. The water in the fracture zones may oxidize magnetite to hematite which implies that a fracture zone may be associated with a magnetic minimum (Henkel and Guzman, 1977). In Fig. 2 the magnetic measurements and the corresponding interpretation from site Gide% are shown. Several magnetic indications are to be seen which correspond to dolerite dykes (Ahlbom et al., 1983c). The high-frequency horizontal loop EM (slingram) is used to indicate the presence of fracture zones. The frequency is made high and the intercoil spacing is made large to obtain a high induction number making it possible to detect fracture zones, which can be considered to be poor sheetlike conductors. The large intercoil spacing also makes the depth of investigation acceptable. Especially in areas with a thin and/or highly resistive overburden, the high frequency EM method is efficient. However, the results obtained in areas covered by conductive clay, have not been satisfactory. In these cases the clay causes anomalies much larger than those which can be expected from fracture zones. An example of the results obtained and the corresponding interpretation at the test site Gide% is given in Fig. 3. In this area the bedrock resistivity is around 10,000 ohm m and overburden resistivity is about 200-1,000 ohm m, the overburden thickness varies from zero to about 5 m. In most
DISTINCT INDICATION WEAK INDICATION FAULT
INTERPRETATION
X80’ Gil @e-
STRIKE AND DIP DIAMOND DRILLED BOREHOLE PERCUSSION DRILLED BOREHOLE
_--
400 m
_
SWEDISH GEoLoGKAl
0
GIDEA
Fig. 2. Magnetic measurements and the corresponding interpretation from site Gide%.
---
-
LEGEND
MAGNETIC
IMAGINARY
LNTERPRETATION
Fig.
3.
Horizontal
loop
EM
(sling-ram)
LEGEND STRONG INDtCATlON X30’ DISTINCT INDICATION Gil@--WEAK INDICATION a-‘- WEAK RESLSTIVITY INDICATION
SLINGRAM
measurement,
imaginary
component
ST RlKE AND DIP DIAMOND DRILLED BOREHOLE PERCUSSION DRILLED BOREHOLE
.
,
,
LpOm
(quadrature)
and
I f
hi’
from site Gide%.
5.00 3.03 I.011 -1.00 -3. w -5.W -1O.W -15.w
IQ.03
15.w
NiJRNALF~LXT
ANPNALI
interpretation
SWEDISH GEOIBGICAL
q
GIDEA
194
cases the anomaly from a fracture zone will be about 3-5% in the imaginary (quadrature) component. Most of the “strong ” indications have been found after drilling to correspond to fracture zones with a width of 5-25 m (Ahlbom et al., 1983c). The resistivity method will also show the presence of good conductors such as fracture zones and mineralizations. The gradient array is used with a large-current electrode separation, thus the depth penetration should be considerable. The distance between the current electrodes is about 1.5 km and the distance between the potential electrodes is 20 m. Indications of conducting minerals have been obtained at depths of about 100 m but no definite indication of deeply situated fracture zones have yet been obtained. When the fracture zones are perpendicular to the direction of the electric field, the resistivity measurements seem to give a better indication of the width of the fracture zones compared to the high-frequency EM measurements. However, in most cases a better resolution of fracture zones is obtained with the high-frequency EM measurements. One advantage with the resistivity method is that it gives a measure of the bedrock resistivity between the fracture zones. Generally, high resistivity corresponds to bedrock with few fractures, The resistivity measurements may thus be used to identify blocks of rock with low fracturing. The resistivity measurements will also be affected by the presence of clay but not to the same extent as the high-frequency EM me~urements. The IP measurements, which are carried out simultaneously and with the same equipment as for the resistivity measurements, are used to discriminate between conductors which are due to mineralizations or fracture zones, The presence of a mineralization will cause both high IP-values and low resistivities while only a low r~sistivity is expected from fracture zones. In some cases the IP measurements give information about the distribution of different rock types and may thus indirectly indicate the presence of fracture zones. The gneissic sites often have disseminations of sulphides and graphite and thus have fairly large background IP values. Seismic refraction me~urements are made in profiles within the area of detailed investigations. The total length of the seismic profiles has normally been 4-6 km. The results are interpreted in terms of overburden thickness, substratum velocity and presence of fracture zones. The seismic method has normally been a good complement to the other methods and, in the case of conductive overburden, a necessary tool for inte~~tation of dubious anomalies which might be due to either fracture zones or mineralizations. To obtain information about the regional geologic structure in the vicinity of the site, profile measurements are made with proton magnetometer, slingram (horizontal loop EM), and VLF. Extensive profile measurements are made where complete airborne geophysical measurements are not available.
195
The objective of the geophysical measurements at the ground surface is to find the location and the orientation of fracture zones within the site. These data are used in the construction of a preliminary geologic and tectonic model of the site. The surface measurements are also used in combination with the geological information for positioning the boreholes that are made to investigate the bedrock and the fracture zones at large depth. GEOPHYSICAL
BOREHOLE
INVESTIGATIONS
AT A SITE
Percussion boreholes are made to investigate the shallow parts of fracture zones. These boreholes have a length of 100-150 m and dip 50-90”. The diameter is 110 mm. The boreholes are used to some extent for hydraulic interference tests. The following geophysical logs are run in these boreholes: borehole deviation; point resistance; natural gamma. The geophysical logs are interpreted together with data on well capacity and drilling speed to determine the intersection of fracture zones in the boreholes. The results from the deviation me~urements indicate that there is a significant curvature of the percussion boreholes. In some areas almost all percussion boreholes deviate significantly downward compared to the expected direction and in some areas upward. An example of the deviation of the boreholes in the horizontal plane is given in Fig. 2 where the boreholes are plotted in their actual position. At each site 7-10 deep diamond boreholes are drilled to investigate the properties of fracture zones at depth. The holes have a diameter of 56 mm, and a length of about 700 m. Most boreholes are drilled with a dip of about 60” and at least one borehole is vertical. A comprehensive investigation program is carried out in the holes including geologic mapping of the cores, geophysical logging, hydraulic water injection tests, piezometric measurements and chemical water analyses. Samples are also taken from the drill cores for determination of several physical parameters and detailed fracture mineral studies. The geophysical data are used to describe the properties of the bedrock in the vicinity of the borehole, such as lithology, hydraulic properties, and water chemistry. The data are also used to determine sections in the boreholes for detailed measurements of hydraulic conductivity and chemical water sampling. The geophysical investigations are an important part of the integrated borehole investigation program to describe the properties of bedrock and to identify and ch~acterize fracture zones at depth. In all core boreholes a standard well logging program is conducted comprising the following methods: borehole deviation; natural gamma; point resistance; resistivity, normal and lateral, array length 1.6 m; self-potential; temperature; salinity (borehole fluid resistivity); induced polarization (made only in a limited number of holes at each site). To investigate the relation between the physical properties of the bedrock and the results obtained with the logs’ petrophysical measurements
196
are made on core samples. The following parameters are measured: density, porosity, resistivity, IP, susceptibility, and remanent magnetization. Since most of the logs used are standard geophysical logs they require little further explanation. Some comments will be made nevertheless on the experience gained as these might be of general interest and on the purpose of each log in the investigation program. The natural gamma log gives information on changes in the mineral composition of the bedrock, specifically the contents of radioactive minerals. The point-resistance log shows the presence of fractures and electrically conductive minerals in the borehole wall. The electrode of the point-resistance log is made very short (5 cm) and the probe is made so as to almost completely fill the borehole. The diameter of the probe is 53 mm which is 3 mm less than the diameter of the borehole. This design makes it possible to obtain a good resolution, and in some cases even single fractures are resolved. The effect of the borehole fluid is reduced by the large diameter of the probe, but in case of saline water (borehole fluid resistivity less than 10 ohm m), the resolution will deteriorate. The resistivity logs (normal and lateral) give information on the presence of fracture zones and conducting minerals. The resistivity probes also have a diameter of 53 mm. This makes it unnecessary to perform borehole fluid corrections in most cases. The electric logs are used together with the results from the core logging to define the width of the fracture zones. The normal resistivity may also be used to determine the porosity (Archie, 1942; Dahknov, 1962). The results obtained from the electric logs in a borehole in Gide”a are shown in Fig. 1. The resistivity of the unfractured rock at Gidei is in the range 50,000-100,000 ohm m. A fracture zone containing clay gives a large resistivity anomaly at a depth of about 380 m. Two minor fracture zones are also seen. Note the high resolution obtained with the pointresistance log and its correlation with the fracture frequency. The self-potential log indicates the presence of electrically conductive minerals (sulphides, graphite) and water flow in or out of the borehole. The self-potential measurements are made with a Cu-CuSO, electrode in the borehole. With these electrodes improved reproducibility is obtained compared to lead electrodes. The temperature log will of course give the temperature of the rock and the temperature gradient. As the temperature is actually measured in the borehole fluid, information is also obtained on water permeable fracture zones and water flow along the borehole. The borehole fluid resistivity log will give the salinity of the groundwater. Salinity variations indirectly indicate the presence of permeable fractures. The salinity is measured with a four-electrode system to minimize errors caused by oxidation of the electrodes In Fig. 5, the borehole logs from the borehole Km 12 in Kamlunge are shown. The upper part of the borehole intersects several fracture zones.
197
Fig. 5. Results obtained from temperature, salinity, single-point resistance and core logging and hydraulic conductivity from the borehole Km 12 at site Kamlunge.
i
4
1
4
3
2
1
LOGGlNG Km 12
BOREHOLE BOREHOLE
KAMLLJNGE
199
The induced polarization log will indicate the presence of electrically conductive minerals (e.g. sulphides, graphite) in the vicinity of the borehole. The zones are clearly indicated on the single-point resistance and the fracture frequency logs. Some of the fracture zones also have high hydraulic conductivity. The permeable zones cause temperature anomalies which are very easily seen on the temperature gradient plot. The temperature log is made only a few days after the drilling is completed, so that stable temperature conditions have not yet been obtained. One effect of the drilling is that warm water from deeper levels is injected into the permeable zones. The warm water will cause a temperature anomaly at the fracture which will decay with time (Drury, 1982). This type of anomaly can be seen at the fracture zones 2 and 3. The permeable fracture zone 4 is associated with a change in salinity. An interesting example of the lithologic information obtained from the geophysical logs is the borehole Fj5 (Fig. 6) in Fj’Zllveden. The dominating rock type in the borehole is a veined gneiss or gneiss migmatite. There are also strata of granite gneiss, which are important because they have a higher hydraulic conductivity than the veined gneiss (Ahlbom et al., 198313). The granite gneiss has higher contents of radioactive mineral, i.e., a higher gamma radiation level, and lower contents of sulphides, i.e., lower IP-effect and higher resistivity, compared to the veined gneiss. The
Fig. 6. Results obtained Fj 5 at site Fjiillveden.
from
resistivity,
gamma,
IP and core
logging
of the borehole
200
granite gneiss at a depth of about 400 m stands out clearly. The logs have been a substantial aid in the core logging and for the identification of the granite gneiss strata in the percussion boreholes. Some of the basic rocks are also identified due to their lower gamma radiation levels. Especially in rock with low content of electrically conductive minerals this combination of logging techniques has been efficient in the localization and characterization of fracture zones. However, at sites where the sulphide content is relatively high, there will be difficulties in discriminating between waterbearing fractures and the presence of conducting minerals. In the evaluation of the borehole data the results from the geological, geophysical and hydrological investigations in adjacent holes and from the surface measurements are correlated to build a three-dimensional model of the site. The orientation, extent and width of the fracture zones are inferred from the data. In some cases the data are ambiguous or too scanty to make possible a reliable model of the site. In such a case complementary investigations are made, e.g. extension of the area measured at the surface, additional boreholes and measurements therein. The mise-‘a-la-massemethod has also been used with some success to find the extent and orientation of fracture zones (Jamtlid et al., 1982). During the execution of the site investigation program there has been careful consideration of what logs to include in the standard investigations. The most obvious complement to the present program is the sonic log. The basic reason why the sonic log has not been used so far, is the lack of readily available equipment for 56-mm holes. However, the introduction of a sonic log in the program will soon be made. Over the last few years of investigations the methods used have changed. The following logs have been considered not to be effective and thus abandoned: induction log (borehole slingram), differential resistance, and VLF (Brotzen et al., 1980). CONCLUSIONS
The geophysical program described has been employed at several sites in granitic and gneissic rocks in Sweden. In the case of thin, resistive overburden the surface investigation program has proven to be effective in the detection of the dipping fracture zones. The methods used are not likely to detect sub-horizontal fracture zones, at least not those at some depth. The geophysical investigations give valuable data for the geological and hydrological characterization of a site. The geophysical logging techniques are normally useful in defining the fracture zones and different lithological units. In a detailed study by Magnusson and Duran (1983), a good correlation between fracture frequency and resistivity has been found. However, there was hardly any correlation between the resistivity and the hydraulic conductivity. The logs which give the best hydraulic infor-
201
mation, are actually the temperature and the salinity fogs, There is a need for new. geophysical logs which can give quantitative hydrologic~ information at a cost essentially lower than the hydraulic-injection tests. In this type of investigations there is a need for obtaining data on the properties of the bedrock and the fracture zones at considerable distance from the borehole. The borehole techniques used in this program normally give info~ation only about a relatively small volume around the borehole (in the order of 1 m). The development of cross-hole geophysical techniques which have high resolution and considerable range is needed. ACKNOWLEDGEMENTS
The financial contribution by the Swedish Nuclear Fuel Supply Co. for the preparation of this paper and the permission to publish these results is appreciated. REFERENCES Ahlbom, K., Carlsson, L. and Olsson, O., 1983a. Final disposal of spent nuclear fuel - geological, hydrogeological and geophysical methods for site characterization. KBS TR, 83-43, Stockholm. Ahlbom, K., Carlsson, L., Carlsten, S-E., Duran, O., Larsson, N-A. and Olsson, O., 1983b. Evaluation of the geological, geophysical and hydrogeological conditions at Fjzlveden. KBS TR, 83-52, Stockholm. Ahlbom, K., Albino, B., Carlsson, L., Nilsson, G., Olsson, O., Stenberg, L. and Timje, H., 1983~. Evaluation of the geological, geophysical and hydrogeological conditions at Gide%. KBS TR, 83-53, Stockholm. Ahlbom, K., Albino, B., Carlsson, L., Danielson, J., Nilsson, G., Olsson, O., Sehlstedt, S., Stejskal, V. and Stenberg, L., 1983d. Evaluation of the geological, geophysical and hydrogeologic~ conditions at Kamlunge. KBS TR, 83-54, Stockholm. Ahlbom, K., Carlsson, L., Gentzschein, B., JZmtlid, A., Olsson, 0. and Tiren, S., 1983e. Evaluation of the geological, geophysical and hydrogeological conditions at Svartboberget. KBS TR, 83-55, Stockholm. Almen, K., Hansson, K., Johansson, B. -E., Nilsson, G., Andersson, O., Wikberg, P. and ahagen, H., 1983. Final disposal of spent nuclear fuel - equipment for site ch~acter~ation. KBS TR, 83-44, Stockholm. Archie, G.E., 1942, The electrical resistivity log as an aid in determining some reservoir characteristics. Trans. A.I.M.E., 146: 54-62. Brotzen, O., Magnusson, K-B. and Duran, O., 1980. Evaluation of geophysical borehole studies. Prav. Rep., 4.14., Stockholm. Dahknov, V.N., 1962. Geophysical we11logging. Color. Sch. Mines, Golden, Cola. Drury, M.J., 1982. Borehole temperature logging for the detection of water flow. In: Geophysical Investigation in Connection with Geological Disposal of Radioactive Waste. OECD/NEA, Ottawa. (Also in: Geoexploration, 22: 231-243.) Henkel, H. and Guzman, M., 1977. Magnetic features of fracture zones. Geoexploration, 15: 173-181. Jgmtlid, A., Magnusson, K-A, and Olsson, O., 1982. Electrical borehole measurements for the mapping of fracture zones in crystalline rock. In: Geophysical Investigation in Connection with Geological Disposal of Radioactive Waste. OECD, Ottawa. Magnusson, K. -A. and Duran, O., 1982. Comparison between core log and hydraulic and geophysical measurements in boreholes. In: Geophysical .Investigation in Connection with Geological Disposal of Radioactive Waste. OECD, Ottawa, 1982. SKBF/KBS, 1983. Final Storage of Spent Nuclear Fuel. SKBF/KBS, Stockholm.