POSIVA 2012-34
Seismic Activity Parameters
of the Olkiluoto Site
Jouni Saari
ÅF-Consult Oy
August 2012
Base maps: ©National Land Survey, permission 41/MML/12
POSIVA OY
Olkiluoto
FI-27160 EURAJOKI, FINLAND
Phone (02) 8372 31 (nat.), (+358-2-) 8372 31 (int.)
Fax (02) 8372 3809 (nat.), (+358-2-) 8372 3809 (int.)
ISBN 978-951-652-215-2
ISSN 1239-3096
Raportin tunnus – Report code
Posiva-raportti – Posiva Report
POSIVA 2012-34
Posiva Oy
Olkiluoto
FI-27160 EURAJOKI, FINLAND
Puh. 02-8372 (31) – Int. Tel. +358 2 8372 (31)
Tekijä(t) – Author(s)
Julkaisuaika – Date
August 2012
Toimeksiantaja(t) – Commissioned by
Jouni Saari, ÅF-Consult Oy
Posiva Oy
Nimeke – Title
SEISMIC ACTIVITY PARAMETERS OF THE OLKILUOTO SITE
Tiivistelmä – Abstract
This study deals with the current and future seismicity associated with the Olkiluoto site. Seismic
belts that participate in the seismic behaviour of the studied site have been identified and the
magnitude-frequency distributions of these belts have been estimated. The seismic activity
parameters of the Olkiluoto site have been deduced in order to forecast the seismicity during the
next 10 000, 50 000 and 100 000 years.
The report discusses the long-term seismicity before the next ice age and the possible earthquakes
induced by the future glaciation. A major assumption of this study has been that future seismicity
will generally resemble the current one. However, when the postglacial seismicity is concerned,
the magnitude-frequency distribution is likely different and the expected maximum magnitude
will be higher. Maximum magnitudes of future postglacial earthquakes have been approximated
by strain release examinations.
The seismotectonic interpretation seems to indicate that the previous postglacial faults in Lapland
have been generated in compressional environment. The orientation of the rather uniform
compression has been NW-SE, which coincide with the current stress field. It seems that, although
the impact of postglacial crustal rebound must have been significant, the impact of plate tectonics
has been dominant.
Seismicity has been examined within the framework of the lineament map, in order to associate
the future significant earthquakes with active fault zones in the Olkiluoto area. The interpretation
includes structures large enough to host an earthquake with magnitude M = 7. The brittle
deformation zone model of the Olkiluoto area is suitable for reactivation analysis of earthquakes
slightly larger than M5. Frequency estimates for time periods of 10 000, 50 000 and 100 000 years
are estimated for magnitudes M > 5 and for magnitudes M > 7.
Due to uncertainty in seismotectonic interpretation, the concept of diffuse seismicity and tectonics
is also applied. This approach assumes that there is not a priori information about the tectonic
structures in the vicinity of the investigated site.
Avainsanat - Keywords
Seismic belts, magnitude-frequency relation, strain release, glaciation, potential active fracture
zones, Olkiluoto.
ISBN
ISSN
ISBN 978-951-652-215-2
Sivumäärä – Number of pages
60
ISSN 1239-3096
Kieli – Language
English
Raportin tunnus – Report code
Posiva-raportti – Posiva Report
POSIVA 2012-34
Posiva Oy
Olkiluoto
FI-27160 EURAJOKI, FINLAND
Puh. 02-8372 (31) – Int. Tel. +358 2 8372 (31)
Tekijä(t) – Author(s)
Julkaisuaika – Date
Elokuu 2012
Toimeksiantaja(t) – Commissioned by
Jouni Saari, ÅF-Consult Oy
Posiva Oy
Nimeke – Title
OLKILUODON ALUEEN SEISMISTÄ AKTIIVISUUTTA KUVAAVIA PARAMETREJÄ
Tiivistelmä – Abstract
Tässä tutkimuksessa arvioidaan Olkiluodon alueen nykyistä ja tulevaa seismisyyttä. Työn aluksi
selvitään Olkiluodon seismisyyteen liittyvät alueelliset vyöhykkeet. Näiden vyöhykkeiden
maanjäristysten magnitudi-lukumäärä relaatiosta johdetaan Olkiluodon alueen seismistä
aktiivisuutta kuvaavat parametrit. Pyrkimyksenä on esittää arvio Olkiluodon alueen seismisestä
aktiivisuudesta seuraavien 10 000, 50 000 ja 100 000 vuoden aikana.
Työssä tarkastellaan pitkän aikavälin seismisyyttä ennen jääkautta sekä tulevan jääkauden
mahdollisesti aiheuttamia maanjäristyksiä. Tutkimuksessa oletetaan, että tuleva seismisyys on
pääpiirteissään nykyisenkaltaista. Kuitenkin, jääkauden jälkeisenä aikana maanjäristysten
magnitudi-lukumäärä jakauma on luultavasti erilainen ja maanjäristykset ovat suurempia.
Jääkauden jälkeisten maanjäristysten suuruuden ylärajaa on arvioitu tarkastelemalla
maanjäristyksissä vapautuvia jännityksiä.
Seismotektonisen tulkinnan mukaan Lapissa olevat edellisen jääkauden jälkeiset siirrokset ovat
syntyneet olosuhteissa, joissa vaakapuristus on ollut vallitsevana. Tämän puristuksen suunta on
ollut yleensä sama kuin nykyisin: luode-kaakko. Näyttäisi siltä, että vaikka jääkauden jälkeisen
maankohoamisen vaikutuksen on täytynyt olla huomattava, laattatektoniikka on myös tuolloin
ollut kallioliikkeiden pääasiallinen syy.
Olkiluodon alueen lineamenttikartalta on pyritty tunnistamaan sellaisia ruhjevyöhykkeitä, joissa
saattaisi tapahtua merkittäviä maanjäristyksiä. Kartalla on rakenteita, jotka ovat riittävän suuria
siihen, että niissä voisi tapahtua magnitudin M = 7 maanjäristys. Olkiluodon alueen hauraista
deformaatiovyöhykkeistä tehty tulkinta soveltuu toistuvuusanalyysiin maanjäristyksille, joiden
magnitudi on hieman yli 5. Työssä esitetään 10 000, 50 000 ja 100 000 vuoden periodeille
toistuvuusarvio magnitudeille M > 5 ja M > 7.
Seismotektonisen tulkinnan epävarmuudesta johtuen tarkastellaan myös diffuusin seismisyyden ja
tektoniikan mahdollisuutta. Tässä lähestymistavassa oletetaan, että tutkimusalueen tektonisista
rakenteista ei ole ennalta mitään tietoa.
Avainsanat - Keywords
Seisminen vyöhyke, magnitudi-lukumäärä suhde, jännityksen vapautuminen seismisesti, jääkausi,
seismisesti aktiivinen ruhjevyöhyke, Olkiluoto.
ISBN
ISSN
ISBN 978-951-652-215-2
Sivumäärä – Number of pages
60
ISSN 1239-3096
Kieli – Language
Englanti
LIST OF ABBREVIATIONS AND KEY WORDS
(Modified mainly from http://earthquakescanada.nrcan.gc.ca/info-gen/glossa-eng.php
maintained by Natural Resources Canada)
Aftershock
An earthquake that occurs after a "main shock" (or larger
earthquake). Aftershocks occur in the same general region as
the "main shock" and result from readjustments of stress at
places along and in the vicinity of the fault zone.
BABEL
BAltic and Bothnian Echoes from the Lithosphere
BFZ
Brittle Fault Zone.
B-L zone
Zone of higher seismic activity that runs from the southern
Bothnian Sea towards Ladoga.
BP
Before present.
CFAZ
Central Finland Active Zone.
Dip-slip fault
A fault in which the relative displacement is along the
direction of dip of the fault plane; the offset is either normal or
reverse.
Earthquake
The sudden release of stored elastic energy caused by the
sudden movement of rocks along a fault. Some of the energy
released is in the form of seismic waves that cause the ground
to shake.
Earthquake swarm
Series of small or moderate magnitude earthquakes without a
distinct major event occurring in a limited area and time.
Epicenter
The point on the earth's surface directly above the focus
(hypocentre) of an earthquake.
Fault
A zone of fractures or breaks in rocks, where movements
occur. Earthquakes often occur along faults because they are
weak zones in the rock.
Fault plane
The plane that most closely coincides with the rupture surface
of a fault.
FENCAT
Earthquake catalogue for Northern Europe compiled and
maintained by the Institute of Seismology of Helsinki
University (http://www.helsinki.fi/geo/seismo/index.html). The
catalogue contains a well-quantified and homogenous
database, with spatial distribution of the epicenters being as
complete as possible.
Foreshock
An earthquake that is smaller than, and precedes, a "main
shock". Foreshocks tend to occur in the same area as the main
shock.
I
Intensity value describes the earthquake effects estimated on
the basis of felt observations. It indicates the local effects and
potential for damage produced by an earthquake on as it affects
humans, animals, structures, and natural objects. These effects
may range from I (not felt) and XII (total damage) on the
Modified Mercalli scale. An earthquake can be described by
one magnitude, but many intensities since the earthquake
effects vary with circumstances such as distance from the
epicenter and local soil conditions.
Intraplate earthquake
Earthquake with its focus within a tectonic plate. Earthquakes
in the Fennoscandian Shield are of this type.
KBS-3
Technology for disposal of high-level radioactive waste
developed in Sweden.
LDF
Lineament determining feature.
kyr
Thousand years.
M
Magnitude is a measure of the amount of energy released
during an earthquake. The data base includes magnitudes
based on macroseismical (MM) and instrumental data (ML). M
is used to represent both magnitude units.
ML
Local magnitude (ML) is based on instrumental recordings and
corresponds the original Richter magnitude. Magnitude and
location of earthquakes in Finland are predominantly based on
instrumental recordings since 1965.
MM
Macroseismic magnitude. Macroseismology studies historical
earthquakes. It is the part of noninstrumental seismology that
investigates written documentary materials testifying of the
effects of local and regional earthquakes. Magnitude and
location of earthquakes in Finland are predominantly based on
macroseismic studies before 1965.
Main shock
The largest earthquake in a "cluster" of earthquakes. Main
shocks are sometimes preceded by "foreshocks", and generally
followed by aftershocks.
Normal fault
A dip-slip fault in which the rock above the fault plane has
moved downward with respect to the rock below.
Oblique faulting
The slip on the fault has components both along the dip and
along the strike of the fault.
Paleoseismology
The study of ancient (prehistoric) earthquakes from their
geological evidences.
Reverse faulting
TA dip-slip fault in which he rock above the fault plane (the
"hanging" wall) moves up and over the rock below ("foot"
wall).
Seismotectonics
The study of earthquakes and their relationships with faults.
Strike-slip fault
A fault whose relative displacement is purely horizontal.
SFQZ
Southern Finland Quiet Zone.
SKB
Swedish Nuclear Fuel and Waste Management Company.
STUK
Finnish Radiation and Nuclear Safety Authority.
Target area
Region within a radius of 100 km from Olkiluoto
Thrust fault
A reverse fault in which the upper rocks above the fault plane
move up and over the lower rocks at an angle of 30° or less.
YVL
Regulatory guides on nuclear safety published and maintained
by STUK.
Å-P-P Zone of higher seismic activity that runs from the Åland archipelago via Paldis to
Pskov in Estonia.
1
TABLE OF CONTENTS
ABSTRACT
TIIVISTELMÄ
LIST OF ABBREVIATIONS AND KEY WORDS
1
INTRODUCTION .................................................................................................... 3
1.1 General ............................................................................................................... 3
1.2 Background and objectives ................................................................................. 3
2
CALCULATION OF EARTHQUAKE PARAMETERS ............................................. 7
2.1 Seismic belts and target area ............................................................................. 7
2.2 Magnitude-Frequency distribution for future earthquakes ................................ 11
3
FUTURE GLACIATION AND SEISMICITY........................................................... 23
3.1 General ............................................................................................................. 23
3.2 Past and future glaciations................................................................................ 24
3.3 Discussion about maximum magnitude ............................................................ 32
4
LINEAMENTS, FAULTS AND SEISMICITY ......................................................... 37
4.1 General ............................................................................................................. 37
4.2 Diffuse seismicity and tectonics ........................................................................ 40
4.3 Lineament maps and brittle fault zone modelling.............................................. 41
5
CONCLUDING REMARKS ................................................................................... 53
REFERENCES ............................................................................................................. 55
2
3
1 INTRODUCTION
1.1 General
Posiva Oy is planning to dispose of spent nuclear fuel in a KBS-3 type repository, to be
constructed at a depth of about 400 m in the crystalline bedrock at Olkiluoto, Finland.
Due to its long-term hazard, the spent fuel has to be isolated from the surface
environment over a prolonged period of time. Safe disposal is achieved by long-term
isolation and containment. In the KBS-3 method, the spent fuel is encapsulated in gas
and water tight copper-iron canisters. The canisters are surrounded by a clay buffer that
protects the canisters from rock movements and from potential detrimental substances
in the groundwater, and the backfill of the deposition tunnel that supports the buffer and
the rock. The host rock shall provide stable and favourable groundwater conditions that
support the longevity of the canisters. Should any of the canisters start leaking, the
repository system as a whole shall hinder or retard the releases of radionuclides to the
biosphere to the level required by the long-term safety criteria. The long-term safety of
the repository shall be demonstrated by means of a safety case.
Posiva is preparing to submit the construction license application for a spent fuel
repository at Olkiluoto by end of the year 2012. This study aims to give input on the
probability of large earthquakes at the Olkiluoto, to recognize the zones that may host
large earthquakes and to support developing the rock suitability criteria. Low seismic
activity at site needs to be demonstrated in order to show that the host rock is stable and
thus acts a natural barrier (YVL D.5 406 and 410). However, the regulations also
require an assessment of the importance of unlikely events impairing long-term safety
by considering the reality, likelihood and possible consequences of each event - rock
movements jeopardizing the integrity of disposal canisters named as one such event
(YVL D.5 314 and 315). Earlier assessments of long-term safety of spent fuel repository
at Olkiluoto (e.g. Vieno & Nordman 1999, Posiva 2010) have considered canister
failure due to rock shear as one way leading to release of radionuclides from the
canister. Olkiluoto is known to be located on a seismically stable area. However, the
occurrence of large earthquakes in the future cannot be excluded, particularly in
association with the retreat of ice sheets. The likelihood for canister failure due to rock
shear is further minimised by applying the rock suitability criteria (RSC). According to
RSC the canisters are emplaced such that they are not in the vicinity of the fault zones
or intersected by large fractures, which may undergo shear movements that have a
potential to break the canister (see e.g. Hellä et al. 2009).
1.2 Background and objectives
Dowding and Rozan (1978) have correlated tunnel damages with surface peak
acceleration (Figure 1-1). Observations of 71 tunnels responding to earthquake motion
were compared. Peak acceleration at surface less than 0.2g did not damage the tunnels.
Damage between 0.2g and 0.5g was only minor and significant only above 0.5g. Figure
1-1 correlates also the damage level to Richter’s magnitude and distance between
epicenter and the tunnel location. The expected maximum magnitude of Finnish
earthquakes is ML = 5.0 (Ahjos et al. 1984), i.e. current seismicity is clearly in “No
Damage Zone” in Figure 1-1.
4
Figure 1-1. Comparison of acceleration and Modified Mercalli Intensity of surface
motion to observed damage to subsurface tunnels. The damage level is also compared
to Richter’s magnitude and distance between earthquake epicenter and tunnel location
(Dowding & Rozan 1978).
On 11 March 2011 an earthquake occurred in Japan (Tohuku, M = 9.0, depth = 32 km
and peak ground acceleration = 3g). No major damage was reported in the urban
subway and railway tunnels by the earthquake. Only small scale leakage of water was
observed from the joint part of underground buildings. The closest underground
structures were 60 - 70 km from the earthquake fault. All in all there is no substantial
collapse of subways by a number of earthquakes in history. One exceptional collapse
occurred at Daikai station in Kobe by the 1995 earthquake (M = 7.2) mainly due to poor
seismic design of columns and shallow (about 5 m) overburden depth (Hanamura 2011).
Tunnel damages appear when tunnel is crossed by a fault or fault fissures. Strong
earthquakes cause little or no damage in tunnels that are well constructed, outside the
epicentral region and away from fault breaks. Within usual range of destructive
earthquake periods, intensity of shaking below ground is less intensive than on the
surface. Damages are mostly found near portals and in near surface tunnels (Hanamura
5
2011). Impacts of the earthquakes on the underground spaces are discussed also in
Posiva 2012a, Section 7.4.
Seismic activity parameters of the Finnish potential repository sites Romuvaara,
Kivetty, Olkiluoto and Hästholmen were assessed (Saari 2000) as a part of the project
"Estimation of rock movements due to future earthquakes at four candidate sites for a
spent fuel repository in Finland". The purpose of the project was to approximate
possible rock movements due to future earthquakes in fractures intersecting the
deposition holes containing spent nuclear fuel. The results of the project were presented
by La Pointe & Hermansson (2002). Further Olkiluoto specific studies include
assessment of stress evolution and fault stability during the Weichselian glaciation by
Lund & Schmidt (2011) and a modelling study estimating the shear movements by Fälth
and Hökmark (2011).
The aim of this study is to update the approximation of the future seismicity related to
the Olkiluoto repository site. The earthquake data used both by Saari (2000) and in this
study are taken from the earthquake catalogue for Northern Europe (FENCAT), which
is compiled and maintained by the Institute of Seismology of Helsinki University
(http://www.helsinki.fi/geo/seismo/index.html). The earthquake catalogue contains a
well-quantified and homogenous database, with spatial distribution of the epicenters
being as complete as possible. The content of the catalogue is described in detail by
Ahjos & Uski (1992). The study by Saari (2000) included seismic history from 1375 to
1998. This study includes data until the end of 2010.
The quality of geophysical data and interpretation has improved significantly after the
previous study of seismicity parameters of Olkiluoto region. New seismic events have
occurred and our knowledge about the tectonic characteristics of the Fennoscandian
Shield and the Olkiluoto region has increased.
The regional seismicity is studied in relation to the seismicity and fault configuration in
the target area, which is a region within a radius of 100 km from Olkiluoto. The task is
divided into two subtasks:
1) Estimation of seismicity during future 100 000 years. That includes separate
examination of periods:
a)
0 - 10 000 years.
b)
10 000 - 50 000 (Period before the next glaciations.)
c)
During deglaciation.
2) Attempt to associate the future earthquakes with the faults of the target area.
The time periods have been selected according to the reference climate scenario applied
in the safety assessment see Figure 1-2. Also, the dose constraints are applied for the
10 000 years period whereas nuclide specific constraints for the releases to environment
(average release of radioactive substances per annum) are applied after that (YVL D.5
306, 311, 312).
6
A 130
110 90 70 50
30
10 0
ka BP 50
70
90
110
130
150
170
ka AP
B Permafrost Ice Sheet Temperate Figure 1-2. A) Schematic representation of the occurrence of permafrost, ice sheets,
and temperate periods during the last glacial cycle (LGC). B) The repetition of the past
glacial cycle after 50,000 years AP onwards years (see Chapter 4 in the Formulation of
Radionuclide Release Scenarios report, Posiva 2012b).
7
2 CALCULATION OF EARTHQUAKE PARAMETERS
2.1 Seismic belts and target area
With the scarce data in FENCAT it is not possible to establish statistically reliable
parameters for the Olkiluoto target area directly. Scaling parameters of larger regional
seismic belts to represent the target area solves this problem.
In Southern Finland earthquakes are sparse and small (M< 5.0). Generally, the majority
of instrumentally located Fennoscandian earthquakes occur in the depth range of 10 - 20
km (Slunga 1991, Ahjos & Uski 1992). The approximated location accuracy of
macroseismically and instrumentally located earthquakes in Finland are 10-100 km and
5-10 km, respectively (Ahjos & Arhe 1983). A preliminary study of instrumentally
located earthquakes in the Bothnian Bay region indicates gradual improvement of the
location accuracy. The calculated mean errors of the locations were about 5.5 km
(before 1990), about 4.5 km (1990-2000) and about 3.5 km (2000-2007). The result is
suggestive but not necessary valid elsewhere in Finland, because of the uneven
distribution of Finnish seismic stations (Valtonen 2012).
Although the number of instrumentally located events increases gradually, still a half of
the events in southern Finland are macroseismically located. Due to the insufficient
location accuracy, it is mainly not possible to accurately associate the epicenters with
certain individual zones of weakness. Instead, seismic belts characterised by different
levels of potential seismicity can be distinguished.
Figure 2-1 presents seismicity within the distance of 500 km from Olkiluoto. The most
active belts of seismicity within the area re the Swedish coast from the Bothnian Sea to
the Bothnian Bay and the northern Bothnian Bay-Kuusamo region (see also Figure 2-3).
Southern Finland, northern Baltic and northwestern Russia are characterised by
relatively low seismicity.
Distribution of larger earthquakes as well as periods of higher seismic activity may
bring out some kind of transmitting mechanism between the earthquakes. Figures 2-2
and 2-3 presents two NW-SE oriented belts of relatively high seismic activity in
southern and central Finland. The northern zone of higher activity runs from the
southern Bothnian Sea towards Ladoga (B-L zone). The other active belt runs from the
Åland archipelago to southeastern Estonia, where it extends to from Paldis to Pskov (ÅP-P zone). The zones are distinguished from their surroundings particularly by the
occurrence of relatively large earthquakes. The period of pronounced seismic activity
from 1920 to 1941 brings out the same seismic belts in southern Finland and elsewhere
as well (see more details in Saari 1998). The general pattern of seismic zones in
southern Finland has remained the same as it was according to data based on the
seismicity before 1995: earthquakes with magnitude M  3.0 occur inside the active
seismic belts. After 1995 the earthquakes in southern Finland have been smaller than
magnitude 3.
8
Figure 2-1. Distribution of the regional seismicity within the distance of 500 km from
Olkiluoto (1375-2010) according to FENCAT. Preliminary data for years 2009 and
2010 are included to official FENCAT.
Olkiluoto is in the Southern Finland quiet zone (SFQZ), but rather close to the
seismically active Å-P-P zone. Therefore the seismicity of the Å-P-P cannot be
disregarded, when the seismicity of Olkiluoto is characterized. The seismicity of the
Olkiluoto target area (100 km from Olkiluoto) describes the actual seismicity in
Olkiluoto more accurately. About 2/3 of the Olkiluoto target area belongs to SFQZ and
about 1/3 to the Å-P-P Seismic Zone. As in the previous study (Saari 2000), it is
assumed that 1/3 of the seismic activity origin from the Å-P-P Seismic Zone and 2/3
from the SFQZ.
Recent studies make it possible to present an alternative and as possible interpretation
(Figures 2-2 and 2-3), which combines seismicity, geology and Moho depth maps
compared to the presentation above which is mainly based on observations of seismic
activity. The south-easternmost part of the B-L zone was associated with the rest of the
B-L zone with long NE-SE oriented shear zones and the period of increased seismic
9
activity 1920 - 1941. According to the interpretation by Dr. A. Korja (Institute of
Seismology, discussions 8.10.2010) the active area in the central Finland does not
necessarily extend as far to SE as the B-L zone. As distinct from the B-L zone this
shorter zone is called here as Central Finland Active Zone (CFAZ). Unlike the B-L
zone, CFAZ zone would remain totally inside the accretionary arc complex of central
and western Finland (Korsman et al. 1997). CFAZ includes still all the largest
earthquakes of the B-L zone and it represents area of high Moho depth (see Figure 2-2).
Figure 2-2. Belts of higher seismic activity (shaded areas). Historical events (13751964) with magnitude M  3.5 and instrumentally located (1965-2010) events with
magnitude M  3.0 are shown by light blue and dark blue filled circles, respectively.
Crosses denote earthquakes during the period 1920-1941. The alternative
interpretation of the active zone in the central Finland is outlined by dashed line.
The quiet zone (SFQZ) between the two active zones and the south-easternmost past of
the B-L zone belong to the accretionary arc complex of southern Finland (Korsman et
al. 1997). In the new interpretation those areas would be merged. The new parts of the
southern Finland’s quiet zone have similar crustal thickness and geological
characteristics as SFQZ. That is called here 2011 model of SFQZ.
10
The seismically active zones seem to be essential elements when the driving
mechanisms of the seismicity of southern Finland are regarded. The zones are
distinguished from their surroundings particularly by the occurrence of relatively large
(M  3.5) earthquakes. Mainly due to these events, 2/3 of the total strain release of the
study area has occurred within the two active seismic zones (Saari 1998).
Figure 2-3. Epicenters of earthquakes in 1375 - 2010 according to the catalogue of the
earthquakes in northern Europe. The alternative interpretation of the active zone in the
central Finland (CFAZ) is outlined by dashed line.
The quiet zone (SFQZ) between the two active zones and the south-easternmost past of
the B-L zone belong to the accretionary arc complex of southern Finland (Korsman et
al. 1997). In the new interpretation those areas would be merged. The new parts of the
southern Finland’s quiet zone have similar crustal thickness and geological
characteristics as SFQZ. That is called here 2011 model of SFQZ.
The seismically active zones seem to be essential elements when the driving
mechanisms of the seismicity of southern Finland are regarded. The zones are
11
distinguished from their surroundings particularly by the occurrence of relatively large
(M  3.5) earthquakes. Mainly due to these events, 2/3 of the total strain release of the
study area has occurred within the two active seismic zones (Saari 1998).
2.2 Magnitude-Frequency distribution for future earthquakes
Owing to the long recurrence interval of Finnish earthquakes, the forecasting of future
seismicity will rely strongly on historical data. There are temporal gaps in FENCAT and
its subrecords representing events inside the seismic belts (Figs. 2-4). Some of these
gaps relate likely to natural variation of seismicity. On the other hand, the lack of
reported earthquake observations from 1917 to 1923 and from 1942 to 1951 in Finland
is probably related to the First and Second World War (Ahjos et al. 1984). In the
subrecords of the Finnish earthquake data catalogue the breaks are even longer.
However, only the period (1942-1951) is disregarded when the annual numbers of
events is approximated.
According to FENCAT, the first known earthquake in Finland occurred in 1610. Since
1750, the number of felt earthquakes in each decade fluctuates but shows a certain
continuity of reporting. The beginning of systematic macroseismic mapping in
Fennoscandia in the 1880’s clearly improved the location accuracy and number of
observations (see e.g. Mäntyniemi & Ahjos 1990 and Mäntyniemi 2008). The
influences of the Lapinjärvi earthquake sequence (1951-1956) and the installation of the
regional seismic network since the middle of 1950's are clearly seen in FENCAT and in
Figures 2-4 as well. The events in Finland have been predominantly instrumentally
located since the mid 1960’s. The improvements instrumentation together with increase
number of seismic station instruments since 1980's caused the gradual increase of the
number of observations in the area of southern Finland. Most of the events smaller than
M=2 and M = 1.5 are recorded during the past 30 years.
As mentioned above, it is assumed that that 1/3 of the seismic activity of the Olkiluoto
target area origin from the Å-P-P Seismic Zone and 2/3 from the SFQZ. Seismicity of
other seismic belts is mainly disregarded in this paper. Differences of the two presented
interpretations of SFQZ are evaluated below (Figure 2-4a and 2-4b).
The number of observed events (133) inside the original SFQZ (Saari 1998) is 23 events
smaller inside the new or extended are of SFQZ. Ten of those events occurred during
period of increased seismicity 1920-1941. That is clearly shown in Figure 2-4b. Because
of the insufficient location accuracy, many of them have the same coordinates and
therefore not distinguishable in Figures 2-2 and 2-3. Otherwise the general pattern of
cumulative seismicity is rather similar. The largest observed events inside the original
and the extended SFQZ zones is the1846 Hauho earthquake (M=3.4).
Earthquake swarms and periods of active seismicity seem to be characteristic of the
south-eastern part of SFQZ. Earthquake swarms can be defined as prolonged series of
small or moderate magnitude without a distinct major event (Uski et al. 2006). Four
sequences are pointed out in Figures 2-4a and 2-4b. FENCAT includes also data located
by three seismic stations of the Imatran Voima Oy (IVO). The seismic stations were in
operation in the Loviisa region, in SE Finland, during the years 1984 - 1988.
12
Figure 2-4a. Cumulative number of observed earthquakes (133 events) in original
SFQZ (Saari 1998). All events and number of events larger than selected magnitudes
are presented (see Figure 2-3).
Figure 2-4b. Cumulative number of observed earthquakes (156 events) inside the
alternative interpretation of SFQZ (2011 model) that covers larger area before (see
Figure 2-3).
13
The largest observed event of the Å-P-P zone is the Osmussaar earthquake (25.10.1976,
ML = 4.9). The Osmussaar earthquake occurred at the western end of the Gulf of
Finland, 5 - 7 km NE of the island of Osmussaar (Estonia). The event was widely felt
around the Gulf of Finland mainly in Estonia and in southern Finland (Figure 5-2).
Under the circumstances prevailing in the southern Finland-Estonia region it was an
exceptionally strong earthquake. On the same day, the event was followed by two
aftershocks (M = 3.5, M = 3.0). Later in November two additional aftershocks occurred
with magnitudes of M = 3.5 and M = 2.5.
The fault type of the Osmussaar earthquake was mainly strike slip. However, there was
also a component of reverse dip slip faulting, which is possibly up to half of the rate of
strike slip component. The direction of the compressional stress given by the fault plane
solution was WNW-ESE (114o). The hypocentral depth was from 10 to 14 km, the fault
radius less than 1 km and the mean relative displacement larger than 3 cm (Slunga
1979). The main shock and the aftershocks were associated with the intersection of
NW-SE and SW-NE trending major fault zones. The rupture direction of the main event
was likely NE or ENE from the epicenter (Saari 1998).
Figure 2-4c. Cumulative number of observed earthquakes (62 events) inside the
seismically active belt Å-P-P (see Figure 2-3). Note the lack of earthquake observations
1935 – 1973.
The seismically active zone B-L delineates the northern border of SFQZ. B-L runs
about 150 km north from Olkiluoto. This subrecord has the most complete and
continuous data set with 149 events and relative short period of missing observations
after 1880 (Figure 2-4d). The largest event (M = 4.6) occurred in the Bothnian Bay in
1909.
14
Figure 2-4d. Cumulative number of observed earthquakes inside the seismically active
belt B-L (see Figure 2-3) north of Olkiluoto (149 events).
Figures 2-4a to 2-4d show that the cumulative number of smaller magnitudes tends to
increase faster closer to the present. The increase rate of larger magnitudes tends to be
more or less constant. The Finnish earthquake data has been considered to be
satisfactory after 1880 (Ahjos et al. 1984). This time window has been generally applied
to the magnitude range 2<M<4. However, for larger events (M>4.0) this period is
considered too short in relation to the recurrence time of events. For those events the
earthquake catalogue is regarded as quite representative since 1750. Due to a relatively
long history of dense population in the area of the Å-P-P belt, the time window between
1820-2010 has been applied (see Fig. 2-4c) to represent the magnitude range 3<M<4.
The number of earthquakes as a function of magnitude for a specific region often
conforms to the equation (Gutenberg & Richter 1944):
Log N = a - b*M
(2-1)
N is the number of earthquakes whose magnitude is  M occurring in a specified time
interval and a and b are constants.
The constant b describes the relative proportion of larger earthquakes to smaller
earthquakes. As b increases, the proportion of large earthquakes decreases. As a
increases, the number of earthquakes increases for earthquakes of all magnitudes.
Figures from 2-5 to 2-9 show the annual number of earthquakes as a function of
magnitude for each of the seismic zones. The least squares fits to the data points
presented in the figures 2-5, 2-6, 2-8 and 2-9 determine the constants a and b in Table 21. The applied magnitude step is generally 0.1 magnitude units, but not always. If the
number of events remains the same when the magnitude increases, only one of the
15
points is accepted. That point is selected by comparing the standard error of least
squares fits of alternative combinations. The alternative with smallest standard error is
accepted. Therefore the magnitude steps in Figures 2-8 and 2-9 are uneven. The above
described quality analyses of the data start from magnitude M = 2.0.
Figure 2-5. Annual frequency of earthquakes versus magnitude (see Eq. 2-1) for the
SFQZ. Earthquake swarms are included.
The b-value and the standard error (b = 1.441  0.074) of SFQZ are significantly larger
than in the neighbouring seismic zones (see Table 2-1). Since 1880, only 51 earthquakes
(M  2.0) have been reported within this belt, and 45 % of those belong to the
Lapinjärvi sequence 1951-1956 (see Fig. 2-4a). The events of this sequence, mainly of
the magnitude 2.4 (13 events), are dominating the magnitude-frequency distribution of
SFQZ. The foreshocks and aftershocks of swarms represent dependent events in the
population of independent events. The sequence contains two internal swarms (1951
and 1952) followed by one event per year from 1953 to 1956. In this study, only the
main shocks of the 1951 (M=2.8) and 1952 (M=3.1) Lapinjärvi swarms and the
individual events in 1953-1956 are included in the SFQZ. The fit to the re-estimated
values is shown in Figure 2-6. This arrangement lowers considerably the b-value but not
the standard errors in comparison to the original data (Table 2-1).
Larger b-values are expected in the area of relatively low seismicity. However, the
result based on original data is probably underestimating the relative number of larger
events. When the swarms are taken into account as described above, the seismicity
parameters of the SFQZ give a more conservative approximation of the seismic hazard
in the area (Figure 2-6).
16
Figure 2-6. Annual frequency of earthquakes versus magnitude for the SFQZ (model
2000). Revised data due to earthquake swarms.
The border of seismically quiet area cannot be determined exactly without bordering
active regions. The original SFQZ was bordered in other directions but not in SE. It was
approximated that SFQZ extends as far as the active zones north and south from it. In
the 2011 model also the south-easternmost past of the B-L zone belong to SFQZ.
As a next step the activity parameters for the proposed new SFQZ (model 2011) were
calculated (see Figure 2-3). The number of events in model 2011 is larger than in model
2000, but also the area larger of the 2011 model (about 161 700 km2) is larger than in
2000 model (125 200 km2). When the activity is scaled to number of events per year
and per square kilometer, there is no significant difference between the two
interpretations (Figure 2-7). The last tests related to the new earthquake (Mäntsälä,
19.3.2011, ML = 2.6) in SFQZ, about 210 km from Olkiluoto. The event was felt widely
in southern Finland and the importance of the Mäntsälä earthquake has been a subject of
common discussion. This event did not change significantly the pattern based on data
until the end of 2010. It was decided that the final analysis is based on the model 2000
and data until 2010.
Gutenberg-Richter constants for the seismically active belts are easier to estimate. The
input data as well as the calculated least square fits for Å-P-P and B-L are presented in
Figures 2-8 and 2-9.
17
Figure 2-7. Comparison of the two models representing SFQZ. Annual activity per km2.
Model 2000: b = 1.27 and model 201:1 b=1.21.
Figure 2-8. Annual frequency of earthquakes versus magnitude for seismic zone Å-P-P.
18
Figure 2-9. Annual frequency of earthquakes versus magnitude for seismic zone B-L.
The parameters a, b and N of the seismic belts are presented in Table 2-1. Constant a/km2
is the value from number of events per year and per square kilometer.
Table 2-1. Annual seismicity parameters (a and b) for seismic zones according to
FENCAT (Fig. 2-3). Constant a/km2 is the value for number of events per year and per
square kilometer.
Seismic Zone
B-L
SFQZ (original data)
SFQZ (re-estimated)
Å-P-P
b
0.752  0.013
1.441  0.074
1.268  0.077
0.715  0.033
a
1.337  0.042
2.728  0.197
2.152  0.207
1.015  0.099
a/km2
-3.760 0.042
-2.370 0.197
-2.946  0.207
-3.885  0.099
The expected number of earthquakes greater or equal to a magnitude M is shown in
Figure 2-10 and in Table 2-2. The values are normalized to number of events per year
and per square km. When the Olkiluoto target area is concerned, it is approximated that
1/3 of the seismic activity origin from the Å-P-P Seismic Zone and 2/3 from the SFQZ.
The re-estimated parameters of SFQZ are applied. This kind of approach follows rather
nicely the behaviour of SFQZ in low magnitudes and the behaviour of Å-P-P in larger
magnitudes. Figure 2-11 gives an impression of the accuracy of the extrapolation
conducted. The estimates are based on standard errors presented in Table 2-1.
19
Table 2-2. Annual number of earthquakes per square kilometer. Estimates for three
seismic zones and the Olkiluoto target area.
M
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
B-L
1.74E-04
7.31E-05
3.08E-05
1.29E-05
5.45E-06
2.29E-06
9.64E-07
4.06E-07
1.71E-07
7.18E-08
3.02E-08
1.27E-08
5.35E-09
2.25E-09
9.46E-10
3.98E-10
1.67E-10
7.05E-11
2.96E-11
SFQZ
1.13E-03
2.63E-04
6.11E-05
1.42E-05
3.30E-06
7.66E-07
1.78E-07
4.13E-08
9.59E-09
2.23E-09
5.18E-10
1.20E-10
2.79E-11
6.49E-12
1.51E-12
3.50E-13
8.13E-14
1.89E-14
4.39E-15
Å-P-P
1.30E-04
5.72E-05
2.51E-05
1.10E-05
4.84E-06
2.13E-06
9.33E-07
4.10E-07
1.80E-07
7.90E-08
3.47E-08
1.52E-08
6.68E-09
2.93E-09
1.29E-09
5.66E-10
2.48E-10
1.09E-10
4.79E-11
Olkiluoto
7.98E-04
1.94E-04
4.91E-05
1.31E-05
3.81E-06
1.22E-06
4.30E-07
1.64E-07
6.64E-08
2.78E-08
1.19E-08
5.15E-09
2.25E-09
9.82E-10
4.30E-10
1.89E-10
8.28E-11
3.64E-11
1.60E-11
Table 2-3 compares the annual numbers of earthquakes number of events per square
kilometer that are based in different studies and regions. This study gives slightly larger
estimate for magnitude 6 or larger than the previous study of the Olkiluoto area (Saari
2000). The annual frequency is slightly (1.4 – 2.1 times) smaller in Olkiluoto than in the
recent studies of Swedish seismicity parameters (Bödvarsson et al. 2006, Hora & Jensen
2005) and in the stable cratonic core regions (Fenton et al. 2006). That is rather
expected when the seismicity in Finland and Sweden concerned (see e.g. Figure 2-1). La
Pointe et al. (1999) predicts smaller annual frequency than other studies.
Table 2-3. Comparison of annual number of earthquakes per square kilometer.
Estimates for magnitude M = 6 according to seismicity studies in Olkiluoto, in different
regions in Sweden and in stable cratonic core (SCC) regions.
4.77E10-9
8.7E10-10
3.18E10-9
4.10E10-9
2.15E10-9
2.25E10-9
Bödvarsson et al. 2006 (Sweden)
La Pointe et al. 1999 (SE Sweden)
Hora & Jensen 2005 (Sweden)
Fenton et al. 2006 (SCC)
Saari 2000 (Olkiluoto)
this study (Olkiluoto)
20
Figure 2-10. Magnitude-frequency distributions of three seismic zones and the
Olkiluoto target area. N = annual number of events whose magnitude is  M per km2.
Figure 2-11. Estimates for upper and lower limits for the Olkiluoto target area
according to the standard deviation of the a- and b-values presented in Table 2-1.
Table 2-4 extends the magnitude frequency values of Table 2-2 to 10 000 year, 50 000
years and 100 000 years. The time intervals are adopted from the interpretation of
climate evolution (Ch 5 in Posiva 2012b) based on Pimenoff et al. (2011).
21
Table 2-4. Number of earthquakes per square kilometre (N) for 10 000 year, 50 000
years and 100 000 years period in the Olkiluoto target area. N is the number of
earthquakes whose magnitude is  M.
M
3
3.5
4
4.5
5
5.5
6
6.5
7
7.5
8
8.5
9
10 000 y
4.30E-03
1.64E-03
6.64E-04
2.78E-04
1.19E-04
5.15E-05
2.25E-05
9.82E-06
4.30E-06
1.89E-06
8.28E-07
3.64E-07
1.60E-07
50 000 y
2.15E-02
8.21E-03
3.32E-03
1.39E-03
5.95E-04
2.58E-04
1.12E-04
4.91E-05
2.15E-05
9.44E-06
4.14E-06
1.82E-06
7.98E-07
100 000 y
4.30E-02
1.64E-02
6.64E-03
2.78E-03
1.19E-03
5.15E-04
2.25E-04
9.82E-05
4.30E-05
1.89E-05
8.28E-06
3.64E-06
1.60E-06
22
23
3 FUTURE GLACIATION AND SEISMICITY
3.1 General
The push from the North Atlantic Ridge in the NW-SE direction seems to be the major
earthquakes generating force (e.g. Slunga 1991, Saari 1992 and Uski et al. 2003). Other
factors of seismicity are glacial rebound and the structural characteristics of the
bedrock. Current seismicity is a combination of these factors. In the future the factors
remain basically the same, but the next ice age will change the way they interact and
increase the expected magnitude extreme.
During the last European Ice Age (Weichselian) the ice mass above the Fennoscandian
Shield made a depression to the Earth’s crust. After the glacial maximum, at about 18
kyr BP, the climate warmed and the glaciers started to melt rapidly. The ice sheet did
not retreat at a constant speed (Pimenoff et al. 2011). When the ice was melted the
isostatic rebound was first very rapid (see Figures 3-1 and 3-4).
In northern Fennoscandia, a series of mainly NNE-SSW trending and SE dipping faultscarps formed within the latest phase of deglaciation, around 9 000 years ago. This area
was concentrated in Lapland measuring some 400 km by 200 km (Muir Wood 1989). In
Finnish Lapland the postglacial faults are 4 - 36 km long and the scarp height is 0 - 12
m. In Swedish Lapland the postglacial movements and earthquakes have been even
larger. The largest estimate of fault length is 150 km and scarp height 35 m (Kujansuu
1960, Lagerbäck 1979 and 1988, Olesen 1984).
The postglacial faults are reverse faults and they are located in old reactivated fracture
zones. The strike direction is perpendicular to the prevailing direction of maximum
horizontal stress. Estimation of the magnitudes related to the largest postglacial
earthquakes in the Finnish Lapland varied from 5.3 to 7.5 (Kuivamäki et al. 1998). The
estimates are based on assumption that the surface fault is a result of a single
earthquake, but those faults can be produced by different combination of smaller
magnitude earthquakes as well.
Current knowledge gives rather limited possibilities to assess the future postglacial
seismicity. In respect of origin and strength, current seismicity and seismic activity
following deglaciation are different.
An important unsettled question is how much of the tectonic forces are actually stored
during the ice age, and how they are released. The duration of the most pronounced
seismicity will presumably be of the same order than after the previous deglaciation, i.e.
few thousand years or less. The release of tectonic forces and sub-ice depression will be
main causes of PG-seismicity, but the strengths and proportions of these forces will
differ from the present state.
An additional uncertain factor is the ice front, its movements and thickness. However,
like before, the retreating ice front will rather likely be under water in the lowland areas.
Therefore the block movements will presumably occur as smaller and "slow
earthquakes" in the main part of Finland, whereas in the higher elevation faults the
movements are more violent.
24
Stable continental interior regions may be characterised by faults that have long period
of quiescence, of 10 000 - 100 000 years or more, punctuated by brief episodes of
activity. Many faults that may be capable of generating large, damaging earthquakes
may never rupture to the surface (so-called blind faults). In addition, in northern
latitudes many if not most geomorphic features associated with prior (10 000 years old)
surface faulting did not survive the erosive effect of late glaciations (Fenton et al. 2006).
Because most, if not all, future earthquakes will occur on existing faults, careful
investigations of all faults near the potential repository site are essential. In a
deterministic seismic hazard assessment each potentially seismogenic structure (i.e.,
fault that have moved recently, and those that may move in the geologically near-future)
must be identified and characterized. This is problematic since: (1) faults as short as 5
km are capable of producing damaging earthquakes; (2) faults dismissed as ‘inactive’
may experience rupture; and (3) even if all surface active faults are identified, the
problem of above mentioned blind faults remains (Fenton et al. 2006). However, this
does not disregard the necessity for deterministic investigations. Combination of both
approaches may illustrate the issue most comprehensively.
3.2 Past and future glaciations
The simulations of future glaciations and their consequences (e.g. Näslund 2006, Lund
& Schmidt 2011 and Pimenoff et al. 2011) follow mainly the patterns of the past
glaciations (mainly Saalian and Weichselian) in Fennoscandia. For example, the ice
history of Weichselian glaciation from 68 kyr BP to present time (Figure 3-1) was
applied in the simulation of stress evolution and fault stability at Olkiluoto during future
glaciation (Lund & Schmidt 2011).
Figure 3-1. Selected maps of ice thickness during the Weichselian in the reference
glacial cycle simulation (Näslund 2006). The ice has its largest extent at 18 kyr BP.
Snapshots of the development of the Baltic Sea and extent of the ice during the
Weichselian illustrate how rapidly the ice retreated (Figure 3-2 and Figure 3-3). At 10.3
kyr BP Fennoscandia is almost ice free. The Fennoscandian ice sheet retreated from the
Olkiluoto area at the end of the Yoldia Sea stage (Figure 3-2b), but the area was
depressed by the weight of the ice sheet, and remained submerged for about 6 kyr. The
isostatic adjustment continued, and the Olkiluoto area emerged from the sea 2.5–3 kyr
BP (Eronen & Lehtinen 1996).The observations indicate that the amount of absolute
25
down warping, due to the past glaciation, was around 800 m (Mörner 1979). It is
estimated that today, some 10 kyr after deglaciation, there is still approximately 50 m of
rebound remaining in central Fennoscandia (Ekman, 1991). In other studies the
remaining potential for isostatic uplift around the current uplift focus has been estimated
to be from 40 m - 130 m (Kakkuri 1985, Fjeldsgaar & Cathles 1991). Vuorela et al.
(2009) reviewed Bothnian sea shore-level data and made an estimate of the remaining
slow uplift at Olkiluoto based on shore-level displacement data as well as using a
derivative method applying crustal thickness and current uplift maps. According to their
results, the remaining uplift at Olkiluoto is 91.5–95.5 m according to the derivative
method and 83.8 m according to the shore-level displacement method. The duration of
the most pronounced uplift was relatively short. Typically, the uplift rate is significantly
higher during the first 2 000 years after the previous deglaciation (Figure 3-4) and
Section 7.3 in Vuorela et al. (2009).
Figure 3-2. Schematic sequence of the development of the Baltic Sea (Pimenoff et al.
2011). a) Baltic Ice Lake, b) Yoldia Sea, c) Ancylus Lake modified from Björck (1995)
and d) Litorina Sea modified from Eronen (1990).
Earth System Models of Intermediate Complexity (EMICs) were used to simulate the
future climate in Olkiluoto (Pimenoff et al. 2011). According to the simulations periods
of glacial climate with ice sheets extreme climatic periods will occur in Fennoscandia
and Olkiluoto during the future 120 kyr.
Soon after the last deglaciation, the crust adjusted itself to the new circumstances
(Figure 3-4). The most pronounced impact of this intense relaxation period occurred in a
periglacial environment. The largest earthquakes in Fennoscandia during the past ten
thousand years have likely occurred during the latest stage of deglaciation or
immediately after it 10 000-8 000 years BP level (see e.g. Lagerbäck 1979 and Muir
Wood 1989). Lack of similar observations after that phase suggests that the seismic
activity has decreased rather close to the current level within a relatively short timeinterval.
26
Figure 3-3. Ice thickness at 15, 14, 13, 12, 11 and 10.5 kyr BP, from left to right, upper
to lower, respectively, from the Näslund (2006) model. The proposed spent fuel
repository site at Olkiluoto has been marked with a green star. Figure from Lund &
Schmidt 2011.
The origin and strength of the reactivation can be related to two main causes. It is
suggested that, consistent with the current aseismicity beneath Greenland and
Antarctica, the ice sheet suppress earthquakes. Johnston (1987, 1989) showed that
earthquakes are suppressed by large ice sheets and discussed strain accumulation under
the ice sheets but did not consider rebound stresses. The "pulse" of seismic activity in
the immediate postglacial period could be explained by the release of stored tectonic
forces through the long lasting standing of ice cover and by the release of stresses from
the relatively sudden and strong glacial unloading. Anomalous shear stresses are
generated when the uniform isostatic rebound is hindered by a steep ice front. The
release of the stored stresses results in that the period of ice front retreating is
characterised by much more seismic activity than the period of its advance. The other
apparent cause is the intense crustal uplift immediately after deglaciation.
Reverse faulting require compressional tectonic environment, which indicates that the
release of stored tectonic forces had a significant role also in Finnish PG-seismicity. It
seems that the impact of postglacial crustal rebound was smaller than the impact of plate
tectonics. The most common seismogenic structure in compressional tectonic
environment is a thrust fault (a low angle dip slip faulting). When the dip of the fault is
steeper (> 45o), as usual in Finland, the reverse or strike slip faultings are expected
(Figure 3-5).
27
Figure 3-4. Shoreline displacement curves in different parts of Finland. Redrawn after
Glückert (1994) by Kuivamäki et al. (1998). Concerning the uplift rates at the target
area, the lowest rate (Hästholmen) was close to Porvoo and the highest rate (Olkiluoto)
was between the Turku and Pohjanmaa curves.
Probably, near the surface, the ice cover is able to stop the vertical movement of the
bedrock and the faulting is suppressed (Muir Wood 1989 and 1992). This explanation
could be applied to dip slip and strike slip faultings with a significant vertical
component, but not to pure strike slip faulting. Current Finnish earthquakes seem to be
mostly of the strike slip type, which indicates a dominantly horizontal movement.
However, also other factors than the stress field, such as the internal friction of the
faults, are involved with the seismicity and vary before, during and after the glaciation.
The results of the study on glacially induced faults stability by Lund & Schmidt (2011)
are briefly reviewed below. Using the combined glacial and tectonic stress field they
estimate how the stability of faults is affected by the glaciation, presenting maps and
vertical cross-sections of the stability field. They limit the analysis to glacially induced
and tectonic stresses, not including stress accumulation during the glaciation as
suggested by Johnston (1987) and Adams (2005).
28
Figure 3-5. Examples of the end member styles of faulting for different slip vectors (Lay
& Wallace 1995). Reverse and thrust faults are prone to slip when the compressional
stress field perpendicular to the fault strike or close to it, whereas for strike slip the
suitable orientation of compression is oblique.
In the fault stability analysis Lund & Schmidt (2011) use two synthetic background
stress fields, one reverse and one strike-slip, and one stress field constructed from local
stress measurements at Olkiluoto. They find that the background stress field is very
important for the resulting stability field. They show that in a reverse state of stress at
9.5 km depth, with a glacially induced pore pressure head of 50 % of the local ice
weight, Olkiluoto would experience reduced fault stability at the end of the glaciation.
In a strike-slip stress state, the stability field is more sensitive to variations in the
direction of the background field, but for their reference field Olkiluoto shows
essentially stable fault conditions.
Examples of fault stability maps calculated in strike slip and reverse synthetic stress
field are shown Figure 3-6 and Figure 3-7. Using a strike-slip background stress Figure
3-6 shows that fault failure is strongly demoted under the central parts of the formerly
glaciated region. The main locus of increased fault stability is in northern Finland and
Sweden, in the region of the endglacial faults, but we see that most of Finland has
increased fault stability. With the reverse background field, Figure 3-7 show that all
earth models predict wide spread fault instability in large parts of Finland, Sweden and
Norway. The highest failure potential is centered on northern Finland and Sweden, in
the area of the large endglacial faults (Lund & Schmidt 2011).
29
Figure 3-6. Maps of fault stability (negative values shown in blue colors are stable) at
10 km depth and 10.2 kyr BP. Strike-slip (SS) synthetic background stress field, R = 0.5,
glacial pore pressure 50 % of the ice weight. Simulation of earth structure models P24,
M14, M118 and L11. R is a parameter to constrain the magnitude of the intermediate
principal stress (Lund & Schmidt 2011).
30
Figure 3-7. Maps of fault stability (negative values shown in blue colours are stable) at
10 km depth and 10.2 kyr BP. Reverse (RF) synthetic background stress field, R = 0.5,
glacial pore pressure 50 % of the ice weight. Simulation of earth structure models P24,
M14, M118 and L11. R is a parameter to constrain the magnitude of the intermediate
principal stress (Lund & Schmidt 2011).
Simulations of future fault stability in Olkiluoto (Lund & Schmidt 2011) indicate:

In a reverse background stress field fault stability is promoted under and outside
the ice sheet. After deglaciation, faults in the central areas under the former ice
sheet show increased instability. In a strike-slip background field, stability is
promoted under the ice sheet but unstable areas develop outside the ice front.
After deglaciation, the area under the former ice sheet remains generally stable,
while areas of instability exist at the former ice front. These results hold at 10
km depth for all tested earth models.
31
 At 9.5 km depth, the models show fault instability in Olkiluoto at the end of
deglaciation in a reverse background stress field, irrespective of the exact
direction of the horizontal stresses. In strike-slip the result varies more with the
direction of the background field, but in our reference field Olkiluoto is stable
during the entire glacial cycle. This depth represents the seismogenic depth in
their study.

At 500 m depth the temporal evolution of the stability fields is similar to the 9.5
km results. Generally, at 9.5 km depth the direction of the faults optimally
oriented for failure follow those determined by the background stress field, with
a rather small range, but at 500 m depth the range of unstable fault orientations
increases significantly and a number of the faults and fractures mapped in
Olkiluoto would become destabilized.

The excess pore pressure produced by the ice sheet has a significant effect on
fault stability.
Based on computer simulations together with acoustic-seismic, sedimentological and
dating methods there are direct and indirect indicators of postglacial paleoseismicity in
the Olkiluoto area and in the adjacent sea areas (Hutri 2007). The average height of the
Holocene sediment faults in the echo-sounding profiles is from 0.5 to 2 m. The bedrock
displacement in the rock has probably not been equal, but less, since unconsolidated
clays are very sensitive to failure. Evidence of post-glacial faulting in the offshore study
area is found only along bedrock fracture zones indication reactivation of the old
fracture zones (Hutri & Kotilainen 2007). According to the simulation the maximum
shear displacements of some essential fracture zones in the area at 500 m depth was
about 3 cm and the maximum permanent shear displacement was about 3 mm (Hutri &
Antikainen 2002).
The age estimation for the events in the Olkiluoto sea area was dated to be from 10.650
to 10.200 years PB, i.e. earlier than in northern Finland. All Holocene faults in the
northern Baltic Sea occur at the same stratigraphical level. Since no younger or repeated
traces of seismic events were found, it corroborates the suggestion that major seismic
events occurred within a short time during and after the last deglaciation (Hutri 2007).
The findings from the offshore areas support the idea based on observations on bedrock
outcrops that the postglacial seismic activity in southern Finland has been less violent
than in northern Finland.
The study by Lindberg (2007) included the mapping of all long fractures and faults at
the shore outcrops of islands surrounding Olkiluoto. Observations were done on 136
outcrops and 619 fractures and faults. Only 30 were true faults. The shortest measured
fractures were 4 m long and the longest 58 m. The average length was 11 m. From the
observed 30 faults 13 were located in a rapakivi area in Kustavi and 17 around
Olkiluoto. In the rapakivi area some mylonites and one clear fracture were deep and old
features, except the fault, which possibly was reactivated also after glaciation. Other
young faults in the rapakivi do not reach deeper than the first horizontal fracture. In the
vicinity of Olkiluoto Island there were no post-glacially active faults, which have
undisputedly been active after glaciation.
32
3.3 Discussion about maximum magnitude
The above presented magnitude-frequency distributions for earthquakes permit
unlimited upper magnitudes. However, in seismic hazard studies it is often assumed that
a certain region has an upper magnitude limit. According to statistical analysis based on
the latter principle, the maximum magnitude of Finnish earthquakes is ML = 5.0 (Ahjos
et al. 1984).
It seems that, under the seismic circumstances prevailing in Finland, the bedrock is able
to release strain without major earthquakes. However, because the period of known
seismicity is much shorter than the forecasting period of seismicity the possibility of
earthquakes larger than M=5.0 cannot be definitely excluded. Usually, in an earthquake
hazard analysis 0.5 magnitude unit is added to the maximum calculated or observed
magnitude, reflecting the assumption that a short catalogue may not have captured the
largest possible earthquake.
However, when the postglacial seismicity is concerned, the basic conditions are
changed and an estimate of maximum magnitude based on current data is not valid. This
means that one should estimate a new maximum magnitude or utilise the data presented
in Table 2-2, where the magnitude-frequency distributions are based on assumption that
the magnitudes of Finnish earthquakes have not any upper limit.
Johnston et al. (1994) reported altogether 15 earthquakes larger than magnitude 7 that
have occurred in stable continental regions. However, they all associated with young
(Mesozoic and Cenozoic) extended continental crust. From the point of view of
Olkiluoto, the closest regions of that kind of bedrock can be found in Norway.
According to their study earthquakes in old Precambrian bedrock have smaller
magnitudes.
Fenton et al. (2006) decided in their study that the upper bound magnitude (Mx) in
stable continental regions, based on worldwide data set (Johnston et al. 1994), would be
Mx = 7 ± 0.2. Estimation of the magnitude related to the largest postglacial earthquake
in the Finnish Lapland is of the same order, from M = 5.3 to M = 7.5 (Kuivamäki et al.
1998).
When Lund & Schmidt (2011) estimated how the stability of faults is affected by the
glaciation, they limited the analysis to glacially induced and tectonic stresses (See
Chapter 3.2). They did not include stress accumulation during the glaciation as
suggested by Johnston (1987) and Adams (2005). It has been shown (Saari 2000) that
the accumulated strains can be large enough to produce postglacial earthquakes larger
than M = 7 also in southern Finland.
The maximum magnitude could be based on calculations of seismic strain release,
which is proportional to the square root of seismic energy release (Richter 1958):
Log(E) = 2.9 + 1.9ML - 0.024 ML2
where energy (E) is expressed in J (joules).
(3-1)
33
Currently the seismic strain release in Finland seems to be constant or close to it (Fig. 38). Major earthquakes dominate the cumulative strain release, but active periods of
smaller earthquakes are also essential elements of the strain release pattern. The
maximum strain release estimate (Fig. 3-8) corresponds to an earthquake with
magnitude 5.3.
Figure 3-8. Cumulative strain release curve for Finnish earthquakes during the period
1880-1980. The major events have been marked with the year of occurrence. S is the
estimate for the maximum strain (Ahjos et al. 1984).
The cumulative strain release curves and the slope of the average accumulation of it for
the Å-P-P zone and SFQZ are presented in Figures 3-9. The accumulation rate of strain
release is rather constant inside SFQZ (Figure 3-9a) except during the Lapinjärvi
earthquake swarm in 1950’s. Figure 3-9b demonstrates how significantly the Osmussaar
earthquake in 1976 increased the accumulated strain release in Å-P-P. Figure 3-8 shows
how the major events, like the Osmussaar earthquake, are strongly responsible to the
strain release in Finland.
The equations of the slopes give an approximate the annual amount of accumulated
unreleased strains per square km within those zones (Å-P-P = 79 400 km2 and SFQZ =
125 000 km2). As in the previous study (Saari 2000), we can assume that about 2/3 of
the Olkiluoto target area belongs to SFQZ and about 1/3 to the Å-P-P Seismic Zone.
Correspondingly, we assume that 1/3 of the accumulated unreleased strain origin from
the Å-P-P Seismic Zone and 2/3 from the SFQZ.
34
Figure 3-9. Cumulative strain release: a) Southern Finland quiet zone and b) Å-P-P
zone. The least squares fit (line) is based on the same data (diamonds) as magnitudefrequency analysis except that the earthquake sequences of SFQZ are included in this
analysis. The 1610 event in included in the analysis of SFQZ and events before 1800
are not included in the analysis of Å-P-P zone.
35
The estimate of accumulated strain over a 100 000 years interval within a distance of
100 km gives one possible approximation of unreleased seismic strain. An earthquake,
that could release the accumulated strain, determines the upper magnitude limit. The
potential maximum magnitude after 100 000 years glacial period within 100 km from
Olkiluoto is ML = 7.9. In practice a single earthquake does not release all the strains
accumulated inside the target area. The assumption that 80 % of accumulated strain is
released in one earthquake would yield about 0.1 magnitude unit smaller estimates. For
50 000 years glacial period the corresponding magnitudes would be ML = 7.5 and ML=
7.4 and for 10 000 years glacial period the corresponding magnitudes would be ML =
7.1 and ML= 6.9.
It is emphasized that the presented magnitude estimates depend strongly on the size of
the selected area. Therefore they are not absolute values, but indication that there is a
potential to accumulate enough strain to produce a large earth earthquake. The 100 km
radius is chosen because it covers the lineament interpretation of the Olkiluoto area
(Kuivamäki 2000). The area is also approximated by means of fault lengths. In stable
continental regions fault lengths of earthquakes of magnitude M = 7.0, 7.5 and 8.0 are
of the order of 40 km, 80 km and 160 km (Leonard 2010, see also Table 4-1). It is
reasonable to assume that the area where the released strain is accumulated is
comparable to the fault length.
The estimated magnitudes support the idea that, if the accumulated tectonic strain is
released shortly after deglaciation, postglacial earthquakes larger than magnitude 7 are
possible in the faults existing Olkiluoto region (see Chapter 4). That is also compatible
with maximum magnitude estimates based on different kind of approaches (Kuivamäki
et al. 1998 and Fenton et al. 2006).
36
37
4 LINEAMENTS, FAULTS AND SEISMICITY
4.1 General
Two fundamental elements of seismotectonics, the orientation of the main stress field
and the location of fracture zones, will remain the same during the next 100 000 years.
The past pattern of changes suggests that in the geological near future of Fennoscandia
variations are to be anticipated in the magnitude rather than in the orientation of stresses
(Muir Wood 1995). The loading and removal of the ice sheet will change the pattern in
some extent.
The stress pattern in Finland has been verified by the analysis of earthquake fault plane
solutions, in situ stress measurements and geodetic information. The current tectonic
stress field is rather consistently in the NW-SE direction. However, in the northernmost
part of Fennoscandia, the stress pattern is less uniform probably due to the increase of
the ridge push from NNE- and N-directions. A comprehensive description of the stress
field in Olkiluoto is presented in the Olkiluoto site description report (Posiva 2012d).
Olkiluoto is situated away from the active plate margins. The push from the North
Atlantic Ridge in the NW-SE direction seems to be the major stress generating
mechanism in Finland (e.g. Heidbach et al. 2010). This is also supported by the regional
in situ data from Olkiluoto and other Finnish sites studied during the site selection
programme. Changes in isostatic load due to glaciations and related isostatic adjustment
and the existence of the brittle fault zones change the stress regime at the site.
Currently, a thrust faulting stress regime is present, i.e. the horizontal stresses are larger
than the vertical stress, H > h > v and the principal stresses are approximately
oriented horizontally and vertically, respectively. The orientation of H at the site is
found to vary slightly with depth and at the repository depth be in the range NW-SE and
E-W. The vertical stress is predominantly well represented by the weight of the
overlying rock mass (Posiva 2012d).
Intraplate earthquakes tend to occur on old zones of crustal weakness reactivated by
present stress field. Locally and regionally the stress field is a combination of plate
boundary forces, glacial rebound and local geology. The bedrock of Finland is heavily
fractured. The earthquakes occur in those reactivated faults and shear zones. Regionally
and locally the most favourably oriented zones of weaknesses are the most potential
structures where stresses are released.
Regional interpretation of magnetic and morphological lineaments around Olkiluoto
revealed that in the whole area, NW-SE (strike 130° – 150°) direction is dominating.
Other important strike directions are 10° – 30°, 50° – 70° and 110° – 130° (Paananen &
Kuivamäki 2007).
Wells and Coppersmith (1994) assembled a comprehensive database of earthquake
source parameters. The database gives values for rupture area, moment magnitude,
average surface displacement and surface rupture length among other parameters. The
data was separated into three groups based upon fault type (normal, reverse and strike
slip faults). Based on this database, empirical relationships can be presented to assess
maximum earthquake magnitudes or dislocation for a particular fault zone or an
38
earthquake source (Wells and Coppersmith 1994). Leonard (2010) expanded their data
set and presented his own relations of rupture length, width, average displacement and
moment release. Figure 4-1 compares the relation of fault area and moment magnitude
of these studies. The curves are rather similar, but the same magnitude gives slightly
smaller rupture area when Leonard’s equation is applied.
Figure 4-1. Data base regressions of the correlation between earthquake magnitude
and rupture area. The solid and dashed lines are regression lines according to Leonard
(2010) and Wells and Coppersmith (1994), respectively. The data points are redrawn
from Wells and Coppersmith (1994). Data from the 3DEC models at Olkiluoto are
indicated by stars (Fälth & Hökmark 2011). The largest faults of the models (BFZ021
and BFZ214) are presented in Figure 4-5 and Table 4-5.
In fault studies it is necessary to remember that faults that may be capable generating
large damaging earthquakes may never rupture to the surface (Fenton et al. 2006). Table
4-1 shows fault dimensions based on equations of Leonard 2010. The table is based on
equations for larger (M >≈ 5) earthquakes in stable continental regions (SCR). It can be
seen that moment magnitudes 5, 7 and 8 can be associated with fault lengths of order of
2.6 km, 40 km and 160 km or larger, respectively. Also it can be seen that the fault
widths related to earthquakes of magnitudes 7.0, 7.5 and 8.0 are 16 km, 25 km and 40
39
km, respectively. Because the majority of Fennoscandian earthquakes seem to occur in
the depth range of 5 - 20 km (Ahjos & Uski 1992, Slunga 1991 and Gregersen et al.
1991), it could be expected that earthquakes of moment magnitude of M ≈ 8 in a
vertical or nearly vertical fault would generate a clear surface rupture. Unfortunately, if
the dip of the fault is less than 30 degrees or less than 15 degrees, the surface rupture is
likely missing, if the focal depth of an earthquake is 20 km or 10 km respectively. The
possibility to observe surface rupture of smaller earthquakes (M < 7) is even smaller.
According to fault plane solutions, the Fennoscandian earthquakes seem to relate to
predominantly vertical faults and fault zones (Slunga 1991, Saari & Slunga 1996, Uski
et al. 2003 and Uski et al. 2006). The studied earthquakes are associated with old
Precambrian faults and shear zones, which have been reactivated (Saari & Slunga 1996,
Uski et al. 2003 and Uski et al. 2006). There are no reported surface ruptures related to
those events.
Table 4-1. Fault dimensions for stable continental regions (SCR) based on equations of
Leonard 2010. MW= Moment magnitude, L = fault rupture length (km), A = fault area
(km2, L*W for rectangular fault, W = Fault width (km)).
Mw
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
L (km) A(km2)
2.6
6.5
5.1
20.4
10.1
64.6
20.2
204.2
40.3
645.7
80.2
2041.7
159.8
6456.5
318.4 20417.4
634.4 64565.4
W (km)
2.5
4.0
6.3
10.0
15.9
25.2
39.9
63.2
100.0
However, integrated analysis of Bouguer anomaly map and the seismic profiles
(BABEL) imply that long nearly vertical, gently dipping and nearly horizontal shear
zones are common in the Gulf of Bothnia region. Interpretation of profiles (e.g. Figure
4-2) show also many horizontal and gently dipping reflective structures, which indicate
anomalous characteristics in the bedrock structure. The length of these structures can be
from few tens to over one hundred kilometres.
Many faults that may be capable of generating earthquakes may never rupture to the
surface (so-called blind faults). However, little or no damage appears in strong
earthquakes of tunnels that are well constructed, outside the epicentral region and away
fault breaks (Hanamura 2011). This emphasizes the importance of earthquake rupture
modelling of the accurately studied fault zones at the Olkiluoto site (e.g. Fälth &
Hökmark 2011).
40
Figure 4-2. Geological interpretation of 300 km long BABEL 1 profile. This, nearly NS
oriented profile, is the closest BABEL profile to the Olkiluoto site. The profile section
that pass by Olkiluoto is south of the mafic nucleus. Dashed lines are interpreted as
shear zones and solid lines are reflective structures (Korja & Heikkinen 2005).
Generally, it is not possible to reliably predict whether any given fault lineament or
subsurface structure will reactivate as a normal, reverse or strike slip fault during some
future earthquake event. This depends on the prevailing stress state. However, the
orientation of the main horizontal stress field and the location of fracture zones, will
remain the same during the future 100 000 years. Therefore, the modelling of future
seismicity could combine the stress pattern and the lineament interpretation of the target
area. When the orientation of the main stress field is in the NW-SE direction, the NWSE and N-S oriented zones of weakness are advantageous for strike slip faulting
whereas reverse faulting is more plausible in fracture zones perpendicular the
orientation of the main stress field.
4.2 Diffuse seismicity and tectonics
This approach represents a concept of regional, diffuse background seismicity and
tectonic structures. It is assumed that zones of different kind of seismicity cannot be
distinguished inside the Fennoscandian Shield. The method also assumes that there is
not a priori information about real location and orientation of tectonic structures that
exist in the vicinity of the investigated site.
The Gutenberg-Richter relation can also be used to assess within which distance from a
certain position an earthquake of a certain magnitude within a certain time is expected.
In order to find the distance to an event of magnitude M or larger from a point in a
region Bödvarsson et al. (2006) introduce an equation.
log d = (bM – a - log 2π – log T)/2.
(4-1)
Where d (km) is a median-probability (50-percentile) distance, constants a and b are
from Gutenberg-Richter relation, M = magnitude and T in years. Constant a is the value
41
from number of events per year and per square kilometer. The equation follows from
the non-normalized form of Gutenberg-Richter relation (Jacob 1997). The equation 4-1
can be presented also in form (see Equation 2-1):
log d = -(log N + log 2π + log T)/2
(4-2)
Where N is the number of events per year and per square kilometer. For Olkiluoto this
N is given in Table 2-2.
The median distance increases with magnitude and decrease with average recurrence
time. The estimated median distances tend to be smaller in Forsmark region than in
Olkiluoto region (see Table 4-2). For example, a magnitude 5 event is expected within
230 km of Forsmark with a recurrence period of 100 years and magnitude 7 earthquake
within 460 km with a recurrence period of 1 000 years. The corresponding distances in
the Olkiluoto region are 365 km (M = 5) and 607 km (M = 7). For larger magnitudes the
relative differences of the value d seem to be smaller. For M5 and M7 the d values are
about 1.6 and 1.3 times larger in Olkiluoto than in Forsmark, correspondingly. In
Olkiluoto the expected median distances for earthquakes with magnitudes M = 5 and M
= 7 for 10 000 years recurrence period are 36 km and 192 km, respectively.
Table 4-2. Median distances d (km) expected for earthquakes with magnitudes M and
recurrence periods T (years) for an unconfined seismic source with uniform cumulative
rate log N (per year per km2) in Forsmark in Sweden (Bödvarsson et al. 2006) and in
Olkiluoto area (This study).
M
3
4
5
6
7
8
Forsmark
T=100
T=1 000
37
12
92
29
230
73
578
183
1453
460
3651
1155
T=100
61
155
365
840
1919
4347
Olkiluoto
T=1 000
19
49
115
266
607
1383
T=10 000
6
15
36
84
192
437
4.3 Lineament maps and brittle fault zone modelling
There are three lineament maps and one brittle fault zone model (BFZ) available in the
Olkiluoto region. The lineament map by Kuivamäki (2000) covers the area within 100
km from Olkiluoto. The other two interpretations cover four map sheets (numbers 1132,
1134, 1141 and 1143) in the southern Satakunta (Paananen & Paulamäki 2007). The
first lineament map covers an area of about 10 km * 15 km and the second map an area
of about 12 km * 12 km in the Olkiluoto surroundings. The brittle fault zone model
covers an area of 18.9 km2 (Posiva 2012d).
The next two analyses are based on the lineament map by Kuivamäki (2000) and on the
BFZ model presented in the Olkiluoto site description report (Posiva 2012d). The
42
interpretation of Kuivamäki (2000) covers an area large enough to include faults
capable to generate earthquakes larger than M >7, which occur in faults from 40 km to
over 100 km. In local scale the BFZ model is selected, because the connection of
seismicity to the fractures in the BFZ model is more probable than to lineaments in the
maps. A lineament is a feature in a landscape, which is likely, but not always, an
expression of an underlying geological structure such as fault. In that sense the
lineament interpretation can be misleading. For example in the Olkiluoto area less than
15% of lineament data was included in the BFZ model (Mattila et al. 2007, Appendix
VI: Assessment of the applicability of lineament data, s. 383-393). The BFZ
interpretation includes brittle fault zones that are capable to produce earthquakes of
magnitude M = 5 (see Table 4-1).
In Finland, as usual in intraplate areas, the principal knowledge of larger earthquakes
generally relies on macroseismic observations, whereas instrumentally located events
are mainly smaller. In both cases the location error is likely to be too large in
comparison to the dimensions and separation of faults. However, within the limits of
location accuracy, some fracture zones can be associated with seismicity (Figure 4-3).
The future 100 000 years will certainly bring out currently unknown active faults. It is
assumed that the most significant earthquakes are related to the largest fault zones or to
the faults associated with present seismicity. These potentially active fault zones are
suggested in the following.
In the lineament map presented by Kuivamäki (2000) the interpreted lineaments were
classified into four size categories (Table 4-3). It is noteworthy that the lineament width
that is observed at the surface is not the same as fault width in vertical direction.
Table 4-3. The size classification of lineaments. Modified after Salmi et al. 1985.
Size category
Width (m) *
I
about 1000 m
II
several hundreds
III
10 - 100 m
IV
<< 10 m
* Estimated width of fracture zone at the surface
Length (km)
tens - hundreds
5 - tens
1-5
<1
Only the I-class and II-class lineaments are included in the presentations. Seismic
activity is more likely associated with the I-class lineaments, but the II-class lineaments
give additional information about the prevailing orientation of regional lineaments. The
numbers of I-class and II-class lineaments in Olkiluoto site is 88 and 320, respectively.
Lineaments longer than 40 km are picked up. Also some lineaments possibly associated
with current seismicity are mentioned.
It is a common approach in seismic hazard studies to assume that, if an earthquake
occurs in a fault, that fault is capable to produce an earthquake of similar size in any
point of this structure (e.g. IAEA 2010). In the Olkiluoto target area, 44 of the I-class
lineaments are more than 40 km long. Most of them are NW-SE oriented. If these
43
lineaments are fractures or fracture zones, they are capable to host an earthquake with
magnitude M = 7 (see Table 4-1).
The target area includes 12 events with a magnitude of M = 0.6 - 3.1 (Table 4-4 and
Figure 4-3). Five of them are historical and 7 instrumentally located earthquakes, which
is likely an indication of different sensitivity during those periods. As a consequence of
improved sensitivity of the seismic network magnitudes are smaller and the frequency
of earthquakes is higher during latest ten or 25 years than before. The closest historical
events occurred about 35 km south (1926, M=3.1) and about 40 km north (1804,
M=2.9) from the site. The closest earthquake (ML = 0.8) occurred in Eurajoki on 29
September 2008 about 12 km from Olkiluoto. The depth of the event was 5.0 km.
According to the recordings of the Posiva’s network, the displacement related to this
event was 1.4 mm. In source calculations, the fault area is approximated by a circle. The
source radius of the Eurajoki earthquake was about 63 m. The earthquake occurred
close to the northern edge of the Laitila rapakivi massif, where NW-SE oriented
lineaments are characteristic (Figure 4-3 and 4-4). One of those II-order lineaments runs
by the epicenter and through the bay just north of the Olkiluoto island.
Table 4-4. Earthquakes less than 100 km from Olkiluoto. Onset time, coordinates,
Magnitude and maximum intensity (Imax). MM = macroseismic magnitude and ML =
Local magnitude.
Date
1804.04.06
1901.10.30
1901.10.30
1925.11.
1926.04.22
1971.10.10
1986.09.05
2007.01.03
2007.02.10
2008.08.29
2009.04.22
2009.04.24
Time
10:50
11:20
23
04:30
05:29:06
20:48:14.5
05:16:18.1
23:38.58.3
10:22:00.6
14:42:21.7
09:06:02.7
o
N
61.6
60.7
60.7
60.6
60.9
61.9
62.03
60.90
60.86
61.235
60.874
60.842
o
E
21.6
22.7
22.7
21.9
21.5
21.9
21.10
21.92
22.28
21.669
21.071
21.137
Magnitude
2.9 MM
2.6 MM
3.0 MM
2.7 MM
3.1 MM
2.5 ML
1.4 ML
1.9 ML
0.6 ML
0.8 ML
1.4 ML
1.4 ML
Imax
IV
III
IV
V
IV
-
Place-name
Pori
Loimaa
Loimaa
Mietoinen
Uusikaupunki
Siikainen
Bothnian Sea
Laitila
Yläne
Eurajoki
Uusikaupunki
Uusikaupunki
Another earthquake that cannot be related to I-order lineaments in Figure 4-3 occurred
3.1.2007 in Laitila (ML = 1.9) about 40 km from Olkiluoto. According to Posiva’s
recordings the fault displacement related to the Laitila event was 1 mm and the source
radius about 43 m. Joint interpretation of recordings of three seismic networks (Posiva,
Finnish and Swedish national networks) was used when the preliminary fault plane
solution of the Laitila earthquake was calculated: The reverse faulting occurred in a
nearly vertical N-S oriented fault which can be associated with mafic dykes in the area
(Figure 4-4). The orientation of compression related to the event was NW-SE.
44
Focal depths are estimated for the six most recent earthquakes occurred since 1986.
Those events have occurred rather close to the ground surface at the depth from 1 km to
5 km.
The closest over 40 km long lineaments are NW - SE oriented lineaments 15 km NE
from Olkiluoto (number 5 in Figure 4-3) and 20 km SW from Olkiluoto (number 36).
Lineament number 36 can be associated with the 1926 earthquake (M=3.1) in
Uusikaupunki.
Figure 4-3. Earthquakes and lineaments of the Olkiluoto target area. I-order
lineaments are shown by red colour. Lineaments longer than 40 km are shown by thick
red line. Lineament numbers refer to the corresponding number in the interpretation by
Kuivamäki (2000). Dashed black lines are II-order lineaments. Lineaments with red
numbering are probably capable host lineament of M =7 earthquake. Light blue dots
are macroseismically (-1964) and dark blue dots (ML>1.5 and open circles (ML<1.5)
are instrumentally (1965-2010) located earthquakes. The solid triangle shows the
location of Olkiluoto. Shaded area denotes the Å-P-P seismic belt.
45
Figure 4-4. Regional seismicity and the bedrock of the Olkiluoto region (Koistinen et
al. 2001). Macroseismically- (before 1965) and instrumentally-located (1965 - 2010)
earthquakes are shown by light and dark blue circles, respectively.
46
NW-SE oriented lineaments 19, 23, 24, 27, 36 and NE-SW oriented lineaments 17, 18,
31, 32, 34, and 41 intersecting are within the active Å-P-P seismic belt (Fig. 4-3, shaded
area) are likely the most potential capable faults for postglacial events as large as M = 7.
Another, less active but probably as capable NW-SE oriented lineaments are northeast
of Olkiluoto (numbers 5-15 and 54). Those above mentioned 28 lineaments (shown with
red numbering in Figure 4-3) are more likely fracture zones or potential host structures
of future seismicity than the other 16 over 40 km long lineaments. They cannot be
disregarded, but this consideration gives an approximation that about 60% of the 40 km
long lineaments could be capable to host an earthquake of M = 7. This percentage is
much higher than found in the lineament study for the BFZ model (Mattila et al. 2007).
Of course there is not any reason to expect that the relative number of true faults to
found lineaments would be the same in different scales, regions and studies. However,
the ratio is rather likely less than one.
The BFZ model covers an area of 18.9 km2 (Posiva 2012d). Along the bay north of the
model area runs the lineament (Figure 4-5), which may relate to the II-class lineament
that was associated with the 2008 (ML = 0.8) Eurajoki microearthquake (Figure 4-3).
The model includes 3D interpretation of one brittle fault zones inside the volume
(Figure 4-6). OL-BFZ214, running along the northern edge of the model area, is
extended outside the model volume. Ten of zones are about 2.5 km long or longer, i.e.
they are capable to host an earthquake of magnitude M = 5 (see Table 4-5 and Figure 46 below).
Figure 4-5. The area and faults of the BFZ model at the surface (Pere et al. 2012). The
faults are about 3 km long or longer. The area is defined by bounding lineament. OLBFZ214 running along the northern edge of the model is extended outside the presented
area (see Figures 4-1 and 4-6 and Table 4-5). LDF = Lineament determining feature.
47
Figure 4-6. All brittle fault zones (upper part) and about 2.5 km long or longer zones
(below) in the BFZ model of the geological model v. 2.1.
From the values of Table 2-4 it is easy to calculate number of earthquakes within any
area larger than 1 km2. However, the selected area should not cover an area with a
radius of more than 150 km from Olkiluoto. At that distance also the seismic zone B-L
should be included in the calculations of Gutenberg-Richter parameters. Table 4-6
shows the number of earthquakes for 10 000 year, 50 000 years and 100 000 years
period inside the area of the BFZ study (18.9 km2) and the area within 100 km from
Olkiluoto (31400 km2).
For earthquakes M>5 the annual frequency within BFZ study area is 2.25E-7. This
frequency (f) is distributed along ten fault zones (Table 4-5 and Figure 4-6) susceptible
to reactivation. The average estimate for annual frequency of reactivation of one of
those faults is then 2.3E-8.
48
Table 4-5. Dimensions of the structures that are about 2.5 km long or longer zones
(below) in the BFZ model (Posiva 2012d). Faults about 3 km long or longer are
presented in Figure 4-5.
STRUCTURE
OL-BFZ020a
OL-BFZ021
OL-BFZ099
OL-BFZ146
OL-BFZ148
OL-BFZ149
OL-BFZ159
OL-BFZ161
OL-BFZ214
OL-BFZ262
LENGTH * DEPTH
in (m*m)
5000*2500
6800*2600
6000*3500
c. 3500*3100
2890*629
2460*760
2750*753
2650*500
13300*2630
8240*3230
Table 4-6. Number of earthquakes (N) for 10 000 year, 50 000 years and 100 000 years
period within an area of 18 900 km2 (BFZ study, Posiva 2012d) and within 100 km from
Olkiluoto. N is the number of earthquakes whose magnitude is  M. Table 2-4 is scaled
to represent the areas of the BFZ study and the lineament study.
M
3
3.5
4
4.5
5
5.5
6
6.5
7
7.5
8
8.5
9
10 000 y
8.1E-02
3.1E-02
1.3E-02
5.3E-03
2.3E-03
9.7E-04
4.2E-04
1.9E-04
8.1E-05
3.6E-05
1.6E-05
6.9E-06
3.0E-06
BFZ study
50 000 y
100 000 y
4.1E-01
8.1E-01
1.6E-01
3.1E-01
6.3E-02
1.3E-01
2.6E-02
5.3E-02
1.1E-02
2.3E-02
4.9E-03
9.7E-03
2.1E-03
4.2E-03
9.3E-04
1.9E-03
4.1E-04
8.1E-04
1.8E-04
3.6E-04
7.8E-05
1.6E-04
3.4E-05
6.9E-05
1.5E-05
3.0E-05
10 000 y
1.3E+02
5.2E+01
2.1E+01
8.7E+00
3.7E+00
1.6E+00
7.1E-01
3.1E-01
1.4E-01
5.9E-02
2.6E-02
1.1E-02
5.0E-03
Radius 100 km
50 000 y
100 000 y
6.7E+02
1.3E+03
2.6E+02
5.2E+02
1.0E+02
2.1E+02
4.4E+01
8.7E+01
1.9E+01
3.7E+01
8.1E+00
1.6E+01
3.5E+00
7.1E+00
1.5E+00
3.1E+00
6.8E-01
1.4E+00
3.0E-01
5.9E-01
1.3E-01
2.6E-01
5.7E-02
1.1E-01
2.5E-02
5.0E-02
SKB (2011, page 466) summarises Swedish annual seismicity parameters within a 5 km
source area (78.5 km2) for earthquakes >M5 (Table 4-7). According to the report the
area includes 30 fault zones susceptible to reactivation. If the result of this study is
scaled to an area within 5 km the number of fault zones capable to host an earthquake of
magnitude M > 5 would be about 40 (Table 4-1 and Leonard 2010).
As in Table 2-3 the annual frequency within a 5 km radius is slightly smaller in
Olkiluoto than in the recent studies of Swedish seismicity parameters (Table 4-7). That
49
is rather expected when the seismicity in Finland and Sweden concerned. Due to the
larger number of potential host faults in Finland this difference is increased when the
annual reactivation frequency of those faults is concerned. The results of La Pointe et al.
(2000, 2002) are about the same order than in this study even when the numbers are
scaled with the numbers of susceptible faults.
In this study the estimate of capable fault length is 2.5 km, whereas in the SKB reports
assumes 3 km long fault for M>5. In Table 4-5 there are six brittle fault zones longer
than 3 km. If that number is used in Table 4-7, the estimated frequency per any fault
susceptible to reactivation (3.7E-8) is slightly smaller than in the studies by Bödvarsson
et al. (2006), Hora & Jensen (2005) and Fenton et al. (2006).
In the evaluation of site suitability for final disposal, both Posiva and SKB have used
the fault length of 3 km as a limiting value for zones which need to be avoided in the
planning of disposal tunnels. This is based on the possibility of a fault zone hosting an
earthquake of a magnitude of M>5 and inducing a shear displacement of 5 cm in large
fractures nearby the rupture plane. This 5 cm is considered as a critical value for the
damage threshold of a disposal canister (Posiva 2012c, SKB 2011). The 3 km size limit
is based on the regression curves given by Wells & Coppersmith (1994).
Table 4-7. Comparison of estimated annual frequency (N 5 km) of earthquakes >M5
within a 5 km radius area in Olkiluoto and in corresponding regions (SKB 2011). In the
second column the frequencies are distributed (f) along the approximated number fault
zones capable to host an earthquake of M>5. SKB (2011) applies number 30 in Sweden
and in stable cratonic core (SCC) regions. The corresponding numbers in Olkiluoto are
about 40 (host fault > 2.5km) and about 25 (host fault > 3 km).
N 5 km
2.4E-6
8.7E-7
2.5E-6
2.0E-6
9.3E-7
9.3E-7
f
7.8E-8
2.9E-8
8.3E-8
6.8E-8
2.3E-8
3.8E-8
Region
Sweden
Sweden
Sweden
SCC regions
Olkiluoto, Finland
Olkiluoto, Finland
Reference
Bödvarsson et al. 2006
Lapointe et al. 2000, 2002
Hora & Jensen 2005
Fenton et al. 2006
This study (host fault > 2.5 km)
This study (host fault >3 km)
The fault length of 2.5 km corresponds on earthquake of magnitude of the order M = 5
(Leonard 2010, Table 4-1), but the frequency estimate in Table 4-7 includes also
earthquakes larger than that. It is also reasonable to assume that the fault should have
some extra length in order to host a rupture length of 2.5 km. Therefore this 3 km fault
length may be a better choice to represent a zone susceptible to reactivation M > 5
earthquake.
One of the zones in the BFZ model (Table 4-5) is over 10 km long i.e. capable to host
an earthquake of magnitude M = 6. In stable continental regions fault lengths of
earthquakes of magnitude M > 7.0 are over 40 km (Table 4-1). It is obvious that the
BFZ model is not suitable for reactivation analysis of earthquakes >M6.
50
For earthquakes M>7 the annual frequency within a 100 km radius from Olkiluoto is
1.4E-5. In the Olkiluoto target area, 44 of the lineaments are more than 40 km long. It
was mentioned above that 28 (about 60%) of those lineaments are more likely
susceptible to reactivation. The average estimate for annual frequency of reactivation of
any of those faults is then 4.8E-7. However, the number of potentially capable faults is
uncertain. Unlike within BFZ-model the true number of faults is not known: 1) some of
the lineaments in Figure 4-3 are maybe not fault zones. 2) The number and
characteristics of the so called blind faults is not known (see Figure 4-2). Although they
seem to be quiet in current conditions some of them might be reactivated in postglacial
conditions. If we assume that the percentage of true faults among the group of mapped
lineaments is about the same found in the lineament study for the BFZ model, i.e. 15 %,
the average estimate for annual frequency of reactivation of any of those faults is about
four times higher, of the order of 2E-6.
The considerations above show the uncertainties when the average frequency of faults
susceptible to reactivation is approximated. In small scale, like in the BFZ model, the
true number of fault zones is rather well known. However, the uncertainty increases
when it is assumed that the texture remains the same when the area is expanded to cover
an area within a 5 km radius or even more. The result is also sensitive to the decision is
the fault length 2.5 km or 3 km represents a potential host fault for an earthquake M > 5.
Those two lengths are used in Table 4-8, in order to approximate the range of average
frequency of M > 5 earthquakes distributed along fault within a 5 km radius.
Table 4-8. Frequency estimates for 10 000 year, 50 000 years and 100 000 years period
from Olkiluoto within a 5km radius. Estimates for magnitudes M > 5 are based on BFZ
study and for magnitudes M > 7 lineament study within a 100 km radius. Column N is
based on magnitude frequency values of Table 2-4. Column f gives the probability for
an earthquake (M>5) per one fault zone based on the approximated upper and lower
number of fault zones susceptible to reactivation (see text).
Period
(y)
10 000
50 000
100 000
N
(M >5)
0.0093
0.047
0.093
f
(M >5)
2.3E-4 ... 4.7E-4
1.2E-3 ... 2.4E-3
2.3E-3 ... 4.7E-3
N
(M >7)
0.00034
0.0017
0.0034
f
(M >7)
1.2E-5 ... 5.0E-5
6.0E-5 ... 2.5E-4
1.2E-4 ... 5.0E-4
51
The same kind of study of brittle fault zones is not available in the scale of over 40 km
long faults, which represent potential host to an earthquake M > 7. Above the present
seismicity and seismic zoning are associated with over 40 km long lineaments. This
yields to an approximation that about 60 % of those lineaments are likely susceptible to
reactivation. That is likely overestimating the number of susceptible faults, but on the
other hand the number of susceptible blind faults is unknown. On the other hand,
according to the lineament study for the BFZ model, the percentage of true faults among
the group of mapped lineaments is about 15 %. Those two percentages are used in Table
4-8, in order to approximate the range of average frequency of M > 7 earthquakes
distributed along fault within a 5 km radius.
52
53
5 CONCLUDING REMARKS
The aim of the study has been to approximate the future seismicity associated with the
investigation site for final disposal of spent nuclear fuel at Olkiluoto.
The study of regional seismicity indicated that the area within a radius of 100 km of the
site contains seismic belts of higher and lower seismic activity. The magnitudefrequency relations of these belts were scaled to represent the seismicity of the target
area.
According to fault plane solutions, the Fennoscandian earthquakes relate to
predominantly vertical faults and fault zones. The studied earthquakes associate with
Precambrian faults and shear zones, which have been reactivated. The current
Fennoscandian seismicity relates generally to NW-SE oriented compressional stress
field. Locally earthquakes occur as a combination of plate boundary forces, glacial
rebound as well as local stress field and geology.
In Finland, the most obvious evidences of postglacial fault movements are found in
western Lapland. All these faults are reverse faults, which require compressional
tectonic environment. It seems that the impact of postglacial crustal rebound was
smaller than the impact of plate tectonics.
The postglacial earthquakes in Lapland were generated in an environment of rather
uniform NW-SE oriented compressional stress field, which indicates that, as the present
seismicity, the seismicity immediately after deglaciation seems to relate strongly to the
push from the North Atlantic Ridge.
A common opinion in seismic hazard studies is that a certain seismotectonic zone has a
characteristic maximum magnitude. The Finnish bedrock seems to be able to release
strain with earthquakes less than M=5.0. However, because the period of known
seismicity is much shorter than the forecasting periods in this study (10 000, 50 000 and
100 000 years) the possibility of larger earthquakes cannot be definitely excluded.
Usually, in an earthquake hazard analysis 0.5 magnitude units is added to the maximum
calculated or observed magnitude, reflecting the assumption that a short catalogue may
not have captured the largest possible earthquake.
According to Fenton et al. (2006) the upper bound magnitude in stable continental
regions could be M = 7 ± 0.2. Magnitude estimate related to the largest postglacial
earthquake in the Finnish Lapland is of the same order (M = 7.5).
The estimate of accumulated strain over a 10 000, 50 000 or 100 000 years interval
within a distance of 100 km gives one possible approximation of unreleased seismic
strain. The accumulated unreleased strains of the target sites over these intervals give
estimates of maximum magnitude: ML = 7.9 (100 000 years), ML = 7.5 (50 000 years)
and ML = 7.1 (10 000 years). In practice a single earthquake does not release all the
strains accumulated inside the target area. The assumption that 80 % of accumulated
strain is released in one earthquake would yield about 0.1 – 0.2 magnitude unit smaller
estimates.
54
It can be assumed that zones of different kind of seismicity cannot be distinguished
inside the Fennoscandian Shield. If regionally diffuse background seismicity and
tectonic structures are expected in Olkiluoto, the median distances for earthquakes with
magnitudes M = 5 and M = 7 for 10 000 years period are 36 km and 192 km,
respectively. However, earthquakes are associated with old Precambrian faults and
shear zones, which have been reactivated.
In stable continental regions rupture lengths of earthquakes of magnitude M > 5.0 and
M > 7.0, are longer than 2.5 – 3.0 km and 40 km, respectively. It is reasonable to
assume that the area where the released strain is accumulated is comparable to the fault
length. In the Olkiluoto area there is a comprehensive brittle fault zone model that is
suitable for magnitudes M >5. The current model is not suitable for reactivation analysis
of earthquakes >M6.
If the BFZ model were larger, the interpretation would be more reliable for magnitudes
M >5 and it might be possible to extend interpretation to larger magnitudes. For this
purpose, it would be also useful to estimate the total length of the brittle shear zone,
when the zone continues over the model edge.
A reliable study of brittle fault zones is not available in the scale of over 10 km long
faults, which represent potential host an earthquake M > 6. Faults longer than 40 km are
potential host an earthquake M > 7. There are 44 lineaments over 40 km long within a
distance of 100 km from Olkiluoto. It is assumed that the number of faults potential to
hosts magnitude M = 7 earthquakes is maybe 15 % - 60 % of the lineaments found in
the lineament interpretation. The closest of those lineaments are NW - SE oriented
lineaments 15 km NE and 20 km SW from Olkiluoto. The latter lineament can be
associated with the 1926 earthquake (M=3.1) in Uusikaupunki.
The low rate of fault makes it difficult to characterize the long-term seismotectonic
behaviour, leading to uncertain estimates of seismic activity. In addition, many faults
that may be capable of generating large, damaging earthquakes may never rupture to the
surface. That adds uncertainty to the interpretation based on lineament data. However,
seismicity can be associated with some over 40 km long lineaments. It is likely that also
the future seismicity is in some extent related those faults. On the other hand, when the
quality of the tectonic model is good, the area of the model is so limited that there are
not any earthquake data available. This emphasizes the importance of earthquake
rupture modelling of reliably studied fault zones at the Olkiluoto site.
55
REFERENCES
Adams, J., 2005. On the probable rate of magnitude > 6 earthquakes close to a Swedish
site during a glacial cycle, in Expert panel elicitation of seismicity following glaciation
in Sweden, edited by S. C. Hora & M. Jensen, no. 2005:20, Swedish Radiation
protection Authority (SSI), Stockholm, Sweden.
Ahjos, T. & Arhe K., 1983. Seismicity of the Loviisa region (Loviisan ympäristön
seismisyys). Institute of Seismology, University of Helsinki. Report T-21, (in Finnish).
Ahjos, T., Saari, J., Penttilä, E. and Korhonen H., 1984. Earthquakes and seismic hazard
in Finland, Engineering Geology, 20, 1-12.
Ahjos, T. & Uski, M., 1992. Earthquakes in northern Europe in 1375-1989.
Tectonophysics, 207:1-23.
Björck, S., 1995. A review of the history of the Baltic Sea, 13.0–8.0 ka BP. Quaternary
International, 27, 19–40.
Bödvarsson, R., Lund, B., Roberts, R., Slunga, R., 2006. Earthquake activity in Sweden.
Study in connection with a proposed nuclear waste repository in Forsmark or
Oskarshamn. SKB R-06-67, Svensk Kärnbränslehantering AB (SKB) Stockholm
Sweden.
Dowding, C.H., Rozan, A., 1978. Damage to rock tunnels from earthquake shaking. J.
Geotech. Eng. Div., ASCE, vol. 104 No. GT2. pp.175-191.
Ekman, M., 1991. Gravity change, geoid change and remaining postglacial uplift of
Fennoscandia, Terra Nova, 3, 390–392.
Eronen, M., 1990. Itämeren kehitys. In Alalammi, P. (ed). Atlas of Finland, Folio 123–
126: Geology, 15–18. National Board of Survey & Geographical Society of Finland,
Helsinki.
Eronen, M. & Lehtinen, K. 1996. Description of the Quaternary geological history of
Romuvaara, Kivetty and Olkiluoto sites (in Finnish with an English abstract). Posiva
Oy, Eurajoki, Finland. Working Report PATU-96-74, 37 p.
Fenton C., Adams J. & Halchuk S, 2006. Seismic hazards assessment for radioactive
waste disposal sites in regions of low seismic activity. Geotechnical and Geological
Engineering, 24, pp 579–592.
Fjeldsgaar, W. & Cathles, L., 1991. The present rate of uplift in Fennoscandia implies a
low-viscosity astenosphere. Terra Nova3, 393-400.
Fälth B. & Hökmark H., 2011. Modelling end-glacial earthquakes at Olkiluoto. Posiva
Oy. Working Report 2011-13.
56
Glückert, G., 1994. Maankohoamisen arvioiminen jääkaudenjälkeisen rannansiirtymisen
avulla. In: Glückert, G.(ed.). Maankohoaminen, neotektoniikka ja Itämeren rannansiirtyminen Suomessa. Publications of the Department of Quaternary Geology University of
Turku 78, 62-76 (in Finnish).
Gregersen S., Korhonen H. & Husebye E. S., 1991. Fennoscandian dynamics: Presentday earthquake activity. Tectonophysics, 189:333-344.
Gutenberg, B. & Richter, C.F., 1944. Frequency of earthquakes in California. Bull.
Seism. Soc. Am., 34, 185-188.
Hanamura, T., 2011. 2011 East Japan earthquake and tsunami and underground
structures. Open session in World Tunnel Congress 2011, Helsinki. 24 May 2011.
Heidbach, O., Tingay, M., Barth, A., Reinecker, J., Kurfe, D. & Müller, B., 2010.
Global crustal stress pattern based on the World Stress Map database release 2008.
Tectonophysics, 482, 1-4, 3-15, doi: 10.1016/j.tecto.2009.07.023.
Hellä, P. (ed.), Ikonen, A., Mattila, J., Torvela, T. & Wikström, L., 2009. RSCprogramme - Interim report. Approach and basis for RSC development, layout
determining features and preliminary criteria for tunnel and deposition hole scale.
Eurajoki, Finland: Posiva Oy. Working Report 2009-29.
Hora S. & Jensen M., 2005. Expert panel elicitation of seismicity following glaciation
in Sweden. SSI Report 2005:20, Statens strålskyddsinstitut (Swedish Radiation
Protection Authority).
Hutri K., 2007. An approach to paleoseismicity in the Olkiluoto (sea) are during the
early Holocene. STUK-A222. Helsinki 2007, 64 pp + Appendices 55 pp.
Hutri K. & Antikainen J., 2002. Modelling of bedrock response to glacial loading at
Olkiluoto site, Finland. Engineering Geology; 67 (1-2), pp. 39-49.
Hutri K. & Kotilainen A., 2007. An acoustic view into Holocene paleoseismicity
offshore southwestern Finland, Baltic Sea. Marine Geology, 238 (1-4), pp. 45-59.
IAEA, 2010. Seisimic Hazards in site evaluation for nuclear installations. Specific
Safety Guide. IAEA Safety Sandarts Series No. SSG-9. International Atomic Energy
Agency, Vienna.
Institute of Seismology of Helsinki University, 2011,
http://www.helsinki.fi/geo/seismo/index.html.
Jacob, K., H., 1997. Scenario earthquakes for urban areas along the Atlantic seaboard of
the United States, in economic consequences of earthquakes: Preparing for the
unexpected, MCEER-SP-0001, January 1997.
Johnston, A. C., 1987. Suppression of earthquakes by large continental ice sheets,
Nature, 330,467–469.
57
Johnston, A.C., 1989. The effects of large ice sheets on earthquake genesis. Gregersen,
S. & Basham, P.M. (Eds.). Earthquakes at North-Atlantic passive margins: Neotectonics
and postglacial rebound. Kluwer Academic Publisher, 501-519.
Johnston, A.C., Coppersmith, K.J., Kanter, L.R. and Cornell, C.A., 1994. The
earthquakes of stable continental areas regions, Volume 1.: Assessment of large
earthquake potential. Electric Power Research Institute. Technical Report TR-102261V1, pp.1 - 368.
Kakkuri, J., 1985. Die Landhebung in Fennoscandien in Lichte der heutigen
Wissenschaft. Zeitschrift für Vermessungwesen 110 (2), 51-59.
Koistinen T., Stephens M.B., Bogatshev V., Nordgulen Ö. and Korhonen J., 2001.
Geological Map of the Fennoscandian Shield, scale 1:2 000 000. Espoo: Trondheim:
Uppsala: Moscow: Geological Survey of Finland, Geological Survey of Norway,
Geological Survey of Sweden: Ministry of Natural Resources of Russia.
Korja A. & Heikkinen P., 2005. The accretionary Svecofennian orogen – Insight from
the BABEL profiles. Precambrian Res. 136, 241-268.
Korsman K, Koistinen T., Kohonen J. Wennervirta M., Engdahl E., Honkamo M,,
Idman H. & Pekkala Y. (Eds.), 1997. Bedrock map of Finland 1:1,000,000. Geological
Survey of Finland, Espoo, Finland.
Kuivamäki, A., Vuorela, P. & Paananen, M., 1998. Indications of postglacial and recent
bedrock movements in Finland and Russian Karelia. Geological Survey of Finland.
Report YST-99.
Kuivamäki, A., 2000. Lineament database of the Finnish potential repository sites for
the calculation of bedrock movements induced by earthquakes. Posiva Oy. Working
Report 2000-12.
Kujansuu R., 1960. Recent faults in Finland. Geologi No 3 (in Finnish).
Lagerbäck R., 1979. Neotectonic structures in northern Sweden. Geologisca
Föreningens i Stockholm Förhandlingar, 100 (1978), 271-278.
Lagerbäck R., 1988. Postglacial faulting and paleoseismicity in Landsjärv area, northern
Sweden. SKB-report: 88-25.
La Pointe, P.R., Cladouhos, T. & Follin, S., 1999. Calculation of displacements on
fractures intersecting canisters induced by earthquakes: Aberg, Beberg and Ceberg
examples. SKB TR-99-03.
La Pointe P. R., Cladouhos T. T., Outters N., Follin S., 2000. Evaluation of the
conservativeness of the methodology for estimating earthquake-induced movements of
fractures intersecting canisters. SKB TR-00-08, Svensk Kärnbränslehantering AB.
58
La Pointe P. R., Cladouhos T., Follin S., 2002. Development, application, and
evaluation of a methodology to estimate distributed slip on fractures due to future
earthquakes for nuclear waste repository performance assessment. Bulletin of the
Seismological Society of America, 92, pp 923–944.
La Pointe, P.R. & Hermansson J., 2002. Estimation of rock movements due to future
earthquakes at four candidate sites for a spent fuel repository in Finland. Posiva report.
POSIVA-2002-02.
Lay T. & Wallace T.C., 1995. Modern global seismology. Academic Press Inc.
Leonard, M., 2010. Earthquake fault scaling: Self-consistent relating of rupture length,
width, average displacement, and moment release. Bulletin of the Seismological Society
of America 100(5A): pp 1971-1988.
Lindberg A., 2007. Search for Glacio-Isostatic Faults in the vicinity of Olkiluoto.
Posiva Oy. Working Report 2007-05.
Lund, B. and Schmidt, P., 2011. Stress evolution and fault stability at Olkiluoto during
the Weichselian glaciations. Posiva Oy. Working Repoert 2011-14.
Mattila, J., Aaltonen, I., Kemppainen, K., Wikström, L., Paananen, M., Paulamäki, S.,
Front, K., Gehör, S., Kärki, A. & Ahokas, T., 2008. Geological Model of the Olkiluoto
Site, Version 1.0. Posiva Oy. 508 p. Working Report 2007-92.
Muir Wood R., 1989. Extraordinary deglaciation reverse faulting in northern
Fennsocandia. Gregersen, S. and Basham, P.W., Earthquakes at Nort-Atlantic passive
margins: Neotectonics and postglacial rebound. Kluwer Academic Publisher, 141-173.
Muir Wood R., 1992. A review of the seismotectonics of Sweden. BEQE Report 43-01R-001, Revision 0, (draft).
Muir Wood, R., 1995. Reconstructing the tectonic history of Fennoscandia from its
margins: The past 100 million years. SKB Technical Report: 95-36, 85 pp.
Mäntyniemi P. and Ahjos T., 1990. A Catalog of Finnish Earthquakes in 1610 - 1990.
Geophysica, 26(2): 17-35.
Mäntyniemi P., 2008. Macroseismology as a component of seismicity assessments in an
intraplate region: Studies of Northern Europe with Emphasis on Finland. Institute of
Seismology, University of Helsinki. Report S – 47.
Mörner, N.-A., 1979. The Fennoscandian uplift and late Cenozoic geodynamics:
geological evidence. GeoJournal, Vol. 3.3.
Natural Resources Canada, 2011. http://earthquakescanada.nrcan.gc.ca/info-gen/glossaeng.php.
59
Näslund, J.-O., 2006. Ice-sheet dynamics, in Climate and climate-related issues for the
safety assessment SR-Can, no. TR-06-23, pp. 33–56, Swedish Nuclear Fuel and Waste
Management Co. (SKB), Stockholm, Sweden.
Olesen, O., 1988. The stuorraguarra fault, evidence of neotectonics in the Precambrian
Finmark, Northern Norway. Norsk Geologisk Tidskrift 68: 107-11.
Paananen M. & Kuivamäki A., 2007. Regional lineament analysis of the southern
Satakunta area. Posiva Oy. Working Report 2007-04, 29p.
Pere, T., Mattila, J., Wikström, L., Lindgren, S., Vaittinen, T. & Ahokas, H., 2012.
Layout determining features, their influence zones and respect distances at the Olkiluoto
site. Posiva report 2012-XX. In prep.
Pimenoff N., Venäläinen A. & Järvinen H., 2011. Climate scenarios for Olkiluoto on a
time scale of 120 000 years. Posiva Oy. Posiva Report 2011-04.
Posiva Oy, 2010. Interim Summary report of the safety case 2009, Posiva Report 201002.
Posiva Oy, 2012a. Complementary considerations report. Posiva Report 2012-12.
Posiva Oy, 2012b. Formulation of Radionuclide Release Scenarios. Posiva Report
2012-08.
Posiva Oy, 2012c. Rock Suitability Classification. Posiva Report 2012-21. in prep.
Posiva Oy 2012d. Olkiluoto site description 2011, Posiva Report 2011-02. in prep.
Richter C. F., 1958. Elementary Seismology. Freeman, San Francisco and London, 768
pp.
Saari J., 1992. A review of the seismotectonics of Finland. Nuclear Waste Commission
of Finnish Power Companies, Report YJT-92-29.
Saari, J., 1998. Regional and local seismotectonic characteristics of the area surrounding
Loviisa nuclear power plant in SE Finland. Institute of Seismology, University of
Helsinki. 248 p. Report 1998.
Saari, J., 2000. Seismic activity parameter of the Finnish potential repository sites.
Posiva report. POSIVA-2000-13.
Saari J. & Slunga R., 1996. Fault plane solutions of microearthquakes in the Loviisa
area in south-eastern Finland. Posiva report. POSIVA-96-02.
Salmi M., Vuorela P. & Kuivamäki A., 1985. Geological site selection studies for the
final disposal of spent nuclear fuel in Finland (Käytetyn ydinpolttoaineen
loppusijoituksen geologiset aluevalintatutkimukset). Nuclear Waste Commission of
Finnish Power Companies. Report YJT-85-27, (in Finnish).
60
SKB, 2011. Long term safety for the final repository for spent nuclear fuel at Forsmark.
Main report of the SR-site project. Volume II, SKB TR-11-01. Svensk
Kärnbränslehantering AB (SKB). Stockholm Sweden.
Slunga R., 1979. Source mechanism of a Baltic earthquake inferred from surface-wave
recordings. Bull. Seism. Soc. Am., 69(6), 1931-1969.
Slunga R. 1991. The Baltic Shield earthquakes. Tectonophysics. Vol. 189: 323-331.
Uski M., Hyvönen T., Korja A. and Airo M.-L., 2003. Focal mechanism of three
earthquakes in Finland and their relation to surface faults. Tectonophysics. 363, pp. 141157
Uski M., Tiira T., Korja A. & Elo S., 2006. The 2003 earthquake swarm in
Anjalankoski, south-eastern Finland. Tectonophysics 422, 55-69.
Valtonen, O., 2012. Location accuracy of instrumentally located earthquakes in the
Bothnian Bay region (manuscript).
Vieno, T. & Nordman, H., 1999. Safety assessment of spent fuel disposal in
Hästholmen, Kivetty, Olkiluoto and Romuvaara. TILA-99, Posiva 99-07, Posiva Oy.
Vuorela, A., Penttinen, T. & Lahdenperä, A-M., 2009. Review of Bothnian Sea shorelevel displacement data and use of a GIS tool to estimate isostatic uplift. Eurajoki,
Finland: Posiva Oy. Working Report 2009-17. 191 p.
Wells, D.L. & Coppersmith, K.J., 1994. New empirical relationships among magnitude,
rupture length, rupture width, rupture area and surface displacement. Bull. Seism. Soc.
Am., vol. 84 (4), 974-1002.
YVL. STUK, 2011. https://ohjeisto.stuk.fi/YVL/?en=on
LIST OF REPORTS
POSIVA-REPORTS 2012
_______________________________________________________________________________________
POSIVA 2012-01
Monitoring at Olkiluoto – a Programme for the Period Before
Repository Operation
Posiva Oy
ISBN 978-951-652-182-7
POSIVA 2012-02
Microstructure, Porosity and Mineralogy Around Fractures in Olkiluoto
Bedrock
Jukka Kuva (ed.), Markko Myllys, Jussi Timonen,
University of Jyväskylä
Maarit Kelokaski, Marja Siitari-Kauppi, Jussi Ikonen,
University of Helsinki
Antero Lindberg, Geological Survey of Finland
Ismo Aaltonen, Posiva Oy
ISBN 978-951-652-183-4
POSIVA 2012-03 Safety Case for the Disposal of Spent Nuclear Fuel at Olkiluoto Design Basis 2012 ISBN 978-951-652-184-1
POSIVA 2012-04
Safety Case for the Disposal of Spent Nuclear Fuel at Olkiluoto Performance Assessment 2012
ISBN 978-951-652-185-8
POSIVA 2012-05
Safety Case for the Disposal of Spent Nuclear Fuel at Olkiluoto Description of the Disposal System 2012
ISBN 978-951-652-186-5
POSIVA 2012-06
Olkiluoto Biosphere Description 2012
ISBN 978-951-652-187-2
POSIVA 2012-07
Safety Case for the Disposal of Spent Nuclear Fuel at Olkiluoto Features, Events and Processes 2012
ISBN 978-951-652-188-9
POSIVA 2012-08
Safety Case for the Disposal of Spent Nuclear Fuel at Olkiluoto Formulation of Radionuclide Release Scenarios 2012
ISBN 978-951-652-189-6
POSIVA 2012-09
Safety Case for the Disposal of Spent Nuclear Fuel at Olkiluoto Assessment of Radionuclide Release Scenarios for the Repository
System 2012
ISBN 978-951-652-190-2
POSIVA 2012-10
Safety case for the Spent Nuclear Fuel Disposal at Olkiluoto - Biosphere
Assessment BSA-2012
ISBN 978-951-652-191-9
POSIVA 2012-11
Safety Case for the Disposal of Spent Nuclear Fuel at Olkiluoto Complementary Considerations 2012
Posiva Oy
ISBN 978-951-652-192-6
POSIVA 2012-12
Safety Case for the Disposal of Spent Nuclear Fuel at Olkiluoto Synthesis 2012
ISBN 978-951-652-193-3
POSIVA 2012-13
Canister Design 2012
Heikki Raiko, VTT
ISBN 978-951-652-194-0
POSIVA 2012-14
Buffer Design 2012
Markku Juvankoski
ISBN 978-951-652-195-7
POSIVA 2012-15
Backfill Design 2012
ISBN 978-951-652-196-4
POSIVA 2012-16
Canister Production Line 2012 – Design, Production and Initial State of
the Canister
Heikki Raiko (ed.), VTT
Barbara Pastina, Saanio & Riekkola Oy
Tiina Jalonen, Leena Nolvi, Jorma Pitkänen & Timo Salonen, Posiva Oy
ISBN 978-951-652-197-1
POSIVA 2012-17
Buffer Production Line 2012 – Design, Production, and Initial State of
the Buffer
Markku Juvankoski, Kari Ikonen, VTT
Tiina Jalonen, Posiva Oy
ISBN 978-951-652-198-8
POSIVA 2012-18
Backfill Production Line 2012 - Design, Production and Initial State of
the Deposition Tunnel Backfill and Plug
ISBN 978-951-652-199-5
POSIVA 2012-19
Closure Production Line 2012 - Design, Production and Initial State of
Underground Disposal Facility Closure
ISBN 978-951-652-200-8
POSIVA 2012-20
Representing Solute Transport Through the Multi-Barrier Disposal
System by Simplified Concepts
Antti Poteri. Henrik Nordman, Veli-Matti Pulkkanen, VTT
Aimo Hautojärvi, Posiva Oy
Pekka Kekäläinen, University of Jyväskylä, Deparment of Physics
ISBN 978-951-652-201-5
POSIVA 2012-21
Layout Determining Features, their Influence Zones and Respect
Distances at the Olkiluoto Site
Tuomas Pere (ed.), Susanna Aro, Jussi Mattila, Posiva Oy
Henry Ahokas & Tiina Vaittinen, Pöyry Finland Oy
Liisa Wikström, Svensk Kärnbränslehantering AB
ISBN 978-951-652-202-2
POSIVA 2012-22
Underground Openings Production Line 2012- Design, Production and
Initial State of the Underground Openings
ISBN 978-951-652-203-9
POSIVA 2012-23
Site Engineering Report
ISBN 978-951-652-204-6
POSIVA 2012-24
Rock Suitability Classification, RSC-2012
ISBN 978-951-652-205-3
POSIVA 2012-25
2D and 3D Finite Element Analysis of Buffer-Backfill Interaction
Martino Leoni, Wesi Geotecnica Srl
ISBN 978-951-652-206-0
POSIVA 2012-26
Climate and Sea Level Scenarios for Olkiluoto for the Next 10,000
Years
Natalia Pimenoff, Ari Venäläinen & Heikki Järvinen, Ilmatieteen laitos
ISBN 978-951-652-207-7
POSIVA 2012-27
Geological Discrete Fracture Network Model for the Olkiluoto Site,
Eurajoki, Finland: version 2.0
Aaron Fox, Kim Forchhammer, Anders Pettersson,
Golder Associates AB
Paul La Pointe, Doo-Hyun Lim, Golder Associates Inc.
ISBN 978-951-652-208-4
POSIVA 2012-28
Safety Case for the Disposal of Spent Nuclear Fuel at Olkiluoto - Data
Basis for the Biosphere Assessment BSA-2012 ISBN 978-951-652-209-1
POSIVA 2012-29
Safety Case For The Disposal of Spent Nuclear Fuel at Olkiluoto Terrain and Ecosystems Development Modelling in the Biosphere
Assessment BSA-2012
ISBN 978-951-652-210-7
POSIVA 2012-30
Safety Case for the Disposal of Spent Nuclear Fuel at Olkiluoto Surface and Near-surface Hydrological Modelling in the Biosphere
Assessment BSA-2012
ISBN 978-951-652-211-4
POSIVA 2012-31
Safety Case for the Disposal of Spent Nuclear Fuel at Olkiluoto Radionuclide Transport and Dose Assessment for Humans in the
Biosphere Assessment BSA-2012
ISBN 978-951-652-212-1
POSIVA 2012-32
Safety Case for the Disposal of Spent Nuclear Fuel at Olkiluoto Dose Assessment for the Plants and Animals in the Biosphere
Assessment BSA-2012
ISBN 978-951-652-213-8
POSIVA 2012-33
Underground Openings Line Demonstrations Stage 1, 2012
ISBN 978-951-652-214-5
POSIVA 2012-34
Seismic Activity Parameters of the Olkiluoto Site
Jouni Saari, ÅF-Consult Oy
ISBN 978-951-652-215-2