Pure appl. geophys. (2008)
DOI 10.1007/s00024-007-0292-6
Ó Birkhäuser Verlag, Basel, 2008
Pure and Applied Geophysics
Radon Changes Associated with the Earthquake Sequence in June 2000
in the South Iceland Seismic Zone
PÁLL EINARSSON,1 PÁLL THEODÓRSSON,2 ÁSTA RUT HJARTARDÓTTIR,1 and
GUðÓN I GUðJÓNSSON2
Abstract—An earthquake sequence at the transform plate boundary in South Iceland, that included two
magnitude 6.5 earthquakes in June 2000, was anticipated on the basis of a centuries-long seismicity pattern in
the area. A program of radon monitoring in geothermal water from drill holes, initiated in 1999, rendered
distinct and consistent variations in radon in association with these events. All seven sampling stations in a
50 9 30 km zone covering the epicentral area showed a consistent pattern. Four types of change could be
identified: 1) Preseismic decrease of radon. Anomalously low values were measured 101–167 days before the
earthquakes. 2) Preseismic increase. Spikes appear in the time series at six stations 40–144 days prior to the
earthquakes. These anomalies were large and unusual if compared to a 17-years history of radon monitoring in
this area. 3) Coseismic step, most likely related to the coseismic change in groundwater pressure observed over
the entire area. 4) Postseismic return of the radon values to the preseismic level about three months later, also
concurrent with groundwater pressure changes.
Key words: South Iceland Seismic Zone, radon, earthquake precursor, co-scismic changes.
1. Introduction
Various studies have shown that an increase in the concentration of radon in
groundwater is an earthquake precursor, see e.g., reviews by HAUKSSON (1981) and KING
(1985), and more recent studies by WAKITA (1996), ROELOFFS (1999), TRIQUE et al.
(1999), and ZMAZEK et al. (2002). Even though numerous examples of premonitory
radon anomalies have been identified and described in the literature, statistical analysis
of the relationship between radon and earthquakes has been difficult because of the lack
of long-time series from a network of recording stations in active seismic or volcanic
areas. A program of radon monitoring was initiated for this purpose in the plate
boundary areas of Iceland in 1977 (HAUKSSON and GODDARD, 1981; HAUKSSON, 1981).
1
Institute of Earth Sciences, University of Iceland, Sturlugata 7, 101 Reykjavı́k, Iceland.
E-mail: [email protected]
2
Science Institute, University of Iceland, Dunhaga 3, 107 Reykjavı́k, Iceland.
P. Einarsson et al.
Pure appl. geophys.,
A network of 11 sampling stations was established, nine of which were located within
or near the South Iceland Seismic Zone; a transform zone where the most damaging
historical earthquakes in Iceland have occurred. Judged from the time pattern of
previous events a sequence of large earthquakes was expected in this zone with high
probability (STEFÁNSSON et al., 1993; EINARSSON et al., 1981). This labor-intensive radon
program gave promising results (JóNSSON and EINARSSON, 1996) but was discontinued in
1993 because of declining funding and deteriorating instruments. Radon monitoring
was resumed in 1999 with new and improved instruments and time-saving sample
preparation (THEODóRSSON, 1996; GUDJóNSSON and THEODóRSSON, 2000) during which the
sample preparation time was reduced from 3 hours to less than 10 minutes. The
expected earthquakes occurred in June 2000 (EINARSSON et al., 2000; STEFáNSSON et al.,
2000) within the network of monitoring stations and after one year of radon
monitoring.
The earthquake sequence occurred along a 90-km stretch of the plate boundary and
contained two magnitude 6.5 (Mx) earthquakes (on June 17 and 21) and several
magnitude 5+ events (see e.g., PEDERSEN et al., 2003; ÁRNADóTTIR et al., 2004). This paper
documents the radon time series and reports on the radon changes detected in association
with these events.
2. The South Iceland Seismic Zone
The South Iceland Seismic Zone is a transform-type plate boundary; a branch of the
mid-Atlantic plate boundary that crosses Iceland (Fig. 1). Plate divergence in the
southern part of Iceland is accommodated by two sub-parallel rift zones: the Western and
the Eastern Volcanic Zones. The gap between them is bridged in the south, near 64°N, by
a zone of high seismic activity, the South Iceland Seismic Zone, which takes up the
transform motion between the Reykjanes Ridge and the Eastern Volcanic Zone
(EINARSSON, 1991). The two rift zones and the transform demarcate a block or a
microplate, the Hreppar microplate. It has been argued that rifting is dying out in the
Western Rift Zone, and is being taken over by the Eastern Rift Zone, or that the partition
of rifting between the rift zones may be uneven and changes with time (SIGMUNDSSON
et al., 1995).
The South Iceland Seismic Zone has been defined by destruction areas of historical
earthquakes, Holocene surface ruptures and instrumentally determined epicenters. It is
oriented E-W and is 10–15 km wide. Destruction areas of individual earthquakes and
surface faulting (Fig. 2) show, however, that each event is associated with faulting on
N-S striking planes, perpendicular to the main zone. The overall left-lateral transform
motion along the zone, i.e., between the Hreppar microplate to the north and the Eurasia
plate to the south, thus appears to be accommodated by right-lateral faulting on many
parallel, transverse faults and counter-clockwise rotation of the blocks between them,
‘‘bookshelf faulting’’ (EINARSSON et al., 1981).
Radon Changes and the South Iceland Earthquakes of June 2000
Figure 1
The active plate boundary of Iceland passes near the center of the Iceland hotspot. The radon program is
centered on the transform zone of South Iceland, SISZ. The Western and Eastern Volcanic Rift Zones are
marked with W and E, respectively. Arrows show the direction of relative plate movements across the plate
boundaries. The rate is 18.5 to 19.5 mm/year.
Earthquakes in South Iceland tend to occur in major sequences in which most of the
zone is affected. These sequences last from a few days to about three years. Each
sequence typically begins with a magnitude 7 event in the eastern part of the zone,
followed by smaller events farther west. Sequences of this type occurred in 1896, 1784,
1732–34, 1630–33, 1389–91, 1339 and 1294. Apart from the historic gap between 1391
and 1630, the sequences thus occur at intervals that range between 45 and 112 years
(EINARSSON et al., 1981), and it has been argued that a complete strain release of the
whole zone is accomplished in about 140 years (STEFÁNSSON and HALLDóRSSON, 1988).
The lengthy time since the last sequence led to a long-term forecast published in 1985
(EINARSSON, 1985), later refined by STEFÁNSSON et al. (1993), of a major earthquake
sequence within the next decades. The original forecast gave a 80% probability for the
occurrence of a major earthquake sequence within the next 25 years (i.e., within the
1985–2010 time window). The magnitude of the first event was estimated in the range
6.3–7.5 and the most likely location was given in the eastern part of the seismic zone.
In the refined version the magnitude range was reduced to 6.3–7.0 and two seismic gaps
were identified, at 20.3°W and 20.7°W. The forecast was fulfilled in June 2000 when
P. Einarsson et al.
Pure appl. geophys.,
Figure 2
The location of the radon sampling stations within the South Iceland Seismic Zone shown with triangles. Thin
lines show Holocene surface fractures, formed in earthquakes before 2000. The two long, thick lines show the
source faults of the two earthquakes of June 17 and 21, 2000 as delineated by aftershocks; the former one on the
easternmost fault, the latter on the western fault. The sense of faulting was right-lateral strike-slip on both faults.
Elevation contours are at 50 m intervals.
two magnitude 6.5 events occurred in the zone; one at 20.37°W and the other at
20.71°W.
3. The 2000 Earthquake Sequence
The earthquakes of 2000 were the largest in the zone since 1896 and 1912. They
occurred within the SIL-network of seismographs operated by the Icelandic Meteorological Office (see e.g., website http://www.vedur.is/, STEFÁNSSON et al. 1993). The sequence
began on June 17 at 15:40 with a magnitude 6.5 event in the eastern part of the zone
(Fig. 2). This immediately triggered a flurry of activity along at least a 90-km-long stretch
of the plate boundary to the west, apparently triggered by the passing S waves from the
first event. Among them was an event with an anomalously low seismic radiation but a
moment equivalent to a magnitude 5.9 (ÁRNADóTTIR et al., 2004). An earthquake of
magnitude 5.7 (mb) followed two minutes later on a small, parallel fault, about 3–4 km to
the west of the first shock. An event of magnitude 4.9 (mb) then occurred on the Reykjanes
Peninsula, 90 km to the west, about 5 minutes after the first shock. Several other
significant shocks also occurred this day along this segment of the plate boundary (PAGLI et
al., 2003). A second mainshock similar in magnitude to the first event occurred about
20 km west of the first one on June 21 at 00:51. It was clearly preceded by a clustering of
microearthquakes along the eventual source fault (STEFÁNSSON et al., 2000). ÁRNADóTTIR
et al. (2003, 2004) present evidence that triggering played a large role in the occurrence of
Radon Changes and the South Iceland Earthquakes of June 2000
events within the sequence, both dynamic triggering by the passing waves and triggering
by regional changes in the Coulomb stress due to faulting. The aftershock distribution,
moment tensor inversions, distribution of surface faulting, and modeling of surface
deformation measured by GPS and InSAR confirm that the mainshocks of the sequence
occurred on N-S striking faults, transverse to the zone itself (CLIFTON and EINARSSON, 2005;
ÁRNADóTTIR et al., 2001; PEDERSEN et al., 2003). The sense of faulting was right-lateral
strike-slip conforming to the model of ‘‘bookshelf faulting’’ for the South Iceland Seismic
Zone. The two mainshocks occurred on pre-existing faults and were accompanied by
surface ruptures consisting primarily of en-échelon tension gashes and push-up structures
(CLIFTON and EINARSSON, 2005). The main zones of rupture were about 15 km long and
fractured the crust down to 10 km depth. The surface faults coincided with the epicentral
distributions of aftershocks. Fault displacements were of the order of 0.1–1 m. Faulting
along conjugate, left-lateral strike-slip faults also occurred, but was less pronounced than
that of the main rupture zones. The maximum fault displacement at depth, determined by
modeling of geodetic data, was 2–2.5 m. The source faults of the two largest earthquakes
are shown in Figure 2. Large hydrological changes were observed in a wide area
surrounding the seismically active zone. Pressure changes in boreholes followed a regular
pattern conforming with crustal stress changes (BJöRNSSON et al., 2001; JóNSSON et al.,
2003). Pressure decreased in areas to the NE and SW of the epicenters but increased in the
quadrants to the NW and SE. These changes were large, but were reversed and
equilibrated in less than three months. A post-earthquake crustal deformation signal was
detected by InSAR that correlates with the water pressure changes (JóNSSON et al., 2003).
4. Previous Radon Studies in South Iceland
The relationship between radon and earthquakes has been studied in this area since
1977, when the first equipment for this purpose was installed. The instruments were
operated until 1993. The radon monitoring network consisted of up to 9 stations. Samples
(0.6 l) of geothermal water were collected from drill holes every few weeks and sent to the
laboratory for radon analysis. The resulting time series varied in length from 3 to 16 years.
Many earthquake-related radon anomalies were identified (JóNSSON and EINARSSON, 1996).
They are represented by both positive and negative excursions from the mean values, and
occur mostly prior to the seismic events, i.e., within a few weeks. For a statistical analysis
of the anomalies and comparison with the seismicity time series, significant earthquakes
were selected according to the criteria of HAUKSSON and GODDARD (1981):
M 2:4 log D 0:43
and M 2;
ð1Þ
where M is the magnitude and D is the distance to a radon monitoring station. Thus 98
independent seismic events were selected. They were in the magnitude range 2–5.8. The
main conclusions were as follows:
P. Einarsson et al.
Pure appl. geophys.,
1.
2.
3.
4.
Radon anomalies were observed before 30 of the significant events.
35% of all observed anomalies were related to seismicity.
80% of the anomalies observed before earthquakes were positive.
If a positive anomaly is detected at one station, the probability of a significant
earthquake occurring afterwards is 38%.
5. Some sampling sites were found to be more sensitive than others. The sensitivity
appears to depend on local geological conditions.
6. A few radon anomalies appeared to be related to eruptive activity of the neighboring
Hekla volcano.
5. Revival of the Radon Monitoring
A new radon program was initiated in 1999 using a new time-saving technique and
an instrument developed at our institute. It is based on a novel liquid scintillation
technique where counting only Bi-218/Po-218 pulse pairs gives high sensitivity with a
simple construction (THEODóRSSON, 1996; GUDJóNSSON and THEODóRSSON, 2000).
Scintillations other than those associated with radon-222 are thus excluded. Samples
of 200 ml are taken from the geothermal drill holes at the sampling sites (Table 1) and
analyzed in the laboratory. About 60% of the radon from the water samples is
transferred to a scintillator in a 15 ml liquid scintillation counting vial by circulating
and bubbling air for four minutes between the two liquids. The scintillation liquid is
mineral oil. The scintillations are subsequently counted in a laboratory-made automatic
sample changer. 226 counts per hour correspond to 1 Bq/l of radon in the water. The
system represents a significant progress in the radon measuring technique where high
sensitivity is needed. The technique also saves considerable time compared to previous
procedures. Sample preparation time was reduced from 3 hours to less than
10 minutes.
Sampling from geothermal wells in the South Iceland Seismic Zone began a year
before the destructive earthquakes of June 2000 occurred. Water samples were taken
Table 1
Location, depth and temperature of geothermal holes sampled for radon
Station
Latitude
Longitude
Depth, m
Teperature, °C
Bakki
Öxnalækur
Selfoss, hole 13
Hlemmiskeið
Flúðir, hole 5
Kaldárholt
Laugaland, hole 3
63°56.6
63°59.1
63°56.8
64°00.6
64°07.7
64°00.2
63°55.0
21°16.6
21°11.3
20°57.5
20°33.2
20°19.5
20°28.8
20°25.0
886
953
500
85
321
38
1100
120
100
& 85
66
94
62
100
Radon Changes and the South Iceland Earthquakes of June 2000
about twice a week from geothermal drill holes at seven sites (Fig. 2, Table 1) and sent
to the Science Institute, University of Iceland, for analysis. These holes, ranging in
depth between 38 m and 1100 m, were mostly the same as used in the earlier radon
program. The June 2000 earthquakes originated within our sampling network. The
nearest station, Laugaland, is only 2 km away from the source fault of the first
earthquake (Fig. 2).
6. The Radon Changes
The radon time series for the seven monitoring stations are shown in Figure 3. The
raw data are plotted as individual points connected by thin lines. The numbers denote
counts of disintegrations per hour in the sample. The thick, broken line shows a sevenpoint running average of the data. The vertical bars give the time of significant
earthquakes in the area. The height of the bar reflects the ‘‘excess magnitude’’ of the
event, ME, defined on the basis of the criteria of equation (1):
ME ¼ M ð2:4 log D 0:43Þ:
ð2Þ
Largest probabilities of radon anomalies are expected prior to events with positive excess
magnitude. The scale in the plots is arbitrary and the relative height of the bars is only
presumed to reflect the likelihood that the respective earthquake is associated with a
radon excursion at that particular station.
The radon variations form a distinct pattern that can be related to the earthquakes. A
typical behavior is seen at the station Hlemmiskeid. The radon values prior to the
earthquakes of June 2000 are relatively stable, varying by less than ± 50% around the
mean. The largest deviations are positive, spike-like excursions that occur 59 and
115 days before the earthquakes. These are the highest values measured at this station
during the two years of operation. Several of the lowest values are measured in a limited
time interval 125–167 days before June 17. A pronounced coseismic step is observed at
this station. The mean value drops by about 50% at the time of the earthquakes. About
three months later the mean value returns to the pre-earthquake level.
The main features of the Hlemmiskeið time series can be identified in the other time
series as well, except at Kaldárholt. The Öxnalækur series has a stable level in the
beginning, low values 124–138 days prior to the earthquakes, and a peak value
114 days before them. At Selfoss the same behavior is observed, but subdued. Low
values occur 124–142 days before, and high values 114 and 144 days before the events.
At Bakki anomalous values are observed, but the background level of radon is very
low. Therefore the negative deviations are more difficult to identify. The values are
apparently low in the time interval 117–134 days before the earthquakes, and 3 out of
the 4 highest values are recorded 57, 50, and 43 days before June 17. At Flúðir the
negative excursions are not prominent, although a large peak is seen 54 days before the
earthquakes. The Laugaland series has well developed negative deviations 101–149
P. Einarsson et al.
Pure appl. geophys.,
prior to the events but positive spikes are not seen, except possibly a high value 58 days
before the events. The Kaldárholt time series is quite irregular and does not display the
same pattern as the other stations. Kaldárholt is located quite close to the source fault of
the June 17 earthquake.
In summary, the preseismic spikes are observed at Öxnalækur, Flúðir, Bakki, and
possibly at Laugaland. The preseismic low is seen at Öxnalækur, Selfoss, Laugaland, and
subdued at Bakki. The coseismic drop is observed at Öxnalækur, Laugaland, and possibly
at Selfoss. At Kaldárholt and Flúðir the sampling was discontinued because of the
coseismic pressure drop in the geothermal system. The radon values had returned to
normal at all stations about three months after the earthquakes. The pressure in the
geothermal systems also returned to normal at about this time.
7. Discussion
The changes in radon concentration at our stations occur on a regional scale as shown
by the similarity between the time series in Figure 3. The distance between the stations
Selfoss and Laugaland is 27 km. This argues for a common cause of the changes. A
meteorological cause is considered highly unlikely. The samples are taken from deep
geothermal boreholes. No evidence for correlation with precipitation has been found.
Seasonal effects are also considered unlikely. Radon time series are available for the
same boreholes for the time period 1977–1993. Seasonal effects were only seen at one of
the stations (Bakki) and only for a part of the observation period (JóNSSON, 1994). All the
boreholes are producing geothermal holes, used for house heating. The production is at a
maximum during the winter months. Therefore, if a seasonal effect is responsible the
radon concentration is expected to be low in the winter and high in the summer. This is
not consistent with the changes observed in 2000 (Fig. 3).
The most robust result of this study is the demonstration of the coseismic drop in
radon concentration and its postseismic return to previous values. Because of its
temporal correlation with observed changes in groundwater pressure, we suggest that
there is a causal relationship between these parameters. We note, however, that the
radon concentration at all stations dropped during the earthquakes, regardless of
whether they were located within areas of increasing or decreasing water pressure. It
would seem that both increasing and decreasing groundwater pressure leads to a
decrease in radon flow from the crustal rocks. This may not be as unreasonable as it
seems at first sight. BJöRNSSON et al. (2001) and JóNSSON et al. (2003) point out that
the pressure variations show a spatial pattern consistent with stress changes due to the
faulting during the earthquakes. Pressure increases in areas where coseismic
volumetric compression occurs in the crustal rocks. Pressure decreases where
volumetric expansion occurs. The volumetric changes and water pressure are related
through the porosity of the rock which in this case is to a large degree due to
fractures. High water pressure is thus caused by the closure of cracks. Since the
Radon Changes and the South Iceland Earthquakes of June 2000
J
1200
F M
A M
J
J
F M
A M
J
J
A S
O
N D
O
N D
Öxnalækur
900
J
A S
600
2000
Hlemmiskeið
300
1500
0
1000
5000
Kaldárholt
4000
500
3000
0
2000
1000
1200
Laugaland
0
900
6000
Flúðir
600
4000
300
2000
0
0
800
Selfoss
600
300
400
200
200
100
Bakki
0
0
50
100
150
200
N day
250
300
350
0
0
50
100
150 200
N day
250
300
350
Figure 3
Time series of radon activity at the seven sampling stations. The activity is given in counts per hour. 226 counts
per hour correspond to 1 Bq/l of radon in the water. The data points are connected by a thin line. The dashed
lines show a seven-point running average of the data. Vertical bars with stars on top give the time of significant
earthquakes in the area. Their heights are proportional to the excess magnitude of the events, as defined by
equation (2) and depend on the earthquake magnitude and epicentral distance to the radon station.
release of radon into the water takes place across fracture walls, the closure of
fractures leads to reduced radon concentrations in the water. In the areas where
coseismic dilatation takes place, increasing pore volume leads to a drop in pressure
which inhibits the flow of water out of the rock. This will also lead to a reduced flux
of radon out of the crustal rocks.
Chemical components or physical parameters other than radon have not been
monitored at our stations. However, a multi-parameter approach is highly desirable for a
more meaningful interpretation of the causes of the changes as shown by the work of
CLAESSON et al. (2004, 2007) in North Iceland.
P. Einarsson et al.
Pure appl. geophys.,
8. Conclusions
Four different patterns can be identified in the radon time series within the South
Iceland Seismic Zone in association with the earthquake sequence of June 2000:
1. Pre-seismic decrease of radon. Anomalously low values were measured in the period
101–167 days before the earthquakes.
2. Preseismic increase. Positive spikes appear in the time series 40–144 days prior to the
earthquakes.
3. Coseismic step. The radon values decrease at the time of the first earthquake. This is
most likely related to the coseismic change in groundwater pressure observed over the
whole area.
4. Postseismic return to preseismic levels about three months after the earthquakes,
probably also linked with the pressure equilibration in the geothermal systems.
In view of the positive results of the project, we are developing and testing a new,
automatic radon instrument, Auto-Radon, based on the same design that continuously
monitors the radon concentration in the geothermal groundwater (THEODóRSSON and
GUDJóNSSON, 2003; JóNSSON et al., 2007). The instruments are located at the drill hole
stations, measuring radon four times a day.
Acknowledgements
The radon programs in South Iceland have been supported by grants from several
agencies, including the Icelandic Research Council, the SEISMIS Project, and the
European Union under the projects PRENLAB and PREPARED. Numerous persons
have participated in this radon project and measurements. We particular like to mention
Gı́sli Jónsson and the attendants of the sampling sites in South Iceland, Guðlaugur
Sveinsson, Stefán Ólafur Ólafsson, Valdimar Þorsteinsson, Vilhjálmur Eirı́ksson,
Guðrún Magnúsdóttir, Olgeir Engilbertsson, and Hannes Bjarnason.
REFERENCES
ÁRNADóTTIR, Th., HREINSDóTTIR, S., GUðMUNDSSON, G., EINARSSON, P., HEINERT, M., and VöLKSEN, C. (2001),
Crustal deformation Measured by GPS in the South Iceland Seismic Zone due to two Large Earthquakes in
June 2000, Geophys. Res. Lett. 28, 4031–4033.
ÁRNADóTTIR, Th., JóNSSON, S., PEDERSEN, R., and GUDMUNDSSON, G. (2003), Coulomb stress changes in the South
Iceland Seismic Zone due to two large earthquakes in June 2000, Geophys. Res. Lett. 30, doi:10.1029/
2002GL016495, no. 5.
ÁRNADóTTIR, Th., GEIRSSON, H., and EINARSSON, P. (2004), Coseismic stress changes and crustal deformation on
the Reykjanes Peninsula due to triggered earthquakes on June 17, 2000, J. Geophys. Res. 109, B09307,
doi:10.1029/2004JB003130.
Radon Changes and the South Iceland Earthquakes of June 2000
BJöRNSSON, G., FLóVENZ, Ó. G., SÆMUNDSSON, K., and EINARSSON, E. M. (2001), Pressure changes in Icelandic
geothermal reservoirs associated with two large earthquakes in June 2000. In Proceedings to Twenty-Sixth
Workshop on Geothermal Reservoir Engineering, Stanford University.
CLAESSON, L., SKELTON, A., GRAHAM, C., DIETL, C., MöRTH, M., TORSSANDER, P., KOCKUM, I. (2004),
Hydrogeochemical changes before and after a major earthquake, Geology 32, 641–644.
CLAESSON, L., SKELTON, A., GRAHAM, C., and MöRTH, M. (in review 2007), The timescale and mechanism of fault
sealing and water-rock interaction after an earthquake, Geofluids.
CLIFTON, A. and EINARSSON, P. (2005), Styles of surface rupture accompanying the June 17 and 21, 2000
earthquakes in the South Iceland Seismic Zone, Tectonophysics 396, 141–159.
EINARSSON, P. (1985), Jarðskjálftaspár (Earthquake prediction, in Icelandic with English summary),
Náttúrufræðingurinn 55, 9–28.
EINARSSON, P. (1991), Earthquakes and present-day tectonism in Iceland, Tectonophysics 189, 261–279.
EINARSSON, P., BJöRNSSON, S., FOULGER, G., STEFÁNSSON, R., and SKAFTADóTTIR, Th. (1981), Seismicity pattern in
the South Iceland Seismic Zone. In Earthquake Prediction - An International Review (eds. D. Simpson and
P. Richards), Am. Geophys. Union, Maurice Ewing Series 4, 141–151.
EINARSSON, P., CLIFTON, A., SIGMUNDSSON, F., and SIGBJöRNSSON, R. (2000), The South Iceland earthquakes of
2000: Tectonic environment and effects, Am. Geophys. Union, Fall Meeting, San Francisco, EOS 81, 890.
GUDJONSSON, G. I. and THEODóRSSON, P. (2000), A compact automatic low-level liquid scintillation system for
Radon in water by pulse pair counting, Appl. Radiation and Isotopes 53, 377–380.
HAUKSSON, E. (1981), Radon content of groundwater as an earthquake precursor: evaluation of worldwide data
and physical basis, J. Geophys. Res. 86, 9397–9410.
HAUKSSON, E. and GODDARD, J. (1981), Radon earthquake precursor studies in Iceland, J. Geophys. Res. 86,
7037–7054.
JóNSSON, S. (1994), Radonmælingar á Suðurlandi (Radon measurements in South Iceland, in Icelandic),
University of Iceland, Faculty of Science, BScThesis, 214 pp.
JóNSSON, S. and EINARSSON, P. (1996), Radon anomalies and earthquakes in the South Iceland Seismic Zone
1977–1993. In Seismology in Europe (ed. Thorkelsson, B. et al.), European Seismol. Commission, Reykjavı́k,
pp. 247–252.
JóNSSON, S., SEGALL, P., PEDERSEN, R., and BJöRNSSON, G. (2003), Post-earthquake ground movements correlated
to pore-pressure transients, Nature 424, 179–183.
JONSSON, G., THEODORSSON, P., and SIGURDSSON, K. (2007), Auto-radon — a new automatic liquid scintillation
system for monitoring Radon in water and air. In: Chalupnik S., Schonhofer, F., and Noakes J, eds. LSC 2005,
Advances in Liquid Scintillation Spectrometry, in press.
KING, C.-Y. (1985), Gas geochemistry applied to earthquake prediction: An overview, J. Geophys. Res. 91,
12,269–12,281.
PAGLI, C., PEDERSEN, R., SIGMUNDSSON, F., and FEIGL, K. L. (2003), Triggered Seismicity on June 17, 2000 on the
Reykjanes Peninsula, SW-Iceland Captured by Radar Interferometry, Geophys. Res. Lett. 30, 1273, 10.1029/
2002GL-015310.
PEDERSEN, R., JóNSSON, S., ÁRNADóTTIR, Th., SIGMUNDSSON, F., and FEIGL, K. L. (2003), Fault slip distribution of
two June 2000 Mw 6.4 earthquakes in South Iceland estimated from joint inversion of InSAR and GPS
measurements, Earth Planet. Sci. Lett. 213, 487–502.
ROELOFFS, E. (1999), Radon and rock deformation, Nature 339, 104–105.
SIGMUNDSSON, F., EINARSSON, P., BILHAM, R., and STURKELL (1995), Rift-transform kinematics in South Iceland:
Deformation from global positioning system measurements, 1986 to 1992, J. Geophys. Res. 100, 6235–6248.
STEFáNSSON, R., BÖðVARSSON, R., SLUNGA, R., EINARSSON, P., JAKOBSDóTTIR, S., BUNGUM, H., GREGERSEN, S.,
HAVSKOV, J., HJELME, J., and KORHONEN, H. (1993), Earthquake prediction research in the South Iceland
Seismic Zone and the SIL Project, Bull. Seismol. Soc. Am. 83, 696–716.
STEFáNSSON, R. and HALLDóRSSON, P. (1988), Strain release and strain build-up in the South Iceland Seismic
Zone, Tectonophysics 152, 267–276.
STEFáNSSON, R., ÁRNADóTTIR, Th., BJöRNSSON, G., GUðMUNDSSON, G. B., HALLDóRSSON, P. (2000), The two large
earthquakes in the South Iceland Seismic Zone in June 2000. A basis for earthquake prediction research, Am.
Geophys. Union, Fall Meeting, San Francisco, EOS 81, 890.
THEODóRSSON, P. (1996), Improved automatic Radon monitoring in groundwater, In Seismology in Europe (eds.
Thorkelsson, B. et al.), European Seismological Commission, Reykjavı́k, pp. 253–257.
P. Einarsson et al.
Pure appl. geophys.,
THEODóRSSON, P. and GUDJONSSON, G. I. (2003), A simple and sensitive liquid scintillation counting system for
continuous monitoring of Radon in water, Advances in Liquid Scintillation Spectrometry, 249–252.
TRIQUE, M., RICHON, P., PERRIER, F., and AVOUAC, J. P. (1999), Radon emanation and electrical potential
variations associated with transient deformation near Reservoir Lakes, Nature 399, 137–140.
WAKITA, H. (1996), Geochemical challenge to earthquake prediction, Proc. National Acad. of Sciences, USA,
Vol. 93, No. 9 (Apr. 30, 1996), pp. 3781–3786.
ZMAZEK, B., ITALIANO, F., ZIVCIC, M., VAUPOTIC, J., KOBAL, I., and MARTINELLI, G. (2002), Geochemical
monitoring of thermal waters in Slovenia: Relationships to seismic activity, Appl. Radiat. Isot. 57, 919–930.
(Received February 15, 2007, revised October 22, 2007, accepted October 24, 2007)
To access this journal online:
www.birkhauser.ch/pageoph