Pure Appl. Geophys.
2016 Springer International Publishing
DOI 10.1007/s00024-016-1461-2
Pure and Applied Geophysics
A Stress Condition in Aquifer Rock for Detecting Anomalous Radon Decline Precursory
to an Earthquake
T. KUO,1 H. KUOCHEN,2 C. HO,3 and W. CHEN4
Abstract—Recurrent groundwater radon anomalous declines
were observed from well measurements in the Antung hot spring
area (eastern Taiwan) prior to five of six earthquakes that occurred
between 2003 and 2011 (Mw range 5.0–6.8). The relationship
between the detectability of radon anomalies and the first motions
of P-waves was investigated. Based on the first motions of P-waves
recorded near the investigated well, a precursory decrease in
groundwater radon can be detected only when the first motion is
compression. No precursory change in groundwater radon concentration was observed for the downward first motion of P-waves.
Key words: Radon-222, Groundwater, Earthquakes, P waves,
Stress.
1. Introduction
Radon concentration variations (Radon-222) in
groundwater have been frequently proposed as
earthquake precursor (Noguchi and Wakita 1977;
Shapiro et al. 1980; Wakita et al. 1980; Igarashi et al.
1995; Papastefanou 2002; Kuo et al. 2006a; Zmazek
et al. 2006; Erees et al. 2007; Yalim et al. 2007;
Namvaran and Negarestani 2013; Hashemi et al.
2013; Tarakçi et al. 2014; Tuccimei et al. 2015;
Attanasio and Maravalle 2016). According to a
worldwide survey (Hauksson 1981), more than 80%
of radon anomalies measured in groundwater prior to
earthquakes shows increases in radon concentration.
Radon-222 is a chemically inert radioactive gas with
a half-life of 3.82 days. The radon concentration in
1
Department of Mineral and Petroleum Engineering, National Cheng Kung University, Tainan, Taiwan. E-mail:
[email protected]
2
Institute of Geophysics, National Central University,
Jhongli, Taiwan.
3
Central Weather Bureau, Taipei, Taiwan.
4
Department of Geosciences, National Taiwan University,
Taipei, Taiwan.
groundwater is proportional to the uranium concentration (Uranium-238) in adjacent rocks. Because of
radon’s short recoil length (3 9 10-8 cm), only
atoms produced at the surface of rock grains are
released to the adjacent groundwater. Therefore, the
concentration of radon in groundwater is largely
dependent on the surface area of adjacent rocks
(Torgersen et al. 1990; Tuccimei et al. 2015). Before
the occurrence of an earthquake, regional stress
increases causing formation of cracks in the rocks.
Thus, prior to earthquakes, both the surface area of
the rocks and the radon concentration in groundwater
increase (Igarashi et al. 1995; Tuccimei et al. 2015).
A few anomalies measured in groundwater manifested decreases in radon concentration (Wakita
et al. 1980; Shapiro et al. 1980; Kuo et al. 2006a).
Anomalous decreases in radon concentration were
observed in groundwater prior to the Japan 1978 IzuOshima-Kinkai earthquake of magnitude M 7.0
(Wakita et al. 1980), prior to the U. S. 1979 Malibu
earthquake of magnitude M 4.6 (Shapiro et al. 1980)
and prior to the Taiwan 2003 Chengkung earthquake
of magnitude Mw 6.8 (Kuo et al. 2006a). The 2003
Mw 6.8 Chengkung was the strongest earthquake near
the Chengkung area in eastern Taiwan since 1951.
Mechanisms and geological conditions for interpreting anomalous decreases in radon prior to
earthquakes are seldom discussed in the literature.
The Antung hot spring is a low-porosity fractured
small aquifer situated in an andesitic block and surrounded by a ductile mudstone of the Lichi mélange
(Chen and Wang 1996). Regarding a physical basis
that explains the anomalous decrease in radon concentration in groundwater prior to the 2003
Chengkung earthquake, a mechanism of radon
volatilization was presented based on radon phase
T. Kuo et al.
Pure Appl. Geophys.
behavior and the geological conditions of the Antung
hot spring (Kuo et al. 2006b).
We have studied recurrent anomalous declines in
the concentration of groundwater radon at the Antung
hot spring in eastern Taiwan since July 2003. The
radon-monitoring well is located about 3 km southeast of the Longitudinal Valley fault (Fig. 1). The
fault ruptured during two 1951 earthquakes of magnitudes M 6.2 and M 7.0 (Hsu 1962). The
Longitudinal Valley fault is part of the eastern
boundary of the present-day plate suture between the
Eurasia and the Philippine Sea plates. Between July
2003 and December 2013, six main earthquakes
occurred near the Antung hot spring (Fig. 1). Anomalous decreases in the concentration of groundwater
radon were observed prior to the 2003 Chengkung
Mw 6.8, 2006 Taitung Mw 6.1 and Mw 5.9, 2008
Antung Mw 5.4, and 2011 Chimei Mw 5.0 earthquakes
(Kuo 2014). Nonetheless, no anomalous decline in
the concentration of groundwater radon was detected
at the radon-monitoring well prior to the 2013
Rueisuei Mw 6.3 earthquake.
A low-porosity fractured aquifer in un-drained
conditions near an active fault is a suitable geological
site to detect precursory declines in groundwater
radon prior to local large earthquakes (Kuo 2014).
Besides, stress conditions in aquifer rock can affect
the detectability of precursory radon declines. The
first motion of the P-waves provides the radiation
pattern of body waves at the time of the earthquake
Figure 1
Map of the epicenters of the earthquakes that occurred on December 10, 2003, April 1 and 15, 2006, February 17, 2008, July 12, 2011,
October 31, 2013 near the Antung hot spring. a Map of Taiwan. b Study area near the Antung hot spring (filled stars mainshocks, filled
triangle radon-monitoring well, filled square seismic station HWA055)
A Stress Condition in Aquifer Rock for Detecting Anomalous Radon Decline…
(nearly instantaneous process). Assuming the doublecouple mechanism, the radiation pattern representing
a source (for normal, thrust or strike slip fault) is
described by four areas (Shearer 2009). Two areas are
characterized by compression (upward first motion
for P-waves at a hypothetical station within this area)
and two areas are characterized by tension (downward first motion for P-waves at a hypothetical
station within this area). We further assume that a
similar stress pattern exists in rocks before and at the
time of the earthquake. The purpose of this work is to
examine the possible relationship between radon
anomalies and the first motions of P-waves recorded
near the radon-monitoring well for the above six
main earthquakes.
2. Materials and Methods
2.1. Geological Setting
The Antung area is in a unique tectonic setting
located at the boundary between the Eurasian and
Philippine Sea plates. Figure 2 shows the geological
map and cross section near the radon-monitoring well
D1 in the Antung area. The Antung hot spring is
situated in an andesitic tuffaceous-sandstone block
(Miocene) which is enclosed within the Paliwan
Formation (Late Pliocene to Pleistocene) of alternating thin-bedded sandstone and shale. The hot spring
is formed nearby an eastward-dipping, high-angle
reverse fault zone which contacts between the Lichi
mélange and the Paliwan Formation. Some hot
springs and mud volcanoes are scattered along the
fault zone, indicating a Quaternary active fault. Four
stratigraphic units are present (Chen and Wang
1996). The Tuluanshan Formation consists of Miocene volcanic units such as lava and volcanic breccia,
as well as tuffaceous sandstone. The Fanshuliao
(Pliocene) and Paliwan (Late Pliocene to Pleistocene)
Formations consist of rhythmic sandstone and mudstone turbidites. The Lichi mélange occurs as a highly
deformed mudstone that is characterized by penetrative foliation visible in outcrop.
Well-developed minor faults and joints are common in the tuffaceous-sandstone block displaying
intensively brittle deformation. It is possible that
these fractures reflect deformation and disruption by
the nearby faults. Ground water flows through the
fault zone and is then diffused into the block along
the minor fractures. The radon-monitoring well D1 is
not artesian, implying a weak recharge to a small
aquifer in un-drained conditions. Hence, geological
evidence suggests that the Antung hot spring at well
D1 is a small low-porosity fractured aquifer in undrained conditions near an active fault.
Under such geological conditions as the Antung
hot spring, two physical processes, rock dilatancy and
water diffusion, are likely to take place. When the
regional stress increases to about half the fracture
stress, rock dilatancy initiates and cracks develop in
aquifer rock (Brace et al. 1966). According to the
dilatancy-diffusion model (Nur 1972; Scholz et al.
1973), the development of new cracks in the aquifer
rock could occur at a rate faster than the recharge of
pore water. In a small aquifer with un-drained
conditions, gas saturation could develop in the rock
cracks. When gas phase develops in aquifer rock, the
radon in groundwater volatilizes into the gas phase
and the radon concentration in groundwater
decreases. The above mechanism is also referred to
as ‘‘in situ radon volatilization’’ (Kuo et al. 2006b).
Carminati et al. (2004) analyzed the migration of
seismicity on thrust and normal faults. The study
reports that, in compressional regimes, fractures
preferentially generate at shallow depths and propagate downward. Doglionia et al. (2011, 2013) devised
a simplified two-layer model to study fault activation
with a fault cross-cutting the brittle upper crust and
the ductile lower crust. The study shows that, along a
thrust fault, the hanging wall is over-compressed
during the interseismic, and should dilate at the
coseismic stage. Figure 2 shows that the Antung hot
spring is situated at the hanging wall along the
Yongfeng Fault and the Longitudinal Valley Fault,
both of which are thrust faults. Dilatant expansion
can develop in aquifer rock precursory to large
earthquakes near Antung.
2.2. Seismic Data
Seismic data (geographical coordinate of epicenter and depth) were taken from earthquake catalogues
of Central Weather Bureau, Taiwan. Focal mechanisms and Mw (moment magnitude scale) were from
T. Kuo et al.
Pure Appl. Geophys.
Figure 2
Geological map and cross section near the radon-monitoring well D1 in the area of Antung hot spring. (B tuffaceous andesitic blocks; filled
black triangle radon-monitoring well D1; Chihshang, or, Longitudinal Valley Fault, ` Yongfeng Fault)
the Global CMT catalog search. Two groups of
earthquakes were selected for this study. For the first
group, we selected all the mainshocks (Mw [ 6.0)
that occurred near the Autng hot spring between
December 10, 2003 and October 31, 2013. The 2003
Mw 6.8 Chengkung, 2006 Mw 6.1 Taitung, and 2013
Mw 6.3 Rueisuei earthquakes are the search results
from the Global CMT catalog. For the second group,
we selected all the mainshocks (Mw [ 5.0) that
occurred between December 10, 2003 and October
31, 2013 with epicenters located on the Longitudinal
Valley fault. The 2006 Mw 5.9 Taitung, 2008 Mw 5.4
A Stress Condition in Aquifer Rock for Detecting Anomalous Radon Decline…
Antung, and 2011 Mw 5.0 Chimei earthquakes are the
search results from the Global CMT catalog. The
focal mechanisms of the above six events are shown
in Fig. 1. Only event 2 (April 1, 2006) is strike-slipfaulting. All the other events are thrust-faulting.
2.3. Radon-monitoring Methods
Discrete samples of groundwater were pumped
and collected from the radon-monitoring well D1
located at the Antung hot spring twice per week for
analysis of radon content. The production interval of
the well ranges from 167 to 187 m below ground
surface. Every sampling started with flushing the
stagnant water in the monitoring well and in the
screen zone. An insufficiently pumped volume represents a major source of error. A minimum of three
well-bored volumes were purged before taking samples for radon measurements. To achieve the above
criterion, a minimum of 50 min purging-time was
required with a pumping rate at around 200 L/min.
Water samples were collected in a 40 mL glass vial
with a TEFLON-lined cap.
It is important to ensure the radon not to escape
during the sampling procedure and the sample transportation and preparation. After collecting a sample,
the sample vial was inverted to check for air bubbles. If
any bubbles were present in the vial, the sample water
was discarded and sampling was repeated. The date
and time of sample collection were recorded. The
samples were stored and transported in a cooler.
Counting radioactivity was done within 4 days.
The liquid scintillation method was adopted to
determine the activity concentration of radon in
groundwater (Noguchi 1964). Radon was partitioned
selectively into a white mineral-oil based scintillation
cocktail (Perkin Elmer) immiscible with the water
samples, and then assayed with a liquid scintillation
counter (LSC). The results were corrected for the
amount of radon decay between sampling and assay.
A calibration factor for the LSC measurements of
7.1 ± 0.1 cpm/pCi was calculated using an aqueous
Ra-226 calibration solution, which is in secular
equilibrium with Rn-222 progeny. For a count time
of 50 min and background less than 6 cpm, a
detection limit below 18 pCi/L was achieved using
the sample volume of 15-ml.
Due to the high background noise of radon time
series (Fig. 3), environmental records such as atmospheric temperature, barometric pressure, and rainfall
were examined to check whether the radon anomaly
could be caused by these environmental factors. Prior
to the 2003 Mw = 6.8 Chengkung and 2006
Mw = 6.1 Taitung earthquakes, radon concentration
in groundwater decreased from background levels of
787 ± 42 and 762 ± 57 pCi/L to minima of 326 ± 9
and 371 ± 9 pCi/L, respectively. There was no heavy
rainfall responsible for the above two radon anomalies. Besides, the atmospheric temperature,
barometric pressure, and rainfall are periodic in
season. It is difficult to explain the above two radon
anomalies by these environmental factors.
3. Results and Discussion
The sequence of events for radon anomalies prior
to the 2003 Chengkung Mw 6.8, 2006 Taitung Mw 6.1
and Mw 5.9, 2008 Antung Mw 5.4, and 2011 Chimei
Mw 5.0 earthquakes can all be characterized into three
stages (Fig. 4a–d). During stage 1, the radon concentration in groundwater is fairly stable; there is an
accumulation of tectonic strain and a slow, steady
increase of regional stress. Well D1 in the Antung hot
spring is completed in a small low-porosity brittle
aquifer surrounded by a ductile formation in undrained conditions. When the regional stress increases under these geological conditions, dilation of
brittle rock masses could occur at a rate faster than
the rate at which groundwater could diffuse or
recharge into the newly created rock cracks (Brace
et al. 1966; Nur 1972; Scholz et al. 1973). During this
stage (stage 2), gas saturation and two phases (vapor
and liquid) develop in the rock cracks. The radon in
groundwater volatilizes into the gas phase and the
radon concentration in groundwater decreases (Kuo
et al. 2006b). Stage 3 starts at the point of minimum
radon concentration when the water saturation in
cracks and pores begins to increase again. During this
stage (stage 3), the radon concentration in groundwater increases and recovers to the previous
background level before the main shock. The above
recurrent groundwater radon anomalies can be
explained based on the dilatancy-diffusion model
T. Kuo et al.
Pure Appl. Geophys.
1200
1100
1000
900
222Rn (pCi/L)
800
700
600
500
400
300
1
2
6.8
6.1
200
100
Pressure (hPa)
0
1040
1020
1000
980
960
Temperature ( oC )
940
35
30
25
20
15
10
5
Rainfall (mm)
0
600
500
400
300
200
100
0
2003/7/1
2004/7/1
2005/7/1
2006/7/1
Year/Month/Day
Figure 3
Comparison of radon concentration at well D1, atmospheric temperature, barometric pressure, and rainfall data (inside open inverted triangles
event number; arrows mainshocks; earthquake magnitude Mw shown beside arrows)
A Stress Condition in Aquifer Rock for Detecting Anomalous Radon Decline…
2003 Mw = 6.8
Chengkung Mainshock
1
65 days
787 ± 42 pCi/L
800
600
400
200
326 ± 9 pCi/L
Stage 1
0
2003/8/1
3
2
2003/9/1
Radon concentration (pCi/L)
56 days
4
700 ± 57 pCi/L
600
400
480 ± 43 pCi/L
200
Stage 1
0
2
2007/10/1 2007/11/1 2007/12/1
2008/1/1
3
2008/2/1
2
61 days
762 ± 57 pCi/L
3
600
400
200
371 ± 9 pCi/L
Stage 1
2
2006/1/1
2006/2/1
3
2006/3/1
(d)
1000
5
2006/4/1
2011 Mw = 5.0
Chimei Mainshock
54 days
752 ± 24 pCi/L
800
600
400
447 ± 18 pCi/L
200
Stage 1
0
2011/3/1
2011/4/1
2
2011/5/1
2011/6/1
3
2011/7/1
2013 Mw = 6.3
Rueisuei Mainshock
(e)
Radon concentration (pCi/L)
800
2008 Mw = 5.4
Antung Mainshock
(c)
1000
1000
0
2005/11/1 2005/12/1
2003/10/1 2003/11/1 2003/12/1
1000
800
Radon concentration (pCi/L)
1000
2006 Mw = 6.1
Taitung Mainshock
(b)
Radon concentration (pCi/L)
Radon concentration (pCi/L)
(a)
R
724 ± 56 pCi/L
800
600
400
200
0
2013/6/1
2013/7/1
2013/8/1
2013/9/1
2013/10/1
Figure 4
Radon concentration data at well D1 in the Antung hot spring prior to a 2003 Chengkung, b 2006 Taitung, c 2008 Antung, d 2011 Chimei, and
e 2013 Rueisuei earthquakes. Green rectangles show radon concentration between the mean radon concentration and three standard deviations
below the mean. Stage 1 is buildup of elastic strain. Stage 2 is development of cracks. Stage 3 is influx of groundwater. (a), (b), (c), and
(d) from Kuo (2014)
(Nur 1972; Scholz et al. 1973) and the mechanism of
in situ radon volatilization (Kuo et al. 2006b).
Physical models relating precursory phenomena
to earthquake occurrence would be useful for interpreting the above recurrent radon anomalies observed
at the Antung hot spring. Based on the physical
processes of rock dilatancy and water diffusion,
Scholz et al. (1973) presented a model to explain a
large class of earthquake precursors such as the ratio
of seismic velocities vp/vs, electrical resistivity, water
flow, geodetic measurements, and number of seismic
events. The model is based on laboratory studies of
rock dilatancy and fracture. The model assumes that
dilatancy precedes the earthquake. The dilatancy-
T. Kuo et al.
Pure Appl. Geophys.
diffusion model consists of three stages. During stage
1, the tectonic stress and strain accumulate. During
stage 2, the stress becomes large enough to cause
rock dilatancy at a rate faster than the rate of water
diffusion. Thus, gas phase develops in the rock cracks
during stage 2. During stage 3, the dilatancy rate
slows down and the rate of water diffusion is faster
than the rate of rock dilatancy. During stage 3, the
water saturation increases and the rock cracks
become saturated again. The Antung hot spring, a
small low-porosity fractured aquifer near an active
fault, is a suitable geological site where two physical
processes, rock dilatancy and water diffusion, are
likely to take place. Thus, the dilatancy-diffusion
model is selected to explain groundwater radon
anomalous declines recurrently observed in the
Antung hot spring area.
As shown in Fig. 4a–d, radon concentration of
well D1 in the Antung hot spring decreased from
background levels of 787 ± 42, 762 ± 57,
700 ± 57, and 752 ± 24 pCi/L to precursory minima
of 326 ± 9, 371 ± 9, 480 ± 43, and 447 ± 18 pCi/
L, respectively, prior to the 2003 Chengkung Mw 6.8,
2006 Taitung Mw 6.1 and Mw 5.9, 2008 Antung Mw
5.4, and 2011 Chimei Mw 5.0 earthquakes. The
observed precursory minimum in radon concentration
decreases as the earthquake magnitude increases.
Consequently, recognizing the v-shaped progression
in three stages becomes easier as the earthquake
magnitude increases (Fig. 4a, b versus c, d). Besides,
we can apply standard deviation to test each radon
anomaly separately. The mean radon concentration
and associated standard deviation are calculated
using radon data from stage 1 for each radon anomaly
(787 ± 42, 762 ± 57, 700 ± 57, and 752 ± 24 pCi/
L for the 2003 Chengkung Mw 6.8, 2006 Taitung Mw
6.1 and Mw 5.9, 2008 Antung Mw 5.4, and 2011
Chimei Mw 5.0 earthquakes, respectively). An
anomaly is defined as a significant deviation from the
mean value, or, three standard deviations below the
mean value. The green rectangles in Fig. 4 show
radon concentration between the mean radon concentration and three standard deviations below the
mean. The precursor times before the earthquake
occurrence range from 58 to 65 days for the above
radon anomalies. Anomalous radon concentrations,
smaller than three standard deviations below the
mean value, appeared during the precursor times as
shown in Fig. 4a–d.
Contrary to the above radon anomalies, the radon
concentration in groundwater observed at well D1
stayed fairly steady prior to the 2013 Mw 6.3 Rueisuei
earthquake that occurred on October 31. As shown in
Fig. 4e, the mean radon concentration and standard
deviation are calculated as 724 ± 56 pCi/L in the
period from June 1, 2013 to October 31, 2013. Figure 4e also shows that no radon data are outside the
green rectangle (smaller than three standard deviations below the mean value). Accordingly, no
anomalous decline in groundwater radon can be
defined at well D1 prior to the 2013 Mw 6.3 Rueisuei
earthquake. Tarakçi et al. (2014) investigated the
relationships between seismic activities and radon
level in Western Turkey. The study reported that, in a
compression seismic area, the radon level tended to
increase before the earthquakes and decrease towards
the time of earthquake occurrence. On the contrary,
in a dilation area, no change in radon levels was
observed. Stress conditions could affect the
detectability of precursory radon declines We will
compare the stress conditions near the Antung hot
spring area before and at the time of the 2013 Rueisuei earthquake with those of the other five
earthquakes.
Figure 5 shows the radon concentration data
between July 1, 2003 and October 31, 2013 at well
D1 in the Antung hot spring. Anomalous radon
minima of 326 ± 9, 371 ± 9, 480 ± 43, and
447 ± 18 pCi/L were observed prior to the 2003
Chengkung Mw 6.8, 2006 Taitung Mw 6.1 and Mw
5.9, 2008 Antung Mw 5.4, and 2011 Chimei Mw 5.0
earthquakes, respectively. We consider the 2006 Mw
5.9 Taitung earthquake that occurred on April 15,
2006 triggered by stress transfer in response to the
2006 Mw 6.1 Taitung earthquake that occurred on
April 1, 2006. The box-and-whisker plot is used on
the right-hand side in Fig. 5. It shows the median
(50th percentile, 767 pCi/L) as a center bar, and the
quartiles (25th and 75th percentiles, 698 and 828 pCi/
L) as a box. The whiskers (441 and 1077 pCi/L)
cover all but the most extreme values in the data set.
Our research effort at the Antung hot spring in eastern Taiwan since 2003 focuses on the anomalous
declines in groundwater radon precursory to large
44.4
1200
40.7
1100
37.0
1000
33.3
900
29.6
800
25.9
22.2
18.5
14.8
222Rn (pCi/L)
222Rn (Bq dm-3)
A Stress Condition in Aquifer Rock for Detecting Anomalous Radon Decline…
1077
828
767
698
700
600
500
441
400
11.1
300
7.4
200
3.7
100
0.0
0
7/1/2003
1
6.8
2
6.1
7/1/2005
R
3
4
5
5.9
5.4
5.0
7/1/2007
7/1/2009
7/1/2011
6.3
7/1/2013
Year/Month/Day
Figure 5
Radon concentration data at well D1 in the Antung hot spring and the box-and-whisker plot (inside open inverted triangles event number;
arrows mainshocks; earthquake magnitude Mw shown beside arrows)
earthquakes. Based on the box-and-whisker plot, a
threshold concentration (441 pCi/L) is estimated to
identify radon anomalies from well D1 data. The
radon minima recorded prior to the 2008 Mw = 5.4
Antung and 2011 Chimei Mw 5.0 earthquakes are
close to the threshold concentration (441 pCi/L) and
can be easily masked by the noisy background. On
the other hand, the radon anomalous minima, recorded precursory to strong earthquakes (Mw [ 6.0), the
2003 Mw = 6.8 Chengkung and 2006 Mw = 6.1
Taitung earthquakes, are well below the threshold
concentration (441 pCi/L) and can be clearly distinguished from the background noise. Unlike the 2003
Mw = 6.8 Chengkung and 2006 Mw = 6.1 Taitung
earthquakes, the radon concentration prior to the
2013 Rueisuei Mw 6.3 earthquake is well above the
threshold concentration (441 pCi/L). Stress conditions near the Antung hot spring area before and at
the time of the earthquake could affect the
detectability of precursory radon declines and the
concentration of anomalous radon minima.
To understand the stress conditions in aquifer rock
near the Antung hot spring area during earthquakes,
we chose the nearest seismic station (HWA055) of
Central Weather Bureau, Taiwan. The distance
between the seismic station and well D1 is 2.29 km.
Figure 6 shows an upward movement of the P-wave
first motion for all those events with
detectable anomalous declines in groundwater radon
(Events 1–5). For Event R without detectable anomalous declines in groundwater radon, the P-wave first
motion is downward. The first motion of the P-waves
indicates the stress condition in rocks, compression or
tension, at the time of the earthquake. For the above
radon anomalies observed at well D1, the precursor
times before the earthquake occurrence range from 58
to 65 days. If a similar stress pattern exists in rocks
before and at the time of the earthquake, the seismological observations described above suggest a
possible relationship between the stress conditions in
aquifer rock and the detectability of precursory radon
declines.
4. Conclusions
With the help of a long-term study of the radon
concentration in groundwater at the Antung hot
spring, suitable geological conditions to site a radonmonitoring well are recommended to detect recurrent
precursory radon declines. Besides, the possible
relationship between radon anomalies and stress
T. Kuo et al.
Pure Appl. Geophys.
Figure 6
Seismic waveforms of Events 1–5 and R recorded in HWA055 station. Gray rectangular area the zoomed region of the seismic waveform of
the P-wave polarity on the right panel for each event. Black arrow P-wave first motion
conditions in aquifer rock are investigated in this
paper. The significance of our research can be summarized as follows.
grateful to Mr. C. Lin of the Antung hot spring for his
kind field assistance.
1. A small low-porosity fractured aquifer in undrained conditions near an active fault is a
suitable geological site to detect precursory declines in groundwater radon prior to local large
earthquakes.
2. For aquifer rock under compressive conditions
before and at the earthquake, a precursory anomalous decrease in the concentration of groundwater
radon can be detected.
3. For aquifer rock under compressive conditions
before and at the earthquake, no precursory
change in the concentration of groundwater radon
was observed.
REFERENCES
Acknowledgements
Supports by the Ministry of Science and Technology
Taiwan (NSC, MOST), Central Geological Surveys,
Industrial
Technology
Research
Institute
(L550001060, N550003318), Radiation Monitoring
Center, and Institute Earth Sciences of Academia
Sinica of Taiwan are appreciated. The authors are
Attanasio, A., & Maravalle, M. (2016). Some considerations
between radon and earthquakes in the crater of L’Aquila. Natural
Hazards, 81, 1971–1979. doi:10.1007/s11069-016-2169-4.
Brace, W. F., Paulding, B. W., Jr., & Scholz, C. H. (1966). Dilatancy in the fracture of crystalline rocks. Journal Geophysical
Research, 71, 3939–3953. doi:10.1029/JZ071i016p03939.
Carminati, E., Doglioni, C., & Barbab, S. (2004). Reverse migration of seismicity on thrusts and normal faults. Earth Science
Reviews, 65, 195–222. doi:10.1016/S0012-8252(03)00083-7.
Chen, W. S., & Wang, Y. (1996). Geology of the Coastal Range,
eastern Taiwan. Geology of Taiwan 7. Central Geological Survey, Taiwan.
Doglionia, C., Barbab, S., Carminatia, E., & Riguzzib, F. (2011).
Role of the brittle–ductile transition on fault activation. Physics
of the Earth and Planetary Interiors, 184, 160–171. doi:10.1016/
j.pepi.2010.11.005.
Doglionia, C., Barbab, S., Carminatia, E., & Riguzzib, F. (2013).
Fault on–off versus coseismic fluids reaction. GSF, 5, 767–780.
doi:10.1016/j.gsf.2013.08.004.
Erees, F. S., Aytas, S., Sac, M. M., Yener, G., & Salk, M. (2007).
Radon concentrations in thermal waters related to seismic events
along faults in the Denizli Basin, Western Turkey. Radiation
Measurements, 42, 80–86. doi:10.1016/j.radmeas.2006.06.003.
Hashemi, S. M., Negarestani, A., Namvaran, M., & Musavi Nasab,
S. M. (2013). An analytical algorithm for designing radon
monitoring network to predict the location and magnitude of
earthquakes. Journal of Radioanalytical and Nuclear Chemistry,
295, 2249–2262. doi:10.1007/s10967-012-2310-0.
A Stress Condition in Aquifer Rock for Detecting Anomalous Radon Decline…
Hauksson, E. (1981). Radon content of groundwater as an earthquake precursor: evaluation of worldwide data and physical
basis. Journal Geophysical Research, 86, 9397–9410. doi:10.
1029/JB086iB10p09397.
Hsu, T. L. (1962). Recent faulting in the Longitudinal Valley of
eastern Taiwan. Geological Society of China, 1, 95–102.
Igarashi, G., Saeki, S., Takahata, N., Sumikawa, K., Tasaka, S.,
Sasaki, Y., et al. (1995). Ground-water radon anomaly before the
Kobe earthquake in Japan. Science, 269, 60–61. doi:10.1126/
science.269.5220.60.
Kuo, T. (2014). Correlating precursory declines in groundwater
radon with earthquake magnitude. Ground Water, 52, 217–224.
doi:10.1111/gwat.12049.
Kuo, T., Fan, K., Kuochen, H., & Chen, W. (2006a). A mechanism
for anomalous decline in radon precursory to an earthquake.
Ground Water, 44, 642–647. doi:10.1111/j.1745-6584.2006.
00219.x.
Kuo, T., Fan, K., Kuochen, H., Han, Y., Chu, H., & Lee, Y.
(2006b). Anomalous decrease in groundwater radon before the
Taiwan M6.8 Chengkung earthquake. Journal of Environmental
Radioactivity, 88, 101–106. doi:10.1016/j.jenvrad.2006.01.005.
Namvaran, M., & Negarestani, A. (2013). Measuring the radon
concentration and investigating the mechanism of decline prior
an earthquake (Jooshan, SE of Iran). Journal of Radioanalytical
and Nuclear Chemistry, 298, 1–8. doi:10.1007/s10967-0132162-7.
Noguchi, M. (1964). New method of radon activity measurement
with liquid scintillator. Radioisotopes, 13, 362–367. doi:10.3769/
radioisotopes.13.5_362.
Noguchi, M., & Wakita, H. (1977). A method for continuous
measurement of radon in groundwater for earthquake prediction.
Journal Geophysical Research, 82, 1353–1357. doi:10.1029/
JB082i008p01353.
Nur, A. (1972). Dilatancy, pore fluids, and premonitory variations
of ts/tp travel times. Bulletin of the Seismological Society of
America, 62, 1217–1222.
Papastefanou, C. (2002). An overview of instrumentantion for
measuring radon in soil gas and groundwaters. Journal of
Environmental Radioactivity, 63, 271–283. doi:10.1016/S0265931X(02)00034-6.
Scholz, C. H., Sykes, L. R., & Aggarwal, Y. P. (1973). Earthquake
prediction: a physical basis. Science, 181, 803–810. doi:10.1126/
science.181.4102.803.
Shapiro, M. H., Melvin, J. D., & Tombrello, T. A. (1980). Automated radon monitoring at a hard-rock site in the southern
California transverse ranges. Journal Geophysical Research, 85,
3058–3064. doi:10.1029/JB085iB06p03058.
Shearer, P. M. (2009). Introduction to seismology. Cambridge, UK:
Cambridge University Press.
Tarakçı, M., Harmanşah, C., Saç, M. M., & İçhedef, M. (2014).
Investigation of the relationships between seismic activities and
radon level in western Turkey. Applied Radiation and Isotopes,
83, 12–17. doi:10.1016/j.apradiso.2013.10.008.
Torgersen, T., Benoit, J., & Mackie, D. (1990). Controls on
groundwater Rn-222 concentrations in fractured rock. Geophys.
Res. Lett., 17, 845–848. doi:10.1029/GL017i006p00845.
Tuccimei, P., Mollo, S., Soligo, M., Scarlato, P., & Castelluccio,
M. (2015). Real-time setup to measure radon emission during
rock deformation: implications for geochemical surveillance.
Geoscientific Instrumentation, Methods and Data Systems, 4,
111–119. doi:10.5194/gi-4-111-2015.
Wakita, H., Nakamura, Y., Notsu, K., Noguchi, M., & Asada, T.
(1980). Radon anomaly: a possible precursor of the 1978 IzuOshima-kinkai earthquake. Science, 207, 882–883. doi:10.1126/
science.207.4433.882.
Yalim, H. A., Sandıkcıoğlua, A., Ünala, R., & Orhunb, Ö. (2007).
Measurements of radon concentrations in well waters near the
Akşehir fault zone in Afyonkarahisar, Turkey. Radiation Measurements, 42, 505–508. doi:10.1016/j.radmeas.2006.12.013.
Zmazek, B., Todorovski, L., Živčić, M., Džeroski, S., Vaupotiča,
J., & Kobala, I. (2006). Radon in a thermal spring: identification
of anomalies related to seismic activity. Applied Radiation and
Isotopes, 64, 725–734. doi:10.1016/j.apradiso.2005.12.016.
(Received May 3, 2016, revised December 13, 2016, accepted December 20, 2016)