Geodesy and Geodynamics 14 (2023) 35e42

Contents lists available at ScienceDirect

Geodesy and Geodynamics

journal homepage: http://www.keaipublishing.com/geog

A review of tidal triggering of global earthquakes

Ruyu Yan a, b, Xiaodong Chen a, b, *, Heping Sun a, b, Jianqiao Xu a, Jiangcun Zhou a
a
State Key Laboratory of Geodesy and Earth's Dynamics, Innovation Academy for Precision Measurement Science and Technology, Chinese Academy of
Sciences, Wuhan 430077, China
b
University of Chinese Academy of Sciences, Beijing 100049, China

a r t i c l e i n f o a b s t r a c t

Article history: Earthquake prediction remains a challenging and difficult task for scientists all over the world. The tidal
Received 19 January 2022 triggering of earthquakes is being proven by an increasing number of investigations, most of which have
Received in revised form shown that earthquakes are positively correlated with tides, and thus, tides provide a potential tool for
30 April 2022
earthquake prediction, especially for imminent earthquakes. In this study, publications concerning the
Accepted 27 June 2022
Available online 12 August 2022
tidal triggering of earthquakes were compiled and analyzed with regard to global earthquakes, which
were classified into three main types: tectonic, volcanic, and slow earthquakes. The results reveal a high
correlation between tectonic earthquakes and tides (mainly for semidiurnal and diurnal tides; 14-day
Keywords:
Tidal triggering
tides) before and after the occurrence of significant earthquakes. For volcanic earthquakes, observa-
Tectonic earthquakes tions of volcanoes on the seafloor and land indicate that volcanic earthquakes in near-shore volcanic
Volcanic earthquakes areas and mid-ocean ridges have a strong correlation with tidal forces, mostly those with semidiurnal
Slow earthquakes and diurnal periods. For slow earthquakes, the periodicity of the tremor duration is highly correlated
Earthquake prediction with semidiurnal and diurnal tides. In conclusion, the tidal triggering of these three types of earthquakes
makes a positive contribution to earthquake preparation and understanding the triggering mechanism,
and thus, the prediction of these types of earthquakes should be investigated. However, there are still
several inadequacies on this topic that need to be resolved to gain a definitiveanswer regarding the tidal
triggering of all earthquakes. The main inadequacies are discussed in this paper from our point of view.
© 2022 Editorial office of Geodesy and Geodynamics. Publishing services by Elsevier B.V. on behalf of
KeAi Communications Co. Ltd. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction complexity of the seismic activity, the applicability of using tides, as

a known quantity, to predict earthquakes has not been fully
The tidal triggering of earthquakes has been debated since the confirmed. The existence of a seismic gap demonstrates that
17th century, when ocean tides were proposed as a cause of earthquakes have a slow stress accumulation and rapid release
earthquakes [1,2]. The periodic activities of tidal forces have been process in underground rocks [12]. The periodic activities of tidal
proven to be the triggering force of earthquakes in many publica- forces are supposed to play an important role during the stress
tions [3e6]; however, several publications have shown that there is accumulation stage, which has a confirmed relationship with
no relationship between tides and earthquakes [7e11]. Due to the earthquake preparation. The peaks of tidal stress are considered to
be the final catalyst during the triggering of earthquakes. A rapid
release of accumulated stress occurs when the stress on the fault
reaches its maximum. After the earthquake occurs, a new round of
* Corresponding author. Innovation Academy for Precision Measurement Science
and Technology, Chinese Academy of Sciences, No.340 Xudong Street, Wuhan, stress accumulation-release starts and seismic gaps appear. In
China. above description, tides are regarded as the dominant factor when
E-mail address: chenxd@apm.ac.cn (X. Chen). the fault stress is in a critical state near the threshold during
Peer review under responsibility of Institute of Seismology, China Earthquake earthquake preparation and triggering. Previous studies have pre-
Administration.
sented abundant observational evidence regarding the tidal trig-
gering of earthquakes using different methods [1,6,13e22].
Based on the abundance of previous investigations, the three
Production and Hosting by Elsevier on behalf of KeAi main methods of exploring the correlation between tides and

https://doi.org/10.1016/j.geog.2022.06.005
1674-9847/© 2022 Editorial office of Geodesy and Geodynamics. Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd. This is an open access article
under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
R. Yan, X. Chen, H. Sun et al. Geodesy and Geodynamics 14 (2023) 35e42

earthquakes are as follows: The first method is to examine the According to the characteristics of their faults, tectonic earth-
relative positions betweenthe Earth and the other planetsin the quakes can be further divided into thrust earthquakes, normal
solar system [13,23]. Celestial gravitation provides a comparatively earthquakes, and strike-slip earthquakes. For these different types
large force that can triggerearthquakes when the Earth and the of tectonic earthquakes, tidal triggering behaves differently [47].
other planets are arranged in a straight line, such as the relation- For thrust earthquakes, the shear stress on the fault plane has a
ship of the lunar phase and seismicitydiscussed by many Chinese very high correlation with the tides [6], especially for earthquakes
researchers [24e26]. The second method is to compute spectrum with depths of 0e40 km [45]. For normal fault types, there is a
analysis through sequences which is usually used for slow earth- significant correlation between the stress tensors of shallow
quakes [27e29]. The third method, which is widely used, is to earthquakes and larger earthquakes [6]. However, for strike-slip
explore the correlation between earthquakes and tidal stress on the faults, nearly no correlation with tides has been found. The
fault (including tidal coulomb failure stress, normal stress or shear reasonable explanation is that thrust and normal faults have
stress) [21,30,31] through statistical tests, including Schuster's test larger tidal stress amplitudes than strike-slip faults [45]. For all
[15,32], the permutation test [33], the Chi-squared test [34], Hi(s- the high-correlation situations, the occurrence of an earthquake is
togram)Cum (ulation) tests [19], and two-rate Poisson model concentrated on the tidal phase that accelerates fault sliding,
[11,35]. Because the general opinion is that earthquakes occur at which indicates that the high correlation is not accidental and is
tidal peaks, the simplest test among these statistical tests is to look reasonable from a physical point of view.
for correlations between tidal peaks and the occurrence of earth- Several researchers have used regional data to investigate the
quakes. Other methods have seldom been used in research on the relationship between earthquakes and tides. High correlations
relationship between earthquakes and tides. between earthquakes and tides have been found not only before
The occurrence of earthquakes is not only related to external but also after major earthquakes. Evidence for the foreshocks of
celestial forces but also to the geological conditions of everysingle major earthquakes has been found in many places around the
earthquake. In order to investigate the relevance of the tidal trig- world. Using Schuster's statistical test, Tanaka explored a series of
gering of earthquakes under different geological conditions, the megathrust earthquakes, including the MW9.1 Tohoku-Oki earth-
publications concerning the tidal triggering of earthquakes are quake in 2011 [15], the MW9.0 Sumatra earthquake in 2004 [48],
discussed according to the different geological types of earthquakes and the MW7.5 South Tonga earthquake in 1982 [32], and it was
in this study. Based on their geological conditions, earthquakes are found that the frequency of the tidal phase was near the peak
generally classified by seismologists into four main types: tectonic where the tidal shear/normal stress could accelerate fault sliding
earthquakes, volcanic earthquakes, induced earthquakes, and slow before the occurrences of these major earthquakes [49]. Significant
earthquakes [36]. Induced earthquakes are mostly influenced by correlations between tidal stresses and earthquakes about 2 years
human activities, such as reservoir water storage [37], hydraulic before the mainshock in Tengchong, China, have been identified
fracturing [38], wastewater injection [39], and oil and gas exploi- [43], indicating that tides may be a potential tool for forecasting
tation [40]. For this reason, it is difficult to make a direct connection earthquakes. Moreover, tides have a clear effecton triggering and/or
between tides and induced seismicity. Therefore, three types of accelerating the rupture of the mainshock-aftershockfault system.
earthquakes, i.e., tectonic earthquakes, volcanic earthquakes, and The aftershocks in the Ning'er area, China [44,50], and Jiaxian,
slow earthquakes, are discussed in this paper. The summary of Wutai, Taiwan region [51], were found to have been triggered by
previous studies on the correlation between tides and these three semidiurnal tides and diurnal tides. In the Hellenic Arc, Greece [52],
types of earthquakes may provide clues to fault ruptures and and Taiwan region [53], it was found that these aftershocks were
earthquake nucleation. Several new ideas for exploring the corre- triggered during the period of lunar tidal variations.
lation between tides and earthquakes are discussed at the end of But even in the same region, diverse results of tidal trigger-
this paper. ingearthquakes were found in some publications. For example,
there are negative and positive results of tidal triggering of earth-
2. Tectonic earthquakes and tides quakes using different time periods earthquake datarecorded in
central Japan [34,54,55]. To a specific earthquake (MW9.1 2011
Tectonic earthquakes, also known as fault earthquakes, are Tohoku-Oki earthquake), Tanaka shows that tidal triggering occurs
caused by tectonic movement in the lithosphere and account for over a decade-long period preceding the Tohoku-Oki earthquake
more than 90% of the world's earthquakes. When the deformation [15]. However, Wangfound that there was no correlation using 23-
exceeds the bearing capacity of the rock, the long-term energy day foreshock sequence of the Tohoku-Oki earthquake [54]. There
accumulated during the tectonic movement is rapidly released, are also similar results found in southern California using different
causing the rocks to vibrate and result in earthquakes. Research on data sequences [56]. Results show that clustering of earthquakes in
the tidal triggering of earthquakes has demonstrated that the tidal aftershock sequences, lower magnitude completeness levels, and
stress caused by the Sun, the Moon, and other celestial bodies is insufficient number of sampling cycles of data sequences may lead
superimposed on tectonic stress and plays a triggering role for to erroneous periodicities [56]. Another possible reason is that
earthquakes on seismogenic faults. Although the tidal stress varies using the Coulomb failure criterion to compute Tidal Coulomb
within a comparatively small magnitudedwith an order of 103 Pa, Failure Stress when calculating earth tidescan make statistics more
which is much smaller than the average stress drop of an earth- physically meaningful [35]. In conclusion, both seismic data and
quake (magnitude of 106 Pa)dthe stress accumulation rate is much physical mechanism of tidal triggering earthquakes are needed to
larger than the tectonic stress [1,41]. Earthquake rates may corre- be well understood.
late directly with the stress rate (especially daily maximum stress Based directly on laboratory experiments, Dieterich obtained
rates) when the stress in the source area is close to the critical level, the mechanism of the model for the nucleation of earthquake slip
therefore, earthquakes are triggered by Earth tides [6,42]. In [57] under periodic loads (earth tides as an example) [58]. Dieter-
conclusion, tides play an important role in the nucleation and ich's model are further confirmed by actual seismic observations
triggering of tectonic earthquakes, which has been proven in [59]. It indicates that the rate of seismicity influenced by tides is
numerous studies [6,43e46]. governed by three aspects: the amplitude of the earth tides, the
Several researchers have used the global earthquakesto normal stress acting on the fault, and the constitutive parameter
calculate the relationship between tectonic earthquakes and tides. that controls the direct velocity dependence of fault slip. Based on
36
R. Yan, X. Chen, H. Sun et al. Geodesy and Geodynamics 14 (2023) 35e42

the constitutive parameter, it is believed that a correlation between systems [59]. After careful consideration of the level of the back-
seismic rate and Earth tides can be detected at normal stresses ground noise, which is modulated by the local wind and meteo-
below around 8 MPa. Consequently, for a simple model of faults rological conditions, no evidence of tides, air pressure, or
with instantaneous instability at the threshold, earthquakes will temperature modulation was found to trigger the VLP events at the
always be expected to occur near the tides' maxima when period- Aso volcano [9]. Another example is the lack of correlations be-
ical tidal loading is shorter than earthquake nucleation time [42]. tween the thermal data and the semidiurnal and fortnightly solid
In conclusion, tides have different effects on the different types Earth tides at the OldoinyoLengai volcano [10]. A reasonable
of tectonic earthquakes. Thrust and normal faults are influenced by explanation for this is that the obtained tidal periods depend on the
tidal stresses, while strike-slip faults are not. Before and after the length of the observation dataset used. For example, a semidiurnal
occurrence of major earthquakes, high correlations between the and fortnightly tidal period may be detected using several months
earthquakes and tides (mainly regarding the periods of semi- of observations. However, the annual tidal period is undetectable in
diurnal, diurnal, and lunar tides) can be found. data with such a short observation duration.
Investigation of the activities of Miyake-Jima volcano [66] and
3. Volcanic earthquakes and tides Axial volcano on the Juan de Fuca Ridge [14,18,20,59,64,67] shows
that the seabed pressure changes caused by ocean tides and the
Before the eruption of volcanoes, the movement of underground peaks in seismic activity are strongly correlated, and a semi-
magma and/or groundwater usually generates earthquakes, which diurnaltidal periodicity is detected inspectrum analysis of the
are often accompanied by phenomena such as surface deformation seismic activity. It should be noted that these volcanic earthquakes
and the emission of gas and heat. There are currently five methods are preferentially triggered during low tides, but earthquakes
of monitoring volcanic activity: seismology, surface deformation, should be triggered at tides’ peaksaccording to the theories of rate-
volcanic gas, image and/or thermal imaging, and infrasound anal- dependent and state-dependent friction [58]. It can be explained by
ysis [60]. Among them, earthquake monitoring is one of the most the complex interplay between magma chamber expansion and
widely used methods. According to their different seismic sources, periodic tidal loading. As the magma chamber expands/deflates in
volcanic earthquakes can be classified into two types: volumetric response to tidal stress, Coulomb stress of opposite sign to tides is
source earthquakes and shear or tensile source earthquakes. correspondingly generated on the fault [59].
Volumetric source earthquakes are dominated by fluids, and the In summary, observations of volcanoes on the seafloor and land
pressure changes induced by the movement of underground ma- indicate that earthquakes in nearshore volcanic areas and mid-
terials or fluid thermal dynamics generate earthquakes. However, ocean ridges have a strong correlation with tidal forces, mostly
shear or tensile source earthquakes are dominated by brittle frac- exhibiting semidiurnal and/or diurnal periods. The fortnightly cy-
tures of the rocks. In addition, for the other classification of volcanic clic behavior of several events may be influenced by Earth tides,
earthquakes, long-period (LP) events, very-long-period (VLP) temperature, and/or atmospheric pressure variations, but re-
events, and volcanic tremors are characteristic of volumetric source searchers have not definitively determined which is the dominant
earthquakes, while volcanic-tectonic (VT) earthquakes are shear or factor.
tensile source earthquakes. Hybrid events have the characteristics
of both these types of sources [61]. Ourstudy of the relationship 4. Slow earthquakes and tides
between volcanic earthquakes and tides is based on these classifi-
cations. Previous studies have shown that tidal forces may trigger Slow earthquakes were considered a rare phenomenon in the
eruptive events [62e64]. 20th century because it was difficult for ordinary seismograph in-
Seven volcanoes (the Miyake-Jima volcano, Axial volcano, struments at the time to record them. In recent decades, a new
CampiFlegrei volcano, Stromboli volcano, Ruapehu volcano, Aso generation ofdensely spaced and highly sensitive earthquake
volcano, andOldoinyoLengai volcano) were investigated in this monitoring networks (including high-sensitivity borehole seis-
study to determine the correlation between volcanoes and tides mometer arrays and continuously recording global positioning
(Table 1 and Fig. 1). Studies on the CampiFlegrei volcano [7], system networks) has made it possible to monitor slow earth-
Stromboli volcano [63], and Ruapehu volcano [65] have revealed quakes [68]. Using data from earthquake monitoring networks,
that these volcanicactivities have a lunar period. Although it is Obara and Kato [69] identified the various types of slow earth-
uncertain which of the external triggers (e.g., rain, pressure, and quakes: low-frequency earthquakes (LFEs), very-low-frequency
tides) is the dominant factor, monitoring the temporal changes in (VLF) earthquakes, deep-low-frequency earthquakes, slow slip
these phenomena and their correlations with earthquakes and events (SSEs), episodic tremors and slips (ETS), and non-volcanic
volcanic activity can help us assess risk scenarios. The relationship tremors (NVTs) [70]. Slow earthquakes mostly occur in the sub-
between volcanic earthquakes and tides is affected by a variety of duction zone. Previous results have revealed that tremor activities
factors, such as changes in magma properties and plumbing are found in the following regions: Taiwan region, Nankai, Cascadia,

Table 1
Representative publications concerning volcanic earthquakes and tides.

No. Volcano Tidal period Publications Eruptiontype

1 Ruapehu volcano fortnightly Girona et al., 2018 [65] eruptions

2 Stromboli volcano fortnightly Sottili and Palladino, 2012 [63] explosive events
3 CampiFlegrei volcano fortnightly, monthly, semiannual, annual Petrosino et al., 2018 [7] volcano tectonic seismicity
4 Miyake-jima volcano semidiurnal Kasahara, 2002 [66] volcanicactivity
5 Axial volcano semidiurnal Sahoo et al., 2021 [64]; Wilcock et al., 2016 [18]; volcanic activity
Tolstoy et al., 2002 [20]; Wilcock, 2001 [14]
6 Aso volcano none Niu and Song, 2021 [9] VLPs
7 OldoinyoLengai volcano none van Manen et al., 2010 [10] variable eruptive activity

37
R. Yan, X. Chen, H. Sun et al. Geodesy and Geodynamics 14 (2023) 35e42

Fig. 1. Distribution of the volcanic earthquakes and slow earthquakes discussed.

Costa Rica, Chile, the San Andreas Fault, Ryukyu, Mexico, Alaska, from solid Earth tides have been found to be responsible for sig-
and New Zealand (Fig. 1) [69,71]. Fig. 2 shows the locations of nificant increases in the tremor rate [79].
various types of slow earthquakes in the cross-section of a sub- Tidal period of S2 (M2, N2, K1, P1, O1, fortnightly, Mm, and semi-
duction zone. The LFEs, shallow VLFs, and NVTs are located shal- annual) is 12.000 h (12.421 h, 12.658 h, 23.934 h, 24.066 h, 25.819 h,
lower than the megathrust earthquakes, while the SSEs and ETSs 13.661 days, 27.555 days, and 182.626 days).
are located deeper than the megathrust earthquakes. Tidal deformations can generate shear stress rates as high as
Slow earthquakes that have occurred in Cascadia [72,73], the ~10 kPa/day at plate interfaces, resulting in a strong triggering
Nankai Trough [16,74], Costa Rica [75], Taiwan region [27], Chile potential [16]. Because the frictional coefficient along a plate
[29], San Andreas [76], and Ryukyu [28] exhibit a high tidal sensi- boundary is relatively small (<0.1) [80], the shear strength of the
tivity, and the periodicities of the durations of the tremors in these fault is very weak [81]. According to the friction theories of rate-
regions are strongly correlated with the semidiurnal and diurnal and state-dependent friction [58], semidiurnal and diurnal tides
tides (Table 2). In Mexico, the tidal sensitivity is not uniform along can trigger SSEs when tidal stress fluctuations are superimposed on
the subduction interface, and the tremors in Ja-Co-Mi are more the plate boundary of a subduction zone.
sensitive to the tidal shear stress than those in Guerrero and Oaxaca
[77]. It has been found that the LFEs along the San Andreas Fault 5. Other types of earthquakes and tides
were modulated by a tidal stress with a 14-day cycle, and the
highest LFE rate occurred during the waxing fortnightly tide [78]. The relationships between tides and tectonic earthquakes, vol-
Not all tectonic earthquakes and volcanic earthquakes are triggered canic earthquakes, and slow earthquakes have been discussed
by tides, but almost all slow earthquakes have been found to be above. However, in addition to the classification based on the cause
triggered by tides. Slow earthquakes are believed to occur on faults of their occurrence, there are other classification schemes for
in the vicinity of the brittle
ductile (seismic
aseismic) transition, earthquakes. Two preliminary schemes adopted by many re-
and they preferentially occur when subjected to tidal shear searchers are the classification based on the magnitude of the
stresses, which promote failure. Very small stress perturbations earthquake and the classification based on the depth of the

Fig. 2. Cross-section of a subduction zone and the locations of the various types of slow earthquakes (an example based on the Nankai subduction zone, from Obara and Kato [69]).

38
R. Yan, X. Chen, H. Sun et al. Geodesy and Geodynamics 14 (2023) 35e42

Table 2
Representative publications concerning slow earthquakes and tides.

No. Location Tidal period Publications Type

1 Taiwanregion Semi-annual, 12 h, 24 h Chen et al., 2018 [27] Tremor events

2 Nankai 12 h, 24 h Yabe et al., 2015 [74]; Nakata et al., 2008 [16] non-volcanic tremors; slow slip events
3 Cascadia 12.4 h, 24e25 h Royer et al., 2015 [72]; Rubinstein et al., 2008 [73] episodic tremors and slips
4 Costa Rica 12 h Walter et al., 2013 [75] Tremors; very-low-frequency earthquakes
5 Chile M2, N2, K1, O1, P1, Mm Gallego et al., 2013 [29] non-volcanic tremors
6 San Andreas 12 h, Fortnightly van der Elst, 2016 [78]; Delorey et al., 2017 [76] low-frequency earthquakes; non-volcanic tremors
7 Ryukyu S2, M2, K1, O1 Nakamuraand Kakazu, 2017 [28] very-low-frequency earthquakes
8 Mexico Unknown Mauryet al., 2018 [77] slow slip events

hypocenter. Accordingly, the relationships between tides and these magnitude and randomness of the earthquakes' occurrence is other
two classification schemes are summarized here, but other types of two key factors in the tidal triggering. This is the reason why we
earthquakes are not discussed in this paper. focused on the relationships between tides and tectonic earth-
Many publications have discussed the relationship between quakes, volcanic earthquakes, and slow earthquakes in this study.
tides and the classification according to the magnitude of earth-
quakes in the last few decades since it is a simple and easy way to 6. Discussion and conclusions
classify earthquakes. The results of previous studies have revealed
that regarding the tidal triggering of earthquakes of various mag- In this study, the effect of tides on tectonic earthquakes, volcanic
nitudes, the degree of tidal influence varies [6,53]. For volcanic earthquakes, and slow earthquakes was investigated. For tectonic
earthquakes, micro-earthquakes (M < 2.5) exhibit a strong corre- earthquakes, thrust and normal faults are more easily influenced by
lation with tidal lows [20]. For tectonic earthquakes, before the tidal stress than strike-slip faults. Before and after the occurrence of
occurrence of a major earthquake, the foreshocks have been found major earthquakes, there is a high correlation between these
to primarily oscillate with the tides during the stress accumulation earthquakes and tides (semidiurnal tides, diurnal tides, and lunar
on the fault. In addition, small earthquakes (2.5 <M < 4) have been tides). For volcanic earthquakes, observations of volcanoes on the
found to be triggered by short-time-scale stress changes [5]. The seafloor and land indicate that earthquakes in nearshore volcanic
aftershocks are not only influenced by the tides but also by external areas and mid-ocean ridges have a strong correlation with tidal
stress such as stress changes related to recent seismic activity (e.g., forces, mostly with semidiurnal and diurnal periods. The fort-
aftershocks) [82] and/or surface loads [83], which makes it difficult nightly cyclic behavior on several time scales may be influenced by
to obtain the relationship between the tides and aftershocks. Earth tides, temperature, and/or atmospheric pressure variations.
Except for reverse faults, for which a relationship has been found However, researchers have not been able to definitively determine
between the shear tidal stress and earthquakes, the relationship which is the main factor. For slow earthquakes, through the stress
between earthquakes (4 <M < 7) and tides was not clear until fluctuations at the plate boundary in a subduction zone, the tidal
recently [5,6]. It has been found that very large earthquakes sensitivity is affected by the fluid in the fault. If the physical
(M  7), such as the 2004 Sumatra earthquake, the 2010 Maule properties of the weak part of a large fault change due to long-term
earthquake, and the 2011 Tohoku-Oki earthquake, often occur near fluid migration and the shear strength in the fault is very weak,
the maximum tidal stress amplitude [84,85]. The reason for this is then tidal forces will trigger an SSE. Inconclusion, not all earth-
that the possibility of small rock fractures expanding into huge quakes are triggered by tides, but based on numerous statistical
ruptures increases as the tidal stress increases [86]. investigations, tides play an important role when the tectonic
To investigate the focal depth of tidal triggered earthquakes, background is in a high-stress state; so, monitoring the changes in
earthquakes need to be classified according to their focal depths. tides and their correlation with earthquakes can help assess risk
Generally, earthquakes are classified into three types according to scenarios. The research results can also provide us with physical
their focal depths. Shallow earthquakes have focal depths of less data for analyzing fault ruptures and clues to the mechanisms and
than 70 km, and account for about 85% of global earthquakes. In- can serve as a powerful constraint for future mechanics models.
termediate earthquakes have focal depths of 70e300 km and ac- Based on results from laboratory experiments [58] and actual
count for about 12%of global earthquakes. Deep earthquakes have seismic observations, influences of tidal triggering earthquakes
focal depths greater than 300 km and account for about 3%of global have three dominant aspects: the amplitude of the earth tides, Tidal
earthquakes [87]. The category that accounts for the largest per- Coulomb Failure Stress (including friction coefficient), and the
centage, i.e., shallow earthquakes, has been used by most re- constitutive parameter of the fault. Earthquakes will always be
searchers to investigate the relationship between tides and expected to occur near the tides’ maxima for a simple model of
earthquakes. The results of these studies demonstrate that the faults with instantaneous instability at the threshold when peri-
correlation between earthquakes and tides is significant for shallow odical tidal loading is shorter than earthquake nucleation time
earthquakes [5,6,14,45]. Earthquakes (focal depths 70 km) exhibit [27,59], which is highly related to the local geological tectonic
indistinct correlations with tides, but some relevance between background [59,88]. The smaller the coefficient of friction (<0.1),
shear tidal stress and earthquakes has been found for normal faults the earthquakes are more likely triggered by tides (eg. slow
and reverse faults [5,6]. One reasonable physical explanation for earthquakes). The irrelevant between seismicity and tides can be
this is that the confinement stress of the fault becomes greater than explained by the constitutive properties of faults. A detectable
that of the tide as the depth increases, and thus, shallow earth- correlation between earthquakes and tidal stress can generally be
quakes are easily triggered by tides [5,42]. expected if the constitutive parameter for natural faults is signifi-
It should be noted that the classifications of earthquakes based cantly smaller than those observed for laboratory faults [80]. Based
on the magnitude of the earthquake and the depth of the focal on the constitutive parameter, it is believed that a correlation be-
source are slightly different for different regions, which is mainly tween seismic rate and Earth tides may be detected at normal
related to the local tectonic background. Completeness of the stresses below around 8 MPa.

39
R. Yan, X. Chen, H. Sun et al. Geodesy and Geodynamics 14 (2023) 35e42

Classification of earthquakes should adopt new methods. Large Conflicts of interest

and small earthquakes, and shallow and deep earthquakes can be
accurately distinguished and classified through the construction of The authors declare that there is no conflicts of interest.
a modern seismic network and precision seismic relocation tech-
nology. However, for tectonic earthquakes, it is more complicated Acknowledgments
to accurately determine the fault types of a series of earthquakes.
Many scholars have used statistics to determine the type of fault, This study was supported by the Strategic Priority Research
but there will inevitably be differences. For example, Tanaka [6] Program of Chinese Academy of Sciences (Grant No. XDB41
defined
120  rake 
60 as a normal fault, while Xu [89] 000000), and the National Natural Science Foundation of China
defined
135  rake 
45 and 45  dip 90 as a normal fault. (Grant No. 42174101, 41974023, 41874094, 41874026).
Several scholars [90,91] have used machine learning to identify
tectonic earthquakes and explosions, and their results have been References
preliminarily verified. Perhaps, in future research, machine
learning methods can be used to identify the fault types of tectonic [1] D. Emter, Tidal Triggering of Earthquakes and Volcanic Events, Tidal Phe-
earthquakes. nomena, Springer, 1997, pp. 3e309.
[2] P.W. Burton, Geophysics - is there coherence between earth tides and
A great deal of attention should be paid to the completeness of earthquakes, Nature 321 (1986) 115.
the earthquake sequence. A comprehensive, detailed, and in-depth [3] P.N. Shebalin, A.A. Baranov, Aftershock rate changes at different ocean tide
study of earthquake sequences can help determine the develop- heights, Front. Earth Sci. 8 (2020).
[4] S. Ide, Y. Tanaka, Controls on plate motion by oscillating tidal stress: evidence
ment trend of seismic activity after an earthquake event or a major from deep tremors in western Japan, Geophys. Res. Lett. 41 (2014)
earthquake. This could deepen our understanding of earthquake 3842e3850.
processes and ensure the scientific nature and stability of research [5] L. Metivier, O. de Viron, C.P. Conrad, et al., Evidence of earthquake triggering
by the solid earth tides, Earth Planet Sci. Lett. 278 (2009) 370e375.
[64,92]. There is an intense debate regarding tidally triggered [6] S. Tanaka, M. Ohtake, H. Sato, Evidence for tidal triggering of earthquakes as
earthquakes, which is due to the inadequacy of earthquake catalogs revealed from statistical analysis of global data, J. Geophys. Res. Solid Earth
[93]. Regarding the completeness of the catalogue, Tang et al. [51] 107 (2002).
[7] S. Petrosino, P. Cusano, P. Madonia, Tidal and hydrological periodicities of
used the matched-filter technique to detect the early aftershocks seismicity reveal new risk scenarios at campi Flegrei caldera, Sci. Rep. 8
after the ML 6.4 Jiaxian earthquake in 2010 and the ML 6.4 Wutai (2018) 13808.
mainshocks in 2012, and they detected approximately three times [8] C. Liu, A.T. Linde, I.S. Sacks, Slow earthquakes triggered by typhoons, Nature
459 (2009) 833e836.
more early aftershocks than are listed in the Central Weather Bu-
[9] J. Niu, T.R.A. Song, The response of repetitive very-long-period seismic sig-
reau's catalog in southern Taiwan region. Tang's results have shown nals at Aso volcano to periodic loading, Geophys. Res. Lett. 48 (2021)
that small early aftershocks frequently occur near or in areas where e2021GL092728.
negative Coulomb and shear stress changes are induced by Earth [10] S.M. van Manen, M. Kervyn, S. Blake, et al., Apparent tidal influence on
magmatic activity at oldoinyo Lengai volcano, Tanzania, as observed in
tides. If we ignore this part of the information, it will be difficult to moderate resolution imaging spectroradiometer (modis) data, J. Volcanol.
explore tidally triggered earthquakes. Therefore, when exploring Geoth. Res. 189 (2010) 151e157.
the correlation between tides and earthquakes, a complete dis- [11] J.E. Vidale, D.C. Agnew, M.J. Johnston, et al., Absence of earthquake correla-
tion with earth tides: an indication of high preseismic fault stress rate,
cussion of earthquakes is indispensable. J. Geophys. Res. Solid Earth 103 (1998) 24567e24572.
Different research methods have been used to obtain different [12] H.F. Reid, The Mechanics of the Earthquake, The California Earthquake of
results regarding tidally triggered earthquakes. There are many April 18, 1906, Report of the State Earthquake Investigation Commission,
1910.
methods of studying the correlation between tides and earth- [13] S.M. Awadh, Solar system planetary alignment triggers tides and earth-
quakes, such as the tidal stress and statistical testing methods quakes, J. Coast Conserv. 25 (2021).
mentioned in the reviews conducted by Li and Jiang [94]and Chen [14] W. Wilcock, Tidal triggering of micro earthquakes on the juan de Fuca Ridge,
Geophys. Res. Lett. 28 (2001) 3999e4002.
[95]; however, we found that the results obtained using different
[15] S. Tanaka, Tidal triggering of earthquakes prior to the 2011 tohoku-oki
methods may be different. For example, different scholars have earthquake (Mw9.1), Geophys. Res. Lett. 39 (2012).
conducted calculations for the Mygdonian Basin [19], the Ionian [16] R. Nakata, N. Suda, H. Tsuruoka, Non-volcanic tremor resulting from the
combined effect of earth tides and slow slip events, Nat. Geosci. 1 (2008)
Islands [96], the Greek arc [52], Santorini Island [97], and other
676e678.
placesusing the Hi(stogram)Cum (ulation) superposition analysis [17] W.A. Sauck, Brawley, California earthquake sequence of january, 1975, and
method [98]. It was found that the earthquakes in these places triggering by earth tides, Geophys. Res. Lett. 2 (1975) 506e509.
were clearly triggered by lunar and semidiurnal tides. However, [18] W.S. Wilcock, M. Tolstoy, F. Waldhauser, et al., Seismic constraints on caldera
dynamics from the 2015 axial seamount eruption, Science 354 (2016)
Schuster's test has also been applied to these regions, and the 1395e1399.
conclusions reached were not as clear [19]. Schuster's test and the [19] M. Contadakis, D. Arabelos, S. Spatalas, Evidence for tidal triggering on the
Hi(stogram)Cum (ulation) method are statistical testing methods, shallow earthquakes of the seismic area of mygdonia basin, north Greece,
Terrestr Stellar Environm (2009) 223e235.
and using different methods may provide different conclusions; [20] M. Tolstoy, F.L. Vernon, J.A. Orcutt, et al., Breathing of the seafloor: tidal
thus, the physical mechanism urgently needs to be investigated. correlations of seismicity at Axial volcano, Geology 30 (2002) 503e506.
Tide can be a potential way for earthquake early warning and [21] X. Wu, W. Mao, Y. Huang, et al., Tidal stress triggering effects of earthquakes
based on various tectonic regions in China and related astronomical char-
probabilistic forecasting. A long-standing question is whether we acteristics, science in China series G: physics, Mech Astron 52 (2009)
can estimate the eventual size of earthquakes when small cores 1271e1283 (in Chinese).
undergo dynamic rupture, tidal stress may be a potential solution. [22] S.R. McNutt, R. Beavan, Volcanic earthquakes at pavlof volcano correlated
with the solid earth tide, Nature 294 (1981) 615e618.
Previous results indicate that there are frequent similarities be- [23] H.-k. Jiang, J.-c. Zheng, Q. Wu, et al., Statistical Features of aftershock dis-
tween different-sized subduction-type earthquakes [99,100]. Large tribution size for moderate and large earthquakes in Chinese mainland, Acta
earthquakes are more probable during periods of high tidal stress Seismol Sin 20 (2007) 158e169 (in Chinese).
[24] P. Du, Correlation of strong earthquakes with lunar phases in north China
according to the previous investigations [86]. It is an approach
and its mechanism, Chin. J. Geophys. 42 (1999) 123e130 (in Chinese).
using small quakes and tidal stress to infer the possibility of a major [25] X. Chen, N. Zhong, J. Ding, The effect of lunar phase of seismicity in north
earthquake in the future. Consequently, knowledge of the tidal China and its significance to earthquake prediction, Earthquake 18 (1998)
stress state in seismic regions can be used to improve probabilistic 325e330.
[26] L. Li, Y. Chen, G. Zhang, Lunar modulation for Mb4.0 earthquakes in the
earthquake forecasting and short-term earthquake prediction, Taiwan area and its vicinity: phenomena and statistics in temporal-spatial
especially for extremely large earthquakes. domain, Earthq. Res. China 17 (2001) 210e220 (in Chinese).

40
R. Yan, X. Chen, H. Sun et al. Geodesy and Geodynamics 14 (2023) 35e42

[27] K.H. Chen, H.-J. Tai, S. Ide, et al., Tidal modulation and tectonic implications of [60] W.H. Lee, P. Jennings, C. Kisslinger, et al., International Handbook of Earth-
tremors in Taiwan, J. Geophys. Res. Solid Earth 123 (2018) 5945e5964. quake & Engineering Seismology, Part A, Elsevier, 2002.
[28] M. Nakamura, K. Kakazu, Tidal sensitivity of shallow very low frequency [61] B.A. Chouet, long-period volcano seismicity: its source and use in eruption
earthquakes in the Ryukyu trench, J. Geophys. Res. Solid Earth 122 (2017) forecasting, Nature 380 (1996) 309e316.
1221e1238. [62] G. Jentzsch, O. Haase, C. Kroner, et al., Mayon volcano, Philippines: some
[29] A. Gallego, R.M. Russo, D. Comte, et al., Tidal modulation of continuous insights into stress balance, J. Volcanol. Geoth. Res. 109 (2001) 205e217.
nonvolcanic seismic tremor in the Chile triple junction region, G-cubed 14 [63] G. Sottili, D.M. Palladino, Tidal modulation of eruptive activity at open-vent
(2013) 851e863. volcanoes: evidence from Stromboli, Italy, Terra. Nova 24 (2012) 233e237.
[30] J. ZHANG, Q.W. XI, L.Z. YANG, et al., A study on tidal force/stress trigerring of [64] S. Sahoo, B. Senapati, D. Panda, et al., Tidal triggering of micro-seismicity
strong earthquakes, Chin. J. Geophys. 50 (2007) 404e411 (in Chinese). associated with caldera dynamics in the juan de Fuca Ridge, J. Volcanol.
[31] Z. Ding, J. Jia, R. Wang, Seismic triggering effect of tidal stress, Acta Seismol. Geoth. Res. 417 (2021).
Sin. (Engl. Ed.) 5 (1983) 172e184 (in Chinese). [65] T. Girona, C. Huber, C. Caudron, Sensitivity to lunar cycles prior to the 2007
[32] S. Tanaka, M. Ohtake, H. Sato, Spatio-temporal variation of the tidal trig- eruption of Ruapehu volcano, Sci. Rep. 8 (2018) 1476.
gering effect on earthquake occurrence associated with the 1982 South [66] J. Kasahara, Tides, earthquakes, and volcanoes, science 297 (2002).
Tonga earthquake of mw 7.5, Geophys. Res. Lett. 29 (2002), 3-1-3-4. [67] Y.J. Tan, F. Waldhauser, M. Tolstoy, et al., Axial seamount: periodic tidal
[33] N. Cadicheanu, M. van Ruymbeke, P. Zhu, Tidal triggering evidence of in- loading reveals stress dependence of the earthquake size distribution (B
termediate depth earthquakes in the vrancea zone (Romania), Nat. Hazards value), Earth Planet Sci. Lett. 512 (2019) 39e45.
Earth Syst. Sci. 7 (2007) 733e740. [68] G.C. Beroza, S. Ide, Slow earthquakes and nonvolcanic tremor, Annu. Rev.
[34] S. Tanaka, M. Ohtake, H. Sato, Tidal triggering of earthquakes in Japan related Earth Planet Sci. 39 (2011) 271e296.
to the regional tectonic stress, Earth Planets Space 56 (2004) 511e515. [69] K. Obara, A. Kato, Connecting slow earthquakes to huge earthquakes, Science
[35] M. Bucholc, S. Steacy, Tidal Stress Triggering of Earthquakes in Southern 353 (2016) 253e257.
California 205 (2016) 681e693. [70] K. Arora, A. Cazenave, E.R. Engdahl, et al., Encyclopedia of Solid Earth
[36] S. Stein, M. Wysession, An Introduction to Seismology, Earthquakes, and Geophysics, Springer Science & Business Media, 2011.
Earth Structure, John Wiley & Sons, 2009. [71] S. Ide, Variety and spatial heterogeneity of tectonic tremor worldwide,
[37] G.R. Foulger, M.P. Wilson, J.G. Gluyas, et al., Global review of human-induced J. Geophys. Res. Solid Earth 117 (2012).
earthquakes, Earth Sci. Rev. 178 (2018) 438e514. [72] A. Royer, A. Thomas, M. Bostock, Tidal modulation and triggering of low-
[38] R.J. Skoumal, M.R. Brudzinski, B.S. Currie, Earthquakes induced by hydraulic frequency earthquakes in northern Cascadia, J. Geophys. Res. Solid Earth 120
fracturing in Poland township, Ohio, Bull. Seismol. Soc. Am. 105 (2015) (2015) 384e405.
189e197. [73] J.L. Rubinstein, M. La Rocca, J.E. Vidale, et al., Tidal modulation of nonvolcanic
[39] W.L. Ellsworth, Injection-induced earthquakes, Science 341 (2013). tremor, Science 319 (2008) 186e189.
[40] T. Van Eck, F. Goutbeek, H. Haak, et al., Seismic hazard due to small- [74] S. Yabe, Y. Tanaka, H. Houston, et al., Tidal sensitivity of tectonic tremors in
magnitude, shallow-source, induced earthquakes in The Netherlands, Eng. Nankai and Cascadia subduction zones, J. Geophys. Res. Solid Earth 120
Geol. 87 (2006) 105e121. (2015) 7587e7605.
[41] T.H. Heaton, Tidal triggering of earthquakes, Geophys. J. Int. 43 (1975) [75] J.I. Walter, S.Y. Schwartz, M. Protti, et al., The synchronous occurrence of
307e326. shallow tremor and very low frequency earthquakes offshore of the nicoya
[42] N.M. Beeler, D.A. Lockner, Why earthquakes correlate weakly with the solid peninsula, Costa Rica, Geophys. Res. Lett. 40 (2013) 1517e1522.
earth tides: effects of periodic stress on the rate and probability of earth- [76] A.A. Delorey, N.J. van der Elst, P.A. Johnson, Tidal triggering of earthquakes
quake occurrence, J. Geophys. Res. Solid Earth 108 (2003). suggests poroelastic behavior on the san Andreas Fault, Earth Planet Sci. Lett.
[43] G. Peng, X. Lei, G. Wang, et al., Precursory tidal triggering and B value vari- 460 (2017) 164e170.
ation before the 2011 M 5.1 and 5.0 Tengchong, China earthquakes, Earth [77] J. Maury, S. Ide, V.M. Cruz-Atienza, et al., Spatiotemporal variations in slow
Planet Sci. Lett. 574 (2021). earthquakes along the Mexican subduction zone, J. Geophys. Res. Solid Earth
[44] C. Xie, X. Lei, X. Zhao, et al., Tidal triggering of earthquakes in the ning’er area 123 (2018) 1559e1575.
of Yunnan province, China, J. Asian Earth Sci. 138 (2017) 477e483. [78] N.J. van der Elst, A.A. Delorey, D.R. Shelly, et al., Fortnightly modulation of san
[45] E.S. Cochran, J.E. Vidale, S. Tanaka, Earth tides can trigger shallow thrust fault Andreas tremor and low-frequency earthquakes, Proceedings of the National
earthquakes, Science 306 (2004) 1164e1166. Academy of Sciences of the United States of America 113 (2016) 8601e8605.
[46] X. Wu, W. Mao, Y. Huang, et al., The triggering effect of tidal stress of [79] A.M. Thomas, R. Bürgmann, D.R. Shelly, et al., Tidal triggering of low fre-
earthquakes in China and its related astronomical characteristics based on quency earthquakes near parkfield, California: implications for fault me-
different tectonic zones, Sci China (Ser G) 39 (2009) 901e912 (in Chinese). chanics within the brittle-ductile transition, J. Geophys. Res. Solid Earth 117
[47] H. Tsuruoka, M. Ohtake, H. Sato, Statistical test of the tidal triggering of (2012).
earthquakes: contribution of the ocean tide loading effect, Geophys. J. Int. [80] A.M. Thomas, R.M. Nadeau, R. Bürgmann, Tremor-tide correlations and near-
122 (1995) 183e194. lithostatic pore pressure on the deep san Andreas Fault, Nature 462 (2009)
[48] S. Tanaka, Tidal Triggering of Earthquakes Precursory to the Recent Sumatra 1048e1051.
Megathrust Earthquakes of 26 December 2004 (Mw9.0), 28 March 2005 [81] H. Houston, Low friction and fault weakening revealed by rising sensitivity of
(Mw8.6), and 12 september 2007 (Mw8.5), Geophys. Res. Lett. 37 (2010). tremor to tidal stress, Nat. Geosci. 8 (2015) 409e415.
[49] B.A.W. Brinkman, M. LeBlanc, Y. Ben-Zion, et al., Probing failure suscepti- [82] R.S. Stein, The role of stress transfer in earthquake occurrence, Nature 402
bilities of earthquake faults using small-quake tidal correlations, Nat. Com- (1999) 605e609.
mun. 6 (2015). [83] J.M. Sauber, B.F. Molnia, Glacier ice mass fluctuations and fault instability in
[50] C. Xie, X. Lei, X. Wu, et al., Effect of tidal stress on fault nucleation and failure tectonically active southern Alaska, Global Planet. Change 42 (2004)
of the 2007 Ms6.4 ning’ Er earthquake, Sci. China Earth Sci. 45 (2015), 1409- 279e293.
1420 (in Chinese). [84] L. Chen, J.G. Chen, Q.H. Xu, Correlations between solid tides and worldwide
[51] C.C. Tang, L.C. Lin, Y. Luo, et al., Possible earth-tide modulations of early af- earthquakes M-S >¼ 7.0 since 1900, Nat. Hazards Earth Syst. Sci. 12 (2012)
tershocks in southern Taiwan, Bull. Seismol. Soc. Am. 109 (2019) 1571e1577. 587e590.
[52] G. Vergos, D.N. Arabelos, M.E. Contadakis, Evidence for tidal triggering on the [85] Z.-r. Li, X.-d. J.E. Zhang, Relationship of tidal stress and large earthquakes,
earthquakes of the Hellenic Arc, Greece, Phys. Chem. Earth 85e86 (2015) Earthquake 31 (2011) 48e57.
210e215. [86] S. Ide, S. Yabe, Y. Tanaka, Earthquake potential revealed by tidal influence on
[53] C.-H. Lin, Y.-H. Yeh, Y.-I. Chen, et al., Earthquake clustering relative to lunar earthquake sizeefrequency statistics, Nat. Geosci. 9 (2016) 834e837.
phases in Taiwan, Terr. Atmos. Ocean Sci. 14 (2003). [87] W. Lowrie, A. Fichtner, Fundamentals of Geophysics, Cambridge University
[54] W. Wang, P. Shearer, No clear evidence for localized tidal periodicities in Press, 2020.
earthquakes in the Central Japan region, J. Geophys. Res. Solid Earth 120 [88] R. Chen, X. Ding, Distribution of local mean lunar times of significant shocks
(2015) 6317e6328. in source region and its vicinity before large earthquake and its explanation,
[55] H. Zhang, C. Xie, R. Yan, et al., Statistical study on the correlation between Chin. J. Geophys. 39 (1996) 224e230 (in Chinese).
seismic activity and tides in Japan, J. Geodesy Geodyn. 39 (2019) 1122e1126 [89] Y.-J. Xu, X.-P. Wu, C.-H. Yan, et al., The Features of tidal coulomb failure
(in Chinese). stresses on various kinds of seismic fault, Chin. J. Geophys. 54 (2011)
[56] S. Hartzell, T. Heaton, The fortnightly tide and the tidal triggering of earth- 756e763 (in Chinese).
quakes, Bull. Seismol. Soc. Am. 79 (1989) 1282e1286. [90] F. Giudicepietro, A.M. Esposito, P. Ricciolino, Fast discrimination of local
[57] J.H. Dieterich, A model for the nucleation of earthquake slip, Earthquake earthquakes using a neural approach, Seismol Res. Lett. 88 (2017)
source mechanics 37 (1986) 37e47. 1089e1096.
[58] J.H. Dieterich, Nucleation and triggering of earthquake slip: effect of periodic [91] S.R. Taylor, Statistical discriminants from two-dimensional grids of regional
stresses, Tectonophysics 144 (1987) 127e139. P/S spectral ratios, Bull. Seismol. Soc. Am. 101 (2011) 1584e1589.
[59] C.H. Scholz, Y.J. Tan, F. Albino, The mechanism of tidal triggering of earth- [92] Z. Han, G. Wang, C. Zhou, A review and discussion on earthquake sequence
quakes at mid-ocean ridges, Nat. Commun. 10 (2019) 1e7. research, Progess in geophysics 18 (2003) 74e78 (in Chinese).

41
R. Yan, X. Chen, H. Sun et al. Geodesy and Geodynamics 14 (2023) 35e42

[93] J. Hao, J. Zhang, Z. Yao, Evidence for diurnal periodicity of earthquakes from Xiaodong Chen was born in Shandong, China, in 1975. He
midnight to daybreak, Natl. Sci. Rev. 6 (2019) 1016e1023. received the B.S. degree from the Shandong University of
[94] J. Li, H. Jiang, A review on tidal triggering of earthquakes, Earthquake 31 Science and Technology, Qingdao, China, in 1998, and the
(2021), 36-47 (in Chinese). M.S. and Ph.D. degrees from the Institute of Geodesy and
[95] X. Chen, The tidal triggering of earthquakes, Progress in Earthquake Sciences Geophysics, Chinese Academy of Sciences, Wuhan, China,
(2021) 145e160 (in Chinese). in 2004 and 2010, respectively.He is currently with the
[96] M.E. Contadakis, D.N. Arabelos, S.D. Spatalas, Evidence for tidal triggering for Innovation Academy for Precision Measurement Science
the earthquakes of the ionian geological zone, Greece, Ann. Geophys. 55 and Technology, Chinese Academy of Sciences. His current
(2012). research interests include the solid Earth tides and lunar
[97] M.E. Contadakis, D.N. Arabelos, G. Vergos, Testing the Recent Santorini tides, gravity field investigation of the Earth and the Moon,
Seismic Activity for Possible Tidal Triggering Effect, EGU General Assembly processing and applications of gravity observations.
Conference Abstracts, 2013, pp. EGU2013eEGU8358.
[98] M. Van Ruymbeke, P. Zhu, N. Cadicheanu, et al., Very weak signals (vws)
detected by stacking method according to different astronomical periodic-
ities (hicum), Nat. Hazards Earth Syst. Sci. 7 (2007) 651e656.
[99] S. Ide, Frequent observations of identical onsets of large and small earth-
quakes, Nature 573 (2019) 112e116.
[100] M.-A. Meier, J. Ampuero, T.H. Heaton, The hidden simplicity of subduction
megathrust earthquakes, Science 357 (2017) 1277e1281.

Ruyu Yan is currently a Ph.D. student at Innovation Heping Sun was born in Jiangsu, China, in 1955. He
Academy for Precision Measurement Science and Tech- received the B.S. degree from the University of Science and
nology, Chinese Academy of Sciences (CAS). Her current Technology of China, Hefei, China, in 1980, and obtained
research mainly focuses on the earthquake tidal triggering. Ph.D. degree from Catholic University of Louvain, Belgium,
in 1995. He is currently academician of Chinese Academy
of Sciences and research professor of Innovation Academy
for Precision Measurement Science and Technology, Chi-
nese Academy of Sciences. His current research interests
include the signal detection, theoretical simulation and
mechanism study of high-precision gravity variations
induced by Earth's micro deformation and geodynamics.

42