Available via license: CC BY 4.0
Content may be subject to copyright.
1
Quaternary surface ruptures of the inherited mature Yangsan fault:
implications for intraplate earthquakes in Southeastern Korea
Sangmin Ha1,2, Hee-Cheol Kang1,3, Seongjun Lee1,4, Yeong Bae Seong2, Jeong-Heon Choi5, Seok-Jin
Kim5,6, Moon Son1
1Department of Geological Sciences, Pusan National University, Busan 46241, South Korea
5
2Department of Geography Education, Korea University, Seoul 02841, South Korea
3Institute of Geohazard Research, Pusan National University, Busan 46241, South Korea
4Department of Geotechnology & Infrastructure, Byucksan Engineering, Seoul 08380, South Korea
5Research Center of Earth and Environmental Sciences, Korea Basic Science Institute, Chungbuk 28119, South Korea
6Conservation Science Division, National Research Institute of Cultural Heritage, Daejeon, 34122, South Korea
10
Correspondence to: Moon Son (moonson@pusan.ac.kr)
Abstract. Earthquake prediction in intraplate regions, such as South Korea, is challenging due to the complexity of fault
zones. This study employed diverse methods and data sources to detect Quaternary surface rupturing along the Yangsan
Fault to improve seismic hazard assessment. Paleoseismic data were analyzed to reveal insights into seismic activity,
displacement, and structural patterns. Observations from five trench sites indicate at least three faulting events during the
15
Quaternary, with the most recent surface rupturing occurring approximately 3,000 years ago. These events resulted in a
cumulative displacement of 3.1–94.0 m and maximum estimated magnitude of 6.7–7.2. The average slip rate of 0.14 mm/yr
suggests a quasi-periodic model with possible recurrence intervals exceeding 10,000 years. The structural patterns imply the
reactivation of a pre-existing fault core with top-to-the west geometry, causing a dextral strike-slip with a minor reverse
component. This study underscores the continuous faulting along the inherited mature fault, the Yangsan Fault, since at least
20
the Early Pleistocene, contributing valuable insights for seismic hazard assessment in the region and offering a broader
understanding of intraplate earthquake dynamics for earthquake prediction.
Short summary. Unlike episodic plate boundary earthquakes, their randomness makes predicting intraplate earthquakes
challenging. This study aids seismic hazard assessment by analyzing paleoearthquake records of the Yangsan Fault in Korea.
25
Five trench sites revealed three Quaternary surface-rupturing earthquakes, the latest 3,000 years ago, with Mw 6.7-7.2. The
right lateral fault has a 0.14 mm/yr slip rate, and a recurrence interval over 10,000 years, continuously active since the
Quaternary.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
2
1 Introduction
30
Earthquake prediction is challenging, and large earthquakes with surface ruptures significantly damage infrastructure (Geller
et al., 1997; Wyss, 1997; Crampin and Gao, 2010; Woith et al., 2018). To predict large earthquakes, it is necessary to identify
the pattern of earthquake occurrences, which can be periodic, quasi-periodic, and random (Shimazaki and Nakata, 1980;
McCalpin, 2009). Earthquakes can be categorized as interplate (plate boundary) or intraplate (mid-continental) based on their
tectonic location (England and Jackson, 2011; Liu and Stein, 2016). Numerous studies have identified relatively "predictable"
35
cycles, patterns, causes, magnitudes, and epicentres in the interplate, where earthquakes are more frequent, leading to improved
earthquake preparedness (e.g., San Andreas Fault, Nankai trough; Powell and Weldon II, 1992; Murray and Langbein, 2006;
Smith and Sandwell, 2006; Uchida and Burgmann, 2019; Obara and Kato, 2024).
Large earthquakes in the intraplate occur less frequently than in the interplate but can have devastating impacts (e.g., 1556
Huaxian earthquake-M 8.0, 1976 Tangshan earthquake-Mw 7.8, 1811-1812 New Madrid earthquake, 2001 Gujarat earthquake-
40
Mw 7.6, 2011 Virginia earthquake-M 5.8; Min et al., 1995; Johnston and Schweig, 1996; Hough et al., 2000; Hough and Page,
2011; Wolin et al., 2012). Large intraplate earthquakes have recently received considerable attention and researchers are
attempting to understand them based on insights obtained from interplate earthquakes (England and Jackson, 2011; Liu and
Stein, 2016; Talwani, 2017). However, intraplate earthquakes exhibit complex spatiotemporal patterns that do not follow
general interplate earthquake pattern models (Liu and Stein, 2016; Talwani, 2017). This can partly be explained by the poor
45
catalogue due to their slow slip rates (<1 mm/year) and long recurrence intervals (>1,000 yr). The primary reason behind the
difficulty in predicting intraplate earthquakes is due to fault zone complexity and far-field stress, resulting in irregular
earthquake behaviour (Liu and Stein, 2016; Talwani, 2017).
To unravel the complexity of fault zones, it is essential to understand the geometry and internal structure of fault zone, along
with the relationship between geometry and the in-situ stress regime, fault kinematics controlled by structure, and the
50
correlation of these kinematics with paleoseismic data. Korea, located on the intraplate, has experienced very few directly
damaging earthquakes since instrumented seismic monitoring began. However, medium-sized earthquakes have caused
damage to people near the epicentres and raised earthquake awareness (e.g., the 2017 Pohang earthquake; Kim et al., 2018).
Despite calls for realistic earthquake preparedness, seismic research in South Korea still remains at an early stage compared
to neighbouring countries on plate boundaries, such as Japan and Taiwan. Basic seismic hazard assessment requires the
55
acquisition of reliable paleoseismic data (Reiter, 1990; Gurpinar, 2005; McCalpin, 2009). The reduction of paleoseismic
uncertainty necessitates an expanded dataset, which in turn demands a significant volume of data. Various methods can be
employed to efficiently detect Quaternary faults and obtain reliable paleoseismic parameters. Although this need has been
recognized, until recently, paleoseismology studies in Korea have mostly been based on a single trench, with a few exceptions
(e.g., Kim et al., 2023). Earthquakes in South Korea appear to follow a pattern of intraplate earthquakes (Kim and Lee, 2023).
60
To unravel this pattern, it is necessary to collect paleoseismic data from multiple faults with Quaternary surface rupturing to
complete the puzzle. In this study, we try to provide clues to the pattern of earthquake behaviour.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
3
This study sought to trace the distribution of the Quaternary surface rupturing during the Holocene and obtain fundamental
paleoseismic data. To achieve this, we applied multiple methods to identify traces of Quaternary faults with higher accuracy
and interpret a highly reliable paleoseismic synthesis from the obtained large amount of data (e.g., five trenches and about ~30
65
numerical ages). We focused on the Byeokgye section (study area; Kim et al., 2022) in the southern part of the northern
Yangsan Fault, a major tectonic line in southeastern Korea, to create geologic maps and select trench sites through fieldwork.
This study aims to reveal 1) robust paleoseismic data, filling gaps in prior surface rupture documentation, 2) Quaternary
reactivation features of intraplate mature fault zones, and 3) the spatiotemporal surface rupturing pattern of the Yangsan Fault,
a significant tectonic feature in East Asia and Korea. The results of this study can be used as essential inputs for seismic hazard
70
assessment and to understand the Quaternary faulting patterns of the Yangsan Fault, which is located intraplate.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
4
Figure 1: Regional study area maps. (a) Tectonic map of East Asia showing plate boundaries and their velocities (modified from
DeMets et al., 1990, 1994; Seno et al., 1993, 1996; Heki et al., 1999; Bird, 2003; O’Nelli et al., 2005; Schellart & Rawlinson, 2010; M.-
C. Kim et al., 2016). The black arrow indicates the moving direction of each plate. (b) Regional geological map of SE Korea (modified
75
from Chang et al., 1990; Kim et al., 1998; Hwang et al., 2007a, 2007b; Son et al., 2015; Song, 2015; Kang et al., 2018). The thick
black line indicates the Yangsan Fault and the 21 km dextral offset of the Yangsan Fault is interpreted by the distribution of a-type
granite. (c) Topographic satellite image map of SE Korea and recent epicentres. The Yangsan fault is divided into two segments
(modified from Kim et al., 2022; 2023b) and the study area located in southern part of the northern segment.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
5
2 Seismotectonic setting
80
2.1 Regional seismotectonic setting
The Korean Peninsula is located on the Amurian Plate, which was once thought to be part of the eastern margin of the Eurasian
Plate (Fig. 1a; e.g., Bird, 2003; Calais et al., 2006; Argus et al., 2010; Ashurkov et al., 2011). The relative frequency of
earthquakes in Korea is low compared with other countries on nearby plate boundaries, such as Japan, China, Taiwan, and the
Philippines (KMA, 2022). While paleoseismology, archaeoseismology, and historical seismicity studies indicate that large
85
(M≥7) earthquakes occurred in the past, medium-sized (M≥5) instrumental earthquakes have been consistently observed since
the onset of seismic instrument measurements (Lee, 1998; Lee and Yang, 2006; Han et al., 2016; Kim et al., 2016; 2018;
2020b). Currently, the Korean Peninsula is under maximum horizontal stress in the E-W to ENE-WSW direction, which is the
result of the combined effects of the subduction of the Pacific Plate and the far-field stress from the collision of the Indian and
Eurasian plates (Park et al., 2006; Kim et al., 2016). GNSS studies show that the Korean Peninsula has a velocity field of about
90
3 cm/yr in the WNW direction (Ansari and Bae, 2020; Sohn et al., 2021; Kim and Bae, 2023). Recent short-term (2 years)
GNSS studies on the Yangsan Fault suggest that the fault appears to be stable, with blocks on both sides of the fault moving
in the same direction and at the same displacement rate (Kim and Bae, 2023).
Paleoseismological studies on the Korean Peninsula since the 1990s revealed that the most Quaternary surface ruptures were
propagated along major structural lines in the southeastern part of the peninsula, such as the Yangsan Fault, Ulsan Fault, and
95
Yeonil Tectonic Line (Figs. 1b and 1c; Kee et al., 2009; Kim et al., 2011; Choi et al., 2012; Kim et al., 2016). Since the 2016
Gyeongju earthquake, the Yangsan Fault has drawn increasing research attention in various fields, including tectonic
geomorphology (Lee et al., 2019; Park and Lee, 2018; Kim and Oh, 2019; Kim et al., 2020d; Kim and Seong, 2021; Hong et
al., 2021), structural geology (Woo et al., 2015; Choi et al., 2017; Cheon et al., 2017, 2019, 2020a; Kim et al., 2017a, b, 2020a,
2021, 2022; Ryoo and Cheon, 2019; Kwon et al., 2020), paleoseismology (Lee et al., 2015; Choi et al., 2019; Cheon et al.,
100
2020b; Ko et al., 2020; Kim et al., 2023), and fault chronography (Yang and Lee, 2012, 2014; Song et al., 2016, 2019; Sim et
al., 2017; Kim and Lee, 2020, 2023). The reported Quaternary surface rupturing were reactivated along the pre-existing fault
surface with various kinematics depending on the relationship between the stress field and the geometry of the pre-existing
structure (Cheon et al., 2020a). The Yangsan Fault, one of the most significant structural lines on the Korean Peninsula, extends
> 200 km and is several hundred meters wide, with a prevailing horizontal displacement of > 20 km (Choi et al., 2017; Cheon
105
et al., 2017, 2019, 2020a, b). The Yangsan Fault underwent multiple deformations with various kinematic senses during the
Cretaceous to Cenozoic (Fig. 1b; Chang et al., 1990; Kim, 1992; Chang and Chang, 1998; Chang, 2002; Hwang et al., 2004,
2007a, b; Choi et al., 2009; Cheon et al., 2017, 2019). The dextral horizontal displacement of the Yangsan Fault is
approximately 25–35 km in the Cretaceous sedimentary rocks (Chang et al., 1990) and 21.3 km in A-type granite on both sides
of the fault (Fig. 1b; Hwang et al., 2004, 2007a, b). The most reported evidence for slip sense of Quaternary surface ruptures
110
along the Yangsan Fault indicate that they have been reactivated with transpressional dextral slip sense under E-W to ENE-
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
6
WSW oriented compressional stress fields (Kim et al., 2011; Choi et al., 2012; Jin et al., 2013; Yang and Lee, 2014; Lee et al.,
2015; Kim et al., 2016; Choi et al., 2019; Cheon et al., 2020a).
2.2 Geological settings of the detailed study area
115
The rocks in the study area are composed of Cretaceous sedimentary, volcanic, granitic, Paleogene alkaline granite, Middle
Miocene sedimentary, and Quaternary sediments (Fig. 2a). Alkaline granite, used as a dextral offset marker for the Yangsan
Fault, is distributed on the west side of the Yangsan Fault zone in the center of the study area and on the east side in Gyeongju
and the offset is 21.3 km (Fig. 1b, Hwang et al., 2004, 2007a, b). The Middle Miocene sedimentary rocks in the eastern part
of the study area are bounded by faults consisting of normal to dextral transfer faults, which form the western boundary of the
120
Pohang Basin (Middle Miocene; Fig. 2a; Son et al., 2015; Song, 2015). Quaternary sediments in the study area are widely
distributed along streams and valleys, and stratigraphic features cut by Quaternary faulting of the Yangsan Fault are observed
in some places (Figs. 2b and 2c).
In the study area, the Byeokgye (outcrop) and Dangu (trench) sites have been reported as Quaternary surface ruptures (Fig. 2;
Ryoo et al., 1999; Kee et al., 2009; Choi et al., 2012; Lee et al., 2015). The NNE-striking fault surface identified at the
125
Byeokgye site cut the Cretaceous acidic and Quaternary sediments. Based on the shear bands within the fault gouge, drag folds
in the footwall, and sub-horizontal striations, it was interpreted that strike-slip occurred after reverse slip (Ryoo et al., 1999).
The most recent earthquake (MRE) of Byeokgye occurred after 7.5±3 ka, the optically stimulated luminescence (OSL) age of
the truncated Quaternary sediments. Dangu is a trench approximately 50 m north of Byoegye that was excavated to obtain
more detailed paleoseismic data (Lee et al., 2015). A fault surface (N10–20°E/75–79°SE) with geometric and kinematic
130
characteristics similar to those of Byeokgye was found in the two trenches. The drag fold, slickenline, and three fault gouge
zones observed in the cross sections indicated that at least three strike-slip events with reverse slip occurred after the deposition
of Quaternary sediments. In addition, the MRE of Dangu using OSL dating, radiocarbon, and archaeological interpretation of
artifacts in the Quaternary sediments is 7.5±3 ka (Lee et al., 2015). Recently, Song et al. (2020) excavated a trench 1 km north
of Byeokgye to identify Quaternary faults and attempted paleoseismological interpretation. We set this trench as Trench 1 for
135
a comprehensive interpretation and summarize the results.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
7
Figure 2: Detailed study area maps. (a) Geological map of the study area (modified from Hwang et al., 2007; Song, 2015). The alkali
granite in the study area is used as a marker of YF's dextral offset. (b) Geomorphological map of the study area (modified from Ha
et al., 2022). A fault trace is recognized as an array of fault scarps and deflected streams. (c) Fault trace map of the study area. Two
140
outcrops and five trenches are located in the fault trace.
3 Method
3.1 Fault surface rupture tracking and trench siting
Ha et al. (2022) reported a 7.6 km fault trace in the study area through detailed topographic analysis using high-resolution (<<
20 cm) LiDAR images (Fig. 2b). They interpreted the main lineament as a fault scarp with a vertical displacement of 2–4 m
145
and a dextral refraction channel with a horizontal displacement of 50–150 m. In the trench on the fault trace, the fault surface
cuts through Quaternary sediments, indicating that the fault trace is Quaternary. We conducted fieldwork along this fault trace,
trenched it at four sites, and identified two natural exposures.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
8
3.2 Numerical dating
3.2.1 Quartz OSL and K-feldspar IRSL dating
150
For luminescence dating, a total of thirty-six samples were collected from the trench section by inserting light-tight stainless-
steel pipes (30 cm long and 5 cm diameter) into sand-rich layers. The sediment samples were then brought into a subdued
darkroom to extract pure quartz and K-feldspar grains. The size fractions of 90–250 μm were first separated by wet sieving.
Subsequently, these fractions were treated with 10% HCl (~ 1 h) and 10% H2O2 (~ 3 h) to remove carbonates and interstitial
organic matter. This was followed by density separation using SPT (sodium polytungstate; ρ=2.61 g/cm3). Pure quartz and K-
155
feldspar extracts were finally prepared by etching the sinking (quartz-rich) and floating (K-feldspar-rich) portions with ~ 40 %
and 10% HF (~ 1 h), respectively. The extracts were then rinsed with 10% HCl to dissolve any fluoride precipitates that may
have formed during HF etching. The purity of quartz extracts (i.e., the absence of K-feldspar contamination) was determined
by comparing infrared-stimulated luminescence (IRSL) and blue-stimulated luminescence (BSL) signals using three natural
and beta-irradiated (20 Gy) aliquots. For all quartz extracts, the IRSL signals were negligible, with IRSL to BSL ratios well
160
below 2 %, indicating that K-feldspar grains were effectively removed through the sample preparation procedure.
The luminescence signals were measured using a conventional Risø reader (Model TL/OSL-DA-20) installed at the Korea
Basic Science Institute (KBSI). The stimulation light sources were Blue-LEDs (470 nm, FWHM 20 nm) for quartz and IR-
LEDs (870 nm, FWHM 40 nm) for K-feldspar. The reader was equipped with a 90Sr/90Y source to deliver beta doses to the
sample position at a calibrated dose rate of 0.150 ± 0.002 Gy/s (Hansen et al., 2015). The signals were detected through a UV
165
filter (U-340) for quartz OSL and blue filter pack (a combination of 4 mm Corning 7-59 and 2 mm Schott BG 39) for K-
feldspar IRSL.
Quartz OSL equivalent dose (De) values were estimated using the single-aliquot regenerative-dose (SAR) procedure (Murray
and Wintle, 2000), with preheat conditions for main regeneration and test doses of 260℃ for 10 s and 220℃ for 0 s,
respectively. Quartz OSL signals in dose saturation (i.e., for samples older than datable age ranges of quartz OSL dating
170
method) and K-feldspar post-IR IRSL signals, which are well known to have higher dose saturation levels than quartz OSL
(Buylaert et al., 2012), were used as samples for De estimation. K-feldspar post-IR IRSL De values were estimated using the
SAR procedure suggested by Buylaert et al. (2009) where, after preheating at 250℃ for 60 s, IRSL signals in K-feldspar were
read at 225℃ for 100 s, immediately followed by lower temperature (50℃) IR stimulation (Hereafter, this signal is referred
to as pIRIR225). The measured fading rates (g2days) of the K-feldspar pIRIR225 signals ranged from 0–2 %/dec and fading
175
corrections were made using the R package ‘Luminescence’ (Calc_FadingCorr; Kreutzer, 2023) to derive final pIRIR225 age
estimates. The quartz OSL and fading-corrected K-feldspar pIRIR225 ages are summarized in Table 2.
3.2.2 Radiocarbon dating
Radiocarbon dating was performed to obtain independent ages and to cross-validate the ages of the sediments. Eight charcoal
samples were collected from sediment profiles of trenches 1, 2, and 4 and analyzed using accelerator mass spectrometry at the
180
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
9
Beta Analytic Radiocarbon Dating Laboratory. The 14C age was calibrated to calendar years using the OxCal 4.3 (Bronk
Ramsey, 2017) and IntCal 20 (Reimer et al., 2020) atmospheric curves.
3.2.3 10Be-26Al isochron burial dating
Isochron-burial dating is a variation of conventional burial dating methods that constrains the time elapsed since the sediment
burial and has been widely used for dating the alluvial sediments of 0.2–5.0 Ma (Granger & Muzikar, 2001). The burial
185
duration was calculated from the difference between the initial 26Al/10Be ratio at the time of burial and the measured ratio of
26Al/10Be (Balco & Rovey, 2008; Erlanger et al., 2012). Unlike traditional simple burial dating, which assumes that the initial
ratio is known or the same as the surface production ratio of 6.8, isochron-burial dating requires multiple samples with their
pre-burial histories (i.e., various initial ratios of 26Al/10Be) to construct an isochron. The initial ratio of 26Al/10Be of the samples
at burial depended on the erosion and production rates in the source basin. Thus, we assumed that all the analyzed samples
190
originated from the same basin under steady-state erosion. The pre-burial (i.e., inherited) concentrations of the gravel at the
time of burial accumulated with a surface production ratio of 6.8 so that it falls on a line in the plot of 26Al-10Be (Lal, 1991;
Granger, 2006; Balco and Rovey, 2008). We used the conventional value of 6.8 for the surface production ratio of 26Al/10Be
for the current study, which can vary depending on longitude, latitude, and altitude (Balco and Rovey, 2008). Without
considering the pre-burial exposure history, the measured nuclide concentrations of gravels at the same depth should fall again
195
on a line in the plot of 26Al-10Be with a mean life (τpb = (1/ (λ26 – λ10 )); ~ 2.07 Ma) because they have the same post-burial
production of 26Al (C26) and 10Be (C10). Thus, the postburial component can be treated as a constant so that it can be modeled,
and the initial ratio of 26Al/10Be (Rinh) can be calculated as
𝑵𝟐𝟔 = (𝑵𝟏𝟎𝑹𝒊𝒏𝒉𝒆−𝒕/𝝉𝒑𝒃 + 𝑪𝟏𝟎𝑹𝒊𝒏𝒉𝒆𝒕/𝝉𝒑𝒃 + 𝑪𝟐𝟔) (1)
The final time (tb) elapsed since burial can be calculated from the decay of the initial isochron (Rinh) to the isochron of the
200
measured samples (Rm):
𝒕𝒃= 𝝉𝒑𝒃 𝐥𝐧(𝑹𝒎/𝑹𝒊𝒏𝒉 ) (2)
3.2.4 ESR dating
ESR dating of fault rocks is a method used to derive the timing of faulting by measuring the ESR signal in the quartz of a fault
gouge (Ikeya et al., 1982; Lee & Schwarcz, 1994). In general, ESR dating has an age detection limit of up to several thousands
205
of ka (e.g., Lee and Yang, 2003) and is thus helpful for direct dating the timing of faulting. The ESR signal of quartz grain s
can be reset when the fault surface is subjected to normal stress of more than 3 MPa and displacements of more than 0.3 m
(Kim and Lee, 2020; 2023). These conditions are usually easier to meet at a depth of more than ~100 m than at the surface.
Therefore, it is likely that ESR signals from fault gouges currently exposed at the surface were reset at depths and then brought
up to the surface (i.e., current position) via uplift. In addition, in the case that the ESR signals were not completely reset during
210
faulting, the ESR ages may indicate the maximum age of MRE.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
10
3.3 Paleo-stress reconstruction
Palaeo-stress reconstruction of Quaternary faulting was conducted using fault-slip data from 23 slickenlines in cross-sections.
The data were analyzed using Wintensor S/W (v.5.8.5; Delvaux & Sperner, 2003). The type of stress regime is expressed
quantitatively using the R' value: R′=R (σ1 is vertical), or 2-R (σ2 is vertical), or 2+R (σ3 is vertical); R=(σ2−σ3)/(σ1−σ3)
215
(Delvaux et al., 1997).
3.4 Displacement and earthquake magnitude estimation
The slickenlines of the main fault surface and the vertical separation of quaternary sediments in the trench section were used
to determine the true displacement of the MRE and the displacement per event. Deriving the true displacement is challenging
when exposed sections are inclined, markers are inclined, or the slip sense is not purely dip-slip or strike-slip (which is almost
220
always the case). In addition, displacements based on fragmentary information, such as bedrock separation and thickness of
Quaternary sediments, can be over- or underestimated by fault slip motion and the possibility of paleo-topographic relief
cannot be ignored. Despite these uncertainties, fault displacement is a necessary factor in earthquake magnitude estimation
and key paleoseismological information; therefore, the process of collecting or estimating fault displacement is indispensable
in paleoseismology. Therefore, correlations based on vertical separation, marker dip angle, angle of outcrop cross-section,
225
fault dip angle, rake of slickenline, etc. are important for estimating the true displacement of a fault (Fig. A1; Xu et al., 2009;
Jin et al., 2013; Lee et al., 2017; Gwon et al., 2021). The method of using their relationship to find the true displacement is
described in detail in Appendix A.
Variables used for earthquake magnitude estimation include average displacement (Kanamori, 1977), maximum displacement
(MD; Bonilla et al., 1984; Wells and Coppersmith, 1994), surface rupture length (Bonilla et al., 1984; Khromovskikh, 1989;
230
Wells and Coppersmith, 1994), rupture area (Wells and Coppersmith, 1994), and surface rupture length × MD (Bonilla et al.,
1984; Mason, 1996). However, in Korea, where rupture traces are difficult to find, it is difficult to use surface rupture length
or rupture area owing to large uncertainties. Thus, we used MD, which is relatively easy to obtain from outcrops and trenches
and more reliable. Many previous studies in Korea have applied the empirical relationship of the MD-moment magnitude (Mw)
presented by Wells and Coppersmith (1994) (Kyung, 2010; Kim & Jin, 2006; Jin et al., 2013; Lee et al., 2017). We also
235
estimated the maximum earthquake magnitude by applying the MD obtained from the trench to the empirical formula. The
rake of slickenlines on the fault surface that underwent Quaternary faulting averages 20° and strike-slip motion was dominant;
therefore, we used a corresponding strike-slip fault type Mw-MD empirical relationship. In addition, we used the MD-surface
rupture length (SRL) empirical relationship to determine the extent to which the derived MD differed from the true
displacement.
240
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
11
4 Results
4.1 Characteristics of Quaternary faulting in the trench
4.1.1 Trench 1
Trench 1 was reported by Song et al. (2020) and Ha et al. (2022) and the findings are summarized as follows:
Trench 1 was located on the main lineament, approximately 1 km north of the Byeokgye site (Fig. 2c). Nine Quaternary
245
sedimentary units and seven east-dipping splay slip surfaces (F1–F7) cutting the units were found in the E-W trending trench
section (Fig. 3). The hanging wall of F6 is the western boundary of pre-existing fault core more than 2.5 m wide and the
footwall of F6 is a 4 m thick Quaternary sedimentary unit overlying the alkali granite. The pre-existing fault core is divided
into two zones based on whether it is related to rupturing. Fault core 1 had a 40 cm wide blue fault gouge cutting Quaternary
sediments and a weakly developed shear band within the fault gouge. Fault core 2, east of fault core 1, did not cut the
250
Quaternary sediments and consisted of a brown to dark gray fault-brecciated zone and a fault gouge zone more than 2 m wide.
White, light-grey, and dark-grey fault gouge bands developed anastomosingly within the fault-brecciated zone. Slickenlines,
observed at the boundary of the fault-brecciated and fault gouge zones, indicate a dextral strike-slip. The stratigraphic features
of the nine units are listed in Table 1, and there are two noteworthy observations: First, the triangular-shaped unit D has a light
brown sandy matrix with good sorting and roundness compared to the surrounding units, despite it is surrounded by the slip
255
surfaces (Fig. 3). It indicates that unit D may have been captured by the horizontal displacement of nearby sediments during
the faulting event. Second, various types of seismogenic soft-sediment deformation structures (SSDs) are developed in units
E and G (Fig. 3, Fig. 10 in Ha et al., 2022). The orientations of slip surfaces ranged from N60°W to N28°E, changing from
NW-to NE-striking to the east. The F6 fault (N01°E/69°SE) cut through unit B and the F7 fault (N28°E/86°SE) is terminated
in unit C. Each of F3–F5 cuts through units D, E, and F, respectively, and is terminated under unit C. F1 and F2 cut through
260
units G, H, and I but not through unit F. The rakes of the slickenline observed on F6 measured 15–55°, indicating a dextral
strike-slip with a reverse component. Three faulting events can be analyzed in terms of the geometry of the sediments and the
kinematics of the slip surface.
OSL/pIRIR225 ages are presented in Table 2. We also conducted 14C and ESR dating on the trench. Three charcoal samples
(1803BYG-01-C, 1803BYG-02-C, and 1803BYG-03-C) were collected from unit E and one charcoal sample (1803BYG-04-
265
C) was collected from unit I for radiocarbon dating. The results of samples 1803BYG-01-C and 1803BYG-02-C are 36,897–
38,420 and 43,802-45,670 Cal yr BP, respectively (Table 3). However, the ages of these two samples were near the upper limit
of the radiocarbon dates and were stratigraphically contradictory, with the lower layer being younger than the upper layer. The
age of 1803BYG-03-C (7,675–7,821 cal yr BP) from the upper part of unit E was inconsistent with the OSL age (1803BYG-
10-O: ~164 ka). It is possible that liquefaction in unit E caused a disturbance in the sediments and that the radiocarbon and
270
OSL dates do not indicate the exact depositional timing. In addition, the radiocarbon age of 41,955–43,292 cal yr BP for the
sample from unit I was near the upper limit of the radiocarbon ages and was thus subject to error. In particular, it was younger
than K-feldspar pIRIR225 (177±7 ka (1803BYG-07-O)) from unit H, making it unlikely that this age was indicative of the
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
12
depositional age of unit I. For ESR dating, 36 fault gouge samples were collected from fault core 2 adjacent to F7, and ESR
dating was performed on 10 of them (1810BYG-01 to 10-E) (Fig. 3). The dates of each sample are presented in Table 4. The
275
weighted average ESR ages of the samples from the same fault viscose band were 245±37 ka (1810BYG-02-E, 1810BYG-10-
E), 406±35 ka (1810BYG-01-E, 1810BYG-05-E, 1810BYG-06-E), 387±26 ka (1810BYG-04-E), and 335±53 ka (1810BYG-
04-E) (Table 4). Samples with dose-saturated ESR signals (1810BYG-03-E, -07-E, and -08-E) were excluded from the
weighted average age calculations. Considering the error, the timing of faulting events using ESR ages at these sites can be
determined to be 245±37 and 406±35 ka. The main fault surface cut unit B and the MRE using the OSL age of unit B was
280
>3.2±0.2 ka (1803BYG-06-O; Table 2, Fig. 3). The geometry and cross-cutting relationship between the Quaternary sediments
and the seven fault surfaces indicated that a pre-existing fault core was reactivated during the Quaternary, resulting in at least
three faulting events (Fig.3; Fig. 9 in Song et al., 2020): the first (antepenultimate earthquake, AE) occurring at <142±4 ka
(1803BYG-12-O), the second (penultimate earthquake, PE) at >17±1 ka (1803BYG-13-O), and the third at the MRE (Table
2, Fig. 3).
285
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
13
Figure 3: (a) Photomosaic of the trench 1 section of the northern wall. The colored circles represent samples for age dating. (b)
Detailed sketch of the trench 1 section. Light grey lines indicate a 1 × 1 m grid.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
14
Table 1. Description of the units in each trench
290
Unit
Trench 1
Trench 2
Trench 3
Trench 4
Trench 5
A
Brown fine sand, lens shape in
unit B, coarsening upward in the
bottom (silt to fine sand)
Reddish brown cobble deposits, A
subangular clast composed of
granitic and sedimentary rocks
with a maximum diameter of 40
cm, poor sorting, charcoal in the
bottom, cover the pre-existing
fault core
Dark brown fine sand-silt
Brown fine sand, have a charcoal
Reddish brown pebble deposits, A
subangular clast composed of
granitic, sedimentary, volcanic
rocks with a maximum diameter of
15 cm, good sorting, matrix-
supported
B
Dark brown cobble, fine sand
matrix, poor roundness, fining
upward in clast (cobble to pebble)
and matrix (sand to fine sand), the
youngest unit cut by fault
Light grey silt-light yellowish
brown fine sand, interbedded two
layers, cut by main fault surface
Light brown boulder-cobble
deposits, colluvial deposits from
mountain slopes, moderate
roundness, decreasing the clast
size toward the west, cover the
main fault surface
Dark brown cobble deposits-sand,
colluvial deposits from mountain
slopes, subangular clast, poor
sorting, the youngest unit cut by
fault
Reddish brown fine sand-slit,
weak horizontal bedding, the
youngest unit cut by fault, no
truncation or deformation after
MRE.
C
Brown medium sand with cobble,
good roundness and sorting
compared to unit b, pinch out in
footwall, cover the pre-existing
fault core
Brown pebble deposits, poor
roundness and sorting, brown
sand matrix, the youngest unit cut
by fault
Light brown boulder deposits-
sand, matrix is coarse sand to
sand, angular clast, poor sorting,
clast mainly composed of granite,
the maximum diameter of a clast
is ~ 120 cm
Reddish brown cobble-pebble
deposits, fine sand matrix, clast
composed of granitic, sedimentary,
volcanic rocks, angular to sub-
angular clast, poor sorting,
containing clasts almost 40%.
Holocene ↑
D
Light brown sand with pebble,
good roundness and sorting
despite adjacent fault, captured by
a triangular shape, with fault
surface
Yellowish-brown pebble deposits,
colluvial wedge, triangular
shaped, angular to subangular
clast, poor sorting in bottom,
fining upward, sand content
increases with distance from the
main fault surface
Brown sand-fine sand, fining
upward
Pleistocene ↓
Light bluish-grey pebble deposits,
light grey fine sand matrix, the
maximum diameter of a clast is ~
20 cm, clast composed of
sedimentary, volcanic rocks,
angular to sub-angular clast, poor
sorting
E
Gray fine sand & brown medium
sand, mixing with gray and brown
parts, SSDS (load structure
dominated; load cats, pillar
structure, sand dike, disturbed
structure, structureless sediments)
Brown cobble-pebble deposits,
subangular clast, poor sorting,
intercalated fine sand to silt
Brown cobble deposits-coarse
sand, alternating sand and gravel,
average diameter of clast is 2-5
cm, subangular to angular,
moderate sorting
Light yellowish brown pebble
deposits, A slit near the fault
surface gradually increases in
grain size as it moves away,
changing to a pebble deposit,
angular to sub-angular clast, good
sorting
F
Grayish-white medium sand
Brown sand-fine sand
Light brown sand-fine sand
G
Gray fine sand with granule,
SSDS (intrusive structure
dominated; ball and pillow, flame
structure, sand dike)
Brown pebble deposits,
subangular clast, poor sorting,
clast composed of granitic,
volcanic, sedimentary rocks
H
Brown gravelly fine sand,
moderate roundness and sorting
Brown sand-fine sand
I
Grayish-white gravelly fine sand,
matrix derived from granite that
basement rock
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
15
Table 2. OSL/pIRIR225 dating results
Trench
number
Dating
method
Sample number
01
02
03
04
05
06
07
08
09
10
11
12
13
14
Trench 1
OSL (ka)
1.3 ± 0.1
9.0 ± 1.0
10 ± 1
4.9 ± 0.3
3.2 ± 0.2
17 ± 1
8.1 ± 0.3
pIRIR225
(ka)
146 ± 8
177 ± 7
151 ± 8
143 ± 7
164 ± 8
155 ± 8
142 ± 7
Target unit
A
B
B
E
B
B
H
G
E
E
G
H
C
B
Trench 2
OSL (ka)
3.2 ± 0.3
3.4 ± 0.4
19 ± 1
pIRIR225
(ka)
Target unit
A
A
A
Trench 3
OSL (ka)
6.4 ± 0.4
pIRIR225
(ka)
173 ± 6
175 ± 5
137 ± 3
Target unit
E
E
D
B
Trench 4
OSL (ka)
1.3 ± 0.1
1.2 ± 0.1
0.15±0.01
0.15±0.01
5.9 ± 0.4
pIRIR225
(ka)
Target unit
B
B
A
A
B
Trench 5
OSL (ka)
10 ± 1
4.8 ± 0.2
2.8 ± 0.1
2.6 ± 0.1
pIRIR225
(ka)
Target unit
C
C
B
B
295
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
16
Table 3. 14C dating results
Sample name
Material dated
13C (‰)
14C age (yr)
Calibrated age
(cal yr BP)a
Probability (%)b
1803BYG-01-C
Charcoal
-25.1
33400±220
38420-36897
95.4
1803BYG-02-C
Charcoal
-27.4
41250±500
45670-43802
95.4
1803BYG-03-C
Sediment
-24.7
6910±30
7821-7675
95.4
1803BYG-04-C
Charcoal
Not analyzed
38560±480
43293-41955
95.4
1810NSR-01-C
Charcoal
-27.8
170±30
0-291
95.4
1810NSR-02-C
Charcoal
-26.3
210±30
0-304
95.4
2009UGR-01-C
Sediment
-24.2
25230±100
29,576-28,966
95.4
2009UGR-02-C
Charcoal
-26.9
160±30
286-0
95.4
2009UGR-03-C
Sediment
-22.5
540±30
634-513
95.4
a Calibration used the database INTCAL13 (Reimer et al., 2013)
b Probability Method (Bronk Ramsey, 2009)
Table 4. ESR dating results
300
Trench
Sample number
01
02
03
04
05
06
07
08
09
10
Byeokgye
421±27
396±24
871±63
1886±319
Trench 1
411±59
245±37
n.d.
387±26
406±35
403±63
n.d.
n.d.
335±53
248±40
Trench 2
790±60
409±41
405±32
n.d.
369±53
330±44
n.d.
n.d.
261±48
Trench 3
425±66
409±52
702±123
n.d.
n.d. = Not Determined
The details of all data are in Kim and Lee (2023)
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
17
4.2.2 Trench 2
Trench 2 was located on the main lineament 1.8 km north of the Byeokgye site (Fig. 2c). A low-angle fault surface
(N02°E/38°SE) cutting the Quaternary deposits was observed in the trench section (Fig. 4). The minimum 6-m-wide fault core
305
of the hanging wall was composed of mature fault rocks. The fault core was divided into fault cores 1 and 2 from the west,
with the yellowish-brown fault core 1 overlaying unit B (Fig. 4). Foliation developed within fault core 1, which abutted the
main fault surface in the upper part of the section. The strikes of the foliation within fault core 1 showed a wide range from
NE to E-W striking in shallow dip parts but were more consistently N-S or NNE striking where the dip increased. The main
fault surface that cut the Quaternary sediments showed a similar change from strike to foliation. The main fault surface had an
310
N-S strike in the lower part of the section and changed to an E-W strike as the dip became shallower. Toward the top of the
section, the dip deepened to regain the N-S striking. This indicated warping of the main fault surface along the pre-existing
structural grains and foliation (Figs. 4b and 4d). Fault core 1, without foliation, was a yellowish-brown fault-brecciated zone
composed of granitic rock breccia and a clayey matrix. The bluish-gray fault core 2 was bounded by a high-angle N-S striking
slip surface with fault core 1. The internal structure of the bluish-gray fault core was divided into four to six alternating gouge
315
and brecciated zones, each of which was bounded by an NNE- or N-S-striking slip surface. Numerous slickenlines (rake ≤ 10˚)
were observed on the slip surfaces in fault core 2, indicating dextral strike-slip movements. The main fault surface cut the
yellowish fault core and unit B and the slickenline revealed a dextral strike-slip with a reverse component. Unit A had a loose
matrix and relatively low consolidation compared to the underlying unit B and had an irregular boundary with the pre-existing
fault core (Table 1). At least one faulting event was identified on the cross-section, which did not reach unit A or the surface
320
(Figs 5c and 5d).
The OSL ages of unit A, which covers the fault, were 3.2±0.3 ka (1810NSR-05) and 3.4±0.4 ka (1810NSR-06) at the southern
wall and 19±1 ka (1810NSR-07) at the northern wall (Table 2). The radiocarbon ages of the charcoal in unit A were 0-291 and
0-304 cal yr BP (Table 3), making them much younger than the OSL age from unit A. Radiocarbon dates do not indicate when
the charcoal was deposited with the sediment but when the tree died after being rooted in the ground. The ESR ages obtained
325
from the fault gouge were higher than the depositional ages of the sediments cut by the fault (Table 4). The ESR ages suggest
that the quartz ESR signal in the fault gouge was not fully initialized during faulting. Therefore, we interpreted the faulting
event to have occurred after the ESR date of 261±48 ka (1810NSR-09-E). Taken together, these results indicate a depositional
age of 3.4±0.4 ka for unit A, which was not cut by the fault, and an ESR age for the fault gouge indicating a faulting event
after 261±48 ka, so the MRE of the fault in Trench 2 was interpreted to be before 3.4±0.4 ka.
330
To estimate the thickness of the Quaternary sediments and the cumulative vertical displacement of the fault, cored sediments
were sampled from the footwall along the main fault surface (Fig. C1). The Quaternary sediments extend to a depth of
approximately 32.8 m, underlain by a granite wash (1.2 m thick) of Paleogene alkali granite, and a subsequent fault damage
zone of the granite exists at its base (Fig. C1). Therefore, the vertical separation caused by Quaternary faulting in Trench 2
was at least 34 m. However, the vertical separation is a paleo-topographic relief difference that may have been caused by the
335
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
18
strike-slip movement of the fault. Cosmogenic 10Be-26Al isochron dating of the granite wash underlying the Quaternary
sediments yielded a burial age of 2,871±593 ka, indicating that the thick Quaternary sediments started to be deposited after
2,871±593 ka (Table 5).
Figure 4: Photomosaic of the trench 2 section of the (a) northern and (b) southern walls. The colored circles represent samples for
340
age dating. Detailed sketch of the trench 2 section of the (c) northern wall and southern wall. White lines indicate a 1 m × 1 m grid.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
19
Table 5. Cosmogenic 10Be-26Al isochron burial dating results
Name
Latitude
Longitude
Elevation
Deptha
Densityb
10Be Conc.
27Al Conc.
26Al/10Bec
Isochron
(˚N, DD)
(˚E, DD)
(m asl)
(m)
(g·cm-3)
(104 atoms g-1)
(104 atoms g-1)
Aged,e (Ma)
NAN001
36.075519
129.256658
104
28
2.0
12.12 ± 0.53
23.39 ± 2.16
1.93
NAN002
36.075519
129.256658
104
28
2.0
2.08 ± 0.08
6.89 ± 0.83
3.31
NAN003
36.075519
129.256658
104
28
2.0
2.41 ± 0.10
10.39 ± 1.08
4.30
2.87 ± 0.59
a Relative depth from the surface.
b Density of unconsolidated sediment.
c Production ratio for the surface is 6.75.
345
d Uncertainties are calculated at the 1σ confidence interval.
e All errors are propagated.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
20
4.2.3 Trench 3
Trench 3 was located on the main lineament extending 3.5 km north of the Byeokgye site (Fig. 2c). Eight Quaternary
sedimentary units and three fault planes were identified in the trench section (Fig. 5, Table 1). The hanging wall of the main
350
fault surface, which cut through Quaternary sediments, was composed of a pre-existing fault core. Excavation revealed a fault
core at least 20 m wide. The pre-existing fault core was divided into gouge and brecciated zones and the brecciated zone was
further divided into black-brown and gray by color. The fault gouge zone that cut the Quaternary sediments was narrower than
5 cm at the bottom of the section and widened to 40 cm at the top of the section; it was divided into an off-white fault zone
and a red fault zone by color and slip surface. The red fault gouge zone was almost entirely composed of clay; however, there
355
were numerous uncrushed quartz and rock fragments within the greyish-white gouge zone. The fault-brecciated zones were
all crushed, with numerous NE-striking veins embedded within the black-brown fault-brecciated zones. The slip surfaces
separating each zone within the fault core were all N-S- to NE-striking. The characteristics of the eight units are shown in
Table 1. Unit D was a colluvial wedge that indicates a paleo-earthquake (Fig. 5, Table 1). Brown sand to fine (units F and H)
and brown gravel (units C, E, and G) deposits were in the trench section. These features can be attributed to environmental
360
factors, such as deposition due to repeated rainfall, flooding, or seismic events due to repeated seismic motion. The slickenline
observed on the main fault surface indicates a dextral strike-slip with a small reverse component. There were at least two
estimated faulting events in this cross-section: event 1, which formed a colluvial wedge, and event 2, which cut the colluvial
wedge (Fig. 5).
The pIRIR225 ages of sample 1903NR1R-02 and 03-O from unit E at the southern wall were 173±6 ka and 175±5 ka,
365
respectively, while the pIRIR225 age of 1903NR1R-08-O from unit D, the colluvial wedge that directly indicated the timing of
the faulting event, revealed that the deposit formed at 137±3 ka (Table 2). Meanwhile, 1903NR1R-10-O from unit B, which
covered the fault, was dated as 6.4±0.4 ka. The youngest ESR age for the fault gouge was 409±52 ka (1903NR1R-02-E, Table
4). However, considering that the quartz ESR signal in the fault zone may not fully reset during faulting, this age implies that
the last faulting event occurred at or after 409±52 ka. The combination of descriptive features and ages suggests a paleo-
370
earthquake that formed the colluvial wedge at 137±3 ka, followed by MRE that cut the wedge between 137±3 and 6.4±0.4 ka.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
21
Figure 5: Photomosaic of the trench 3 section of the (a) northern wall and (b) southern wall. The colored circles represent samples
for age dating. Detailed sketch of the trench 3 section of the (c) northern and southern walls. White lines indicate a 1 × 1 m grid.
Black dot line reveals bedding trace.
375
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
22
4.2.4 Trench 4
Trench 4 was located on an NE-trending branch lineament from the main lineament that extends 6.3 km north of the Byeokgye
site (Fig. 2c). In the trench section, there were five Quaternary sedimentary units cut by one main fault surface (Fig. 6, Table
1). The hanging wall of the main fault surface that cut the Quaternary sediments contained a pre-existing fault core that was at
380
least 5 m wide during excavation. The pre-existing fault core was divided into two zones based on the color and type of fault
rock. Fault core 1 is likely of alkaline granite origin and consisted of a reddish-brown gouge zone and a light-yellow breccia
zone cutting through Quaternary sediments. The fault gouge zone was an equal mixture of gouge and breccia and the breccia
was mostly composed of alkaline granite. In the area adjacent to the Quaternary sediments, a 20-cm-wide fault gouge zone
developed, which was distinguished by yellowish-brown and reddish-brown gouges. The fault breccia zone was dominated by
385
30–50 cm alkaline granite breccias with dark gray to black fault gouges along both edges of the zone. Fault core 2, shown in
gray, was dominated by a mixture of fault gouge and fault breccia. Slip surfaces of the N-S strike were developed within the
fault core. In some sections within the fault core, a 30-cm-wide, dark gray fault gouge zone developed and the boundary of the
fault gouge zone consisted of N-S striking slip surfaces containing slickenlines indicating dextral strike-slip. Units A and B
showed horizontal to sub-horizontal bedding, whereas the bedding of units C-F tilted to the west, with dips of up to 50° adjacent
390
to the main fault surface and becoming shallower toward the west (Fig. 6, Table 1). The difference in bedding indicated an
angular unconformity between unit B and units C–F. The slickenline observed on the main fault surface indicates a dextral
strike-slip with a small reverse component. At least two faulting events were inferred from the cross-section: event 1 (PE),
which caused units C–F to tilt after they were deposited, and MRE, during the deposition of unit B, after an angular
unconformity (Fig. 6).
395
We collected five samples from the northern wall of units A and B and four from units D and F for dating. The OSL age of
5.9±0.4 ka was 2009UGR-09-O, which was cut by the fault. For the remaining four samples that were not cut by the fault, the
oldest OSL age was 1.3±0.1 ka, obtained for 2009UGR-05-O (Table 2). The fault event occurred during the continuous
deposition of unit B. Samples were collected from units F (2009UGR-01-C) and A (2009UGR-02-C) for radiocarbon dating
(Table 3). The radiocarbon age of the charcoal from 2009UGR-02-C was 160±30 cal yr BP, which agreed with the OSL age
400
of 0.15±0.01 ka of the sediment containing the charcoal (2009UGR-07-O) and strongly indicated that unit A was deposited at
this time. The MRE for this trench, derived from comprehensive dating analyses, indicates 1.3±0.1–5.9±0.4 ka.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
23
Figure 6: Photomosaic of the trench 4 section of the (a) northern and (b) southern walls. The colored circles represent samples for
age dating. Detailed sketch of the trench 4 section of the (c) northern and southern walls. Grey lines indicate a 1 × 1 m grid.
405
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
24
4.2.5 Trench 5
Trench 5 was located 40 m north of Trench 4. The trench section contained five quaternary sedimentary units, cut by one main
fault surface (Fig. 7). The overall appearance of the cross-section was similar to that of Trench 4. The hanging wall of the fault
that cut the Quaternary sediments consisted of a pre-existing fault core at least 20 m wide. The pre-existing fault core was
410
derived from alkaline granite, as shown in Fig. 6. The fault core was divided into a fault gouge zone and a fault breccia zone
bounded by an N-S to NNE striking slip surface. The host rock of the light-yellow fault breccia zone was interpreted as alkaline
granite and contained a 5–10 cm wide black fault gouge. Where it abuts the Quaternary sediments, a 10 cm wide light gray
fault gouge developed, which changed to a yellowish grey fault gouge with yellow clay mixed toward the top. A 20 cm-wide
fissure filling was unveiled near the main fault surface and the slip surface within the fault core showed multiple slickenlines,
415
indicative of dextral strike slips. Compared to Trench 4, units A–C in Trench 5 were considered to have the same deposits as
units A and B in Trench 4 and units D and E in Trench 5 were interpreted to have the same deposits as units C–F in Trench 4.
Units D and E in Trench 5 showed westward-dipping bedding like units C–F in Trench 4, whereas units A–C had subhorizontal
bedding. The reddish-brown sediments in the upper part of Trench 5 were thicker than those in Trench 4, indicating that a
more accurate MRE could be identified compared with Trench 4. The slickenline observed on the high-angle main fault surface
420
indicated a dextral strike-slip with a small reverse component. The number of faulting events was estimated at least two based
on the same angular unconformity as in Trench 4. Event 1 caused units D and E tilting, which cut them. Event 2 occurred
during the deposition of unit C, which failed to cut into unit B.
The OSL ages of 2010UGR-03-O and 04-O from unit B on the south slope were 2.8±0.1 and 2.6±0.1 ka, respectively, and
those of 2010UGR-01-O and 02-O from the fault-cut unit C were 10±1 and 4.8±0.2 ka, respectively (Table 2). Trench 5 yielded
425
a tighter MRE range of 2.8±0.1–4.8±0.2 ka than the MRE of Trench 4.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
25
Figure 7: Photomosaic of the trench 5 section of the (a) northern and (b) southern walls. The colored circles represent samples for
age dating. Detailed sketch of the trench 5 section of the (c) northern and southern walls. Grey lines indicate a 1 × 1 m grid.
430
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
26
4.3 Paleo-stress reconstruction
The slickenlines found in the trench were divided into those in the main fault surface that cut the Quaternary sediments and
those in the pre-existing fault core. For the reconstruction of the paleo-stress field, twenty kinematic data along with the
geometry of the fault planes and slickenlines were collected and analyzed using Wintensor S/W (v.5.8.5) (Delvaux & Sperner,
2003). Based on the slickenlines of the main fault surface, the analysis yielded a maximum horizontal stress (σHmax) in the
435
ENE-WNW direction (R´=1.62; Delvaux et al., 1997; Fig. 8), which agrees with the current stress field on the Korean Peninsula
(Kim et al., 2016). The slickenlines in the pre-existing fault core indicated NE-SW compression, which was consistent with
the stress field that causes dextral strike-slip deformation, which is one of the major movements of the Yangsan Fault (Cheon
et al., 2019, 2020a). The reconstructed paleo-stress indicated that the dextral strike-slip with a small reverse component
identified in the main fault surface occurred in an ENE-WSW compressional stress regime.
440
Figure 8: Fault slip data in the fault zone (lower-hemisphere, equal-area projection). Convergent and divergent arrowheads
represent contraction (σHmax) and horizontal stretching (σHmin) directions, respectively. The principal stress axes σ1 (circles), σ2
(triangles), and σ3 (squares) are projected. R´=2-R (σ2 is vertical) [Delvaux et al., 1997; R=(σ2-σ3)/(σ1-σ3)].
4.4 Displacement and earthquake magnitude estimation
445
The results calculated using the marker, vertical separation of each trench, and Eq. (A1) are listed in Table 6. In our previous
study, the true displacement of the MRE at the Dangu site was determined to be 2.64 m (Lee et al., 2016). For each faulting
event in Trench 1, the displacement per event according to the event horizon was 0.33–1.81 m, the true displacement of the
MRE was 0.61–2.81 m, and the cumulative displacement based on the unconformity of the bedrock and Quaternary sediments
was 3.1–8.8 m. Using the bedrock and Quaternary sediments unconformity identified by corings in Trench 2 as a marker, the
450
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
27
cumulative true displacement was 94 m. The MRE cutting the colluvial wedge in Trench 3 had a true displacement of 3.29 m.
However, when considering the overall interpretation, only the MRE and AE, but not the PE, are recognized in Trench 3 (Figs.
5 and 9). The displacement cutting the colluvial wedge likely reflects the displacement of the missing PE as well as the MRE,
which is supported by the long interval between the wedge (unit D) and the deposit covering the wedge (unit B). Thus, it is
reasonable to exclude the calculated displacement as it is unlikely to be the displacement of the MRE. The true displacement
455
of the MRE in Trench 4 and 5 was 0.86 m and 1.17–2.34 m, respectively, using the lower boundary of units B and C as markers.
Combining the results from each trench, the true displacement of MRE in the study area was 0.64–2.81 m and the cumulative
displacement was 3.1–94 m. The displacement per event was similar, between 0.33–1.1 m for PE and AE (event 1, 2), but the
trench showed a higher displacement for the MRE (event 3).
We assumed the maximum earthquake magnitude by applying the MD (0.64–2.81 m) of the MRE. The maximum magnitude
460
of the MRE was estimated to be 6.7–7.2. Furthermore, the surface rupture length (SRL) of 7.6 km obtained from the
topographic analysis (Ha et al., 2022) was applied to the MD-SRL empirical relationship. The calculated MD was 0.21 m,
which was smaller than the true displacement of the MRE (0.64–2.81 m), suggesting that the actual surface rupture length in
the study area exceeds 7.6 km, although this was not confirmed by the current topography. The earthquake magnitude was
estimated from the seismic SSDs in the trench cross-sections (units E and G in Trench 1; unit F in Trench 3). In unit E, the
465
clastic dike varied in size from approximately 20 to 50 cm, whereas in unit G, a ball-and-pillow structure of more than 30 cm
developed (Song et al., 2020; Ha et al., 2022). Atkinson et al. (1984) reported that liquefaction phenomena, including SSDs
above a certain size in sediments of shallow lake or fluvial origin, occur when the minimum earthquake magnitude exceeds
5.5. Based on this, the estimated earthquake magnitude of these SSDs structures may vary depending on the depositional
environment and substrate characteristics. However, it was estimated to be at least 5.5, which is consistent with the magnitude
470
of the inferred empirical relationship.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
28
Table 6. Fault displacement of study area
Dangua
Trench 1b
Trench 2
Trench 3
Trench 4
Trench 5
Sv
(m)
α
(˚)
γ
(˚)
St
(m)
Sv
(m)
α
(˚)
γ
(˚)
St
(m)
Sv
(m)
α
(˚)
γ
(˚)
St
(m)
Sv
(m)
α
(˚)
γ
(˚)
St
(m)
Sv
(m)
α
(˚)
γ
(˚)
St
(m)
Sv
(m)
α
(˚)
γ
(˚)
St
(m)
MRE
0.67
79
15
2.64
0.49
69
17-
56
0.64-
1.81
1.1
42
30
3.29
0.25
86
17
0.86
0.4-
0.8
84
20
1.17-
2.34
Marker
Unit D
Unit C
Unit D
Unit B
Unit C
PE
(event 2)
0.31
75
17-
56
0.39-
1.1
Marker
Unit G
AE
(event 3)
0.22
53
17-
56
0.33-
0.94
Marker
Unit H
Cumulative
displacement
2.4
69
17-
56
3.1-
8.8
34
38
36
94
Marker
Quaternary deposit thickness
amodified from Lee et al., 2015
bmodified from Song et al., 2020
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
29
5 Discussion
475
5.1 Interpretation of Paleoseismic data
5.1.1 MRE and number of surface faulting
We determined the MRE and number of faulting events of the study area based on each trench site, considering previous
studies, the age of Quaternary deposits and fault gouges, the geometry and cross-cutting relationship between faults and
sediments, and the kinematics in the trench section (see the heading 4.1). The results for each trench were synthesized to
480
estimate the MRE, number of earthquakes (faulting events), and timing of earthquakes in the study area (Fig. 9). First, the
MRE was 2.8±0.1–3.2±0.2 ka, which is the time of overlap in several trenches. Considering the error range, the surface rupture
must have occurred approximately 3,000 years ago. Considering the minimum number of earthquakes estimated for each
trench as the maximum, at least three faulting events may have occurred in the study area, as determined from the Dangu site
and Trench 1. The timings of the remaining two prior earthquakes, excluding the MRE, were quantified by combining the age
485
and interpretation of each trench. The penultimate earthquake (PE) occurred in 17±1–75±3 ka, which was determined using
the youngest age of the PE (unit C) from Trench 1 and the MRE from the Byeokgye site (Fig. 9). The antepenultimate
earthquake (AE) was from 137±3–142±4 ka, constrained by the age of the lowermost sediments cut by the fault in Trench 1
and the colluvial wedge in Trench 3. On the other hand, clustering the ESR ages at each trench with an error range suggested
at least three separate earthquakes at 245±37, 404±10, and 817±10 ka (Fig. 9). These ages indicate the time elapsed since the
490
ESR signal was set to zero by the faulting at depth, but it needs much care to use as a timing indicator of the faulting event as
the ESR signal was not completely bleached. Nevertheless, the faulting patterns recognized from clustering in several trenches
indicated that the study area experienced at least three earthquakes in addition to those that cut Quaternary sediments.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
30
Figure 9: Age distribution graph of the study area. The graph reveals the timing of the paleo earthquake based on trench
495
interpretation and age overlap. The most recent earthquake in the study area occurred probably around 3,000 years ago.
5.1.2 Quaternary slip rate and recurrence interval
The slip rate is an expression of the average displacement of a fault over a period, which numerically shows how quickly
energy (stress) accumulates in a fault zone and is used as an important input parameter in seismic hazard assessment (Liu et
500
al., 2021). The slip rate in the study area was calculated based on the age and displacement of each trench. The slip rate was
calculated using the MRE and MRE displacement only, and the cumulative slip rate was calculated using the cumulative
displacement and the oldest age in the trench section and drilling core. The slip rate calculated using the MRE can be considered
maximum because it uses the shortest time for the last earthquake. The cumulative slip rate reflects trends over a relatively
longer period than the MRE; therefore, it may show a slip rate closer to the mean but is likely to be an underestimate. The
505
minimum overall slip rate was 0.02 mm/yr, maximum was 0.38 mm/yr, and average was 0.14 mm/yr (Fig. 10). The most
reliable slip rate of each trench was 0.04–0.11 mm/yr, which is equal to the slip rate of the MRE obtained for Trench 1, where
the second and third of the three earthquakes were identified. The long-time interval and displacement of Trench 3 included
the MRE from the AE so it is reasonable to consider the slip rate of Trench 3 to be a cumulative slip rate (Fig. 10). The slip
rates from the MRE ranged from 0.20–0.35 mm/yr, suggesting a maximum slip rate that can be derived for the study area. The
510
cumulative slip rates derived for each trench indicate a range of 0.02–0.06 mm/yr. The difference in slip rates can be inferred
in two ways. First, slip rates during the Quaternary in the study area averaged 0.02–0.06 mm/yr but may have accelerated in
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
31
the recent Holocene due to changes in the surrounding tectonic setting, such as changes in the thickness of the subducting
plates and increases or decreases in far-field stress, but also due to local factors such as seasonal climate, fluid inflow, and
increased stress accumulation on faults. Second, the discontinuous distribution of Quaternary sediments may have led to an
515
overestimation of the slip rate using the MRE. There are two distinct types of sediments in the trench section: light brown,
relatively coarse-grained sediments of mid-to-late Pleistocene age, which were observed to be tilted in the vicinity of the fault,
and dark brown, relatively coarse-grained, nearly horizontal Holocene sediments (Table 1, Figs. 3-7). The exact absolute time
interval between these two deposits is unknown; however, there was unconformity and the MRE mostly cut Holocene
sediments (<10,000 years). The unconformity in deposition is likely to have missed the earthquakes between the two periods
520
and the MRE cut through younger sediments (Sadler effect; Sadler, 1999), causing the maximum slip rate to be overestimated.
The recurrence interval in the study area was also estimated. Based on the minimum of three earthquakes estimated from the
number of faulting events, the interval between the MRE and PE was 14–72 ka and between the PE and AE was 62–129 ka
(Fig. 9). The recurrence interval inversely calculated using the slip rate (RI=event per displacement/slip rate; Wallace, 1970)
was approximately 9.5 ka using the slip rate of the MRE (0.21 mm/yr) and the event per displacement (average 2 m) and
525
approximately 23 ka using the long-term slip rate (0.031 mm/yr) and the displacement per event (average 0.7 m) of the PE and
AE. Although no clear recurrence interval has been presented in Korean paleoseismological studies, Kim and Lee (2023) used
ESR ages to suggest that the Yangsan Fault follows a quasiperiodic model and has a recurrence interval of approximately 100
ka, which is closely related to the interval of interglacial sea-level loading over the Late Quaternary. Determining the
recurrence interval and earthquake periodicity model of the intraplate is difficult. Earthquakes occur in a regular pattern along
530
the boundary in an interplate; however, in an intraplate, they often occur randomly, depending on the heterogeneous and
complex interior structure (Liu and Stein, 2016). Long recurrence intervals of 400 ka have been reported for intraplate
(Williams et al., 2017); therefore, we can make a conservative estimate that the recurrence interval of the study area is over
10,000 years. However, the recurrence interval may be shorter if the complexity of the fault zone is combined with external
factors, such as increased seismicity on the Korean Peninsula after the Tohoku Earthquake (Hong et al., 2015; 2018) or changes
535
in slip rates.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
32
Figure 10: Slip rate graph of the study area. Blue circles indicate slip rate using displacement and age of MRE. Pink circles represent
the slip rate using cumulative displacement and age. Dash lines show the trend line and average value of each slip rate. The slip rate
in the study area is expected to be between 0.031 and 0.21 mm/yr.
540
5.2 Structural patterns of Quaternary reactivation of the Yangsan Fault
In all trenches of the study area, the hanging wall of the main fault surface maintained a pre-existing fault core. The exposure
of the fault core varied from trench to trench but showed several common features. First, only Holocene sediments were present
in the hanging wall of the fault core, with no Middle Pleistocene sediments observed. This indicates that reverse faulting has
occurred continuously since at least the Middle Pleistocene. Second, NNE to N-S striking slip surfaces with high-angle dips
545
were present within the fault core, and slickenlines developed on these slip surfaces, indicating dextral strike-slip with rakes
of 10° or less. The main fault surface, which cut Quaternary sediments, dictated E-W compression; however, most shear
fractures and slip surfaces in the fault core indicated NE-SW compression. Third, the internal structure of the fault core was
composed of two or more zones delimited by the slip surface, each showing different types of fault rocks and colors depending
on the host rock. Fault rocks of the Cretaceous sedimentary origin were mainly composed of thick, purple-colored fault gouge
550
zones and the captured sandstones were greyish–white lenticular fault breccia or brecciated zones. The Cretaceous volcanic
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
33
rocks mainly comprised dark blue–grey fault breccia zones and fault gouges with a thickness of over 10 cm on both sides of
the fault breccia zone. Fault rocks of the Cretaceous granite origin were mainly observed as white fault breccia zones
characterized by numerous veins and a clayey matrix due to hydrothermal alteration. In the case of alkaline granites, bright
orange fault breccia mainly developed, and a 2–3 cm fault gouge was observed surrounding the lenticular fault breccia. The
555
descriptive features of the fault core internal structure and the occurrence of slickenlines and shear surfaces indicative of NE-
SW compression within the fault core are similar to the reported fault core features of the Yangsan Fault (Cheon et al., 2017,
2019; Kim et al., 2022). Fieldwork and previous studies revealed that the alkaline granite in the study area is a dextral offset
marker of the Yangsan Fault (Hwang et al., 2004, 2007a, b), and vertical cores from the footwall of Trench 2 revealed that the
basement rock is alkaline granite. In addition, Kim et al. (2022) conducted inclined corings and microstructural studies in the
560
vicinity of Trench 1 and identified a fault damage zone, undeformed wall rock, and a fault core approximately 25 m wide
within the alkaline granite of the footwall. Multiple lines of evidence demonstrate that the pre-existing fault core distributed
on the trenches is the main fault core zone of the Yangsan Fault cutting the alkali granite and that the western boundary of the
main fault core was reactivated during the Quaternary. The fault slip surface where the alkali granite contacts the main fault
core suggests that it was in a more slip-prone state (e.g., Low coefficient of friction; Woo et al., 2015) during the Quaternary
565
than other slip surfaces within the fault core. The NE-SW compression shown by the slip surfaces and shear fractures within
the pre-existing fault core is also consistent with a stress field that generates dextral strike-slip movement, which is the major
deformation of the Yangsan Fault (Cheon et al., 2017, 2019). Taken together, these results demonstrate that the western
boundary of the fault core within the Yangsan fault zone has been reactivated as a dextral strike-slip with a small reverse
component since at least the Early Pleistocene, causing surface rupture in the study area.
570
Given that the present-day ENE-WSW stress field acting on the Korean Peninsula has existed since 5 Ma (Kim et al., 2016),
it is reasonable to infer that the study area has been continuously faulted with the same kinematics since the beginning of the
Quaternary. The hanging wall of the main fault surface that cuts the Quaternary sediments is composed of a pre-existing fault
core not only in the study area but also in other Quaternary fault sites along the Yangsan Fault. In all reported Quaternary fault
sites, except for one (Bogyeongsa site) north of the study area, the fault core was distributed in the hanging wall and had a top-
575
to-the west geometry (Kyung, 2003; Ko et al., 2022; Lee et al., 2022; Kim et al., 2023). The top-to-west geometry is also
confirmed topographically. Where the Quaternary faults are located north of the study area, the average elevation is higher,
and the relief is more mountainous on the east side where the hanging wall is located. In the Bogyeongsa section, which does
not have this similarity, the topography is uplifted to the west and is the only Quaternary fault of the Yangsan Fault with a top-
to-the east geometry. In the southern part of the study area, the pre-existing fault core constitutes a hanging wall up to Mihori,
580
Ulju-gun (MH, IBN trench in Kim et al., 2023), located in southern Yangsan Fault (Kim et al., 2023). However, the Quaternary
fault sites south of Mihori show different deformation patterns from those to the north. In the Inbo trench (IB), which is closest
to the IBN trench, fault planes developed between unconsolidated sediments (Cheon et al., 2020a), these features are also
present in other fault sites of the southern Yangsan Fault (Choi et al., 2012). The deformation pattern of the Quaternary faulting
of the northern Yangsan Fault is top to the west, with the main fault core and unconsolidated sedimentary layers abutting the
585
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
34
main fault surface, while the Quaternary faulting of the southern Yangsan Fault is characterized by the development of the
main fault surface between unconsolidated sedimentary layers. The Mihori area, which has been suggested as the boundary
between the central and southern Yangsan faults (Choi et al., 2017), is a point location where the trend of the Yangsan Fault
changes on the surface. The fault-line valley was relatively wide south of Mihori and narrowed as it passed through the Mihori
area. In addition, the distribution of aftershocks was concentrated in this area during the 2016 Gyeongju earthquake and the
590
geometry of surface geological surveys and faults suggests that this area is prone to deformation (Kim et al., 2017). Taken
together, the topographic, structural, seismic, and paleoseismic features of the Mihori area suggest a high probability of large
earthquakes or future earthquakes.
Figure 11: MRE of Yangsan fault. Green stars indicate where MREs have been identified due to the presence of Quaternary sediment
595
ages. GS: Gasan (Lim et al., 2022), IB: Inbo (Cheon et al., 2020a), IBN: Inbonorth, MH: Miho, WS: Wolsan (Kim et al., 2023), BG:
Bangok (Lee, 2023), YG: Yugye (Kyung, 2003), BGS: Bogyeongsa (Lee et al., 2022), GOS: Goesi (Ko et al., 2022), PH: Pyoenghae
(Choi et al., 2012), OG: Ogok (Han et al., 2021). "No obvious proof" refers to areas where sites that cut unconsolidated sediments
are present but not dated.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
35
5.3 MRE and activity for each segment of the Yangsan Fault
600
The MRE of the Yangsan Fault was examined section-by-section by synthesizing previous studies (Fig. 11). First, the MREs
of the Ogok, Pyeonghae, and Goesi sites in the northernmost region of the northern Yangsan Fault were both Late Pleistocene
and Holocene (Choi et al., 2012; Han et al., 2021; Ko et al., 2022). The Yeongdeok area, which extends from the Goesi trench
to the northern part of the Bogyeongsa site, has several reported Quaternary fault sites but no clear evidence of Quaternary
faulting because they either cut unconsolidated sediments with no age constraint or do not cut unconsolidated sediments and
605
have only ESR ages of the fault gouge (Choi et al., 2012). The Bogyeongsa site is between the Middle Pleistocene and Holocene
(154±13–0.9±0.1 ka; Lee et al., 2022) and the MRE of the Yugye site located at the northern Yangsan faults is after Holocene
(AD 646), making it the youngest MRE in the entire Yangsan fault (Kyung, 2003). The MRE of the Bangok site, closest to the
study area, and the study area both occurred after the Holocene (Lee, 2023). The Angang area beyond the study area has no
reports of Quaternary faulting and at the Wolsan site, in the northernmost part of the southern Yangsan Fault, the MRE is Late
610
Pleistocene. The Miho, Inbonorth, and Inbo sites of the southern Yangsan Fault are all Late Pleistocene and the southernmost
part of the Yangsan Fault, the Gasan site, is after the Late Pleistocene (Lim et al., 2021). Looking at the MRE trends by area,
the northern part of the Bogyeongsa site experienced MREs after the Late Pleistocene, the section between Yugye and
Byeokgye experienced surface rupture in the Holocene, and the southern Yangsan fault from the Wolsan site to the Gasan site
experienced MREs mostly in the Late Pliocene. At first glance, it may appear that the section between Yugye and the study
615
area was more active with Holocene MREs but the timing of the MREs alone was inadequate to assess which zone was more
active. For example, Holocene surface rupture may indicate that the fault has a short recurrence interval; conversely, a fault
with an older MRE may have a higher potential for future earthquakes because recent faulting events may have released stress
accumulation. Deviation in the timing of MREs at different sites or sections can be a direct indication of the fault motion that
caused the surface rupture; however, uncertainty in age constraints due to the presence or absence of sediments must be
620
considered. The MRE is an important factor in activity assessment because if it has a recurrence interval, the elapsed time can
be calculated, which can be used to predict future earthquakes. However, the activity considered using MRE alone can be
misleading; therefore, it is necessary to derive recurrence intervals and elapsed time to accurately predict earthquakes based
on paleoseismic data. Ultimately, activity assessment should consider not only paleoseismic data but also topographic,
structural, seismic, and geodetic aspects.
625
6 Conclusion
We conducted surface geological surveys, trenching, coring, OSL/IRSL dating, radiocarbon dating, ESR dating, and
cosmogenic nuclides dating along the lineament extracted from topographic analysis using high-resolution LiDAR imagery at
the Yangsan Fault. The primary aim was to identify the surface ruptures that occurred during the Quaternary period, derive
structural features, and gather paleoseismic data for the Quaternary fault. The main conclusions are as follows:
630
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
36
(1) A comprehensive analysis, combining two previously reported Quaternary fault sites (Byeokgye and Dangu), two newly
discovered natural outcrop sites, five trenches, coring surveys, and dating of unconsolidated sediments, charcoal, and fault
rocks, revealed a 7.6 km Quaternary surface rupture along the Yangsan Fault. (2) The Quaternary faults in the study area
exhibited an east-dipping geometry with an N-S to NNE strike and moved as a dextral strike-slip with a reverse slip component
during the Quaternary. (3) At least three faulting events were interpreted from the trench sections, with an MRE of
635
approximately 3,000 years ago. The MRE displacement ranged from 0.64 to 3.29 m, cumulative displacement from 3.1 to 94
m, displacement per event before the MRE from 0.33 to 1.1 m, and maximum magnitude using the MD from Mw 6.7–7.2. The
slip rate was 0.03–0.21 mm/yr, with a recurrence interval of at least 10,000 years. (4) The Quaternary structural features
indicate that during the Quaternary period, the western boundary of the main fault core of the Yangsan Fault was reactivated
under ENE-WSW compressional stress. The top-to-west geometry is consistent with that observed throughout the Yangsan
640
Fault. (5) The southern part of the northern Yangsan Fault, including the study area, documents a Holocene MRE that is
younger than those documented in the southern Yangsan Faults.
Surface rupturing has been observed along most sections of the Yangsan fault and in the study area, and basic paleoseismic
studies continue to be reported (Fig. 11; Lee et al., 2022; Kim et al., 2023). The challenge is that Korea's geologic diversity
results in a large number of faults that need to be studied (e.g., Chugaryeong Fault). Only then will it be possible to properly
645
understand the seismicity pattern of Korea and the earthquakes in the low active intraplate regions. Although the study results
are only for a small 10 km section of the Yangsan fault with a 200 km extension, the San Andreas fault, which has high-
resolution paleoseismic data from more than 200 trenches, also started out in the same way. In particular, studying seismic
hazards on the Korean Peninsula, with its urbanization, cultivated land, slow slip rates, long recurrence intervals, and fast
erosion rates, is a challenge. This research will hopefully contribute to the research on earthquake hazards in Korea, which is
650
just beginning to advance. Furthermore, this study contributes valuable insights for seismic hazard assessment in the region
and offers a broader understanding of intraplate earthquake dynamics, thereby aiding earthquake prediction efforts.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
37
Appendices
Appendix A. Calculation of true displacement
655
The true displacement (St) can be calculated using a trigonometric function that considers the vertical displacement (Sv), fault
dip angle (α), rake (γ), and their relationships (Fig. A1; Eq. A1). Assume that the attitude of the marker in the cross-section at
each trench is nearly horizontal in three dimensions and the angle (β) of the cross-section is nearly vertical, then the two factors
are perfectly horizontal and vertical, respectively. Thus, the vertical separation (Svm) and vertical displacement (Sv) measured
in the cross-section are equal.
660
Therefore,
𝑺𝒗𝒎 = 𝑺𝒗, 𝑺𝒎=𝑺𝒗
𝐬𝐢𝐧 𝜶 , 𝑺𝒕=𝑺𝒎
𝐬𝐢𝐧 𝜸 (A1)
We calculate true displacement (St) using Eq. (A1) for vertical separation (Svm) of the marker measured in the trench section,
as shown in Table 6.
665
Figure A1: (a) Schematic diagram showing how to calculate true displacement. St: true displacement, Sv: vertical displacement, Sm:
dip separation, α: dip of fault surface, β: dip of cut slope, γ: rake of the striation (modified from Xu et al., 2009). (b and c) Photographs
showing the measured vertical separation of the trenches 1 and 5. Svm: vertical separation.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
38
Appendix B. Description of the Outcrop 1 and 2
670
Outcrop 1 is located 30 m south of the Byeokgye site (Fig. 2c). A high-angle fault cut the Cretaceous volcanic rocks and
unconsolidated sediments interbedded with gravel deposits and sand layers (Fig. B1a). The fault trace was partially recognized
on the surface based on the lithological distribution. The fault core consisted of a 20 cm-wide fault breccia with a fault surface
attitude of N12°E/84°NW, and no clear slickening was observed (Fig. B1b). Outcrop 2, on the extension of the main lineament,
is located 2.7 km north of the Byeokgye site (Fig. 2c). The outcrop contained highly weathered purple sedimentary rock
675
overlying unconsolidated sediments (Fig. B1c). The NNE-striking fault surface dipped toward the east with a low angle. The
purple sedimentary rock of the hanging wall was mostly clayey and may be either a pre-existing purple fault gouge or
weathered purple Cretaceous sedimentary rock. The fault core adjacent to the unconsolidated deposits consisted of a 2–3 cm
thick foliated fault gouge (Fig. B1d). The foliation is subparallel to the slip surface. The unconsolidated gravel deposits have
poor sorting, and angular clasts consist of sandstone, granitic, and volcanic rocks. No clear slickenlines were observed, owing
680
to highly weathered outcrop conditions. We could not date either outcrop because the clast-rich sediments were not suitable
for OSL dating and did not include organic material for 14C dating.
Figure B1: Photographs showing the outcrops of the fault which that cut the unconsolidated sediments. (a) Fault trace on the surface.
(b) NNE striking fault cut the boundary between volcanic rocks and gravel and sand deposits. The fault rock is composed of breccia.
685
(c) Weathered pre-existing fault rocks thrust gravel deposits along the low-angle fault surface. (d) Close-up photograph of the fault
surface. A 2-3 cm thick fault gouge forms a boundary between two different rocks.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
39
Appendix C. Coring at the Trench 2
We cored the footwall of Trench 2. Unconsolidated sand and gravel deposits alternated up to 24 m below the surface. From
690
24 m to 32.8 m, the material consisted of a granite wash, at the base of which a 1.2 m thick weathering zone was recognized.
From 34 m onwards, the bedrock is clearly identified as alkaline granite (Fig. C1). We performed 10Be-26Al isochron burial
dating on the granite wash immediately above the basement rock.
Figure C1: Photograph showing drilled core. The lowermost of the unconsolidated sediments, the granite wash, directly overlies
695
bedrock. Cosmogenic 10Be-26Al isochron dating samples were taken from this granite wash.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
40
Appendix D. Detailed OSL, IRSL, and 10Be-26Al isochron burial age dating results
The graphs of the detailed results of 1-2 representative samples per trench for OSL and IRSL. We selected samples from
sediments cut into the MRE and from sediments overlying the MRE. The graphs include decay curves, probability density
700
graphs, and 10Be-26Al isochron graphs. The tables also include dose rates, equivalent doses and luminescence ages of the
samples from each trench.
Figure D1: Representative quartz OSL decay curves (a, b) and probability density graph (c, d) of Trench 1.
705
Figure D2: Representative quartz OSL decay curves (a, b) and probability density graph (c, d) of Trench 2.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
41
Figure D3: Representative quartz OSL/IRSL decay curves (a, b) and probability density graph (c, d) of Trench 3.
710
Figure D4: Representative quartz OSL decay curves (a, b) and probability density graph (c, d) of Trench 4.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
42
Figure D5: Representative quartz OSL decay curves (a, b) and probability density graph (c, d) of Trench 5.
715
Figure D6: 10Be-26Al isochron graph of coring samples in Trench 2.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
43
Table D1: Dose rates, equivalent doses and luminescence ages of the samples from the Trench 1
Method
Sample code
Dose ratea
(Gy/ka)
Water contentb
(%)
Equivalent
dose
(Gy)
Fading rate
(g2days, %/dec.)
Aliquots used
(n)
agec
(ka, 1σ SE)
Quartz
OSL
1803BYG-01-O
3.26 ± 0.08
13
4.2 ± 0.2
-
24
1.3 ± 0.1
1803BYG-02-O
3.54 ± 0.09
16
33 ± 1
-
24
9 ± 1
1803BYG-03-O
3.00 ± 0.08
17
31 ± 1
-
24
10 ± 1
1803BYG-05-O
3.16 ± 0.08
14
16 ± 1
-
24
4.9 ± 0.3
1803BYG-06-O
3.37 ± 0.09
12
11 ± 1
-
24
3.2 ± 0.2
1803BYG-13-O
4.02 ± 0.11
13
68 ± 4
-
24
17 ± 1
1803BYG-14-O
3.31 ± 0.09
10
27 ± 1
-
24
8.1 ± 0.3
K-feldspar
pIRIR225
1803BYG-04-O
3.90 ± 0.09
18
537 ± 23
0.8 ± 0.3
12
146 ± 8
1803BYG-07-O
4.36 ± 0.10
16
733 ± 31
0.7 ± 0.2
12
177 ± 7
1803BYG-08-O
3.99 ± 0.09
18
558 ± 18
1.1 ± 0.3
24
151 ± 8
1803BYG-09-O
4.21 ± 0.10
22
561 ± 21
1.0 ± 0.3
12
143 ± 7
1803BYG-10-O
3.95 ± 0.10
20
581 ± 20
1.5 ± 0.3
12
164 ± 8
1803BYG-11-O
4.12 ± 0.09
18
594 ± 19
1.0 ± 0.4
12
155 ± 8
1803BYG-12-O
4.63 ± 0.11
15
613 ± 21
1.0 ± 0.3
12
142 ± 7
a Wet total dose rates. The activities of each radionuclide were converted to dose rates using the data presented by Liritzis et al.
720
(2013). Cosmic ray dose rates were estimated using the method suggested by Prescott and Hutton (1994). The internal dose rate of
K-feldspar was calculated assuming internal K content of 12.5 % and by using the methods proposed by Mejdahl (1987) and
Readhead (2002). For deriving the total dose rates absorbed by K-feldspar, the internal dose rate of 0.736±0.04 Gy/ka was added to
the external ones of each sample.
b Present water content.
725
c Fading-corrected ages in the case of K-feldspar
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
44
Table D2: Dose rates, equivalent doses and luminescence ages of the samples from the Trench 2
Method
Sample code
Dose ratea
(Gy/ka)
Water contentb
(%)
Equivalent dose
(Gy)
Aliquots used
(n)
agec
(ka, 1σ SE)
Quartz
OSL
1810NSR-05-O
3.09 ± 0.08
18
9.9 ± 0.8
23
3.2 ± 0.3
1810NSR-06-O
3.46 ± 0.09
14
12 ± 1
22
3.4 ± 0.4
1810NSR-07-O
2.55 ± 0.06
19
47 ± 2
24
19 ± 1
a Wet total dose rates. The activities of each radionuclide were converted to dose rates using the data presented by Liritzis et al.
(2013). Cosmic ray dose rates were estimated using the method suggested by Prescott and Hutton (1994).
730
b Present water content.
Table D3: Dose rates, equivalent doses and luminescence ages of the samples from the Trench 3
Method
Sample code
Dose ratea
(Gy/ka)
Water contentb
(%)
Equivalent
dose
(Gy)
Fading rate
(g2days, %/dec.)
Aliquots used
(n)
agec
(ka, 1σ SE)
Quartz
OSL
1903NS1R-10-O
3.67 ± 0.09
23
23 ± 1
-
24
6.4 ± 0.4
K-feldspar
pIRIR225
1903NS1R-02-O
4.01 ± 0.09
18
606 ± 22
1.8 ± 0.6
12
173 ± 11
1903NS1R-03-O
4.11 ± 0.09
16
600 ± 16
2.4 ± 0.4
12
175 ± 9
1903NS1R-08-O
4.55 ± 0.10
15
595 ± 18
0.6 ± 0.3
12
137 ± 6
a Wet total dose rates. The activities of each radionuclide were converted to dose rates using the data presented by Liritzis et al.
(2013). Cosmic ray dose rates were estimated using the method suggested by Prescott and Hutton (1994). The internal dose rate of
735
K-feldspar was calculated assuming internal K content of 12.5 % and by using the methods proposed by Mejdahl (1987) and
Readhead (2002). For deriving the total dose rates absorbed by K-feldspar, the internal dose rate of 0.736±0.04 Gy/ka was added to
the external ones of each sample.
b Present water content.
c Fading-corrected ages in the case of K-feldspar
740
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
45
Table D4: Dose rates, equivalent doses and luminescence ages of the samples from the Trench 4
Method
Sample code
Dose ratea
(Gy/ka)
Water contentb
(%)
Equivalent dose
(Gy)
Aliquots used
(n)
agec
(ka, 1σ SE)
Quartz
OSL
2009UGR-05-O
3.11 ± 0.08
12
3.9 ± 0.1
24
1.3 ± 0.1
2009UGR-06-O
2.97 ± 0.08
11
3.6 ± 0.1
24
1.2 ± 0.1
2009UGR-07-O
3.44 ± 0.09
12
0.5 ± 0.1
24
0.15 ± 0.03
2009UGR-08-O
3.41 ± 0.09
13
0.5 ± 0.1
24
0.15 ± 0.03
2009UGR-09-O
2.75 ± 0.07
12
16.1 ± 0.9
24
5.9 ± 0.4
a Wet total dose rates. The activities of each radionuclide were converted to dose rates using the data presented by Liritzis et al.
(2013). Cosmic ray dose rates were estimated using the method suggested by Prescott and Hutton (1994).
745
b Present water content.
Table D5: Dose rates, equivalent doses and luminescence ages of the samples from the Trench 5
Method
Sample code
Dose ratea
(Gy/ka)
Water contentb
(%)
Equivalent dose
(Gy)
Aliquots used
(n)
agec
(ka, 1σ SE)
Quartz
OSL
2010UGR-01-O
3.19 ± 0.08
14
32 ± 3
23
10 ± 1
2010UGR-02-O
3.47 ± 0.09
14
17 ±1
24
4.8 ± 0.2
2010UGR-03-O
2.77 ± 0.07
21
7.9 ± 0.3
24
2.8 ± 0.1
2010UGR-04-O
2.96 ± 0.07
17
7.6 ± 0.2
24
2.6 ± 0.1
a Wet total dose rates. The activities of each radionuclide were converted to dose rates using the data presented by Liritzis et al.
(2013). Cosmic ray dose rates were estimated using the method suggested by Prescott and Hutton (1994).
750
b Present water content.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
46
Data availability. All detailed age dating data are available in Tables 2-5, D1-D5. Questions or requests for original
photomosaic of trenches can be sent to the corresponding author.
Author contributions. Conceptualization: SH, HCK, Funding acquisition: SM, Investigation: SM, HCK, SL, YBS, JHC, SJK,
755
Project administration: SM, Supervision: SM, Validation: YBS, JGC, SJK, Visualization: SM, SJK, Writing – original draft
preparation: SH, Writing – review & editing: SH, YBS, JHC, SM,
Competing interests. The authors declare that they have no conflicts of interest.
760
Financial support.
This research was supported by a grant (2022-MOIS62-001(RS-2022-ND640011)) of National Disaster Risk Analysis and
Management Technology in Earthquake funded by Ministry of Interior and Safety (MOIS, Korea).
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
47
References
765
Ansari, K. and Bae, T.S.: Contemporary deformation and strain analysis in South Korea based on long-term (2000–2018)
GNSS measurements, Int. J. Earth Sci. (Geol. Rundsch.), 109, 391–405, https://doi-org-ssl.oca.korea.ac.kr/10.1007/s00531-
019-01809-4, 2020.
Argus, D.F., Gordon, R.G., Heflin, M.B., Ma, C., Eanes, R.J., Willis, P., Peltier, W.R., and Owen, S.E.: The angular velocities
of the plates and the velocity of the Earth’s centre from space geodesy, Geophys. J. Int., 18, 1-48, https://doi:10.1111/j.1365-
770
246X.2009.04463.x, 2010.
Ashurkov, S.V., San'kov, V.A., Miroschnichenko, A.I., Lukhnev, A.V., Sorokin, A.P., Serov, M.A., and Byzov, L.M.: GPS
geodetic constrains on the kinematics of the Amurian Plate, Russ. Geol. Geophys., 52, 239-249,
https://doi.org/10.1016/j.rgg.2010.12.0172, 2011.
Atkinson, G.M., Finn, W.D.L. and Charlwood, R.G.: Simple computation of liquefaction probability for seismic hazard
775
applications, Earthq. Spectra, 1, 107-123, https://doi.org/10.1193/1.1585259, 1984.
Balco, G., and Rovey, C.W.: An isochron method for cosmogenic-nuclide dating of buried soils and sediments. Am. J. Sci.,
308, 1083–1114, https://doi.org/10.2475/1 0.2008.02, 2008.
Bird, P.: An updated digital model of plate boundaries, Geochem. Geophy. Geosy., 4,
1027, https://doi.org/10.1029/2001GC000252, 2003.
780
Bonilla, M. G., Mark, R. K., and Lienkaemper, J. J.: Statistical relations among earthquake magnitude, surface rupture length,
and surface fault displacement, B. Seismol. Soc. Am., 74, 2379–2411, https://doi.org/10.1785/BSSA0740062379,1984.
Bronk Ramsey, C.: Methods for summarizing radiocarbon datasets, Radiocarbon, 59, 1809-1833,
https://doi.org/10.1017/RDC.2017.108, 2017.
Buylaert, J.P., Jain, M., Murray, A.S., Thomsen, K.J., Thiel, C., and Sohbati, R.: A robust feldspar luminescence dating method
785
for Middle and Late Pleistocene sediments, Boreas, 41, 435–451, https://doi.org/10.1111/j.1502-3885.2012.00248.x, 2012.
Buylaert, J.P., Murray, A.S., Thomsen, K.J., and Jain, M.: Testing the potential of an elevated temperature IRSL signal from
K-feldspar, Radiat. Meas., 44, 560–565, https://doi.org/10.1016/j.radmeas.2009.02.007, 2009.
Calais, E., Dong, L., Wang, M., Shen, Z., and Vergnolle, M.: Continental deformation in Asia from a combined GPS solution,
Geophys. Res. Lett., 33, L24319, https://doi.org/ 10.1029/2006GL028433, 2006.
790
Chang, C.-J., and Chang, T.W.: Movement History of the Yangsan Fault based on Paleostress Analysis, J. Eng. Geo., 8, 35-
49, 1998.
Chang, C.-J.: Structural characteristics and evolution of the Yangsan fault, SE Korea, Ph.D. thesis, Kyungpook National
University, Korea, 2002.
Chang, K.H., Woo, B.G., Lee, J.H., Park, S.O., and Yao, A.: Cretaceous and early Cenozoic stratigraphy and history of eastern
795
Kyongsang Basin, S. Korea, J. Geol. Soc. Korea, 26, 471-487, 1990.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
48
Cheon, Y., Cho, H., Ha., S., Kang, H.-C., Kim, J.-S., and Son, M.: Tectonically controlled multiple stages of deformation
along the Yangsan Fault Zone, SE Korea, since Late Cretaceous, J. Asian Earth Sci., 170, 188-207,
https://doi.org/10.1016/j.jseaes.2018.11.003, 2019.
Cheon, Y., Choi, J.-H., Choi, Y., Bae, H., Han, K.-H., Son, M., Choi, S.-J., and Ryoo, C.-R.: Understanding the distribution
800
and internal structure of the main core of the Yangsan Fault Zone: Current trends and future work, J. Geol. Soc. Korea, 56, 61-
640, https://doi.org//10.14770/jgsk.2020.56.5.619, 2020a.
Cheon, Y., Choi, J.-H., Kim, N., Lee, H., Choi, I., Bae, H., Rockwell, T.K., Lee, S.R., Ryoo, C.-R., and Choi, H.: Late
Quaternary transpressional earthquakes on a long-lived intraplate fault: A case study of the Southern Yangsan Fault, SE Korea,
Quatern. Int., 553, 132-143, https://doi.org/10.1016/j.quaint.2020.07.025, 2020b.
805
Cheon, Y., Ha, S., Lee, S., Cho, H, and Son, M.: Deformation features and history of the Yangsan Fault Zone in the Eonyang-
Gyeongju area, SE Korea, J. Geol. Soc. Korea, 53, 95-114, http://doi.org/10.14770/jgsk.2017.53.1.95, 2017.
Choi, J.-H., Kim, Y.-S., and Klinger, Y.: Recent progress in studies on the characteristics of surface rupture associated with
large earthquakes, J. Geol. Soc. Korea, 53, 129-157, https://doi.org/10.14770/jgsk.2017.53.1.129, 2017.
Choi, J.-H., Yang, S.-J., and Kim, Y.-S.: Fault zone classification and structural characteristics of the southern Yangsan fault
810
in the Sangcheon-ri area, SE Korea, J. Geol. Soc. Korea, 45, 9-28, 2009.
Choi, S.-J., Ghim, Y.S., Cheon, Y., and Ko, K.: The First Discovery of Quaternary Fault in the Western Part of the South
Yangsan Fault-Sinwoo Site, Econ. Environ. Geol., 52, 251-258, http://doi.org/10.9719/EEG.2019.52.3.251, 2019.
Choi, S.-J., Jeon, J.S., Song, G.Y., Kim, H.C., Kim, Y.H., Choi, B.Y., Chwae, W.C., Han, J.G., Ryoo, C.-R., Sun, C.-G., Jun,
M.-S., Kim, G.-Y., Kim, Y. B., Lee, H. J., Shin, J. S., Lee, Y. S., and Kee, W.-S.: Active fault map and seismic hazard map,
815
National Emergency Management Agency, Seoul, Seoul, 939 pp., 2012.
Crampin, S. and Gao, Y.: Earthquakes can be stress-forecast, Geophy. J. Int., 180, 1124-1127, https://doi.org/10.1111/j.1365-
246X.2009.04475.x, 2010.
Delvaux, D., and Sperner, B.: New aspects of tectonic stress inversion with reference to the TENSOR program, in: New
Insights into Structural Interpretation and Modelling, edited by Nieuwland, D. A., Geol. Soc., London Spec. Publ., 212, 75-
820
100, https://doi.org/10.1144/GSL.SP.2003.212.01.06, 2003.
Delvaux, D., Moeys, R., Stapel, G., Petit, C., Levi, K., Miroshnichenko, A., Ruzhich, V., and San'kov, V.: Paleostress
reconstructions and geodynamics of the Baikal region, Central Asia, Part 2. Cenozoic rifting, Tectonophysics, 282, 1-38,
https://doi.org/10.1016/S0040-1951(97)00210-2, 1997.
DeMets, C., Gordon, R.G., Argus, D.F., and Stein, S.: Current plate motions, Geophys. J. Int., 101, 425-478,
825
https://doi.org/10.1111/j.1365-246X.1990.tb06579.x, 1990.
DeMets, C., Gordon, R.G., Argus, D.F., and Stein, S.: Effect of recent revisions to the geomagnetic reversal time scale on
estimates of current plate motions, Geophys. Res. Lett., 21, 191-2,194, https://doi.org/10.1029/94GL02118, 1994.
England, P., and Jackson, J.: Uncharted seismic risk, Nat. Geosci., 4, 348–349, https://doi.org/10.1038/ngeo1168, 2011.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
49
Erlanger, E.D., Granger, D.E., and Gibbon, R.J.: Rock uplift rates in South Africa from isochron burial dating of fluvial and
830
marine terraces, Geology, 40, 1019–1022. https://doi.org/10.1130/G33172.1., 2012.
Geller, R. J., Jackson, D. D., Kagan, Y. Y., and Mulargia, F.: Earthquakes Cannot Be Predicted, Science, 275, 1616-1616,
https://doi.org/10.1126/science.275.5306.1616, 1997.
Granger, D. E.: A review of burial dating methods using 26Al and 10Be, in: In Situ-Produced Cosmogenic Nuclides and
Quantification of Geological Processes: Geological Society of America Special Paper 415, edited by: Siame, L. L., Bourlès,
835
D. L., and Brown, 1–16, https://doi.org/10.1130/2006.2415(01), 2006.
Granger, D.E., and Muzikar, P.F.: Dating sediment burial with in situ-produced cosmogenic nuclides: Theory, techniques, and
limitations, Earth Planet. Sc. Lett., 188, 269-281, https://doi.org/10.1016/S0012-821X(01)00309-0, 2001.
Gurpinar, A.: The importance of paleoseismology in seismic hazard studies for critical facilities, Tectonophysics, 408, 23-28,
https://doi.org/10.1016/j.tecto.2005.05.042, 2005.
840
Gwon, O., Park, K., Naik, S. P., Shin, H.-C., and Kim, Y. -S.: A study on the characteristics of fault activity in the southern
part of the Ulsan fault using paleoseismic method, J. Geol. Soc. Korea, 57, 109-121,
https://doi,org/10.14770/jgsk.2021.57.2.109, 2021.
Gwon, S., Cho, S., Park, D., Choi, W.-H., Nakao, R., Tanaka, T., and Inoue, D.: Microscopic analysis of fault rock using X-
ray computed tomography from the Wolsan trench in the middle part of Yangsan fault, SE Korea, J. Geol. Soc. Korea, 56,
845
273-284, https://doi.org/10.14770/jgsk.2020.56.2.273, 2020.
Ha, S., Son, M., and Seong, Y.B.: Active Fault Trace Identification Using a LiDAR High-Resolution DEM: A Case Study of
the Central Yangsan Fault, Korea, Remote Sens., 14, 4838, https://doi.org/10.3390/rs14194838, 2022.
Han, J.-W., Lee, S., Ha, S., Lee, C. H., Seong, Y. B., Kang, H.-C., Kim, M.-C., Gil, T., and Son, M.: A preliminary study on
the Quaternary faulting along the northern part of Yangsan Fault and Yeongdeok Fault, in: 76th Annual Meeting of the
850
Geological Society of Korea and 2021 Fall Joint Conference of the Geological Sciences, Jeju, Korea, 26-28 October 2021,
317, 2021.
Han, M., Kim, K.-H., Son, M., Kang, S.Y., and Park, J.-H.: Location of recent micro earthquakes in the Gyeongju area,
Geophysics Geophys. Explor., 19, 97-104, https://doi.org/10.7582/GGE.2016.19.2.097, 2016.
Hansen, V., Murray, A., Buylaert, J.-P., Yeo, E.Y., and Thomsen, K.: A new irradiated quartz for beta source calibration,
855
Radiat. Meas., 81, 123–127, https://doi.org/10.1016/j.radmeas.2015.02.017, 2015.
Heki, K., Miyazaki, S., Takahashi, H., Kasahara, M., Kimata, F., Miura, S., Vasilenko, N.F., Ivashchenko, A., and An, K.-D.:
The Amurian Plate motion and current plate kinematics in eastern Asia, J. Geophys. Res., 104, 29, 147-29, 155,
https://doi.org/10.1029/1999JB900295, 1999.
Hong, T. K., Lee, J., and Houng, S. E.: Long-term evolution of intraplate seismicity in stress shadows after a megathrust, Phys.
860
Earth Planet. In., 245, 59–70, https://doi. org/10.1016/j.pepi.2015.05.009, 2015.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
50
Hong, T. K., Lee, J., Park, S., and Kim, W.: Time-advanced occurrence of moderate-size earthquakes in a stable intraplate
region after a megathrust earthquake and their seismic properties, Sci. Rep., 8, 13331, https://doi.org/10.1038/s41598-018-
31600- 5, 2018.
Hong, Y., OH J.-S., Hong S.-C., and Shin J.: Geomorphological Development and Fault Activity of the Central-Southern
865
Yangsan Fault (I): Developmental Characteristics and Distribution of the Quaternary Landforms, J. Korean Geomorphological
Assoc., 28, 67-81. https://doi.org/10.16968/JKGA.28.1.67, 2021.
Hough, S., and Page, M.: Toward a consistent model for strain accrual and release for the New Madrid seismic zone, central
United States, J. Geophys. Res., 116, B03311. http://dx.doi.org/10.1029/2010JB007783, 2011.
Hough, S.E., Armbruster, J.G., Seeber, L., and Hough, J.F.: On the modified Mercalli intensities and magnitudes of the 1811–
870
1812 New Madrid earthquakes, J. Geophys. Res., 105(B10), 23,839–23,864, https://doi.org/10.1029/2000JB900110, 2000.
Hwang, B.-H., Lee, J. D., and Yang, K.: Petrological study of the granitic rocks around the Yangsan Fault: Lateral displacement
of the Yangsan Fault, J. Geol. Soc. Korea, 40, 161–178, 2004.
Hwang, B. H., Lee, J. D., Yang, K., and McWilliams, M.: Cenozoic strike-slip displacement along the Yangsan Fault, southeast
Korean Peninsula, Int. Geol. Rev., 49, 768–775, https://doi.org/10.2747/0020-6814.49.8.768, 2007a.
875
Hwang, B. H., McWilliams, M., Son, M., and Yang, K.: Tectonic implication of A-type granites across the Yangsan Fault,
Gigye and Gyeongju areas, southeast Korean Peninsula, Int. Geol. Rev., 49, 1094–1102, https://doi.org/10.2747/0020-
6814.49.12.1094, 2007b.
Ikeya, M., Miki, T., and Tanaka, K.: Dating of a Fault by Electron Spin Resonance on Intrafault Materials, Science, 215, 1392-
1393, https://doi.org/10.1126/science.215.4538.1392, 1982.
880
Jin, K., Kim, Y.-S., Kang, H.C. and Shin, H.C.: Study on developing characteristics of the Quaternary Gusan Fault in Uljin,
Gyeongbuk, Korea, J. Geol. Soc. Korea, 49, 197-207, https://doi.org/10.14770/jgsk.2013.49.2.197, 2013.
Johnston, A.C., and Schweig, E.S.: The enigma of the New Madrid earthquakes of 1811–1812, Annu. Rev. Earth Pl. Sc., 24,
339-384, https://doi.org/10.1146/annurev.earth.24.1.339, 1996.
Kanamori, H.: The energy release in great earthquakes, J. Geophys. Res., 82, 2981–2987,
885
https://doi.org.10.1029/JB082i020p02981, 1977.
Kang, H.-C., Cheon, Y., Ha, S., Seo, K., Kim, J.-S., Shin, H. C., and Son, M.: Geology and U-Pb Age in the Eastern Part of
Yeongdeok-gun, Gyeongsangbuk-do, Korea, J. Petrological Soc. Korea, 27, 153-171,
https://doi.org/10.7854/JPSK.2018.27.3.15, 2018.
Kee, W.-S., Kim, Y.-H., Lee, H.-J., Choi, D.-L., Kim, B.-C., Song, K.-Y., Koh, H.-J., Lee, S.R., Yeon, Y.G., Hwang, S.-H.,
890
Sung, N.-H.: South eastern fault variable research and DB Construction, Korea Institute of Geoscience and Mineral Resources,
Daejeon, 327 pp., 2009.
Khromovskikh, V. S.: Determination of magnitudes of ancient earthquakes from dimensions of observed
seismodislocations, Tectonophysics, 166, 269-280, https://doi.org/10.1016/0040-1951(89)90219-9, 1989.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
51
Kim, C.-M., Cheon, Y., Lee, T.-H., Choi, J.-H., Ha, S., and Jeong, J. O.: Long-Term Weakening Processes and Short-Term
895
Seismic Slip Behavior of an Intraplate Mature Fault Zone: A Case Study of the Yangsan Fault, SE Korea, J. Geophys. Res-
Sol Ea, 127, e2021JB023154, https://doi.org/10.1029/2021JB023154, 2022.
Kim, C.-M., Ha, S., and Son, M.: Evidence of coseismic slip recorded by Quaternary fault materials and microstructures,
Naengsuri, Pohang, J. Geol. Soc. Korea, 56, 175-192, https://doi.org/10.14770/jgsk.2020.56.2.175, 2020a.
Kim, C.-M., Jeong, J. O., Gu, D., and Han, H.: Identification of materials in principal slip zones of faults by X-ray diffraction
900
analysis using a small amount of sample, J. Geol. Soc. Korea, 53, 873-883, http://doi.org/10.14770/jgsk.2017.53.6.873, 2017a.
Kim, D.-E., and Seong. Y.B.: Cumulative Slip Rate of the Southern Yangsan Fault from Geomorphic Indicator and Numerical
Dating, J. Kor. Geogr. Soc., 56, 201-213. https://doi.org/10.22776/KGS.2021.56.2.201, 2021.
Kim, D. E., and Oh. J.-S.: Landform Classification Using Geomorphons on the Middle Yangsan Fault System, Southeastern
Korea, J. Korean Geogr. Soc., 54, 493-505, 2019.
905
Kim, D.H., Hwang, J.H., Park, K.H., and Song, K.Y.: Explanatory Note of The Pusan Sheet, 1:1:250,000, Korea Institute of
Energy and Resources, Daejeon, 1998.
Kim, H.-U., and Bae, T.-S.: Monitoring of Possible Activities of Yangsan Fault Zone Using GNSS, Appl. Sci., 13, 1862,
https://doi.org/10.3390/app13031862, 2023.
Kim, I.-S.: Origin and tectonic evolution of the East Sea (Sea of Japan) and the Yangsan fault system: a new synthetic
910
interpretation, J. Geol. Soc. Korea, 28, 84-109, 1992.
Kim, K.-H., Ree, J.-H., Kim, Y., Kim, S., Kang, S.Y., and Seo, W.: Assessing whether the 2017 Mw 5.4 Pohang earthquake in
South Korea was an induced event, Science, 360, 1007-1009, https://doi.org/10.1126/science.aat6081, 2018.
Kim, K.-H., Seo, W., Hang, J., Kwon, J., Kang, S.Y., Ree, J.-H., Kim, S., and Liu, K.: The 2017 ML 5.4 Pohang earthquake
sequence, Korea, recorded by a dense seismic network, Tectonophysics, 774, 228306,
915
https://doi.org/10.1016/j.tecto.2019.228306, 2020b.
Kim, M.-C., Jung, S., Yoon, S., Jeong, R.-Y., Song, C.W., and Son, M.: Neotectonic crustal deformation and current stress
field in the Korean Peninsula and Their Tectonic Implications: a review, J. Petrological Soc. Korea, 25, 169–193,
https://doi.org/10.7854/JPSK.2016.25.3.169, 2016.
Kim, M.-J., and Lee, H.-K.: Long-term patterns of earthquakes influenced by climate change: Insights from earthquake
920
recurrence and stress field changes across the Korean Peninsula during interglacial periods, Quaternary Sci. Rev., 321, 108369,
https://doi.org/10.1016/j.quascirev.2023.108369, 2023
Kim, M.-J., and Lee, H.-W.: ESR dating of fault gouge - review, J. Geol. Soc. Korea, 56, 175-192,
https://doi.org/10.14770/jgsk.2020.56.2.211, 2020.
Kim, N., Choi, J.-H., Park, S.-I., Lee, T.-H. and Choi, Y.: Cumulative offset analysis of the Central-Southern Yangsan Fault
925
based on topography of Quaternary fluvial terrace, J. Geol. Soc. Korea, 56, 135-154,
https://doi.org/10.14770/jgsk.2020.56.2.135, 2020c.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
52
Kim, N., Park, S.-I., and Choi, J.-H.: Internal architecture and earthquake rupture behavior of a long-lived intraplate strike–
slip fault: A case study from the Southern Yangsan Fault, Korea, Tectonophysics, 816, 229006,
https://doi.org/10.1016/j.tecto.2021.229006, 2021.
930
Kim, T., Choi, J.-H., Cheon, Y., Lee, T.-H., Kim, N., Lee, H., Kim, C.-M., Choi, Y., Bae, H., Kim, S.-G., Ryoo, C.-R., and
Klinger, Y.: Correlation of paleoearthquake records at multiple sites along the southern Yangsan Fault, Korea: Insights into
rupture scenarios of intraplate strike-slip earthquakes, Tectonophysics, 854, 229817,
https://doi.org/10.1016/j.tecto.2023.229817, 2023.
Kim, Y.-S. and Jin, K.: Estimated earthquake magnitude from the Yugye Fault displacement on a trench section in Pohang,
935
SE Korea, J. Geol. Soc. Korea, 42, 79-94, 2006.
Kim, Y.-S., Jin, K., Choi, W.-H., and Kee, W.-S.: Understanding of active faults: A review for recent researches, J. Geol. Soc.
Korea. 47, 723-752, 2011a.
Kim, Y.-S., Kihm, J.-H., and Jin, K.: Interpretation of the rupture history of a low slip-rate active fault by analysis of
progressive displacement accumulation: an example from the Quaternary Eupcheon Fault, SE Korea, J. Geol. Soc., 168, 273–
940
288, https://doi.org/10.1144/0016-76492010-088, 2011b.
Kim, Y.-S., Kim, T., Kyung, J. B., Cho, C. S., Choi, J.-H., Choi, C. U.: Preliminary study on rupture mechanism of the 9.12
Gyeongju Earthquake, J. Geol. Soc. Korea, 53, 407-422, https://doi.org.10.14770/jgsk.2017.53.3.407, 2017b.
Ko, K., Choi, S.-J., Lee, T.-H., Gihm, Y. S., Kim, C.-M., Kim, K., and Cheon, Y.: A multidisciplinary approach to
characterization of the mature northern Yangsan fault in Korea and its active faulting, Mar. Geophys. Res., 43, 21,
945
https://doi.org/10.1007/s11001-022-09486-w, 2022.
Korea Meteorological Administration: 2022 yearbook, Korea Meteorological Administration, Seoul, 416 pp., 2022. Retrieved
from https://www.kma.go.kr (archived on 6/March/2023).
Kreutzer, S.: calc_FadingCorr(): Apply a fading correction according to Huntley and Lamothe (2001) for a given g-value and
a given tc. Function version 0.4.3, in Luminescence, edited by Kreutzer, S., Burow, C., Dietze, M., Fuchs, M.C., Schmidt, C.,
950
Fischer, M., Friedrich, J., Mercier, N., Philippe, A., Riedesel, S., Autzen, M., Mittelstrass, D., Gray, H.J., and Galharret, J.,
Comprehensive Luminescence Dating Data Analysis R package version 0.9.23, available at https://CRAN.R-
project.org/package=Luminescence (last access: January 2024), 2023.
Kyung, J. B.: Paleoseismological study and evaluation of maximum earthquake magnitude along the Yangsan and Ulsan Fault
Zones in the Southeastern Part of Korea, Geophysics Geophys. Explor., 13, 187-197, 2010.
955
Kyung, J.B.: Paleoseismology of the Yangsan Fault, southeastern part of the Korean Peninsula, Ann. Geophys., 46, 983-996,
http://hdl.handle.net/2122/999, 2003.
Lal, D.: Cosmic ray labeling of erosion surfaces: in situ nuclide production rates and erosion models, Earth Planet. Sc. Lett.,
104, 424–439, https://doi.org/10.1016/0012-821X(91)90220-C, 1991.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
53
Lee, C. H., Seong Y. B., Oh, J.-S., Kim. D. E.: Tectonic Geomorphology on Yugye-Bogyeongsa Area of Yangsan Fault Zone,
960
J. Korean Geomorphological Assoc., 26, 93-106, https://doi.org/10.16968/JKGA.26.1.93, 2019.
Lee, H.-K., and Schwarcz, H. P.: ESR dating of the subsidiary faults in the Yangsan Fault System, Korea, Quaternary Sci.
Rev., 20, 999–1003, https://doi.org/10.1016/S0277-3791(00)00055-X, 2001.
Lee, H.K., and Yang, J.S.: ESR dating of the Eupchon fault, south Korea, Quat. Geochronol., 2, 392–397,
https://doi.org/10.1016/j.quageo.2006.04.009, 2003.
965
Lee, J., Rezaei, S., Hong, Y., Choi, J.-H., Choi, J.-H., Choi, W.-H., Rhee, K.-W., and Kim, Y.-S.: Quaternary fault analysis
through a trench investigation on the northern extension of the Yangsan fault at Dangu-ri, Gyungju-si, Gyeongsangbuk-do, J.
Geol. Soc. Korea, 51, 471-485, https://doi.org.10.14770/jgsk.2015.51.5.471, 2015.
Lee, K.: Historical earthquake data of Korean. J. Korea Geophys. Soc. 1, 3-22, 1998.
Lee, K., and Yang, W.-S.: Historical Seismicity of Korea, B.Seismol. Soc. Am., 96, 846-855,
970
https://doi.org/10.1785/0120050050, 2006.
Lee, S.: Characteristics of the Quaternary Faults in SE Korea: Insight from the Yangsan Fault and Gokgang Fault, Ph.D. thesis,
Pusan National University, Korea, 184 pp., 2023.
Lee, S., Han, J., Ha, S., Lim, H., Seong, Y. B., Choi, J.-H., Lee, C. H., Kim, S.-J., Kang, H.-C., Kim, M.-C., Lim, H., and Son,
M.: Characteristics of the Quaternary faulting detected along the Yangsan Fault in Yugye- and Jungsan-ri, northern Pohang
975
City, J. Geol. Soc. Korea, 58, 427-443, http://doi.org.10.14770/jgsk.2022.58.4.427, 2022.
Lee, Y., Cheon, Y., Ha, S., Kang, H.-C., Choi, J.-H., and Son, M.: Geometric and kinematic characteristics of the Quaternary
fault at Seooe site, in Goseong-gun, Gyeongsangnam-do, J. Geol. Soc. Korea, 53, 115-
127, https://doi.org/10.14770/jgsk.2017.53.1.115, 2017.
Lim, H., Ha, S., Ryoo, C.-R., Lee, T.-H., and Son, M.: Evidence of surface rupture in the southernmost part of the Yangsan
980
fault and deterministic seismic hazard assessment: in the Busan metropolitan area, in: 76th Annual Meeting of the Geological
Society of Korea and 2021 Fall Joint Conference of the Geological Sciences, Jeju, Korea, 26-28 October 2021, 275, 2021.
Liu, J., Ren, Z., Min, W., Ha, G., and Lei, J.: The advance in obtaining fault slip rate of strike slip fault-A review. Earthq. Res.
Adv., 1, 100032. https://doi.org/10.1016/j.eqrea.2021.100032, 2021.
Liu, M., and Stein, S.: Mid-continental earthquakes: Spatiotemporal occurrences, causes, and hazards, Earth-Sci. Rev., 162,
985
364–386, https://doi.org/10.1016/j.earscirev.2016.09.016, 2016.
Mason, D.B.: Earthquake magnitude potential of the Intermountain Seismic Belt, USA, from surface-parameter scaling of late
Quaternary faults. B. Seismol. Soc. Am., 86, 1487–1506, https://doi.org/10.1785/BSSA0860051487, 1996.
McCalpin, J.P.: Application of Paleoseismic Data to Seismic Hazard Assessment and Neotectonic Research, in:
Paleoseismology, 2nd ed., edited by McCalpin, J.P., Academic Press-Elsevier, Burlington, USA, 1–106,
990
https://doi.org/10.1016/S0074-6142(09)95009-4, 2009.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
54
Min, Z., Hu, G., Jiang, X., Liu, S., and Yang, Y.: Catalog of Chinese Historic Strong Earthquakes From 23 AD to 1911,
Seismological Press, Beijing, China, 514 pp., 1995.
Murray, A. S., and Wintle, A. G.: Luminescence dating of quartz using an improved single-aliquot regenerative-dose protocol,
Radiat. Meas., 32, 57-73, https://doi.org/10.1016/S1350-4487(99)00253-X, 2000.
995
Murray, J., and Langbein, J.: Slip on the San Andreas Fault at Parkfield, California, over Two Earthquake Cycles, and the
Implications for Seismic Hazard, B.Seismol. Soc. Am., 96, S283–S303, https://doi.org/10.1785/0120050820, 2006.
O'Neill, C., Müller, D., and Steinberger, B.: On the uncertainties in hot spot reconstructions and the significance of moving
hot spot reference frames, Geochem. Geophy. Geosy., 6, https://doi.org/10.1029/2004GC000784, 2005.
Obara, K., and Kato, A.: Connecting slow earthquakes to huge earthquakes. Science, 353, 6296,
1000
https://doi.org/10.1126/science.aaf1512, 2016.
Park, C., and Lee, G.: Analysis on Fault-Related Landforms in the Gyeongju Area of the Yangsan Fault Valley, J. Korean
Geomorphological Assoc., 25, 19-30, https://doi.org.-/10.16968/JKGA.25.1.19, 2018.
Park, Y., Ree, J.-H., and Yoo, S.-H.: Fault slip analysis of Quaternary faults in southeastern Korea, Gondwana Res., 9, 118–
125, https://doi.org/10.1016/j.gr.2005.06.007, 2006.
1005
Powell, R. E., and Weldon II, R. J.: Evolution of the San Andreas fault, Annu. Rev. Earth Pl. Sc., 20, 431-468,
https://doi.org/10.1146/annurev.ea.20.050192.002243, 1992.
Reimer, P., Austin, W., Bard, E., Bayliss, A., Blackwell, P., Bronk Ramsey, C., and Talamo, S.: The IntCal20 Northern
Hemisphere Radiocarbon Age Calibration Curve (0–55 cal kBP), Radiocarbon, 62, 725–
757, https://doi.org/10.1017/RDC.2020.41, 2020.
1010
Reiter, L.: Earthquake Hazard Analysis: Issue and Insights. Columbia University Press, New York, USA, 254 pp., 1990.
Ryoo, C.-R., and Cheon, Y.: Characteristics of the Main Fault Zone Developed Along Yangsan Fault: On the Outcrop of
Cheonjeon-ri, Dudong-myeon, Ulju-gun, Ulsan, Korea, J. Petrological Soc. Korea, 28, 347-357,
https://doi.org/10.7854/JPSK.2019.28.4.347, 2019.
Ryoo, C.-R., Lee, B.-J., Cho, D.-L., Chwae, U.-C., Choi, S.-J., and Kim J.-Y.: Quaternary fault of Dangu-ri in Gyeongju
1015
Gangdong-myeon: Byeokgye Fault, in: Spring Joint Conference on East-North Asia Ore deposit of Cause of Formation, The
Korean Society of Economic and Environmental Geology/The Korean Society of Mineral and Energy Resources
Engineers/Korean Society of Earth and Exploration Geophysicists, 1999.
Sadler, P.: The Influence of Hiatuses on Sediment Accumulation Rates, GeoResearch Forum, 5, 15-40, 1999.
Schellart, W. P., and Rawlinson, N.: Convergent plate margin dynamics: New perspectives from structural geology, geophysics
1020
and geodynamic modelling, Tectonophysics, 483, 4-19, https://doi.org/10.1016/j.tecto.2009.08.030, 2010.
Seno, T., Sakurai, T., and Stein, S.: Can the Okhotsk plate be discriminated from the North American plate?, J. Geophys. Res.-
Sol. Ea., 101(B5), 11305-11315, https://doi.org/10.1029/96JB00532, 1996.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
55
Seno, T., Stein, S., and Gripp, A. E.: A model for the motion of the Philippine Sea plate consistent with NUVEL‐1 and
geological data, J. Geophys. Res.- Sol. Ea., 98(B10), 17941-17948, https://doi.org/10.1029/93JB00782, 1993.
1025
Shimazaki, K., and Nakata, T.: Time-predictable recurrence model for large earthquakes, Geophys. Res. Lett., 7, 279-282,
https://doi.org/10.1029/GL007i004p00279, 1980.
Sim, H., Song, Y., Son, M., Park, C., and Choi, W.: Reactivated Timings of Yangsan Fault in the Northern Pohang Area,
Korea, Econ. Environ. Geol., 50, 97-104, https://doi.org/10.9719/EEG.2017.50.2.97, 2017.
Smith, B. R., and Sandwell, D. T.: A model of the earthquake cycle along the San Andreas Fault System for the past 1000
1030
years, J. Geophys. Res., 111, B01405. https://doi.org/10.1029/2005JB003703, 2006.
Sohn, D.-H., Choi, B.-K., Kim, S., Park, S.-C., Lee, W.-J., and Park, P.-H.: Decaying Post-Seismic Deformation Observed on
the Korean Peninsula Following the 2011 Tohoku-Oki Earthquake, Sensors, 21, 4493, https://doi.org/10.3390/s21134493,
2021.
Son, M., Song, C.W., Kim, M.-C., Cheon, Y., Cho, H., and Sohn, Y.K.: Miocene tectonic evolution of the basins and fault
1035
systems, SE Korea: dextral, simple shear during the East Sea (Sea of Japan) opening, J. Geol. Soc. London, 172, 664-680,
https://doi.org/10.1144/jgs2014-079, 2015.
Song, C.W.: Study on the evolution of the Miocene Pohang Basin based on its structural characteristics, Ph.D. thesis, Pusan
National University, Korea, 146 pp., 2015.
Song, Y., Ha, S., Lee, S., Kang, H.-C., Choi, J.-H., and Son, M.: Quaternary structural characteristics and paleoseismic
1040
interpretation of the Yangsan Fault at Dangu-ri, Gyeongju-si, SE Korea, through trench survey, J. Geol.Soc. Korea, 56, 155–
173, https://doi.org/10.14770/jgsk.2020.56.2.155, 2020.
Song, Y., Park, C., Sim, H., Choi, W., and Son, M.: Reactivated Timings of Yangsan Fault in the Sangcheon-ri Area, Korea,
Econ. Environ. Geol., 49, 97-104, https://doi.org/10.9719/EEG.2016.49.2.97, 2016.
Song, Y., Sim, H., Hong, S., and Son, M.: K-Ar Age-dating Results of Some Major Faults in the Gyeongsang Basin: Spatio-
1045
temporal Variability of Fault Activations during the Cenozoic Era, Econ. Environ. Geol., 52, 449-457,
https://doi.org/10.9719/EEG.2019.52.5.449, 2019.
Talwani, P.: On the nature of intraplate earthquakes, J. Seismol., 21, 47-68, https://doi.org/10.1007/s10950-016-9582-8, 2017.
Uchida, N., and Burgmann, R.: Repeating Earthquakes, Annu. Rev. Earth Pl. Sc., 47, 305–332,
https://doi.org/10.1146/annurev-earth-053018-060119, 2019.
1050
Wells, D. L., and Coppersmith, K. J.: New empirical relationships among magnitude, rupture length, rupture width, rupture
area, and surface displacement, B. Seismol. Soc. Am., 84, 974–1002, https://doi.org/10.1785/BSSA0840040974, 1994.
Williams, R.T., Goodwin, L.B., Sharp, W.D. and Mozley, O.S.: Reading a 400,000-year record of earthquake frequency for
an intraplate fault, P. Natl. Acad. Sci. USA, 114, 4893–4898. https://doi.org/10.1073/pnas.1617945114, 2017.
Woith, H., Petersen G.M., Hainzl, S., and Dahm, T.: Review: Can Animals Predict Earthquakes?, B. Seismol. Soc. Am., 108,
1055
1031–1045, https://doi.org/10.1785/0120170313, 2018.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
56
Wolin, E., Stein, S., Pazzaglia, F., Meltzer, A., Kafka, A., and Berti, C.: Mineral, Virginia, earthquake illustrates seismicity of
a passive-aggressive margin, Geophys. Res. Lett., 39, L02305, https://doi.org/10.1029/2011GL050310, 2012.
Woo, S., Lee, H., Han,R., Chon, C.-M., Son, M., and Song, I.: Frictional properties of gouges collected from the Yangsan
Fault, SE Korea. J. Geol. Soc. Korea, 51, 569-584, https://doi.org//10.14770/jgsk.2015.51.6.569, 2015.
1060
Wyss, M.: Cannot Earthquakes Be Predicted?, Science, 278, 487-490, https://doi.org/10.1126/science.278.5337.48, 1997.
Xu, S., Nieto-Samaniego, A.F., and Alaniz-À lvarez, S.A.: Quantification of true displacement using apparent displacement
along an arbitrary line on a fault plane, Tectonophysics, 467, 107-118, https://doi.org/10.1016/j.tecto.2008.12.004, 2009.
Yang, J.-S., and Lee, H.-K.: ESR dating of fault gouge from the Gacheon 1 site on the Yangsan fault zone, J. Geol. Soc.
Korea, 48, 459-472, 2012.
1065
Yang, J.-S., and Lee, H.-K.: Quaternary Fault Activity of the Yangsan Fault Zone in the Samnam-myeon, Ulju-gun, Ulsan,
Korea, Econ. Environ. Geol., 47, 17-27, https://doi.org/10.9719/EEG.2014.47.1.17, 2014.
https://doi.org/10.5194/egusphere-2024-1696
Preprint. Discussion started: 8 August 2024
c
Author(s) 2024. CC BY 4.0 License.
